JP5366988B2 - Exhaust gas purification system for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an exhaust emission control system for an internal combustion engine capable of constantly maintaining a NOx purification rate in a selective reduction catalyst to near its maximum during just after beginning of the start of the internal combustion engine through running. <P>SOLUTION: The exhaust emission control system includes an oxide catalyst and CSF provided in an exhaust pipe, and the selective reduction catalyst provided on a downstream side than the oxide catalyst and CSF in the exhaust pipe, for selectively reducing NOx in exhaust. In NO<SB POS="POST">2</SB>-NOx ratio optimization control for controlling a NO<SB POS="POST">2</SB>-NOx ratio of the exhaust flowing into the selective reduction catalyst toward an optimum value for maximizing a NOx purification rate in the selective reduction catalyst, an ECU prohibits performance of the NO<SB POS="POST">2</SB>-NOx ratio maximization control until predetermined time elapses from the beginning of the start of the engine or if the temperature of the exhaust system is below predetermined temperature, and permits performance of the NO<SB POS="POST">2</SB>-NOx ratio optimization control after the predetermined time has elapsed from the beginning of the start of the engine, or if the temperature of the exhaust system is the predetermined temperature or above. <P>COPYRIGHT: (C)2012,JPO&amp;INPIT

Description

  The present invention relates to an exhaust gas purification system for an internal combustion engine. More specifically, the present invention relates to an exhaust gas purification system for an internal combustion engine including a selective reduction catalyst (Selective Catalytic Reduction Catalysts) that selectively reduces nitrogen oxide (NOx) in exhaust gas in the presence of a reducing agent.

Conventionally, as one of exhaust purification systems for purifying NOx in exhaust, a system in which a selective reduction catalyst for selectively reducing NOx in exhaust with a reducing agent such as ammonia (NH ₃ ) is provided in the exhaust passage has been proposed. ing. For example, in the exhaust purification system of urea addition type, the NH ₃ by supplying urea water which is a precursor of NH ₃ from the upstream side of the selective reduction catalyst, thermal decomposition or hydrolysis in the exhaust heat from the urea water The NOx in the exhaust gas is selectively reduced by this NH ₃ . In addition to such a urea addition type system, for example, a system in which NH ₃ is generated by heating a NH ₃ compound such as ammonia carbide and this NH ₃ is directly added has also been proposed. Hereinafter, a urea addition type system will be described.

It is known that the NOx purification rate in the selective reduction catalyst varies depending on the ratio of nitrogen monoxide (NO) and nitrogen dioxide (NO ₂ ) constituting NOx in the inflowing exhaust gas. More specifically, when the NO ₂ -NOx ratio of the inflowing exhaust gas (the molar ratio of NO ₂ to NOx in which NO and NO ₂ are combined) is 0.5, that is, the ratio of NO and NO ₂ is 1: 1. It becomes the maximum when it is.

Patent Document 1 proposes an exhaust emission control device in which the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is set to 0.5 in order to maximize the performance of such a selective reduction catalyst. Yes. In this exhaust purification device, NO _{2 of} exhaust flowing into the selective reduction catalyst is controlled by feedforward control of the EGR amount, fuel injection timing, and the like by searching a predetermined map from the operating state of the internal combustion engine. -Ensure that the NOx ratio is always maintained at 0.5.

JP 2008-231950 A

As described above, in the exhaust gas purification device of Patent Document 1, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is maintained at 0.5 in order to keep the NOx purification rate in the selective reduction catalyst high. However, the optimum value of the NO ₂ -NOx ratio is not always 0.5.

For example, when HC adheres to the selective reduction catalyst or HC is contained in the exhaust gas newly flowing into the selective reduction catalyst, a reaction represented by the following formula (1) occurs on the selective reduction catalyst. However, NO ₂ in the exhaust is consumed, and on the contrary, NO increases.

Therefore, when the HC purification performance of the oxidation catalyst is high and in an ideal state, the amount of HC flowing into the selective reduction catalyst is very small. Therefore, the NO ₂ -NO x ratio is set to around 0.5 to reduce the NO x purification rate. In order to maximize the NOx purification rate from the presence of HC as described above, when the temperature of the oxidation catalyst is low or the deterioration progresses and the HC purification performance is low. It is necessary to make the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst larger than 0.5 (for example, about 0.65).

For example, when an exhaust gas purification filter that collects soot in the exhaust gas is provided on the upstream side of the selective reduction catalyst, a CRT (Continuously Regenerating Trap) reaction as shown in the following formula (2) occurs, and the oxidation catalyst The generated NO ₂ will return to NO. For this reason, the NO ₂ -NOx ratio that maximizes the NOx purification rate in the selective reduction catalyst becomes substantially more uncertain.

As described above, although the optimum value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst does not always remain at 0.5, the above-mentioned Patent Document 1 is sufficient for this point. Has not been studied.

  The present invention has been made in view of the above-described problems, and is an internal combustion engine exhaust purification system that can always maintain the NOx purification rate in the selective reduction catalyst in the vicinity of its maximum from immediately after the start of the internal combustion engine to running. The purpose is to provide.

In order to achieve the above object, the present invention provides an oxidation catalyst (for example, an after-mentioned oxidation catalyst) provided in an exhaust passage (for example, an after-mentioned exhaust pipe 11) of an internal combustion engine (for example, an after-mentioned engine 1, 1A, 1B, 1C). Catalyst 21 and CSF 22), and a selective reduction catalyst (for example, selective reduction catalyst 23 described later) provided downstream of the oxidation catalyst in the exhaust passage and selectively reducing NOx in the exhaust. An exhaust gas purification system for an internal combustion engine (for example, an exhaust gas purification system 2, 2A, 2B, 2C described later) is provided. The exhaust purification system controls the NO ₂ -NOx ratio corresponding to the ratio of NO ₂ to NO x in the exhaust flowing into the selective reduction catalyst toward an optimum value that maximizes the NO x purification rate in the selective reduction catalyst. NO ₂ -NOx ratio optimization control (for example, control based on the NO ₂ sensor feedback mode described later, control based on the NO sensor feedback mode, control based on the feedforward control mode) (for example, ECU 3 described later) , 3A, 3B, 3C, 3D), the control means until the predetermined time elapses after starting the internal combustion engine or the temperature of the exhaust system of the internal combustion engine is less than the predetermined temperature in some cases it prohibits execution of the NO ₂ -NOx ratio optimization control, a predetermined time from the start of the start-up of the internal combustion engine After spent, or the temperature of the exhaust system of the internal combustion engine when it is higher than a predetermined temperature to allow the execution of the NO ₂ -NOx ratio optimization control.

In the present invention, a selective reduction catalyst is provided downstream of the oxidation catalyst, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is set to an optimum value that maximizes the NOx purification rate in the selective reduction catalyst. Control means for executing the NO ₂ -NOx ratio optimization control to be controlled is provided. That is, in the present invention, the NO ₂ -NOx ratio optimization control is appropriately executed, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, for example, by increasing or decreasing the amount of NO ₂ oxidized and generated by the oxidation catalyst. Can be controlled toward the optimum value, and the NOx purification rate in the selective reduction catalyst can be maximized.
By the way, immediately after the start of the internal combustion engine or when the temperature of the exhaust system is low, the temperature of the oxidation catalyst has not yet reached the activation temperature, so that a relatively large amount of HC cannot be oxidized by the oxidation catalyst and flows into the selective reduction catalyst. It is thought that it adheres and adheres. Since the the number of HC or attached or flowing into the selective reduction catalyst, so that the the NO ₂ in the exhaust flowing into the selective reduction catalyst as described above decreased NO increases, NO ₂ -NOx ratio The optimum value of is greatly deviated. In contrast, in the present invention, such NO ₂ is in the period in which the optimum value of -NOx ratio is considered to deviate leave prohibits the execution of the NO ₂ -NOx ratio optimization control described above, the start of the internal combustion engine The execution of the NO ₂ -NOx ratio optimization control is permitted after a predetermined time has elapsed since the start of the engine, or after the temperature of the exhaust system becomes equal to or higher than the predetermined temperature. Thereby, the NO ₂ -NOx ratio is controlled to a value different from the actual optimum value, and conversely, the NOx purification rate in the selective reduction catalyst can be prevented from deteriorating.

FIG. 13 shows the HC and CO purification rates of the oxidation catalyst in the traveling vehicle, the NO ₂ generation efficiency in the oxidation catalyst (the ratio of the amount of NO ₂ flowing out from the oxidation catalyst to the amount of NO flowing into the oxidation catalyst), is a diagram showing changes in the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst. In FIG. 13, the broken line indicates a case where an oxidation catalyst having a high oxidation performance is used, and the alternate long and short dash line indicates a case where an oxidation catalyst having a low oxidation performance is used. Further, the solid line in FIG. 13 shows ideal characteristics that can efficiently purify all HC, CO, and NOx in the exhaust during and after the engine is warmed up.

As shown in FIG. 13, when an oxidation catalyst with high oxidation performance is used, the rise of the HC and CO purification rates during warm-up immediately after engine startup is greater than when an oxidation catalyst with low oxidation performance is used. It can be done quickly and the CO and HC purification rate after warm-up can be increased. For this reason, in order to increase the CO and HC purification rate, it is generally preferable to use an oxidation catalyst having high oxidation performance. That is, the characteristics of the HC and CO purification rates when using an oxidation catalyst with high oxidation performance are consistent with the ideal characteristics during and after warm-up.
In addition, when an oxidation catalyst with high oxidation performance is used, the rise of NO ₂ generation efficiency during warm-up immediately after engine startup can be accelerated, and therefore the NO ₂ -NOx ratio flowing into the selective reduction catalyst can be quickly reduced to 0.5. It can be raised to an optimal value in the vicinity. Therefore, the characteristics of NO ₂ production efficiency and NO ₂ -NOx ratio in the case of using an oxidation catalyst having high oxidation performance coincide with the ideal characteristics during warm-up.

However, when an oxidation catalyst having a high oxidation performance is used, the NO ₂ generation efficiency after warm-up becomes too high, and the NO ₂ -NOx ratio greatly exceeds the optimum value in the vicinity of 0.5. As a result, NOx A purification rate will fall. Therefore, after warm-up, the NO ₂ production efficiency and the NO ₂ -NOx ratio are closer to ideal characteristics when using an oxidation catalyst with low oxidation performance. That is, if priority is given to quick warm-up, it is preferable to use an oxidation catalyst with high oxidation performance. If priority is given to improving the NOx purification rate after warm-up, it is preferable to use an oxidation catalyst with lower oxidation performance. preferable.

As described above, even if an oxidation catalyst with high oxidation performance is used or an oxidation catalyst with low oxidation performance is used, both the HC and CO purification rates and the NO ₂ generation efficiency are ideal characteristics indicated by solid lines. Therefore, simply changing the specifications such as the noble metal loading, cell density, volume, and noble metal composition of the oxidation catalyst and adjusting its oxidation performance during and after warm-up immediately after engine startup In addition, it is considered difficult to efficiently purify all HC, CO, and NOx.
On the other hand, according to the present invention, after using an oxidation catalyst having high oxidation performance, until a predetermined time elapses after the start of the internal combustion engine or when the temperature of the exhaust system is lower than the predetermined temperature, By prohibiting execution of the NO ₂ -NOx ratio optimization control, warm-up can be speeded up. Then, after a predetermined time has elapsed since the start of the internal combustion engine, or when the temperature of the exhaust system becomes equal to or higher than the predetermined temperature, the execution of the NO ₂ -NOx ratio optimization control is permitted, so that as for the use of high oxidation performance oxidation catalyst NO ₂ generation efficiency reduces the NO ₂ generation efficiency from the state became slightly excessive in _as, optimize the NO 2 -NOx ratio _(NO 2 -NOx ratio Can be reduced toward the optimum value), and the NOx purification rate in the selective reduction catalyst can be kept high.
According to the present invention, by prohibiting or permitting the execution of the NO ₂ -NOx ratio optimization control at an appropriate time as described above, NO ₂ − with ideal characteristics that cannot be achieved only by preparing the oxidation catalyst. The NOx ratio can be changed, and therefore, the NOx purification rate in the selective reduction catalyst can always be kept high from immediately after the start of the internal combustion engine until the vehicle is running.

In this case, the exhaust purification system further includes NO ₂ detection means for detecting NO _{2 in} the exhaust downstream of the selective reduction catalyst in the exhaust passage, and in the NO ₂ -NOx ratio optimization control, When the detected value by the NO ₂ detecting means is larger than a predetermined value, the NO ₂ -NOx ratio is reduced toward the optimum value (for example, processing at times t2 to t3 and t4 to t5 in FIG. 7 described later, FIG. 20 at times t2 to t3 and t4 to t5, and processes at times t2 to t3 and t4 to t5 in FIG.

For example, when the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is at the optimum value, the NOx purification rate in the selective reduction catalyst is maximized, so that both NO and NO ₂ are hardly discharged downstream. On the other hand, when the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst becomes larger than the optimum value and the NOx purification rate is lowered, the exhaust gas flowing into the selective reduction catalyst is in a NO ₂ excessive state, and the downstream side is NO 2 ₂ will be discharged. Therefore, according to the present invention, the fact that the exhaust gas flowing into the selective reduction catalyst becomes excessive in NO ₂ and the NO ₂ -NOx ratio becomes larger than the optimum value is that the detection value by the NO ₂ detection means becomes larger than the predetermined value. In the NO ₂ -NOx ratio optimization control, the NO ₂ -NOx ratio is reduced toward the optimum value accordingly, and as a result, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced. It can be controlled to vibrate between an optimum value and a predetermined value larger than the optimum value. As described above, according to the present invention, by controlling the NO ₂ -NOx ratio in the vicinity of the optimum value in a feedback manner based on the detected value by the NO ₂ detecting means, the operating state, operating condition, operating history of the internal combustion engine. In addition, the NOx purification rate of the selective reduction catalyst can be maintained in the vicinity of the maximum regardless of the deterioration state of the oxidation catalyst or the selective reduction catalyst.

In this case, in the NO ₂ -NOx ratio optimization control, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is directed toward the optimum value by increasing the amount of NO discharged from the internal combustion engine. It is preferable to reduce.

In the present invention, the amount of NO discharged from the internal combustion engine is increased and the selective reduction is performed under the condition that the detected value by the NO ₂ detecting means is larger than the predetermined value, that is, under the condition that the NO ₂ -NOx ratio is larger than the optimum value. The NO ₂ -NOx ratio of the exhaust gas flowing into the catalyst is reduced toward its optimum value. Thereby, although the amount of NOx discharged from the internal combustion engine becomes large, the NOx purification rate in the selective reduction catalyst can be maintained in the vicinity of the maximum, and as a result, the amount of NOx discharged outside the system can be greatly reduced. In the exhaust purification device of Patent Document 1 described above, when the NO ₂ -NOx ratio is reduced, control is performed to reduce the amount of NOx discharged from the internal combustion engine, whereas in the present invention, as described above, conversely The present invention is different from the invention of Patent Document 1 in that the amount of NOx discharged from the internal combustion engine is increased and the NO ₂ -NOx ratio is maintained in the vicinity of the optimum value.

In this case, the exhaust gas purification system recirculates a part of the exhaust gas flowing through the exhaust passage to an intake passage (for example, an intake pipe 12 described later) of the internal combustion engine (for example, a high pressure EGR device 26 described later). In the NO ₂ -NOx ratio optimization control, it is preferable that the amount of NO exhausted from the internal combustion engine is increased by decreasing the amount of EGR corresponding to the amount of exhaust gas recirculated by the EGR device. .

According to the present invention, by increasing the NO amount discharged from the internal combustion engine by decreasing the EGR amount, new hardware for increasing the NO amount is added, or complicated combustion control is performed. Without this, the NO ₂ -NOx ratio can be maintained in the vicinity of the optimum value.

In this case, the ratio of the amount of NO ₂ flowing out from the oxidation catalyst to the amount of NO flowing into the oxidation catalyst is defined as NO ₂ generation efficiency, and the NO ₂ -NOx ratio optimization control is exhausted from the internal combustion engine. The reduction effect of the NO ₂ generation efficiency due to the increase in the amount of NOx discharged from the internal combustion engine is greater than the increase effect of the NO ₂ generation efficiency due to the decrease in the amount of HC and CO that is produced. Further, it is preferable to set a combustion parameter correlated with the combustion state of the internal combustion engine.

According to the present invention, in the NO ₂ -NOx ratio optimization control, the NO ₂ generation efficiency lowering effect due to the NOx amount increasing than the NO ₂ generation efficiency increasing effect due to the HC amount and CO amount decreasing. As a result, the NO ₂ production efficiency in the oxidation catalyst is lowered, thereby reducing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, and consequently in the selective reduction catalyst. The NOx purification rate is maintained near the maximum. Here, the combustion parameter refers to all parameters correlated with the combustion state of the internal combustion engine, such as the fuel injection amount, the fuel injection timing, the supercharging pressure, and the EGR amount.

In this case, in the NO ₂ -NOx ratio optimization control, when the detected value (Vno2) by the NO ₂ detecting means is equal to or smaller than a predetermined value (Vno2_th), the amount of NO discharged from the internal combustion engine is decreased. (For example, processing after time t1 to t2, t3 to t4, t5 in FIG. 7, processing after time t1 to t2, t3 to t4, t5 in FIG. 20, time t1 to t2, t3 to t4, t5 in FIG. Subsequent processing) is preferred.

As described above, in the NO ₂ -NOx ratio optimization control, the NO amount discharged from the internal combustion engine is suppressed by increasing the NO amount discharged from the internal combustion engine, but the NO amount discharged from the internal combustion engine. If the amount increases excessively, the amount of NOx discharged outside the system may also start to increase. Therefore, according to the present invention, when the detected value by the NO ₂ detecting means is equal to or less than the predetermined value, the NO amount discharged from the internal combustion engine is thus reduced by reducing the NO amount discharged from the internal combustion engine. It can prevent becoming too much.

In this case, the control means determines the degree of deterioration of the selective reduction catalyst. If it is determined that the degree of deterioration is small, the control means prohibits execution of the NO ₂ -NOx ratio optimization control and sets the EGR amount. When the fuel consumption of the internal combustion engine is set to be improved and it is determined that the degree of deterioration is large, it is preferable to permit execution of the NO ₂ -NOx ratio optimization control.

In the present invention, the degree of deterioration of the selective reduction catalyst is determined, and when it is determined that the degree of deterioration is small, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is not accurately controlled to the optimum value. Therefore, it is determined that the NOx purification rate can be maintained high, execution of the NO ₂ -NOx ratio optimization control is prohibited, and the EGR amount is set so that the fuel consumption of the internal combustion engine is improved. When it is determined that the degree of deterioration of the catalyst is large, the execution of the NO ₂ -NOx ratio optimization control is permitted, and the NO ₂ -NOx ratio optimization control is executed as necessary. As a result, it is possible to prevent NO ₂ -NOx ratio optimization control from being performed excessively and trying to keep the NOx purification rate higher than necessary, thereby preventing fuel consumption from deteriorating.

  In this case, it is preferable that the control means determines the degree of deterioration of the selective reduction catalyst based on a correction value (Kegr_no2) of a target value (Gegr_cmd) for the EGR amount from a predetermined reference value (Gegr_map).

According to the present invention, the target value for the EGR amount is corrected from the predetermined reference value, that is, the parameter used for changing the target value of the EGR amount from the reference value in the NO ₂ -NOx ratio optimization control. By determining the degree of deterioration of the selective reduction catalyst based on this, the degree of deterioration can be determined without adding a new device such as a sensor.

In this case, in the NO ₂ -NOx ratio optimization control, the air-fuel ratio of the internal combustion engine is changed to a richer side, and the oxygen concentration of the exhaust gas is reduced to thereby reduce the NO ₂ − of the exhaust gas flowing into the selective reduction catalyst. It is preferable to reduce the NOx ratio toward the optimum value.

In the present invention, the air-fuel ratio of the air-fuel mixture of the internal combustion engine is changed to a richer side when the detected value by the NO ₂ detecting means is larger than a predetermined value, that is, under the condition that the NO ₂ -NOx ratio is larger than the optimum value. Then, by reducing the oxygen concentration of the exhaust gas, the ratio of NO oxidation in the oxidation catalyst is reduced, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward its optimum value. As a result, the NOx purification rate in the selective reduction catalyst can be maintained in the vicinity of the maximum, and the amount of NOx discharged outside the system can be greatly reduced.

In this case, the ratio of the amount of NO ₂ flowing out from the oxidation catalyst to the amount of NO flowing into the oxidation catalyst is defined as NO ₂ generation efficiency, and the NO ₂ -NOx ratio optimization control is exhausted from the internal combustion engine. that than increasing effect of the NO ₂ generation efficiency due to the NOx amount decreases, the oxygen concentration of the exhaust gas is lowered and the NO ₂ generation efficiency due to the HC amount and CO amount exhausted from the internal combustion engine is increased It is preferable to set a combustion parameter correlated with the combustion state of the internal combustion engine so that the lowering effect is greater.

According to the present invention, in the NO ₂ -NOx ratio optimization control, the oxygen concentration is reduced and the HC amount and the CO amount are lower than the effect of increasing the NO ₂ generation efficiency due to the reduction of the NOx amount discharged from the internal combustion engine. by but setting the combustion parameter so it is larger in the effect of lowering NO ₂ generation efficiency by increasing, as a result reduce the NO ₂ generation efficiency of the oxidation catalyst as, thereby the exhaust flowing into the selective reduction catalyst The NO ₂ -NOx ratio is reduced toward the optimum value, and the NOx purification rate in the selective reduction catalyst is maintained in the vicinity of the maximum. In the present invention, in order to reduce the NO ₂ generation efficiency, the amount of NOx reduced by the selective reduction catalyst can be reduced by reducing the amount of NOx discharged from the internal combustion engine. The amount of reducing agent supplied to the reduction catalyst can also be suppressed.

In this case, in the NO ₂ -NOx ratio optimization control, when the detected value (Vno2) by the NO ₂ detecting means is equal to or less than a predetermined value (Vno2_th), the air-fuel ratio of the air-fuel mixture of the internal combustion engine is further reduced to the lean side. It is preferable to change to.

According to the present invention, when the detected value by the NO ₂ detecting means is less than or equal to the predetermined value, the consumption of unnecessary fuel is suppressed by changing the air-fuel ratio of the air-fuel mixture of the internal combustion engine to a leaner side. Can do.

  In this case, it is preferable that the control means changes the air-fuel ratio of the air-fuel mixture according to at least one of a fuel injection parameter, a supercharging pressure, and an EGR amount corresponding to the amount of exhaust gas recirculated by the EGR device.

According to the present invention, new hardware can be obtained by changing the air-fuel ratio of the air-fuel mixture by at least one of the fuel injection parameter, the supercharging pressure, and the EGR amount corresponding to the amount of exhaust gas recirculated by the EGR device. The NO ₂ -NOx ratio can be maintained near the optimum value without adding.

In this case, the control means determines the degree of deterioration of the selective reduction catalyst. When it is determined that the degree of deterioration is small, the control means prohibits execution of the NO ₂ -NOx ratio optimization control and It is preferable to allow the execution of the NO ₂ -NOx ratio optimization control when the fuel ratio of the internal combustion engine is set to be improved and it is determined that the degree of deterioration is large.

In the present invention, when the degree of deterioration of the selective reduction catalyst is determined and it is determined that the degree of deterioration is small, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is not accurately controlled to the optimum value, It is determined that the NOx purification rate can be maintained high, the execution of the NO ₂ -NOx ratio optimization control is prohibited, and the air-fuel ratio of the air-fuel mixture is set so that the fuel consumption of the internal combustion engine is improved. If it is determined that the degree of deterioration is large, the execution of the NO ₂ -NOx ratio optimization control is permitted, and the NO ₂ -NOx ratio optimization control is executed as necessary. As a result, it is possible to prevent NO ₂ -NOx ratio optimization control from being performed excessively and trying to keep the NOx purification rate higher than necessary, thereby preventing fuel consumption from deteriorating.

  In this case, the control means determines the degree of deterioration of the selective reduction catalyst based on a correction value (Daf_no2) from a predetermined reference value (AF_map) of the target value (AF_cmd) for the air-fuel ratio of the air-fuel mixture. Is preferred.

In the present invention, based on the correction value for changing the target value of the air-fuel ratio of the air-fuel mixture from the reference value, that is, the parameter used for changing the air-fuel ratio in the NO ₂ -NOx ratio optimization control, By determining the degree of deterioration, the degree of deterioration can be determined without adding a new device such as a sensor.

In this case, in the NO ₂ -NOx ratio optimization control, the temperature of the oxidation catalyst is decreased in a region below the temperature at which the NO oxidation efficiency is maximized (temperature region [Tdoc_L, Tdoc_scr_opt]). It is preferable to reduce the NO ₂ -NOx ratio of the exhaust gas flowing into the reduction catalyst toward the optimum value.

In the present invention, the temperature of the oxidation catalyst is set to the temperature at which the NO oxidation efficiency becomes maximum when the detected value by the NO ₂ detecting means is larger than a predetermined value, that is, under the condition that the NO ₂ -NOx ratio is larger than the optimum value. By reducing the ratio within the following region and reducing the rate at which NO is oxidized in the oxidation catalyst, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward its optimum value. As a result, the NOx purification rate in the selective reduction catalyst can be maintained in the vicinity of the maximum, and the amount of NOx discharged outside the system can be greatly reduced.

In the NO ₂ -NOx ratio optimization control, it is preferable to reduce the temperature of the oxidation catalyst by decreasing at least one of the after injection amount and the post injection amount of the internal combustion engine.

According to the present invention, the temperature of the oxidation catalyst is lowered by reducing at least one of the after injection amount and the post injection amount, so that the NO ₂ -NOx ratio is close to the optimum value without adding new hardware. Can be maintained.

In this case, in the NO ₂ -NOx ratio optimization control, the combustion parameters correlated with the combustion state of the internal combustion engine, the CO and HC exhausted from the internal combustion engine have been reduced in oxidation capacity as the temperature decreased. It is preferable to set the amount to be equal to or less than the amount that can be treated with an oxidation catalyst.

In the present invention, when the NO ₂ -NOx ratio optimization control is executed and the temperature of the oxidation catalyst is lowered, the CO and HC discharged from the internal combustion engine may be the oxidation catalyst whose oxidation ability is reduced as the temperature is lowered. Set the combustion parameters so that they are less than the amount that can be processed. As a result, it is possible to prevent the CO and HC purification rates from being lowered by maintaining the NOx purification rate in the vicinity of the maximum. In addition, fuel efficiency can be improved by setting the combustion parameters so that the amount of CO and HC discharged from the internal combustion engine is reduced. In addition, under the setting of combustion parameters that reduce the amount of CO and HC emitted from the internal combustion engine, it is considered that the amount of NOx emitted from the internal combustion engine will not increase excessively. The amount of reducing agent consumed for purifying NOx in the catalyst can also be suppressed.

In this case, it is preferable that the NO ₂ -NOx ratio optimization control increases the temperature of the oxidation catalyst when the detected value (Vno2) by the NO ₂ detecting means is equal to or lower than a predetermined value (Vno2_th).

According to the present invention, when the detected value by the NO ₂ detecting means is equal to or lower than the predetermined value, the amount of NO flowing into the selective reduction catalyst becomes excessive by increasing the temperature of the oxidation catalyst, and the NOx in the selective reduction catalyst is increased. On the contrary, it is possible to prevent the purification rate from decreasing.

In this case, the lower limit temperature of the region with respect to the temperature of the oxidation catalyst is, when the temperature of the oxidation catalyst is lowered from the temperature at which the NO oxidation efficiency is maximized to the lower limit temperature in the NO ₂ -NOx ratio optimization control, Setting is made such that the NOx purification rate improvement effect by reducing the NO ₂ -NOx ratio toward its optimum value is greater than the NOx purification rate reduction effect by the temperature of the selective reduction catalyst decreasing. It is preferred that

According to the present invention, by setting as described above the minimum temperature at the time of lowering the temperature of the oxidation catalyst in NO ₂ -NOx ratio optimization control, selective catalytic reduction by the execution of the NO ₂ -NOx ratio optimization control On the other hand, it is possible to prevent the NOx purification rate from decreasing.

In this case, the control means determines the degree of deterioration of the selective reduction catalyst, and when it is determined that the degree of deterioration is small, the execution of the NO ₂ -NOx ratio optimization control is prohibited, and the degree of deterioration is large. When it is determined, it is preferable to allow execution of the NO ₂ -NOx ratio optimization control.

In the present invention, when the degree of deterioration of the selective reduction catalyst is determined and it is determined that the degree of deterioration is small, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is not accurately controlled to the optimum value, It is determined that the NOx purification rate can be maintained high, and execution of the NO ₂ -NOx ratio optimization control is prohibited. If it is determined that the degree of deterioration is large, the execution of the NO ₂ -NOx ratio optimization control is permitted, and the NO ₂ -NOx ratio optimization control is executed as necessary. As a result, it is possible to prevent NO ₂ -NOx ratio optimization control from being performed excessively and trying to keep the NOx purification rate higher than necessary, thereby preventing fuel consumption from deteriorating.

  In this case, the control means may determine the degree of deterioration of the selective reduction catalyst based on a correction value (Dt_no2) of a target value (Tdoc_cmd) with respect to the temperature of the oxidation catalyst from a predetermined reference value (Tdoc_scr_opt). preferable.

In the present invention, the selective reduction catalyst is based on a correction value for changing the target value of the oxidation catalyst temperature from the reference value, that is, a parameter used for changing the temperature of the oxidation catalyst in the NO ₂ -NOx ratio optimization control. By determining the degree of deterioration, it is possible to determine the degree of deterioration without adding a new device such as a sensor.

In this case, in the NO ₂ -NOx ratio optimization control, the temperature of the oxidation catalyst is increased in a region (temperature region [Tdoc_scr_opt, Tdoc_H]) equal to or higher than the temperature at which the NO oxidation efficiency is maximized. It is preferable to reduce the NO ₂ -NOx ratio of the exhaust gas flowing into the reduction catalyst toward the optimum value.

In the present invention, the temperature of the oxidation catalyst is set to the temperature at which the NO oxidation efficiency becomes maximum when the detected value by the NO ₂ detecting means is larger than a predetermined value, that is, under the condition that the NO ₂ -NOx ratio is larger than the optimum value. By raising within the above region and reducing the rate of oxidation of NO in the oxidation catalyst, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward its optimum value. As a result, the NOx purification rate in the selective reduction catalyst can be maintained in the vicinity of the maximum, and the amount of NOx discharged outside the system can be greatly reduced.

In this case, in the NO ₂ -NOx ratio optimization control, it is preferable to increase the temperature of the oxidation catalyst by increasing at least one of the after injection amount and the post injection amount of the internal combustion engine.

According to the present invention, the temperature of the oxidation catalyst is increased by increasing at least one of the after injection amount and the post injection amount, so that the NO ₂ -NOx ratio is close to the optimum value without adding new hardware. Can be maintained.

In this case, in the NO ₂ -NOx ratio optimization control, it is preferable that the temperature of the oxidation catalyst is lowered when the detected value (Vno2) by the NO ₂ detecting means is equal to or less than a predetermined value (Vno2_th).

According to the present invention, when the detected value by the NO ₂ detecting means is not more than the predetermined value, the amount of NO flowing into the selective reduction catalyst becomes excessive by lowering the temperature of the oxidation catalyst, and NOx in the selective reduction catalyst On the contrary, it is possible to prevent the purification rate from decreasing.

In this case, the upper limit temperature (Tdoc_H) of the region with respect to the temperature of the oxidation catalyst is increased from the temperature at which the NO oxidation efficiency is maximized to the upper limit temperature in the NO ₂ -NOx ratio optimization control. Sometimes, the effect of improving the NOx purification rate by reducing the NO ₂ -NOx ratio toward its optimum value is greater than the effect of reducing the NOx purification rate by increasing the temperature of the selective reduction catalyst. It is preferable to set as follows.

According to the present invention, by setting the upper limit temperature at the time of raising the temperature of the oxidation catalyst in the NO ₂ -NOx ratio optimization control as described above, the selective reduction catalyst is executed by executing the NO ₂ -NOx ratio optimization control. On the other hand, it is possible to prevent the NOx purification rate from decreasing.

In this case, the control means determines the degree of deterioration of the selective reduction catalyst, and when it is determined that the degree of deterioration is small, the execution of the NO ₂ -NOx ratio optimization control is prohibited, and the degree of deterioration is large. When it is determined, it is preferable to allow execution of the NO ₂ -NOx ratio optimization control.

According to the present invention, it is possible to prevent the NO ₂ -NOx ratio optimization control from being performed excessively and trying to keep the NOx purification rate higher than necessary, thereby deteriorating fuel consumption.

In this case, when the control means continues to reduce the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the detection value (Vno2) by the NO ₂ detection means falls below the deterioration determination threshold value (Vno2_JD_th). It is preferable to determine the degree of deterioration of the selective reduction catalyst based on the determined timing.
Alternatively, in this case, when the control means continues to increase the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the detected value (Vno2) by the NO ₂ detection means is a deterioration determination threshold value (Vno2_JD_th). It is preferable to determine the degree of deterioration of the selective reduction catalyst based on a timing exceeding

In the present invention, when the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is continuously increased, the selective reduction catalyst is based on the timing when the detected value by the NO ₂ detection means falls below or exceeds the deterioration judgment threshold. The degree of deterioration is determined. As a result, the degree of deterioration can be determined without adding a new device such as a sensor.

In this case, the control means determines the degree of deterioration of the oxidation catalyst, and when it is determined that the degree of deterioration is small, the control means permits execution of the NO ₂ -NOx ratio optimization control and determines that the degree of deterioration is large. In this case, it is preferable to prohibit execution of the NO ₂ -NOx ratio optimization control.

In the present invention, the degree of deterioration of the oxidation catalyst is determined. When it is determined that the degree of deterioration is small, execution of the NO ₂ -NOx ratio optimization control is permitted. When the degree of deterioration in the oxidation catalyst is small, the NO ₂ generation efficiency in the oxidation catalyst is sufficient, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst tends to be excessive in NO _{2 with} respect to the optimum value. since, by running in conjunction with NO ₂ -NOx ratio optimization control this can be controlled to the optimum value of the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst.
If it is determined that the degree of deterioration of the oxidation catalyst _prohibits the execution of the NO 2 -NOx ratio optimization control. The NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst due to the decrease in the NO oxidation efficiency in the oxidation catalyst and the increase in the amount of HC flowing into the selective reduction catalyst due to the decrease in the HC oxidation efficiency in the oxidation catalyst is NO Since it tends to be excessive, prohibiting the execution of the NO ₂ -NOx ratio optimization control can prevent an excessive decrease in the NO ₂ -NOx ratio.

In this case, the exhaust purification system further includes NO detection means (for example, a NO sensor 43C described later) for detecting NO in the exhaust downstream of the selective reduction catalyst in the exhaust passage, and the NO ₂ -NOx. In the ratio optimization control, it is preferable to increase the NO ₂ -NOx ratio toward the optimum value when the detected value (Vno) by the NO detecting means is larger than a predetermined value (Vno_th).

For example, when the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is at an optimum value in the vicinity of 0.5, the NOx purification rate in the selective reduction catalyst is maximized, and therefore both NO and NO ₂ are downstream. When the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is smaller than the above optimum value and the NOx purification rate is lowered while the exhaust gas flowing into the selective reduction catalyst is reduced to a NO excessive state, NO will be discharged to the side. Therefore, according to the present invention, it is detected that the detected value by the NO detecting means is larger than the predetermined value when the exhaust gas flowing into the selective reduction catalyst becomes excessive in NO and the NO ₂ -NOx ratio becomes smaller than the optimum value. In the NO ₂ -NOx ratio optimization control, the NO ₂ -NOx ratio is increased toward the optimum value accordingly, and as a result, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst becomes the optimum value. And a predetermined value smaller than the optimum value can be controlled to vibrate. As described above, according to the present invention, by controlling the NO ₂ -NOx ratio in the vicinity of the optimum value in a feedback manner based on the detected value by the NO detecting means, the operating state, operating conditions, operating history of the internal combustion engine, In addition, the NOx purification rate in the selective reduction catalyst can be maintained in the vicinity of the maximum regardless of the deterioration state of the oxidation catalyst or the selective reduction catalyst.

In this case, in the NO ₂ -NOx ratio optimization control, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is directed toward the optimum value by reducing the amount of NO discharged from the internal combustion engine. It is preferable to increase.
In this case, the exhaust purification system further includes an EGR device that recirculates a part of the exhaust gas flowing through the exhaust passage to the intake passage of the internal combustion engine, and in the NO ₂ -NOx ratio optimization control, the EGR device It is preferable to reduce the NO amount discharged from the internal combustion engine by increasing the EGR amount corresponding to the amount of exhaust gas recirculated.
In this case, in the NO ₂ -NOx ratio optimization control, the air-fuel ratio of the air-fuel mixture of the internal combustion engine is changed to a leaner side, and the oxygen concentration of the exhaust gas flowing into the oxidation catalyst is increased, whereby the selective reduction It is preferable to increase the NO ₂ -NOx ratio of the exhaust gas flowing into the catalyst.
In this case, it is preferable that the control means changes the air-fuel ratio of the air-fuel mixture according to at least one of a fuel injection parameter, a supercharging pressure, and an EGR amount corresponding to the amount of exhaust gas recirculated by the EGR device.
In this case, in the NO ₂ -NOx ratio optimization control, the temperature of the oxidation catalyst is raised within a region below the temperature at which the NO oxidation efficiency is maximized, so that NO _{2 of} the exhaust gas flowing into the selective reduction catalyst is increased. It is preferable to increase the -NOx ratio toward the optimum value.
In this case, in the NO ₂ -NOx ratio optimization control, it is preferable to increase the temperature of the oxidation catalyst by increasing at least one of the after injection amount and the post injection amount of the internal combustion engine.
In this case, in the NO ₂ -NOx ratio optimization control, the temperature of the oxidation catalyst is reduced in a region equal to or higher than the temperature at which the NO oxidation efficiency is maximized, so that NO _{2 of} the exhaust gas flowing into the selective reduction catalyst is reduced. It is preferable to increase the -NOx ratio toward the optimum value.
In this case, in the NO ₂ -NOx ratio optimization control, it is preferable to reduce the temperature of the oxidation catalyst by reducing at least one of the after injection amount and the post injection amount of the internal combustion engine.
According to the above invention, the same effects as the above-described invention using the NO ₂ detecting means instead of the NO detector.

In this case, the exhaust purification system includes NO ₂ detection means (for example, a NO ₂ sensor 43 described later) for detecting NO _{2 in} the exhaust downstream of the selective reduction catalyst in the exhaust passage, and the selective reduction catalyst. An estimation means (for example, a feed NOx estimator 34D and a NO ₂ -NOx ratio estimator 35D described later) for calculating an estimated value (Rscr_no_nox) of the NO ₂ -NOx ratio of the exhaust gas flowing into the engine, and the NO ₂ detection means Correction means (for example, a model corrector 36D described later) for correcting the estimated value (Rscr_no_nox) of the NO ₂ -NOx ratio based on the detected value (Vno2), and the control means includes the NO ₂ -NOx. The estimated value of the ratio converges in the vicinity of the optimum value (Rscr_no_nox_cmd) so that the N of exhaust flowing into the selective reduction catalyst Parameters to increase or decrease the ₂ -NOx ratio (Kegr_scr, Dt_no2, Daf_no2) preferably comprises a _NO 2 -NOx ratio controller for determining (e.g., feedforward _NO 2 -NOx ratio controller 31D below).

In the present invention, an estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is calculated, and this estimated value converges in the vicinity of the optimum value that maximizes the NOx purification rate in the selective reduction catalyst. A parameter for increasing or decreasing the NO ₂ -NOx ratio is determined. Further, NO ₂ detection means for detecting NO _{2 in} the exhaust downstream from the selective reduction catalyst is provided, and correction means for correcting the estimated value of the NO ₂ -NOx ratio based on the detected value is provided. Thereby, the estimated value of the NO ₂ -NOx ratio can be corrected in response to an unexpected change such as the deterioration of the oxidation catalyst or the inflow or adhesion of HC to the selective reduction catalyst.

In this case, the correction means, when the detection value of the NO ₂ detecting means exceeds a predetermined threshold value is corrected in the increasing side an estimate of the NO ₂ -NOx ratio, the detection value of the NO ₂ detecting means When it is less than or equal to the threshold value, it is preferable to correct the estimated value to the decreasing side.

According to the present invention, when the detected value of the NO ₂ detecting means exceeds a predetermined threshold value, the estimated value of the NO ₂ -NOx ratio is corrected to the increasing side, and when the detected value is equal to or less than the threshold value, NO ₂ by modifying the decreasing side estimates of -NOx ratio, it is possible to correct the estimated value of the _NO 2 -NOx ratio as the deviation of the actual _NO 2 -NOx ratio is eliminated.

In this case, NO detection means (for example, NO sensor 43C described later) for detecting NO in the exhaust downstream of the selective reduction catalyst in the exhaust passage, and NO ₂ -NOx of the exhaust flowing into the selective reduction catalyst An estimation unit that calculates an estimated value of the ratio; and a correction unit that corrects the estimated value of the NO ₂ -NOx ratio based on the detection value of the NO detection unit; and the control unit includes the NO ₂ -NOx It is preferable to provide a NO ₂ -NOx ratio controller that determines a parameter for increasing or decreasing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst so that the estimated value of the ratio converges in the vicinity of the optimum value.

In the present invention, an estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is calculated, and this estimated value converges in the vicinity of the optimum value that maximizes the NOx purification rate in the selective reduction catalyst. A parameter for increasing or decreasing the NO ₂ -NOx ratio is determined. In addition, NO detection means for detecting NO in the exhaust downstream of the selective reduction catalyst is provided, and correction means for correcting the estimated value of the NO2-NOx ratio based on the detected value is provided. Thereby, the estimated value of the NO 2 -NO x ratio can be corrected in response to an unexpected change such as deterioration of the oxidation catalyst or inflow or adhesion of HC to the selective reduction catalyst.

In this case, when the detection value of the NO detection means exceeds a predetermined threshold value, the correction means corrects the estimated value of the NO ₂ -NOx ratio to the decreasing side, and the detection value of the NO detection means is the threshold value. In the case of the following, it is preferable to correct the estimated value to the increasing side.

According to the present invention, when the detected value of the NO detecting means exceeds a predetermined threshold value, the estimated value of the NO ₂ -NOx ratio is corrected to the decreasing side, and when the detected value is equal to or smaller than the threshold value, NO ₂ -NOx. by modifying the increasing side an estimate of the ratio, it is possible to correct the estimated value of the NO ₂ -NOx ratio as the deviation of the actual NO ₂ -NOx ratio is eliminated.

In this case, the NO ₂ -NOx ratio controller sets a value (Eff_no_nox) that causes a delay in the estimated value (Rscr_no_nox) of the NO ₂ -NOx ratio calculated by the estimating means to the vicinity of the optimum value (Rscr_no_nox_cmd). It is preferable to converge.

According to the present invention, the intentional delay of the estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is converged to the vicinity of the optimum value, so that Considering the storage effect of NO and NO ₂ , it is possible to maintain the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst at a substantially effective value.

In this case, the estimation means searches for a predetermined map based on the input including the amount of NOx discharged from the internal combustion engine and the temperature of the oxidation catalyst, and thereby the exhaust flowing into the selective reduction catalyst. It is preferable to calculate an estimated value of the NO ₂ -NOx ratio.

In the present invention, the amount of NOx discharged from the internal combustion engine and the temperature of the oxidation catalyst, that is, parameters that greatly affect the NO ₂ generation efficiency in the oxidation catalyst are selected with high accuracy by including them in the input when searching the map. An estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the reduction catalyst can be calculated.

In this case, the estimation means calculates an estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst based on a neural network configured by connecting a plurality of neurons that output in accordance with a predetermined function. It is preferable.

In the present invention, the NOx amount is estimated based on a neural network that is excellent in reproducibility of nonlinear dynamic characteristics, so that the NO ₂ amount in the exhaust shows a nonlinear behavior, for example, during a transient state. This can be estimated with high accuracy.

  In this case, the input to the neural network includes parameters correlated with the oxygen concentration of the exhaust (for example, Gair, Gegr, Gfuel described later) and parameters correlated with the amount of unburned HC in the exhaust (for example, described later). Gfuel, Gpost, Gpilot, θpost, θmain, etc.) and a soot deposition amount (for example, Ms, which will be described later) on a filter (for example, CSF22 which will be described later) provided in the exhaust passage. It is preferable.

According to the present invention, there is a great influence on the parameters correlated to the oxygen concentration of exhaust gas and the amount of unburned HC in the exhaust gas, and the soot accumulation amount on the filter provided in the exhaust passage, that is, the NO ₂ generation efficiency in the oxidation catalyst. By including the given parameter in the input to the neural network, the estimated value of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst can be calculated with high accuracy.

In this case, the exhaust purification system further includes an EGR device that recirculates a part of the exhaust gas flowing through the exhaust passage to the intake passage of the internal combustion engine, and the NO ₂ -NOx ratio controller flows into the selective reduction catalyst. As a parameter for increasing / decreasing the NO ₂ -NOx ratio of exhaust gas, the correction value (Kegr_no2) of the target value (Gegr_cmd) of the EGR amount corresponding to the amount of exhaust gas recirculated by the EGR device from the predetermined reference value (Gegr_map) ) Is preferably determined.
In this case, the NO ₂ -NOx ratio controller uses a predetermined reference value (Tdoc_cmd) of the target value (Tdoc_cmd) of the temperature of the oxidation catalyst as a parameter for increasing or decreasing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst. It is preferable to determine a correction value (Dt_no2) from Tdoc_scr_opt).
In this case, the NO ₂ -NOx ratio controller has a predetermined air-fuel ratio target value (AF_cmd) of the air-fuel mixture of the internal combustion engine as a parameter for increasing or decreasing the NO2-NOx ratio of the exhaust gas flowing into the selective reduction catalyst. It is preferable to determine a correction value (Daf_no2) from the reference value (AF_map).

According to the present invention, by correcting the target value of the EGR amount, the target value of the temperature of the oxidation catalyst, or the target value of the air-fuel ratio of the air-fuel mixture, new hardware is added or complex combustion control is performed. The NO ₂ -NOx ratio can be increased or decreased without being lost.

It is a mimetic diagram showing composition of an engine and its exhaust purification system concerning a 1st embodiment of the present invention. It is a figure which shows the relationship between the temperature of an oxidation catalyst, and the NO oxidation efficiency in an oxidation catalyst. CO amount in each portion of the oxidation catalyst and CSF, HC amount is a diagram showing the amount of NO and NO ₂ amount. Is a diagram showing characteristics of the NOx purification rate for _NO 2 -NOx ratio of the selective reduction catalyst. It is a figure which shows the structure of the control block which concerns on the determination of the command value with respect to the lift amount of a high pressure EGR valve. And NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst is a diagram showing the relationship between the ₂ amount and the amount of NO on the downstream side NO selective reduction catalyst. The output value of the NO ₂ sensor in the case of actuating the _NO 2 -NOx ratio controller NO ₂ feedback mode is a time chart showing EGR correction coefficient, and the change in the target EGR amount. _NO 2 -NOx ratio at each part of the exhaust pipe, NO amount, NO ₂ amount is a diagram showing the amount of HC and CO amounts. If reduced EGR correction coefficient, or feed amount of NO in the case of increasing, NO ₂ -NOx ratio, and it is a diagram showing changes in the downstream side of the NO ₂ content of the selective reduction catalyst. Is a graph showing changes in EGR correction coefficient and NO ₂ sensor output values during the execution of the catalyst degradation determination mode. Is a graph showing changes in EGR correction coefficient and NO ₂ sensor output values during the execution of the catalyst degradation determination mode. Is a time chart showing an example of a mode switching procedure in the NO ₂ -NOx ratio controller. And HC and CO purification rates of the oxidation catalyst in a moving vehicle, and NO ₂ generation efficiency in the oxidation catalyst is a graph showing changes in the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst. It is a flowchart which shows the procedure which determines the target EGR amount. It is a flowchart which shows the procedure which determines the target EGR amount. NO ₂ is a simulation result at the time of non-operation of the sensor feedback mode. NO ₂ is a simulation result at the time of operation of the sensor feedback mode. And air-fuel ratio of the mixture, and NO ₂ amount on the downstream side of the oxidation catalyst and CSF, is a diagram showing the relationship between NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst. It is a block diagram which shows the structure of the engine exhaust gas purification system which concerns on 2nd Embodiment of this invention, and its ECU. 6 is a time chart showing changes in the output value of the NO ₂ sensor, the air-fuel ratio correction coefficient, and the target air-fuel ratio when the NO ₂ -NOx ratio controller is operated in the NO ₂ sensor feedback mode. Oxygen concentration in each portion of the exhaust _pipe, NO 2 -NOx ratio, NO amount, NO ₂ amount, HC amount and CO amount, and a diagram showing the air-fuel ratio of the mixture. Is a time chart showing an example of a mode switching procedure in the NO ₂ -NOx ratio controller. And the temperature of the oxidation catalyst, and NO ₂ amount on the downstream side of the oxidation catalyst and CSF, is a diagram showing the relationship between NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst. It is a block diagram which shows the structure of the engine exhaust gas purification system which concerns on 3rd Embodiment of this invention, and its ECU. The output value of the NO ₂ sensor in the case of actuating the NO ₂ -NOx ratio controller NO ₂ sensor feedback mode, temperature correction amount, and a time chart showing the changes in the target oxidation catalyst temperature. And the NOx purification rate of the selective reduction catalyst _is a diagram showing the relationship between the temperature of the NO 2 -NOx ratio and oxidation catalyst. Oxygen concentration in each portion of the exhaust _pipe, NO 2 -NOx ratio, NO amount, NO ₂ amount, HC amount and CO amount, and a diagram showing the air-fuel ratio of the mixture. Is a time chart showing an example of a mode switching procedure in the NO ₂ -NOx ratio controller. It is a block diagram which shows the structure of the engine exhaust gas purification system which concerns on 4th Embodiment of this invention, and its ECU. It is a time chart which shows the change of the output value of a NO sensor, the EGR correction coefficient, and the target EGR amount when the NO ₂ -NOx ratio controller is operated in the NO sensor feedback mode. It is a figure which shows the structure of the control block which concerns on determination of an EGR valve command value among the control blocks comprised by ECU3D of the exhaust gas purification system which concerns on 7th Embodiment of this invention. It is a figure which shows the specific example of the map which determines the reference value with respect to feed NOx amount estimated value. It is a figure which shows the specific example of the map which determines a correction coefficient based on an EGR rate. It is a figure which shows the specific example of the map which determines the reference value with respect to NO oxidation efficiency based on the estimated value of exhaust volume, and feed NOx amount estimated value. It is a figure which shows the specific example of the map which determines a correction coefficient based on oxidation catalyst temperature. It is a figure which shows the setting table of a switching function setting parameter. It is a time chart which shows the change of the correction coefficient determined by the model corrector. It is a simulation result when the oxidation catalyst and CSF are new and the feedforward control mode and the model corrector are deactivated. It is a simulation result in the case where the oxidation catalyst and the CSF are new and the feedforward control mode is operated while the model corrector is inactive. It is a simulation result when the oxidation catalyst and CSF are new and the feedforward control mode and the model corrector are operated together. It is a simulation result in the case where the oxidation catalyst and CSF are deteriorated products, and the feedforward control mode and the model corrector are deactivated. It is a simulation result in the case where the oxidation catalyst and CSF are deteriorated products and the feedforward control mode is operated while the model corrector is inactive. It is a simulation result in the case where the oxidation catalyst and the CSF are deteriorated products, and the feedforward control mode and the model corrector are operated. It is a figure which shows the neural network structure of a feed NOx estimator. It is a figure which shows a sigmoid function.

[First Embodiment]
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic diagram showing a configuration of an internal combustion engine (hereinafter referred to as “engine”) 1 and its exhaust purification system 2 according to the present embodiment. The engine 1 is a lean burn operation type gasoline engine or diesel engine, and is mounted on a vehicle (not shown).

The exhaust purification system 2 includes an oxidation catalyst 21 provided in the exhaust pipe 11 of the engine 1, a CSF (Catalyzed Soot Filter) 22 provided in the exhaust pipe 11 for collecting soot in the exhaust, and provided in the exhaust pipe 11. A selective reduction catalyst 23 for purifying NOx in the exhaust gas flowing through the exhaust pipe 11 in the presence of NH ₃ as a reducing agent, and an ammonia precursor upstream of the selective reduction catalyst 23 in the exhaust pipe 11. A urea water injection device 25 for supplying urea water, a high pressure EGR device 26 for returning a part of the exhaust gas flowing through the exhaust pipe 11 into the intake pipe 12, and an electronic control unit (hereinafter referred to as "ECU") 3 It is comprised including.

  The high pressure EGR device 26 includes a high pressure EGR pipe 261 and a high pressure EGR valve 262. The high pressure EGR pipe 261 connects the intake pipe 12 to the upstream side of the oxidation catalyst 21 in the exhaust pipe 11. The high pressure EGR valve 262 is provided in the high pressure EGR pipe 261 and controls the amount of exhaust gas recirculated through the high pressure EGR pipe 261 (hereinafter referred to as “EGR amount”). The high pressure EGR valve 262 is connected to the ECU 3 via an actuator (not shown), and the opening degree (lift amount) is electromagnetically controlled by the ECU 3.

The oxidation catalyst 21 is provided in the exhaust pipe 11 immediately below the engine 1 and upstream of the CSF 22 to oxidize and purify HC and CO in the exhaust, and also oxidizes NO in the exhaust and converts it into NO ₂ . To do.

FIG. 2 is a diagram showing the relationship between the temperature of the oxidation catalyst and the NO oxidation efficiency in the oxidation catalyst. Here, the NO oxidation efficiency refers to the ratio of the amount of NO ₂ that is oxidized and flows out by this oxidation catalyst to the amount of NO that flows into the oxidation catalyst, and can also be referred to as NO ₂ generation efficiency. As shown in FIG. 2, the NO oxidation efficiency in the oxidation catalyst shows a convex characteristic with respect to the temperature of the oxidation catalyst, and in the example shown in FIG. 2, NO is oxidized most efficiently in the vicinity of 300 ° C. ing. That is, the NO oxidation efficiency in the oxidation catalyst is lowered when the temperature of the oxidation catalyst is lower than the optimum value (300 ° C. in the example of FIG. 2), and is lowered even when the temperature is higher than the optimum value. In contrast, the oxidation efficiency of CO and HC in the oxidation catalyst basically has a characteristic of increasing with the temperature of the oxidation catalyst. That is, the higher the temperature of the oxidation catalyst, the higher the oxidation efficiency of CO and HC.

  Returning to FIG. 1, the CSF 22 is provided downstream of the oxidation catalyst 21 and upstream of the selective reduction catalyst 23 in the exhaust pipe 11. When the exhaust gas passes through the fine holes in the filter wall, the CSF 22 collects soot mainly composed of carbon in the exhaust gas by depositing soot on the surface of the filter wall and the holes in the filter wall. In addition, since an oxidation catalyst is applied to the filter wall, the filter wall has a function of oxidizing CO, HC, and NO in the exhaust similarly to the oxidation catalyst 21 described above.

In addition, you may share a function by an upstream side and a downstream side by using what has a different noble metal composition by the upstream oxidation catalyst 21 and the downstream CSF22. For example, by using a mixture of Pt and Pd for the upstream side oxidation catalyst 21, the oxidation performance of HC and CO at low temperatures is improved, and a small amount of Pd is mixed in the downstream CSF 22 and Pt is mainly used. By using the above, NO oxidation performance (that is, NO ₂ generation performance) can be improved.

FIG. 3 is a diagram showing the CO amount, HC amount, NO amount, and NO ₂ amount in each part of the oxidation catalyst and CSF.
As shown in FIG. 3, CO, HC and NO contained in the exhaust discharged from the engine are oxidized in the process of passing through the oxidation catalyst and CSF, respectively. Decrease. Further, since NO ₂ is generated by oxidizing NO, the amount of NO ₂ increases as it goes from the upstream side to the downstream side.

In addition, in the oxidation catalyst and CSF having almost the same function as this oxidation catalyst, the oxidation reaction of CO, HC and NO has a priority in the order of CO, HC and NO. That is, in the process in which the exhaust gas containing CO, HC and NO passes through the oxidation catalyst and CSF, CO is oxidized first (ie, most upstream), then HC is oxidized, and finally (ie, most) downstream side) NO is is oxidized NO ₂ is generated. That is, NO in exhaust gas is oxidized after NO of CO and HC in exhaust gas is generated, so that NO ₂ is generated. Therefore, if a lot of CO and HC are contained in exhaust gas, The NO oxidation efficiency tends to decrease before the CO and HC oxidation efficiency in the catalyst and CSF decreases.
In general, in an oxidation catalyst or CSF, as the space velocity of exhaust gas, that is, the passing amount (g / s) per unit time of substances to be oxidized (CO, HC, NO) increases, the oxidation efficiency decreases. To do. Furthermore, as described above, NO in the exhaust is oxidized on the most downstream side of the oxidation catalyst and CSF. Therefore, if the volume of exhaust exhausted from the engine increases, the CO and HC oxidation efficiency decreases. In addition, the NO oxidation efficiency tends to decrease.

  Returning to FIG. 1, the urea water injection device 25 includes a urea water tank 251 and a urea water injection valve 253. The urea water tank 251 stores urea water, and is connected to the urea water injection valve 253 via a urea water supply path 254 and a urea water pump (not shown). The urea water tank 251 is provided with a urea water level sensor 255. The urea water level sensor 255 detects the water level of the urea water in the urea water tank 251 and supplies a detection signal substantially proportional to the water level to the ECU 3. The urea water injection valve 253 is connected to the ECU 3 and is operated by a control signal from the ECU 3 to inject an amount of urea water corresponding to the control signal into the exhaust pipe 11.

The selective reduction catalyst 23 selectively reduces NOx in the exhaust in an atmosphere in which a reducing agent such as NH ₃ exists. Specifically, when urea water is injected by the urea water injection device 25, the urea water is thermally decomposed or hydrolyzed by the heat of the exhaust to generate NH ₃ as a reducing agent. The produced NH ₃ is supplied to the selective reduction catalyst 23, and NOx in the exhaust is selectively reduced by these NH ₃ .

The reaction formulas showing the reduction reaction of NO and NO ₂ in the selective reduction catalyst 23 are as shown in the following formulas (3-1), (3-2), and (3-3). The reaction shown in Formula (3-1) is a reaction that simultaneously reduces NO and NO ₂ in the exhaust gas, and the reaction shown in Formula (3-2) is a reaction that reduces only NO in the exhaust gas, The reaction shown in Formula (3-3) is a reaction that reduces only NO _{2 in} the exhaust.

In the selective reduction catalyst, the reactions shown in the above formulas (3-1) to (3-3) proceed, so that NO and NO ₂ in the exhaust are reduced by NH _3. Is considered to change according to the NO ₂ -NOx ratio.

For example, when the NO ₂ -NOx ratio is 0.5, the molar ratio of NO to NO _{2 in} the exhaust gas is 1: 1. Therefore, the selective reduction catalyst mainly performs the reaction represented by the above formula (3-1). proceed.

When the NO ₂ -NOx ratio is smaller than 0.5, that is, when there is more NO than NO ₂ , NO that cannot be reduced only by the reaction shown in the above formula (3-1) remains. NO is reduced by the progress of the reaction shown in the above formula (3-2). Therefore, when the NO ₂ -NOx ratio is smaller than 0.5, as the NO ₂ -NOx ratio decreases, the degree of progress of the reaction shown in the above equation (3-1) decreases, and the above equation (3-2) The degree of progress of the reaction shown in FIG.

On the other _hand, if NO 2 -NOx ratio is greater than 0.5, that is, if there are more NO ₂ than NO, although NO ₂ where only reactions represented by the above formula (3-1) can not be reduced remains, the Excess NO ₂ is reduced by the progress of the reaction shown in the above formula (3-3). Therefore, when the NO ₂ -NOx ratio is larger than 0.5, as the NO ₂ -NOx ratio increases, the progress of the reaction shown in the above equation (3-1) decreases, and the above equation (3- The degree of progress of the reaction shown in 3) is increased.

FIG. 4 is a graph showing the characteristics of the NOx purification rate with respect to the NO ₂ -NOx ratio in the selective reduction catalyst. The solid line shows the characteristic of the NOx purification rate of the new selective reduction catalyst, and the broken line shows the characteristic of the NOx purification rate of the deteriorated selective reduction catalyst. In the selective reduction catalyst, NO and NO ₂ are reduced by the three different reactions as described above. Therefore, the NOx purification rate in the selective reduction catalyst is NO ₂ -NOx in the inflowing exhaust gas as shown in FIG. It will change according to the ratio.
That is, the NOx purification rate in the selective reduction catalyst exhibits an upwardly convex characteristic so that it becomes maximum when the NO ₂ -NOx ratio of the inflowing exhaust gas is 0.5, regardless of the degree of deterioration. . In addition, when the ratio of decrease in the NOx purification rate when the NO ₂ -NOx ratio deviates from the optimum value is compared between the case where the NOx purification rate is larger than the optimum value and the case where it is smaller, the rate of decrease in the NOx purification rate is larger.
Further, the rate of decrease in the NOx purification rate when the NO ₂ -NOx ratio deviates from the optimum value is when the degree of deterioration of the selective reduction catalyst is small (solid line in FIG. 4) and when it is large (broken line in FIG. 4). If the degree of deterioration is larger, the rate of decrease in the NOx purification rate is larger. That is, when the degree of deterioration of the selective reduction catalyst is small, the NOx purification rate is substantially constant regardless of the NO ₂ -NOx ratio of the exhaust gas, whereas when the degree of deterioration of the selective reduction catalyst is large, the NOx purification rate. The rate varies greatly depending on the NO ₂ -NOx ratio of the exhaust.

Returning to FIG. 1, the selective reduction catalyst 23 has a function of reducing NOx in the exhaust with NH ₃ generated from urea water, and also has a function of storing the generated NH ₃ by a predetermined amount. Hereinafter, the amount of NH ₃ stored in the selective reduction catalyst 23 is referred to as a storage amount, and the amount of NH ₃ that can be stored in the selective reduction catalyst 23 is referred to as a maximum storage capacity. The NH ₃ stored in this way is also consumed as appropriate in the reduction of NOx in the exhaust. For this reason, as the storage amount increases, the NOx purification rate in the selective reduction catalyst 23 increases. On the other hand, when the storage amount reaches the storage capacity and the selective reduction catalyst 23 becomes saturated, the NOx purification rate also reaches the maximum value, but the surplus NH ₃ that is not used for NOx reduction is downstream of the selective reduction catalyst 23. NH ₃ slip is generated. In order to prevent the NH ₃ discharged to the downstream side of the selective reduction catalyst 23 from being discharged outside the system in this way, a slip suppression catalyst 24 is provided on the downstream side of the selective reduction catalyst 23. Examples of the slip suppression catalyst 24 include an oxidation catalyst that oxidizes NH ₃ slipped from the selective reduction catalyst 23 and decomposes it into N ₂ and H ₂ O, or stores slipped NH ₃ or reduces NOx in exhaust gas. A selective reduction catalyst for use in the above can be used.

In order to detect the state of the exhaust purification system 2, the ECU 3 includes an exhaust temperature sensor 41, an NH ₃ sensor 42, a NO ₂ sensor 43, a crank angle position sensor 14, an accelerator opening sensor 15, a urea water remaining amount warning lamp 16, In addition, a catalyst deterioration warning lamp 17 and the like are connected.

  The exhaust temperature sensor 41 detects the exhaust temperature downstream of the oxidation catalyst 21 and the CSF 22 and supplies a signal substantially proportional to the detected value to the ECU 3. The ECU 3 calculates the temperature Tscr of the selective reduction catalyst 23 and the temperature Tdoc of the oxidation catalyst 21 based on the detected value of the exhaust temperature sensor 41.

The NH ₃ sensor 42 detects the concentration or amount of exhaust ammonia between the selective reduction catalyst 23 and the slip suppression catalyst 24 in the exhaust pipe 11, and supplies a signal Vnh 3 that is substantially proportional to the detected value to the ECU 3. The NO ₂ sensor 43 detects the concentration or amount of NO _{2 in} the exhaust pipe 11 immediately below the selective reduction catalyst 23, and supplies the ECU 3 with a signal Vno2 that is substantially proportional to the detected value.

  The crank angle position sensor 14 detects the rotation angle of the crankshaft of the engine 1, generates a pulse at every crank angle, and supplies the pulse signal to the ECU 3. The ECU 3 calculates the rotational speed NE of the engine 1 based on this pulse signal. The crank angle position sensor 14 further generates a cylinder identification pulse at a predetermined crank angle position of the specific cylinder and supplies it to the ECU 3.

  The accelerator opening sensor 15 detects a depression amount (hereinafter referred to as “accelerator opening”) AP of an accelerator pedal (not shown) of the vehicle, and supplies a detection signal substantially proportional to the detected accelerator opening AP to the ECU 3. The ECU 3 calculates the required engine load TRQ of the engine 1 according to the accelerator opening AP and the rotational speed NE.

  The urea water remaining amount warning lamp 16 is provided, for example, on a meter panel of the vehicle, and lights up when the remaining amount of urea water in the urea water tank 251 is less than a predetermined remaining amount. As a result, the driver is warned that the remaining amount of urea water in the urea water tank 251 has decreased.

  The catalyst deterioration warning lamp 17 is provided, for example, on a meter panel of the vehicle, and is turned on when a catalyst deterioration determination value DET_SCR_AGD described later becomes “3”. As a result, the driver is warned that the selective reduction catalyst 23 is in a deteriorated state.

  The ECU 3 shapes an input signal waveform from various sensors, corrects a voltage level to a predetermined level, converts an analog signal value into a digital signal value, and a central processing unit (hereinafter, “ CPU ”). In addition, the ECU 3 includes a storage circuit that stores various calculation programs executed by the CPU and calculation results, an output circuit that outputs control signals to the engine 1, the high pressure EGR valve 262, the urea water injection valve 253, and the like. Is provided.

  FIG. 5 is a diagram showing a configuration of a control block related to determination of a command value (hereinafter referred to as “EGR valve command value”) Legr_cmd for the lift amount of the high pressure EGR valve 262 among the control blocks configured in the ECU 3.

In addition to the control block relating to the determination of the EGR valve command value Legr_cmd as shown in FIG. 5, the ECU 3 includes, for example, a urea water injection control, that is, a control block that determines the urea water injection amount from the urea water injection valve 253. Is formed. More specifically, in the urea water injection control, while estimating the storage amount and maximum storage capacity of the selective reduction catalyst 23, the downstream of the selective reduction catalyst 23 is maintained so that this storage amount is maintained in the vicinity of the maximum storage capacity. The injection amount of urea water is determined based on the detection value of the NH ₃ sensor 42 provided on the side. Thus, by maintaining the storage amount in the vicinity of the maximum storage capacity, it is possible to keep the NOx purification rate in the selective reduction catalyst 23 high while minimizing the NH ₃ slip from the selective reduction catalyst 23. The detailed algorithm of the urea water injection control as described above is described in detail in, for example, International Publication No. 2009/128169 by the applicant of the present application, and therefore, detailed description thereof is omitted here.

As shown in FIG. 5, the control block relating to the determination of the EGR valve command value Legr_cmd includes a NO ₂ -NOx ratio controller 31, a reference EGR amount map value calculation unit 32, and an EGR controller 33. .

According to this control block, the target EGR amount Gegr_cmd is obtained by multiplying the reference EGR amount Gegr_map calculated by the reference EGR amount map value calculation unit 32 by the EGR correction coefficient Kegr_no2 calculated by the NO ₂ -NOx ratio controller 31. (See the following formula (4)). The EGR valve command value Legr_cmd is calculated by the EGR controller 33 so that an estimated value for the EGR amount (hereinafter referred to as “EGR amount estimated value”) Gegr_hat matches the target EGR amount Gegr_cmd.
The target EGR amount Gegr_cmd is defined not only by multiplying the reference EGR amount Gegr_map by the EGR correction coefficient Kegr_no2 but also by adding the EGR correction coefficient Kegr_no2 to the reference EGR amount Gegr_map as shown in the equation (4). May be.

  Here, the symbol (k) is a symbol indicating the discretized time, and indicates that the data is detected or calculated every predetermined control period. That is, when the symbol (k) is data detected or calculated in the current control cycle, the symbol (k-1) indicates that the data is detected or calculated in the previous control cycle. In the following description, the symbol (k) is omitted as appropriate.

The reference EGR amount map value calculation unit 32 determines a reference EGR amount Gegr_map by searching a predetermined map based on the engine speed NE and the required engine load TRQ. The map in the reference EGR amount map value calculation unit 32 has a NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst 23 in excess of NO ₂ in consideration of a balance of fuel consumption, soot amount, feed NOx amount, and the like. It is preferable that the setting is made so as to make a difference.

The NO ₂ -NOx ratio controller 31 operates in one of the following four control modes, and calculates the EGR correction coefficient Kegr_no2 for correcting the map value Gegr_map described above based on an algorithm that is different in each control mode. .
1. NO ₂ sensor feedback mode 2. Catalyst deterioration judgment mode 3. Fuel economy priority mode NO ₂ Generation Priority Mode Hereinafter, the procedure for calculating the EGR correction coefficient Kegr_no2 by these four control modes will be described in order.

<NO ₂ sensor feedback mode>
In the NO ₂ sensor feedback mode, the NO ₂ -NOx ratio controller 31 controls the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst toward an optimum value that maximizes the NOx purification rate. More specifically, as _NO 2 -NOx ratio of the exhaust flowing into the selective reduction catalyst is averagely maintained near its optimum value, the EGR correction coefficient Kegr_no2 based on the output value Vno2 of NO ₂ sensor decide.

In this mode, the output value Vno2 of the NO ₂ sensor proportional to the NO ₂ concentration in the exhaust gas immediately below the selective reduction catalyst is converted into an output deviation E_Vno2 defined by the following equation (5). That is, the output deviation E_Vno2 of the NO ₂ sensor becomes “0” when the output value Vno2 is equal to or smaller than the predetermined NO ₂ detection threshold value Vno2_th, and when the output value Vno2 is larger than the NO ₂ detection threshold value Vno2_th, the deviation (Vno2 −Vno2_th).

Hereinafter, as will be described in detail, the NO ₂ detection threshold value Vno2_th is a threshold value used for determining the presence of NO ₂ on the downstream side of the selective reduction catalyst based on the output value Vno2 of the NO ₂ sensor. Ideally, it should be set to a value slightly larger than “0”. However, in consideration of the influence of solid dispersion, aging, interference gas, etc. of the NO ₂ sensor, downstream of the selective reduction catalyst. The value is set such that it can be reliably determined that NO ₂ is discharged to the side.

Here, the difference between the state in which the output deviation E_Vno2 is 0 and the state in which the output deviation E_Vno2 is a positive value other than 0 will be described with reference to FIG.
FIG. 6 is a diagram showing the relationship between the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst and the NO ₂ amount and NO amount on the downstream side of the selective reduction catalyst. In FIG. 6, the solid line indicates the relationship between the NO ₂ -NOx ratio, the NO amount, and the NO ₂ amount when the selective reduction catalyst is in an ideal state. Here, the ideal state of the selective reduction catalyst is that no HC adheres to or flows into the selective reduction catalyst, and therefore the NOx purification rate is maximum when the NO ₂ -NOx ratio becomes approximately 0.5. The state which becomes. The broken line indicates the relationship between the NO ₂ -NOx ratio, the NO amount, and the NO ₂ amount when the amount of HC flowing into the selective reduction catalyst is large. For example, if the HC inflow to the selective reduction catalyst increases due to deterioration of the upstream oxidation catalyst, NO ₂ is consumed and NO increases as shown in the above formula (1). The NO ₂ -NOx ratio that maximizes the rate is greater than the ideal state. The thick line indicates the NO ₂ amount, and the thin line indicates the NO amount.

As shown by the thick solid line in FIG. 6, the NO ₂ amount on the downstream side of the selective reduction catalyst in an ideal state is approximately 0.5 or less where the NO ₂ -NOx ratio is the maximum of the NOx purification rate. _, the next approximately 0 regardless of the NO 2 -NOx _ratio, when NO 2 -NOx ratio is approximately 0.5 or _more, increases as NO 2 -NOx ratio increases. Further, as shown by a thick broken line in FIG. 6, the NO ₂ amount on the downstream side of the selective reduction catalyst in a state where the HC inflow amount is large is approximately 0.6 in which the NO ₂ -NOx ratio becomes the maximum of the NOx purification rate. the case is _less, next approximately 0 regardless of the NO 2 -NOx _ratio, when NO 2 -NOx ratio is approximately 0.6 or _more, increases as NO 2 -NOx ratio increases.
That is, the amount of NO ₂ on the downstream side of the selective reduction catalyst does not depend on the state, and the NO ₂ -NOx ratio exceeds the optimum value at which the NOx purification rate becomes maximum, that is, the exhaust gas flowing into the selective reduction catalyst Increased when NO _{2 is} excessive.

In addition, as shown by a thin solid line in FIG. 6, the NO amount on the downstream side of the selective reduction catalyst in an ideal state is approximately 0.5 or more at which the NO ₂ -NOx ratio becomes the maximum of the NOx purification rate. in this _case, it becomes substantially 0 regardless of the NO 2 -NOx _ratio, when NO 2 -NOx ratio is approximately 0.5 or _less, increases as NO 2 -NOx ratio decreases. Further, as shown by a thin broken line in FIG. 6, the NO ₂ amount on the downstream side of the selective reduction catalyst in a state where the amount of HC inflow is large is approximately 0.6 in which the NO ₂ -NOx ratio becomes the maximum of the NOx purification rate. the case is greater than or _equal, it becomes substantially 0 regardless of the NO 2 -NOx _ratio, when NO 2 -NOx ratio is approximately 0.6 or _less, increases as NO 2 -NOx ratio decreases.
That is, the amount of NO on the downstream side of the selective catalytic reduction catalyst does not depend on the state, and the NO ₂ -NOx ratio falls below the optimum value at which the NOx purification rate becomes maximum, that is, the exhaust flowing into the selective catalytic reduction catalyst is NO. It increases when it becomes excessive.

From the above, in the above equation (5), the state in which the output deviation E_Vno2 becomes “0” means that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is close to the optimum value at which the NOx purification rate becomes maximum. It can be determined that there is an excessive NO value that is smaller than the optimum value.
On the other hand, the state where the output deviation E_Vno2 becomes a positive value can be determined that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is in a state where the NO _{2 is} excessively larger than the optimum value at which the NOx purification rate is maximized. .

In the NO ₂ sensor feedback mode, the EGR correction coefficient Kegr_no2 is calculated based on the following equations (6), (7), and (8) using the output deviation E_Vno2 having the above meaning.

As shown in Expression (8), the EGR correction coefficient Kegr_no2 is set between the upper limit value and the lower limit value, with the upper limit value being “1” and the lower limit value being Kegr_no2_L.
The feedback gain Ki_no2 in the equation (7) is set to a negative value. Thereby, when the output deviation E_Vno2 becomes a positive value, that is, when the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst becomes a state of excessive NO ₂ , the EGR correction coefficient Kegr_no2 is gradually decreased. Can do.
Equation (6) first subtraction amount Dkegr_DEC in, when the output deviation E_Vno2 is turned to a positive value from "0", i.e. only definitive EGR correction when the output value Vno2 of NO ₂ sensor exceeds the NO ₂ detection threshold Vno2_th This corresponds to the amount of change of the coefficient Kegr_no2, and is set to a negative value. Further, the return amount Dkegr_INC in the equation (6) is the state where the output deviation E_Vno2 is “0”, that is, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is at an optimum value or is in an excessive NO state. This corresponds to the amount of change in the EGR correction coefficient Kegr_no2 in a certain case, and is set to a positive value.

7, the above equation (5) to the output value of the NO ₂ sensor in the case of actuating the _NO 2 -NOx ratio controller defined NO ₂ sensor feedback mode (8) Vno2, EGR correction coefficient Kegr_no2, and the target It is a time chart which shows the change of EGR amount Gegr_cmd.

Between times t1 to t2, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. In this case, the EGR correction coefficient Kegr_no2 increases toward the upper limit by the return amount Dkegr_INC set in Expression (6). As a result, the target EGR amount Gegr_cmd gradually increases so as to approach the map value Gegr_map, and as a result, the NO amount discharged from the engine gradually decreases as compared to the case where the EGR correction coefficient Kegr_no2 is not increased. .

Next, at time t2, the output value Vno2 of NO ₂ sensor exceeds the NO ₂ detection threshold Vno2_th. At this moment, the EGR correction coefficient Kegr_no2 is reduced by the initial subtraction amount Dkegr_DEC set by the equation (6). Thereby, the target EGR amount Gegr_cmd is instantaneously changed to a smaller value so as to be away from the map value Gegr_map. Thereafter, from time t2 the output value Vno2 of NO ₂ sensor to the time t3 below NO ₂ detection threshold Vno2_th, EGR correction coefficient Kegr_no2, as shown in equation (7), decreased by amount proportional to the output deviation E_Vno2 To do. As a result, the target EGR amount Gegr_cmd decreases further away from the map value Gegr_map, and as a result, the NO amount discharged from the engine gradually increases as compared with the case where the EGR correction coefficient Kegr_no2 is not decreased. .

Between times t3 to t4, and time t5 and later, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. Therefore, since the EGR correction coefficient Kegr_no2 and the target EGR amount Gegr_cmd during this period exhibit the same qualitative behavior as from the time t1 to the time t2, detailed description thereof is omitted. The time until t4 to t5, the output value Vno2 of NO ₂ sensor is greater than the NO ₂ detection threshold Vno2_th. Accordingly, during this period, the EGR correction coefficient Kegr_no2 and the target EGR amount Gegr_cmd behave qualitatively from the time t2 to t3, and detailed description thereof is omitted.

  Next, a process of decreasing the target EGR amount Gegr_cmd so as to be away from the map value Gegr_map (time t2 to t3, t4 to t5 in FIG. 7) is executed in response to the output deviation E_Vno2 becoming a positive value. The effect by doing will be described with reference to FIG.

FIG. 8 is a diagram showing the NO ₂ -NOx ratio, the NO amount, the NO ₂ amount, the HC amount, and the CO amount in each part of the exhaust pipe. In FIG. 8, the broken line indicates an example of a conventional technique that continues to adopt the map value Gegr_map as the target EGR amount Gegr_cmd, and the solid line indicates the target EGR amount Gegr_cmd in response to the output deviation E_Vno2 having a positive value. The example of this embodiment reduced so that it may leave | separate from map value Gegr_map is shown.

First, in the conventional technique shown by the broken line, the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst is in a state of greatly surpassed NO ₂ excessive optimum value near 0.5, NO ₂ is that has not been purified It will be discharged downstream of the selective reduction catalyst.

On the other hand, in the present embodiment, when the output deviation E_Vno2 is a positive value and is in a state of excessive NO ₂ , the EGR amount is reduced by decreasing the target EGR amount Gegr_cmd away from the map value Gegr_map. . When the EGR amount decreases, the NO amount discharged from the engine (hereinafter referred to as “feed NO amount”) increases as compared to the conventional method, the HC amount and CO amount discharged from the engine decrease, and the exhaust volume becomes smaller. To increase. Note that the amount of NO ₂ discharged from the engine is very small compared to the amount of feed NO, and does not change much depending on the amount of EGR.

When the feed NO amount increases and the exhaust volume increases, as described above, the CO and HC oxidation efficiency in the oxidation catalyst and CSF does not change, and the NO ₂ production efficiency first decreases. As a result, compared to the conventional method, the amount of NO (residual NO) flowing into the selective reduction catalyst on the downstream side without being oxidized by the oxidation catalyst and CSF increases (residual NO amount), and is generated by the oxidation catalyst and CSF and becomes the selective reduction catalyst. The amount of inflowing NO ₂ decreases.

As described above, in the present embodiment, the amount of NO flowing into the selective reduction catalyst is increased and the amount of NO ₂ is decreased compared to the conventional method in which the NO ₂ -NOx ratio is significantly higher than the optimum value in the vicinity of 0.5. As a result, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward an optimum value in the vicinity of 0.5, and as a result, both the NO amount and NO ₂ amount discharged from the selective reduction catalyst are suppressed. it can.

In this embodiment, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward the optimum value by reducing the EGR amount. However, the method of reducing the NO ₂ -NOx ratio is an EGR method. It is not limited to adjusting the amount. As shown in FIG. 8, the amount of NOx discharged from the engine increases more than the effect of increasing the NO ₂ generation efficiency in the oxidation catalyst and CSF due to the decrease in the amount of HC and CO discharged from the engine. The NO ₂ -NOx ratio is reduced by setting combustion parameters correlated with the combustion state of the engine so that the effect of lowering the NO ₂ production efficiency in the oxidation catalyst and CSF is greater. Also good. Examples of the combustion parameter include a fuel injection amount, fuel injection timing, a supercharging pressure, and an EGR amount.

Next, the setting policy of the initial subtraction amount Dkegr_DEC and the return amount Dkegr_INC in the above equation (6) will be described with reference to FIG.
FIG. 9 shows the amount of feed NO when the EGR correction coefficient Kegr_no2 is decreased from “1” to “0” or increased from “0” to “1”, and the NO of exhaust flowing into the selective reduction catalyst. ₂ -NOx ratio, and is a diagram showing changes in the downstream side of the NO ₂ content of the selective reduction catalyst. In FIG. 9, the solid line indicates when the EGR correction coefficient Kegr_no2 is decreased, and the broken line indicates when the EGR correction coefficient Kegr_no2 is increased.

The amount of NO ₂ on the downstream side of the selective reduction catalyst, that is, the NO ₂ purification rate in the selective reduction catalyst has a hysteresis characteristic as shown in FIG.
For example, when the correction coefficient Kegr_no2 is decreased from “1” to “0” and the EGR amount is decreased, the feed NO amount increases, and accordingly, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst becomes 0. The NOx purification rate in the selective reduction catalyst increases. However, while changing the correction coefficient Kegr_no2 from “1” to “0”, in the region indicated by Δ in FIG. 9, the NOx purification due to the increase in the feed NO amount rather than the NOx purification rate increasing effect in the selective reduction catalyst. The lowering effect of the rate wins, and as a result, the NO ₂ amount on the downstream side of the selective reduction catalyst temporarily increases.
On the other hand, when the correction coefficient Kegr_no2 is increased from “0” to “1”, the NO ₂ amount on the downstream side of the selective reduction catalyst is qualitative as shown in the broken line in FIG. Shows different behavior. That is, when the correction coefficient Kegr_no2 is increased, the amount of NO ₂ on the downstream side of the selective reduction catalyst is not temporarily increased even when the correction coefficient Kegr_no2 is passing through the region Δ. This is because the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is more than 0.5 even when passing through the region Δ due to the NO storage effect generated in the selective reduction catalyst by some mechanism or the NO ₂ storage effect. This is considered to be because the effective NO ₂ -NOx ratio in the selective reduction catalyst is maintained in the vicinity of 0.5 in spite of the increase in the value of NO.

For the selective reduction catalyst having the above hysteresis characteristics with respect to the correction coefficient Kegr_no2, the initial subtraction amount DKegr_DEC and the return amount Dkegr_INC are maintained so that the NO ₂ purification rate is maintained high even when the correction coefficient Kegr_no2 is decreased or increased. Is preferably set as follows.
That is, in order to prevent the NO ₂ purification rate from temporarily deteriorating when the correction coefficient Kegr_no2 is decreased, the initial subtraction amount Dkegr_DEC can instantaneously pass through the region Δ where the NO ₂ purification rate is temporarily degraded. Set to the correct value.
When the correction coefficient Kegr_no2 is increased, the correction coefficient Kegr_no2 is gradually increased over a longer period of time than the above decrease so that the NO storage effect or the NO ₂ storage effect as described above can be reliably achieved. Is preferred. For this reason, the return amount Dkegr_INC is set to a value such that the increase in the correction coefficient Kegr_no2 becomes moderate.

  As described above, in the present embodiment, the initial subtraction amount Dkegr_DEC and the return amount Dkegr_INC are fixed values. However, the present invention is not limited to this, and the NOx on the upstream side of the engine speed, load, exhaust system temperature, selective reduction catalyst You may change according to quantity.

<Catalyst deterioration judgment mode>
Returning to FIG. 5, in the catalyst deterioration determination mode, the NO ₂ -NOx ratio controller 31 determines a catalyst deterioration determination value DET_SCR_AGD indicating the deterioration degree of the selective reduction catalyst 23.
More specifically, in the catalyst deterioration determination mode, the EGR correction coefficient Kegr_no2 is changed based on the following formulas (9) and (10), and the EGR correction coefficient Kegr_no2 at this time and the output value Vno2 of the NO ₂ sensor are used. Then, the deterioration determination parameter J_SCR is updated by the following equation (11), and the catalyst deterioration determination value DET_SCR_AGD is determined according to the magnitude of the deterioration determination parameter J_SCR.

FIG. 10 is a diagram showing changes in the EGR correction coefficient Kegr_no2 and the output value Vno2 of the NO ₂ sensor when the catalyst deterioration determination mode is executed.
In the catalyst deterioration determination mode, as shown in the following formulas (9) and (10), the initial value Kegr_no2_temp (0) is set to “1”, and from there, the decrease amount Dkegr_JD_DEC (<0) is added by EGR. The correction coefficient Kegr_no2 is decreased from “1” to the lower limit value Kegr_no2_L.

When the EGR correction coefficient Kegr_no2 is decreased from “1” in this way, as described with reference to FIG. 9, the amount of feed NO increases from the state of excessive NO ₂ , and the NO of exhaust flowing into the selective reduction catalyst _{The 2-} NOx ratio gradually approaches the optimum value, and the NO ₂ amount on the downstream side of the selective reduction catalyst once increases and then decreases.
On the other hand, when the degradation of the selective reduction catalyst proceeds, as described with reference to FIG. 4, the NOx purification rate decreases according to the deviation of the NO ₂ -NOx ratio from the optimum value. Therefore, when the EGR correction coefficient Kegr_no2 is decreased from “1” from the state of excessive NO ₂ and the feed NO amount is continuously increased, the timing when the output value Vno2 of the NO ₂ sensor falls below the predetermined deterioration determination threshold value Vno2_JD_th is It is considered that the selective reduction catalyst becomes slower as the deterioration progresses.

Therefore, in this catalyst deterioration determination mode, as shown in the following equation (11), the deterioration determination parameter J_SCR is fixed by the value of the EGR correction coefficient Kegr_no2 when the output value Vno2 of the NO ₂ sensor falls below the deterioration determination threshold value Vno2_JD_th. To do.

Since the degradation determination parameter J_SCR obtained in this way is considered to decrease as the degradation of the selective reduction catalyst proceeds, the degradation determination parameter J_SCR and a predetermined threshold value J_SCR_AGD are expressed as shown in the following equation (12). To determine the catalyst deterioration determination value DET_SCR_AGD indicating the degree of deterioration of the selective reduction catalyst. That is, when J_SCR is “1”, the catalyst deterioration determination value DET_SCR_AGD is set to “1” which means that the selective reduction catalyst is almost new. When J_SCR is less than 1 and greater than or equal to the threshold value J_SCR_AGD, the catalyst deterioration determination value DET_SCR_AGD is set to “2” which means that the selective reduction catalyst has not deteriorated so much and is normal. When J_SCR is smaller than the threshold value J_SCR_AGD, the catalyst deterioration determination value DET_SCR_AGD is set to “3”, which means that the selective reduction catalyst is in a state where deterioration has progressed. The initial value of the catalyst deterioration determination value DET_SCR_AGD is “0”.

In addition to determining the degree of deterioration by reducing the EGR correction coefficient Kegr_no2 as shown in the above equations (9) to (12), the EGR is reversed as shown in FIG. 11 and the following equations (13) to (16). The degree of deterioration can also be determined by increasing the correction coefficient Kegr_no2.
In this case, as shown in the following formulas (13) and (14), the initial value Kegr_no2_temp (0) is set to the lower limit value Kegr_no2_L, and from there, the increment DKegr_JD_INC (> 0) is added, thereby the EGR correction coefficient Kegr_no2 Is increased from the lower limit value Kegr_no2_L to “1”.

Then, as shown in the following equation (15), the deterioration determination parameter J_SCR is fixed by the value of the EGR correction coefficient Kegr_no2 when the output value Vno2 of the NO ₂ sensor exceeds the deterioration determination threshold value Vno2_JD_th, and the following equation (16) As shown in FIG. 4, the catalyst deterioration determination value DET_SCR_AGD is determined by comparing the deterioration determination parameter J_SCR and the threshold value J_SCR_AGD.

  Note that, when the degree of deterioration is determined by increasing the EGR correction coefficient Kegr_no2 by the above formulas (13) to (16), the increase amount DKegr_JD_INC is ensured so that the storage effect described with reference to FIG. Is preferably set to a value sufficiently smaller than the absolute value of the decrease amount DKegr_JD_DEC in the above equation (9), and the EGR correction coefficient Kegr_no2 is preferably increased gradually.

<Fuel consumption priority mode>
Returning to FIG. 5, in the fuel efficiency priority mode, the NO ₂ -NOx ratio controller 31 does not depend on the output value Vno ₂ of the NO ₂ sensor so that the fuel efficiency is improved as compared with the execution of the NO ₂ feedback mode, and the EGR correction coefficient Kegr_no 2. Is set to the fuel efficiency priority EGR correction coefficient Kegr_no2_opt (see the following equation (17)).

  In the case of a gasoline engine, basically, increasing the EGR amount tends to improve fuel efficiency. Therefore, this fuel efficiency priority EGR correction coefficient Kegr_no2_opt is set to a value close to “1”, for example. On the other hand, in the case of a diesel engine, the fuel efficiency tends to improve when the amount of EGR is decreased. Therefore, the fuel efficiency priority EGR correction coefficient Kegr_no2_opt is set to a value close to the lower limit value Kegr_no2_L, for example.

<NO ₂ generation priority mode>
In the NO ₂ generation priority mode, the NO ₂ -NOx ratio controller 31 generates a large amount of NO ₂ in the oxidation catalyst 21 and the CSF 22 so that the amount of NO ₂ in the exhaust gas flowing into the selective reduction catalyst 23 increases. As shown in (18), the EGR correction coefficient Kegr_no2 is set to “1”.

As described above, the NO ₂ -NOx ratio controller can be operated in four different control modes including the NO ₂ sensor feedback mode, the catalyst deterioration determination mode, the fuel consumption priority mode, and the NO ₂ generation priority mode. It has become. Next, a preferable time for executing each mode will be described.

First, as described with reference to FIG. 4, when the selective reduction catalyst is a new product, the NOx purification rate is maintained high regardless of the NO ₂ -NOx ratio. Therefore, NOx purification ratio of the selective reduction catalyst without accurately controlling the NO ₂ -NOx ratio to the optimum value is high, the effect of NO ₂ sensor feedback mode is small.
Therefore, when the catalyst deterioration determination value DET_SCR_AGD is “1” and it can be determined that the selective reduction catalyst has not deteriorated, the NO ₂ sensor feedback mode is prohibited and the fuel efficiency priority mode (the above formula (17)) is set. It is preferable to improve the fuel efficiency as compared with the execution of the NO ₂ sensor feedback mode. In addition, when the catalyst deterioration determination value DET_SCR_AGD is “2” or “3”, that is, when it can be determined that the deterioration of the selective reduction catalyst has progressed to some extent, it is preferable to permit execution of the NO ₂ sensor feedback mode.

Further, when the deterioration degree of the oxidation catalyst and the CSF is determined and it is determined that the deterioration degree is small, the execution of the NO ₂ sensor feedback mode is permitted. When it is determined that the deterioration degree is large, the NO ₂ sensor It is preferable to prohibit the execution of the feedback mode and execute the NO ₂ generation priority mode, for example. Here, as a method of determining the degree of deterioration of the oxidation catalyst or CSF, for example, a conventionally known method such as a method based on the measured value of the oxygen storage capacity of these oxidation catalyst and CSF, which is considered to decrease with the progress of deterioration. Is used. In addition, when the degree of deterioration of the oxidation catalyst and CSF is small and the oxygen storage capacity is large, the delay in the downstream oxygen concentration change with respect to the oxygen concentration change in the upstream exhaust gas is considered to increase, so this oxygen storage capacity is, for example, It can be measured from the time difference between the outputs of the air-fuel ratio sensors provided on the upstream side and the downstream side. The determination of the degree of deterioration based on this method is limited to the case where the oxidation catalyst contains oxygen storage ability such as ceria (also called a three-way catalyst). When targeting an oxidation catalyst that does not contain oxygen storage capacity, for example, the temperature rise control is performed by comparing the rise pattern of the oxidation catalyst temperature during the temperature rise control immediately after engine startup with a preset reference pattern. It is also possible to detect that the temperature increase rate of the oxidation catalyst at that time has decreased, and to determine the degree of deterioration based on this.

FIG. 12 is a time chart showing an example of a mode switching procedure in the NO ₂ -NOx ratio controller.
In the example shown in FIG. 12, the NO ₂ -NOx ratio controller is operated in the order of the NO ₂ generation priority mode, the catalyst deterioration determination mode, and the NO ₂ sensor feedback mode after starting the engine at time “0”. Show.
As described above, since the execution of the NO ₂ sensor feedback mode is determined according to the value of the catalyst deterioration determination value DET_SCR_AGD, as shown in FIG. 12, the catalyst deterioration determination is performed before the execution of the NO ₂ sensor feedback mode is determined. It is preferable to execute the mode and determine the catalyst deterioration determination value DET_SCR_AGD. However, this catalyst deterioration determination value DET_SCR_AGD is unlikely to change frequently, so it is not necessary to perform it every time the engine is started. In this case, for example, the execution of the NO ₂ sensor feedback mode may be determined based on the previous execution result of the catalyst deterioration determination mode.

FIG. 13 is a diagram showing changes in the HC and CO purification rates of the oxidation catalyst, the NO ₂ generation efficiency in the oxidation catalyst, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst in a traveling vehicle. In FIG. 13, the broken line indicates a case where an oxidation catalyst having a high oxidation performance is used, and the alternate long and short dash line indicates a case where an oxidation catalyst having a low oxidation performance is used. Further, the solid line in FIG. 13 shows ideal characteristics that can efficiently purify all HC, CO, and NOx in the exhaust during and after the engine is warmed up.

As shown in FIG. 13, when an oxidation catalyst with high oxidation performance is used, the rise of the HC and CO purification rates during warm-up immediately after engine startup is greater than when an oxidation catalyst with low oxidation performance is used. It can be done quickly and the CO and HC purification rate after warm-up can be increased. For this reason, in order to increase the CO and HC purification rate, it is preferable to use an oxidation catalyst having high oxidation performance. That is, the characteristics of the HC and CO purification rates when using an oxidation catalyst with high oxidation performance are consistent with the ideal characteristics during and after warm-up.
In addition, when an oxidation catalyst with high oxidation performance is used, the rise of NO ₂ generation efficiency during warm-up immediately after engine startup can be accelerated, and therefore the NO ₂ -NOx ratio flowing into the selective reduction catalyst can be quickly reduced to 0.5. It can be raised to an optimal value in the vicinity. Therefore, the characteristics of NO ₂ production efficiency and NO ₂ -NOx ratio in the case of using an oxidation catalyst having high oxidation performance coincide with the ideal characteristics during warm-up.

However, when an oxidation catalyst having a high oxidation performance is used, the NO ₂ generation efficiency after warm-up becomes too high, and the NO ₂ -NOx ratio greatly exceeds the optimum value in the vicinity of 0.5. As a result, NOx A purification rate will fall. Therefore, after warm-up, the NO ₂ production efficiency and the NO ₂ -NOx ratio are closer to ideal characteristics when using an oxidation catalyst with low oxidation performance.

As described above, even if an oxidation catalyst with high oxidation performance is used or an oxidation catalyst with low oxidation performance is used, both the HC and CO purification rates and the NO ₂ generation efficiency are ideal characteristics indicated by solid lines. Therefore, simply changing the specifications such as the noble metal loading, cell density, volume, and noble metal composition of the oxidation catalyst and adjusting its oxidation performance during and after warm-up immediately after engine startup In addition, it is considered difficult to efficiently purify all HC, CO, and NOx.

Therefore, in the present embodiment, as shown in FIG. 12, the execution of the NO ₂ sensor feedback mode is prohibited during the period from the start of the engine start until the predetermined time elapses (during warming up). The NO ₂ generation priority mode is executed. That is, during the warm-up, the NO ₂ generation priority mode is executed, and the NO ₂ generation efficiency is quickly raised together with the HC and CO purification rates. After a predetermined time has elapsed since the start of the engine (after warming up), the execution of the NO ₂ sensor feedback mode is permitted, and the NO ₂ -NOx ratio is maintained at an optimum value in the vicinity of 0.5. In addition, the NOx purification rate is kept high together with the HC and CO purification rates.
As described above, the NO ₂ sensor feedback mode is similarly prohibited and the NO ₂ generation priority mode is executed when the temperature of the oxidation catalyst is lower than the activation temperature in addition to the warming up immediately after the engine is started. Then, when the temperature of the oxidation catalyst is equal to or higher than its activation temperature, execution of the NO ₂ sensor feedback mode may be permitted.

  14 and 15 are flowcharts showing a procedure for determining the target EGR amount Gegr_cmd by the ECU configured as described above.

First, in S1, it is determined whether or not the urea water injection device is in a failed state. If this determination is NO and the urea water injection device is in a normal state, the process proceeds to S2. In S2, it is determined whether or not the high-pressure EGR device is in a failed state. If this determination is NO and the high-pressure EGR device is in a normal state, the process proceeds to S3. In S3, it is determined whether various sensors such as an NH ₃ sensor, a temperature sensor, and an NO ₂ sensor are in a failed state. If this determination is NO and all of the above sensors are normal, the routine proceeds to S4. Further, when it is determined as YES in any of these S1 to S3, that is, when it is determined that any one of the urea water injection device, the high pressure EGR device, and the sensor is in a failed state. Moves to S5, forcibly sets the target EGR amount Gegr_cmd to “0”, and then moves to S17.
In S4, a reference EGR amount Gegr_map is determined by searching a predetermined map based on parameters such as engine speed and required engine load, and the process proceeds to S6.

In S6, it is determined whether or not a predetermined warm-up time set for warming up the oxidation catalyst to the activation temperature has elapsed since the engine was started. If this determination is NO and the engine is warming up, the process proceeds to S7.
In S7, after determining the EGR correction coefficient Kegr_no2 in the NO ₂ generation priority mode, the process proceeds to S16. In the NO ₂ generation priority mode, the EGR correction coefficient Kegr_no2 is set to “1” (see the above equation (18)), and as a result, the map value Gegr_map is adopted as the target EGR amount Gegr_cmd.

If the determination in S6 is YES, and if it is after warm-up, the process proceeds to S8 to determine whether or not the selective reduction catalyst is in an active state. More specifically, it is determined whether or not the temperature Tscr of the selective reduction catalyst is higher than a threshold value Tscr_act (for example, 250 ° C.) set for determining the active state. When the determination in S8 is NO and the selective reduction catalyst is not in the active state, the process also proceeds to S7, and the EGR correction coefficient Kegr_no2 is determined in the NO ₂ generation priority mode as in the warm-up.

If the determination in S8 is YES and the selective reduction catalyst is in the active state, the process moves to S9, whether the storage amount of the selective reduction catalyst is sufficient, or whether NH ₃ slip has occurred. Is determined. More specifically, whether or not the storage amount is sufficient can be determined based on whether or not the ratio of the estimated value of the storage amount to the estimated value of the maximum storage capacity is a predetermined value (for example, 20% or more). Whether or not NH ₃ slip has occurred is determined based on whether or not the output value Vnh3 of the NH ₃ sensor is equal to or greater than a predetermined threshold value. If the determination in S9 is NO and the NOx purification performance of the selective reduction catalyst is not sufficient, the process also proceeds to S7, and the EGR correction coefficient Kegr_no2 is determined in the NO ₂ generation priority mode, as in the warm-up.

  When the determination in S9 is YES and the NOx purification performance of the selective reduction catalyst is sufficient, the process proceeds to S10, and it is determined whether or not the degree of deterioration of the selective reduction catalyst has been determined. More specifically, the determination can be made based on whether or not the value of the catalyst deterioration determination value DET_SCR_AGD is other than the initial value “0” indicating that the catalyst deterioration determination mode is not executed.

  If the determination in S10 is NO and the degree of deterioration of the selective reduction catalyst has not yet been determined since the engine was started, the process proceeds to S11, and in the catalyst deterioration determination mode, the EGR correction coefficient Kegr_no2 and the catalyst deterioration determination value DET_SCR_AGD are After determining, the process proceeds to S16. More specifically, the EGR correction coefficient Kegr_no2 and the catalyst deterioration determination value DET_SCR_AGD are determined based on the above equations (9) to (12) (or equations (13) to (16)).

  When the determination in S10 is YES and the degree of deterioration of the selective reduction catalyst has been determined, the process proceeds to S12 to determine the degree in more detail. In S12, it is determined whether or not the selective reduction catalyst is in a new state, that is, whether or not the catalyst deterioration determination value DET_SCR_AGD is “1”. If the determination in S12 is YES and the selective reduction catalyst is in a new state, the process proceeds to S13, and after determining the EGR correction coefficient Kegr_no2 in the fuel efficiency priority mode, the process proceeds to S16. In the fuel efficiency priority mode, the EGR correction coefficient Kegr_no2 is set to the fuel efficiency priority EGR correction coefficient Kegr_no2_opt (see the above formula (17)).

If the determination in S12 is NO, and therefore the selective reduction catalyst is in a state that has deteriorated to some extent, the process proceeds to S20, where the deterioration degree of the oxidation catalyst and CSF is determined, and whether or not the deterioration degree is large is determined. . If the determination in S20 is YES and it is determined that the degree of deterioration of the oxidation catalyst and CSF is large, execution of the NO ₂ sensor feedback mode is prohibited, and the process proceeds to S7 to determine the EGR correction coefficient Kegt_no2 in the generation priority mode. To do.

If the determination in S20 is NO, and therefore the deterioration of the oxidation catalyst and CSF has not progressed significantly and the selective reduction catalyst has deteriorated to some extent, the current engine operating state is determined as NO ₂ sensor feedback. To determine whether or not the mode is suitable for operating the mode, the process proceeds to S14. In S14, for example, it is determined whether or not the exhaust volume is large. As described with reference to FIG. 3, the NO ₂ generation efficiency in the oxidation catalyst and CSF decreases as the exhaust volume increases. Therefore, when the determination in S14 is YES and the exhaust volume is in a large state (high load operation state), the process proceeds to S7 to prevent an excessive decrease in NO ₂ generation efficiency, and in the NO ₂ generation priority mode. An EGR correction coefficient Kegr_no2 is determined.

If the determination in S14 is YES and the exhaust volume is small, the process proceeds to S15, and after determining the EGR correction coefficient Kegr_no2 in the NO ₂ sensor feedback mode, the process proceeds to S16. More specifically, the EGR correction coefficient Kegr_no2 is determined based on the above formulas (5) to (8).

  In S16, the target EGR amount Gegr_cmd is determined by multiplying the map value Gegr_map by the EGR correction coefficient Kegr_no2 determined in each mode, and then the process proceeds to S17. In S17, it is determined whether or not the selective reduction catalyst is in a deteriorated state, that is, whether or not the catalyst deterioration determination value DET_SCR_AGD is “3”. If this determination is YES, the process proceeds to S18, the process is terminated after the catalyst deterioration warning lamp is turned on, and if this determination is NO, this process is immediately terminated.

Next, the effect of the NO ₂ sensor feedback mode will be examined with reference to the simulation results shown in FIGS.
FIG. 16 shows a simulation result when the NO ₂ sensor feedback mode is not operated, more specifically, when the EGR correction coefficient Kegr_no2 is forcibly continuously set to “1”.
FIG. 17 is a simulation result when the NO ₂ sensor feedback mode is activated. 16 and 17, respectively, from the upper stage, the exhaust volume, the EGR amount, the feed NO amount, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the NOx purification rate in the selective reduction catalyst, and the NO ₂ sensor An output value Vno2 and an EGR correction coefficient Kegr_no2 are shown.

As shown in FIG. 16, since the EGR correction coefficient Kegr_no2 is forcibly kept at “1”, the target EGR amount Gegr_cmd matches the map value Gegr_map. As the engine is intermittently applied with a high load, the exhaust volume and the feed NO amount increase at the same timing. As a result, the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst is larger than 0.5. _Although the state is excessively shifted from the state of ₂ excess to the state of NO excess that is smaller than 0.5, the state is excessively NO _{2 on} average. For this reason, the NOx purification rate of the selective reduction catalyst has a value lower than the original maximum value, and NO ₂ that could not be purified downstream of the selective reduction catalyst is intermittent even though the feed NO amount is small. Are exhausted.

On the other hand, as shown in FIG. 17, in the NO ₂ sensor feedback mode, the EGR correction coefficient Kegr_no2 is changed in a sawtooth shape between “1” and the lower limit value based on the output value Vno2 of the NO2 sensor. The EGR amount Gegr_cmd is set to a value less than or equal to the map value Gegr_map. Accordingly, the amount of feed NO is controlled more than the result shown in FIG. 16, but as a result, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is an optimum value on average. It moves in the vicinity of a certain 0.5, and the NOx purification rate of the selective reduction catalyst is maintained high. For this reason, compared with the result shown in FIG. 16, although the amount of feed NO is large, the amount of NO ₂ on the downstream side of the selective reduction catalyst is greatly suppressed. As described above, the effect of the NO ₂ sensor feedback mode was verified.

[Modification of First Embodiment]
Next, a modification of the first embodiment will be described.
In the first embodiment, the catalyst deterioration determination mode is set separately from the NO ₂ sensor feedback mode, and the deterioration degree of the selective reduction catalyst is determined by operating the NO ₂ -NOx ratio controller in the catalyst deterioration determination mode. . On the other hand, in the present modification, the degree of deterioration is determined based on the EGR correction coefficient Kegr_no2 during operation in the NO ₂ sensor feedback mode without separately setting the catalyst deterioration determination mode as described above. Different from the first embodiment.

As described above, the NO ₂ sensor feedback mode, with NO ₂ sensor to NO ₂ is detected, the state of gradual NO ₂ excessive by increasing the EGR correction coefficient Kegr_no2, then, NO ₂ in NO ₂ sensor Until the EGR correction coefficient Kegr_no2 is decreased. On the other hand, as the deterioration of the selective reduction catalyst proceeds, the NOx purification rate decreases according to the deviation of the NO ₂ -NOx ratio from the optimum value. Thus, for the selective reduction catalyst deterioration has progressed, with NO ₂ sensor to NO ₂ is not detected, that is, until _NO 2 -NOx ratio becomes close to the optimum value NOx purification rate increases significantly EGR correction The coefficient Kegr_no2 needs to be reduced. That is, when the controller is operated in the NO ₂ sensor feedback mode with respect to the selective reduction catalyst in which the deterioration has progressed, the minimum value of the EGR correction coefficient Kegr_no2 (see the asterisk in FIG. 7) is considered to be small.

Therefore, in the present modification, while changing the EGR correction coefficient Kegr_no2 in the NO ₂ feedback mode, the following is provided for the EGR correction coefficient Kegr_no2 when the output value Vno2 of the NO ₂ sensor falls below the deterioration determination threshold value Vno2_JD_th. By performing statistical processing as shown in Expression (19), a deterioration determination parameter J_SCR that is inversely proportional to the progress of the deterioration degree of the selective reduction catalyst is calculated. Here, the filtering coefficient Kjd_scr is set between “0” and “1”, for example, “0.995”.

Then, by comparing the thus obtained deterioration determination parameter J_SCR and the threshold value J_SCR_AGD, the catalyst deterioration determination value DET_SCR_AGD is determined as shown in the following equation (20).

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to the drawings. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

In the first embodiment, as described with reference to FIG. 3, focusing on the fact that the oxidation of NO in the oxidation catalyst and CSF has a lower priority than CO and HC, the feed NO amount and the exhaust volume are increased or decreased. Then, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is controlled in the vicinity of an appropriate value. On the other hand, in the second embodiment, attention is paid to the fact that the NO ₂ production efficiency in the oxidation catalyst and CSF also changes depending on the oxygen concentration of the exhaust gas.

FIG. 18 is a diagram showing the relationship among the air-fuel ratio of the air-fuel mixture, the amount of NO ₂ downstream of the oxidation catalyst and CSF, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst. FIG. 18 shows a case where the feed NO amount is made constant by appropriately adjusting parameters such as the fuel injection timing and the injection pattern that are unrelated to the air-fuel ratio of the air-fuel mixture.
As shown in FIG. 18, the air-fuel ratio is changed from the lean side to the rich side in the region on the lean side from the stoichiometric state with the feed NO amount kept constant, thereby reducing the oxygen concentration of the exhaust gas flowing into the oxidation catalyst and the CSF. As a result, the ratio of NO oxidized by the oxidation catalyst and CSF decreases (that is, NO ₂ production efficiency decreases), and the NO ₂ -NOx ratio decreases. This means that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst can be adjusted by the air-fuel ratio of the air-fuel mixture, more directly, the oxygen concentration of the exhaust gas flowing into the oxidation catalyst and CSF.

FIG. 19 is a block diagram showing the configuration of the exhaust purification system 2A of the engine 1A and its ECU 3A according to the present embodiment made by paying attention to the relationship between the air-fuel ratio of the air-fuel mixture and the NO ₂ -NOx ratio.

  In order to detect the state of the exhaust purification system 2A, an oxygen concentration sensor 45A is connected to the ECU 3A. The oxygen concentration sensor 45A detects the oxygen concentration of exhaust gas downstream of the oxidation catalyst 21 and CSF 22 in the exhaust pipe 11, that is, the air-fuel ratio of the exhaust gas, and supplies a signal AF_act substantially proportional to the detected value to the ECU 3A.

The method for adjusting the air-fuel ratio of the air-fuel mixture differs between a gasoline engine and a diesel engine.
In the case of a gasoline engine, the air-fuel ratio of the air-fuel mixture can be adjusted by increasing or decreasing the amount of fresh air by throttling.
In the case of a diesel engine equipped with a supercharger, the EGR amount, the combustion fuel injection amount corresponding to the fuel injection amount related to main injection and after injection, the post injection amount corresponding to the fuel injection amount related to post injection, and the supercharging The air-fuel ratio of the air-fuel mixture can be adjusted by pressure or the like. The main injection is a fuel injection that is executed at a predetermined timing from the intake process to the expansion process, and the after injection is a fuel injection that is executed after the main injection. Post-injection is fuel injection executed at a predetermined timing between the expansion process and the intake process. For example, when the EGR amount is increased, the air-fuel ratio of the air-fuel mixture tends to be rich, and conversely when it is decreased, the air-fuel ratio of the air-fuel mixture tends to be leaner. When the after injection amount or the post injection amount is increased, the air-fuel ratio of the air-fuel mixture tends to be rich, and conversely, when it is decreased, it tends to be lean. In addition, if the timing of main injection or after injection is retarded, the combustion efficiency decreases, so that it is necessary to increase the combustion fuel injection amount in order to maintain the same engine output torque. On the contrary, when these timings are advanced, there is a tendency to be lean.
Hereinafter, the engine 1A is a diesel engine, and an example in which the combustion fuel injection amount Gcomb, the post injection amount Gpost, the target boost pressure Boost_cmd, and the target EGR amount Gegr_cmd are determined as parameters for adjusting the air-fuel ratio of the air-fuel mixture. explain.

As shown in FIG. 19, the control block relating to the determination of the parameters (Gcomb, Gpost, Boost_cmd, and Gegr_cmd) relating to the adjustment of the air-fuel ratio of the air-fuel mixture includes the NO ₂ -NOx ratio controller 31A and the reference air-fuel ratio map value calculation. The unit 32A and an air-fuel ratio controller 33A are included.

According to this control block, the target air-fuel ratio AF_cmd, which is the target value for the air-fuel ratio of the exhaust downstream of the oxidation catalyst 21 and CSF 22, is set to the reference target air-fuel ratio AF_map calculated by the reference air-fuel ratio map value calculation unit 32A. It is calculated by adding the air-fuel ratio correction coefficient Daf_no2 calculated by the NO ₂ -NOx ratio controller 31A (see the following formula (21)). The combustion fuel injection amount Gcomb, the post injection amount Gpost, the target boost pressure Boost_cmd, and the target EGR amount Gegr_cmd are set so that the output value AF_act of the oxygen concentration sensor 45A matches the target air-fuel ratio AF_cmd. Is calculated by

The reference air-fuel ratio map value calculation unit 32A determines a reference target air-fuel ratio AF_map by searching a predetermined map based on the engine speed NE and the required engine load TRQ. The map in the reference air-fuel ratio map value calculation unit 32A is similar to the first embodiment in consideration of the balance of the fuel consumption, the soot amount, the feed NOx amount, etc., and the NO _{2 of} the exhaust gas flowing into the selective reduction catalyst. it is preferred that -NOx ratio is set to be NO ₂ excessive slightly.

Similar to the NO ₂ -NOx ratio controller 31 of the first embodiment, the NO ₂ -NOx ratio controller 31A operates in one of the following four types of control modes, and is based on an algorithm that is different in each control mode. An air-fuel ratio correction coefficient Daf_no2 for correcting the map value AF_map is calculated.
1. NO ₂ sensor feedback mode 2. Catalyst deterioration judgment mode 3. Fuel economy priority mode NO ₂ Generation Priority Mode Hereinafter, the procedure for calculating the air-fuel ratio correction coefficient Daf_no2 in these four control modes will be described in order.

<NO ₂ sensor feedback mode>
In the NO ₂ sensor feedback mode, the NO ₂ -NOx ratio controller 31A outputs the NO ₂ sensor output so that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is maintained in the vicinity of the optimum value on average. An air-fuel ratio correction coefficient Daf_no2 is determined based on the value Vno2.

Similarly to the first embodiment, this embodiment uses the output deviation E_Vno2 defined by the following equation (22), and further calculates the air-fuel ratio correction coefficient Daf_no2 based on the following equations (23), (24), and (25). calculate.

  As shown in Expression (25), the air-fuel ratio correction coefficient Daf_no2 is set between the upper limit value and the lower limit value with the upper limit value being “0” and the lower limit value being Daf_no2_L. The feedback gain Ki_af_no2 in the equation (24) is set to a negative value. The initial subtraction amount DDaf_DEC in equation (23) is set to a negative value, and the return amount DDaf_INC is set to a positive value.

Figure 20 shows the above formula (22) to (25) the output value of the NO ₂ sensor in the case of actuating the _NO 2 -NOx ratio controller defined NO ₂ sensor feedback mode Vno2, the air-fuel ratio correction coefficient Daf_no2 and, It is a time chart which shows the change of target air fuel ratio AF_cmd.

Between times t1 to t2, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. In this case, the air-fuel ratio correction coefficient Daf_no2 increases toward the upper limit by the return amount DDaf_INC set in Expression (23), and the target air-fuel ratio AF_cmd gradually increases so as to approach the map value AF_map. As a result, the air-fuel ratio of the air-fuel mixture is changed to a leaner side, and as a result, the oxygen concentration of the exhaust gas gradually increases as compared to the case where the air-fuel ratio correction coefficient Daf_no2 is not increased.

Next, at time t2, the output value Vno2 of NO ₂ sensor exceeds the NO ₂ detection threshold Vno2_th. At this moment, the air-fuel ratio correction coefficient Daf_no2 is decreased by the initial subtraction amount DDaf_DEC set by the equation (23). As a result, the target air-fuel ratio AF_cmd is instantaneously changed to a smaller value so as to depart from the map value AF_map. Thereafter, from time t2 the output value Vno2 of NO ₂ sensor to the time t3 below NO ₂ detection threshold Vno2_th, the air-fuel ratio correction coefficient Daf_no2, as shown in equation (24), each amount that is proportional to the output deviation E_Vno2 Decrease. Thereby, the target air-fuel ratio AF_cmd becomes smaller so as to be away from the map value AF_map. As a result, the air-fuel ratio of the air-fuel mixture is changed to a richer side. As a result, the oxygen concentration of the exhaust gas gradually decreases as compared with the case where the air-fuel ratio correction coefficient Daf_no2 is not decreased.

Between times t3 to t4, and time t5 and later, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. Accordingly, the air-fuel ratio correction coefficient Daf_no2 and the target air-fuel ratio AF_cmd during this period exhibit the same qualitative behavior as during the period from the time t1 to the time t2, and thus detailed description thereof is omitted. The time until t4 to t5, the output value Vno2 of NO ₂ sensor is greater than the NO ₂ detection threshold Vno2_th. Accordingly, the air-fuel ratio correction coefficient Daf_no2 and the target air-fuel ratio AF_cmd during this period exhibit qualitatively the same behaviors from the time t2 to t3, and detailed description thereof will be omitted.

  Next, in accordance with the output deviation E_Vno2 becoming a positive value, a process of decreasing the target air-fuel ratio AF_cmd so as to be away from the map value AF_map (time t2 to t3, t4 to t5 in FIG. 20) is executed. The effect by doing will be described with reference to FIG.

FIG. 21 is a diagram showing the oxygen concentration, the NO ₂ -NOx ratio, the NO amount, the NO ₂ amount, the HC amount and the CO amount, and the air-fuel ratio of the air-fuel mixture in each part of the exhaust pipe. In FIG. 21, the broken line indicates an example of the conventional technique that continues to adopt the map value AF_map as the target air-fuel ratio AF_cmd, and the solid line indicates the target air-fuel ratio AF_cmd in response to the output deviation E_Vno2 having a positive value. The example of this embodiment reduced so that it may leave | separate from map value AF_map is shown.

First, in the conventional technique shown by the broken line, the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst is in a state of greatly surpassed NO ₂ excessive optimum value near 0.5, NO ₂ is that has not been purified It will be discharged downstream of the selective reduction catalyst.

On the other hand, in the present embodiment, when the output deviation E_Vno2 becomes a positive value and is in a state of excessive NO ₂ , the target air-fuel ratio AF_cmd is decreased away from the map value AF_map, whereby the air-fuel ratio of the air-fuel mixture is decreased. To a richer side to reduce the oxygen concentration in the exhaust. Note that enrichment of the air-fuel ratio of the air-fuel mixture is performed, for example, by increasing the combustion fuel injection amount Gcomb, the post injection amount Gpost, and the EGR amount Gegr_cmd. For this reason, as the air-fuel ratio of the air-fuel mixture becomes richer, the HC amount and CO amount of the exhaust gas flowing into the oxidation catalyst and the CSF increase, and conversely the NO amount decreases.

When the oxygen concentration in the exhaust gas decreases, the oxidation efficiency of NO in the oxidation catalyst decreases. In addition, as the air-fuel ratio of the air-fuel mixture becomes richer, the amount of HC and CO whose oxidation priority is higher than that of NO increases, so that the oxidation efficiency of NO further decreases. Therefore, the NO ₂ production efficiency in the oxidation catalyst is lowered.

As described above, in contrast to the conventional method in which the NO ₂ -NOx ratio is significantly higher than the optimum value in the vicinity of 0.5, in this embodiment, the air-fuel ratio of the air-fuel mixture is changed to a richer side to lower the oxygen concentration. As a result, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward an optimum value in the vicinity of 0.5, and as a result, both the NO amount and NO ₂ amount discharged from the selective reduction catalyst are suppressed. Can do.

In the present embodiment, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward the optimum value by changing the air-fuel ratio of the air-fuel mixture to the rich side and lowering the oxygen concentration of the exhaust gas. However, the method for reducing the NO ₂ -NOx ratio toward the optimum value is not limited to this. As shown in FIG. 21, the oxygen concentration in the exhaust gas is reduced and exhausted from the engine rather than the effect of increasing the NO ₂ production efficiency in the oxidation catalyst and CSF due to the decrease in the amount of NOx exhausted from the engine. By setting combustion parameters correlated to the combustion state of the engine so that the NO ₂ production efficiency reduction effect in the oxidation catalyst and CSF due to the increase in the HC amount and the CO amount becomes larger, NO _{The 2-} NOx ratio may be reduced.

By the way, in the first embodiment, the NO ₂ generation efficiency is lowered by reducing the EGR amount in a lean state where the air-fuel ratio of the air-fuel mixture is sufficiently away from the stoichiometry, whereas in the second embodiment, the EGR is reduced. By increasing the amount and enriching the air-fuel ratio of the air-fuel mixture to the lean side of the stoichiometry and close to the stoichiometry, the NO ₂ generation efficiency is lowered. As described above, in the first embodiment and the second embodiment, the direction of change in the EGR amount when the NO ₂ generation efficiency is reduced is reversed, but this is because the air-fuel ratio of the air-fuel mixture as a premise is greatly different. It is not contradictory.

<2. Catalyst deterioration judgment mode>
Returning to FIG. 19, in the catalyst deterioration determination mode, the NO ₂ -NOx ratio controller 31 A determines a catalyst deterioration determination value DET_SCR_AGD that indicates the deterioration degree of the selective reduction catalyst 23.
More specifically, when the air-fuel ratio correction coefficient Daf_no2 is decreased from the upper limit value “0” toward the lower limit value Daf_no2_L to continuously change the air-fuel ratio of the air-fuel mixture to the rich side, the output of the NO ₂ sensor The catalyst deterioration determination value DET_SCR_AGD is determined based on the timing when the value Vno2 falls below the deterioration determination threshold value Vno2_JD_th. In this way, the algorithm for determining the catalyst deterioration determination value DET_SCR_AGD while continuing to change the air-fuel ratio correction coefficient Daf_no2 is a parameter for the EGR correction coefficient Kegr_no2 in the equations (9) to (12) in the first embodiment. It can be constructed by substituting the air-fuel ratio correction coefficient Daf_no2.
Conversely, when the air-fuel ratio correction coefficient Daf_no2 is increased from the lower limit value Daf_no2_L toward the upper limit value “0” to continue to change the air-fuel ratio of the air-fuel mixture to the lean side, the output value of the NO ₂ sensor The catalyst deterioration determination value DET_SCR_AGD can also be determined based on the timing when Vno2 exceeds the deterioration determination threshold value Vno2_JD_th. This algorithm can be constructed by replacing the parameter related to the EGR correction coefficient Kegr_no2 in the equations (13) to (16) in the first embodiment with the air-fuel ratio correction coefficient Daf_no2.

<Fuel consumption priority mode>
In the fuel efficiency priority mode, the NO ₂ -NOx ratio controller 31A outputs the output value of the NO ₂ sensor so that the fuel efficiency is improved as compared with the execution of the NO ₂ feedback mode, that is, the air-fuel ratio of the air-fuel mixture becomes leaner. Regardless of Vno2, the air-fuel ratio correction coefficient Daf_no2 is set to the fuel efficiency priority air-fuel ratio correction coefficient Daf_no2_opt (see the following equation (26)).

<NO ₂ generation priority mode>
In the NO ₂ generation priority mode, the NO ₂ -NOx ratio controller 31A generates the following equation (27) so that a large amount of NO ₂ is generated by the oxidation catalyst and CSF, and the amount of NO ₂ in the exhaust gas flowing into the selective reduction catalyst increases. ), The air-fuel ratio correction coefficient Daf_no2 is set to “0”.

As described above, the NO ₂ -NOx ratio controller can be operated in four different control modes including the NO ₂ sensor feedback mode, the catalyst deterioration determination mode, the fuel consumption priority mode, and the NO ₂ generation priority mode. It has become. The preferred time for executing each mode is the same as in the first embodiment.
That is, when the catalyst deterioration determination value DET_SCR_AGD is “1” and it can be determined that the selective reduction catalyst has not deteriorated, execution of the NO ₂ sensor feedback mode is prohibited and the fuel efficiency priority mode (see the above formula (26)). It is preferable to improve the fuel efficiency compared to when the NO ₂ sensor feedback mode is executed. In addition, when the catalyst deterioration determination value DET_SCR_AGD is “2” or “3”, that is, when it can be determined that the deterioration of the selective reduction catalyst has progressed to some extent, it is preferable to permit execution of the NO ₂ sensor feedback mode.
Further, when the deterioration degree of the oxidation catalyst and the CSF is determined and it is determined that the deterioration degree is small, the execution of the NO ₂ sensor feedback mode is permitted. When it is determined that the deterioration degree is large, the NO ₂ sensor It is preferable to prohibit the execution of the feedback mode and execute the NO ₂ generation priority mode, for example.

FIG. 22 is a time chart illustrating an example of a mode switching procedure in the NO ₂ -NOx ratio controller. In the example shown in FIG. 22, the NO ₂ -NOx ratio controller is operated in the order of the NO ₂ generation priority mode, the catalyst deterioration determination mode, and the NO ₂ sensor feedback mode after starting the engine at time “0”. Show.
As shown in FIG. 22, during the period from when the engine starts to when a predetermined time elapses (during warm-up), the execution of the NO ₂ sensor feedback mode is prohibited, and instead the NO ₂ generation priority mode is set. Run. That is, during the warm-up, the NO ₂ generation priority mode is executed, and the NO ₂ generation efficiency is quickly raised together with the HC and CO purification rates. After a predetermined time has elapsed since the start of the engine (after warming up), the execution of the NO ₂ sensor feedback mode is permitted, and the NO ₂ -NOx ratio is maintained at an optimum value in the vicinity of 0.5. In addition, the NOx purification rate is kept high together with the HC and CO purification rates.
As described above, the NO ₂ sensor feedback mode is similarly prohibited and the NO ₂ generation priority mode is executed when the temperature of the oxidation catalyst is lower than the activation temperature in addition to the warming up immediately after the engine is started. Then, when the temperature of the oxidation catalyst is equal to or higher than its activation temperature, execution of the NO ₂ sensor feedback mode may be permitted.
Accordingly, as described with reference to FIG. 13 in the first embodiment, HC, CO, and NOx can all be efficiently purified during warm-up immediately after engine startup and after warm-up.

[Modification of Second Embodiment]
Next, a modification of the second embodiment will be described.
In the present modification, the degree of deterioration is determined based on the air-fuel ratio correction coefficient Daf_no2 during operation in the NO ₂ sensor feedback mode without separately setting the catalyst deterioration determination mode as in the second embodiment.
More specifically, while changing the air-fuel ratio correction coefficient Daf_no2 in the NO ₂ feedback mode, the above formula is applied to the air-fuel ratio correction coefficient Daf_no2 when the output value Vno2 of the NO ₂ sensor falls below the deterioration determination threshold value Vno2_JD_th. By performing the same statistical processing as in (19), a deterioration determination parameter J_SCR that is inversely proportional to the progress of the deterioration degree of the selective reduction catalyst is calculated, and this parameter is compared with the threshold in the same manner as in the above equation (20) Thus, the catalyst deterioration determination value DET_SCR_AGD is determined.

[Third Embodiment]
Next, a third embodiment of the present invention will be described with reference to the drawings. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

In the first embodiment, attention is paid to the fact that oxidation of NO in the oxidation catalyst and CSF has a lower priority than CO and HC. In the second embodiment, the NO ₂ generation efficiency in the oxidation catalyst and CSF varies depending on the oxygen concentration of the exhaust gas. Focused on that. On the other hand, in the third embodiment, attention is paid to the fact that the NO ₂ production efficiency in the oxidation catalyst or CSF varies depending on the temperature.

FIG. 23 is a diagram showing the relationship between the temperature of the oxidation catalyst, the NO ₂ amounts on the downstream side of the oxidation catalyst and CSF, and the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst. FIG. 23 shows a case where the temperature of the oxidation catalyst is changed so that the feed NO amount becomes constant.
As shown in FIG. 23, when the temperature of the oxidation catalyst is increased, but also rises NO ₂ generation efficiency of the oxidation catalyst and CSF, it becomes more than a certain temperature, generating NO ₂ for reaction NO ₂ returns to again NO is generated Efficiency also decreases again. Hereinafter, as described in detail, in this embodiment, the target temperature of the oxidation catalyst is changed between the upper limit value Tdoc_scr_opt and the lower limit value Tdoc_L, thereby controlling the NO ₂ -NOx ratio in the vicinity of the optimum value. .

FIG. 24 is a block diagram showing the configuration of the exhaust purification system 2B of the engine 1B and its ECU 3B according to the present embodiment made by paying attention to the relationship between the temperature of the oxidation catalyst 21 and the NO ₂ -NOx ratio.

The temperature of the oxidation catalyst 21 and the CSF 22 can be adjusted by the main injection timing, the after injection timing, the main injection timing, the after injection timing, and the like. For example, when the main injection amount, the after injection amount, and the post injection amount are increased, the temperatures of the oxidation catalyst 21 and the CSF 22 increase, and conversely, when these amounts are decreased, they tend to decrease. Further, when the main injection timing and after injection timing are retarded, the temperatures of the oxidation catalyst 21 and the CSF 22 increase, and conversely, when advanced, they tend to decrease.
Hereinafter, an example in which the post injection amount Gpost is determined as a parameter for adjusting the temperatures of the oxidation catalyst 21 and the CSF 22 will be described.

As shown in FIG. 24, the control block relating to the determination of the post injection amount Gpost includes a NO ₂ -NOx ratio controller 31B, a reference post injection amount map value calculation unit 32B, and a temperature controller 33B. .

According to this control block, the target temperature Tdoc_cmd of the oxidation catalyst 21 is calculated by adding the temperature correction amount Dt_no2 calculated by the NO2-NOx ratio controller 31B to the optimum temperature Tdoc_scr_opt described later (the following equation (28) )reference). The post injection amount Gpost is calculated by adding the post injection amount correction value DGpost calculated by the temperature controller 33B to the reference post injection amount Gpost_map calculated by the reference post injection amount map value calculation unit 32B ( (See the following formula (29)).

Incidentally, it is known that the NOx purification rate of the selective reduction catalyst varies depending on the temperature. More specifically, the NOx purification rate of the selective reduction catalyst is equal to the temperature in the same manner as the NO ₂ production efficiency of the oxidation catalyst and CSF shows a convex characteristic with respect to the temperature (see FIG. 23). On the other hand, it exhibits an upwardly convex characteristic and is therefore maximum at a given temperature. Therefore, the above-described optimum temperature Tdoc_scr_opt of the oxidation catalyst is the temperature of the oxidation catalyst in a state where the downstream selective reduction catalyst is at a temperature at which the NOx purification rate becomes maximum. Further, the optimum temperature Tdoc_scr_opt that maximizes the NOx purification rate of the downstream selective reduction catalyst by adjusting the specifications of the oxidation catalyst and CSF and the layout of the exhaust system, as shown in FIG. It can be approximately equal to the temperature at which efficiency is maximized.

The reference post injection amount map value calculation unit 32B calculates a reference post injection amount Gpost_map by searching a predetermined map based on the engine speed NE and the required engine load TRQ. The map in the reference post-injection amount map value calculation unit 32B is such that the temperature of the oxidation catalyst 21 is maintained at the above-mentioned optimum temperature Tdoc_scr_opt while taking into account the fuel consumption, the soot amount, the feed NOx amount, and the like. Similarly to the first embodiment, it is preferable that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is set so as to be excessively NO ₂ .

The temperature controller 33B calculates the post injection amount correction value DGpost based on the following equation (31) so that the deviation E_tdoc between the oxidation catalyst temperature Tdoc and the target temperature Tdoc_cmd shown in the following equation (30) becomes “0”. To do.

The NO ₂ -NOx ratio controller 31B operates in one of the following three control modes, and calculates a temperature correction amount Dt_no2 for correcting the above-described optimum temperature Tdoc_scr_opt based on different algorithms in each control mode. .
1. NO ₂ sensor feedback mode 2. Catalyst deterioration judgment mode NO ₂ Generation Priority Mode Hereinafter, the procedure for calculating the temperature correction amount Dt_no2 by these three control modes will be described in order.

<NO ₂ sensor feedback mode>
NO The ₂ sensor feedback _mode, NO 2 -NOx ratio controller 31B, as _NO 2 -NOx ratio of the exhaust flowing into the selective reduction catalyst is maintained in the vicinity of the optimum value, the output value Vno2 of NO ₂ sensor Based on this, the temperature correction amount Dt_no2 is determined.

Similarly to the first embodiment, this embodiment uses the output deviation E_Vno2 defined by the following equation (32), and further calculates the temperature correction amount Dt_no2 based on the following equations (33), (34), and (35). To do.

As shown in the equation (35), the temperature correction amount Dt_no2 is set between the upper limit value and the lower limit value with the upper limit value being “0” and the lower limit value being “Tdoc_L-Tdoc_scr_opt”. Therefore, from equation (28), the upper limit value of the target temperature Tdoc_cmd is the optimum temperature Tdoc_scr_opt, and the lower limit value is Tdoc_L.
The feedback gain Ki_no2 in the equation (34) is set to a negative value. The initial subtraction amount DDt_DEC in equation (33) is set to a negative value, and the return amount DDaf_INC is set to a positive value.

Figure 25 shows the above formula (32) - (35) the output value of the NO ₂ sensor in the case of actuating the _NO 2 -NOx ratio controller defined NO ₂ sensor feedback mode Vno2, temperature correction amount Dt_no2, and the target It is a time chart which shows the change of oxidation catalyst temperature Tdoc_cmd.

Between times t1 to t2, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. In this case, the temperature correction amount Dt_no2 increases by the return amount DDt_INC set in the equation (33) toward the upper limit value “0”, and the target oxidation catalyst temperature Tdoc_cmd gradually increases so as to approach the optimum temperature Tdoc_scr_opt. . As a result, the post injection amount is corrected to the increase side, and as a result, the temperature of the oxidation catalyst gradually increases as compared to the case where the temperature correction amount Dt_no2 is not increased.

Next, at time t2, the output value Vno2 of NO ₂ sensor exceeds the NO ₂ detection threshold Vno2_th. At this moment, the temperature correction amount Dt_no2 is reduced by the initial subtraction amount DDt_DEC set by the equation (33). As a result, the target oxidation catalyst temperature Tdoc_cmd is instantaneously changed to a smaller value so as to be away from Tdoc_scr_opt. Between the subsequent time t2 the output value Vno2 of NO ₂ sensor to the time t3 below NO ₂ detection threshold Vno2_th, temperature correction amount Dt_no2, as shown in equation (34), decreases by amount proportional to the output deviation E_Vno2 . Thereby, the target oxidation catalyst temperature Tdoc_cmd becomes smaller so as to be further away from Tdoc_scr_opt. As a result, the post injection amount is corrected to the decreasing side, and as a result, the temperature of the oxidation catalyst gradually decreases as compared with the case where the temperature correction amount Dt_no2 is not decreased.

Between times t3 to t4, and time t5 and later, the output value Vno2 of NO ₂ sensor is less NO ₂ detection threshold Vno2_th. Accordingly, the temperature correction amount Dt_no2 and the target oxidation catalyst temperature Tdoc_cmd during this period exhibit qualitatively the same behavior as during the period from the time t1 to the time t2, and thus detailed description thereof is omitted. The time until t4 to t5, the output value Vno2 of NO ₂ sensor is greater than the NO ₂ detection threshold Vno2_th. Accordingly, the temperature correction amount Dt_no2 and the target oxidation catalyst temperature Tdoc_cmd during this period exhibit qualitatively the same behaviors from the time t2 to t3, and detailed description thereof will be omitted.

FIG. 26 is a diagram showing the relationship between the NOx purification rate in the selective reduction catalyst, the NO ₂ -NOx ratio, and the temperature of the oxidation catalyst. In FIG. 26, the NOx purification rate when the temperature of the oxidation catalyst is set to the optimum temperature Tdoc_scr_opt is indicated by a solid line, and the NOx purification rate when the temperature of the oxidation catalyst is set to the target temperature Tdoc_cmd lower than the optimum temperature Tdoc_opt is indicated by a broken line.
As described above, the NOx purification rate in the selective reduction catalyst becomes maximum when the temperature of the oxidation catalyst is at the optimum temperature Tdoc_scr_opt. Therefore, basically, the target temperature Tdoc_cmd of the oxidation catalyst is set to this optimum temperature Tdoc_scr_opt. Is done. However, even if kept at optimum temperature Tdoc_scr_opt, for example in FIG. 26, when the _NO 2 -NOx ratio as indicated by the white circle is the state of NO ₂ excessive, the NOx purification rate decreases significantly. In such a case, even if the target temperature Tdoc_cmd of the oxidation catalyst is lowered, the NO ₂ generation efficiency of the oxidation catalyst is lowered, and the NO ₂ -NOx ratio is reduced to the vicinity of the optimum value as shown by an asterisk in FIG. It is possible to increase the NOx purification rate by making it.
From this, the lower limit value Tdoc_L of the target temperature Tdoc_cmd defined by the above equation (35) is NOx due to the decrease in the temperature of the selective reduction catalyst when the temperature of the oxidation catalyst is decreased from Tdoc_scr_opt to Tdoc_L. It is preferable that the effect of increasing the NOx purification rate by reducing and optimizing the NO ₂ -NOx ratio is set larger than the effect of reducing the purification rate.

  Next, in accordance with the output deviation E_Vno2 becoming a positive value, a process of reducing the target oxidation catalyst temperature Tdoc_cmd so as to be away from Tdoc_scr_opt (time t2 to t3, t4 to t5 in FIG. 25) is executed. The effect of this will be described with reference to FIG.

FIG. 27 is a diagram showing the oxygen concentration, the NO ₂ -NOx ratio, the NO amount, the NO ₂ amount, the HC amount and the CO amount, and the air-fuel ratio of the air-fuel mixture in each part of the exhaust pipe. In FIG. 27, the broken line indicates an example of the conventional method in which the optimum temperature Tdoc_scr_opt is continuously adopted as the target oxidation catalyst temperature Tdoc_cmd, and the solid line indicates the target oxidation catalyst temperature Tdoc_cmd according to the output deviation E_Vno2 having a positive value. An example of this embodiment is shown in which the temperature is lowered so as to be away from the optimum temperature Tdoc_scr_opt.

First, in the conventional method indicated by the broken line, the NO ₂ -NOx ratio of the inflowing exhaust gas greatly exceeds the optimum value in the vicinity of 0.5, even though the selective reduction catalyst has a temperature at which the NOx purification rate is maximized. In a state of excessive NO ₂ , NO ₂ is not completely purified and is discharged downstream of the selective reduction catalyst.

In contrast, in the present embodiment, when the output deviation E_Vno2 becomes a positive value in a state of NO ₂ excessive by lowering away the target temperature Tdoc_cmd from the optimum temperature Tdoc_scr_opt, the post injection amount Gpost It corrects more to the weight loss side and lowers the temperature of the oxidation catalyst and CSF. Here, by reducing the post injection amount Gpost, the HC amount and CO amount of the exhaust gas flowing into the oxidation catalyst are decreased, and conversely, the NO amount is slightly increased. Further, when the temperature of the oxidation catalyst and CSF is lowered, NO ₂ generation efficiency in these oxidation catalyst and CSF decreases.

As described above, in contrast to the conventional method in which the NO ₂ -NOx ratio greatly exceeds the optimum value in the vicinity of 0.5, in this embodiment, the post injection amount is corrected to the decrease side, and the temperatures of the oxidation catalyst and CSF are adjusted. By lowering, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is reduced toward an optimum value in the vicinity of 0.5, and as a result, both the NO amount and NO ₂ amount discharged from the selective reduction catalyst are suppressed. be able to.

  In the present embodiment, the post-injection amount Gpost is corrected to the decrease side, and the temperatures of the oxidation catalyst and CSF are reduced. However, the method of reducing the temperatures of the oxidation catalyst and CSF is not limited to this. In order to lower the temperatures of the oxidation catalyst and the CSF, not only the post injection amount but also the after injection amount may be corrected to the decrease side, for example. Further, by reducing the temperature of the oxidation catalyst, not only the NO oxidation efficiency but also the oxidation efficiency of HC and CO are reduced. For this reason, in this embodiment, the combustion parameters correlated with the combustion state of the engine are set so that the CO and HC exhausted from the engine are less than the amount that can be treated with an oxidation catalyst whose oxidation ability has decreased with a decrease in temperature and CSF. It is preferable to set so that

<2. Catalyst deterioration judgment mode>
Returning to FIG. 24, in the catalyst deterioration determination mode, the NO ₂ -NOx ratio controller 31B determines a catalyst deterioration determination value DET_SCR_AGD indicating the deterioration degree of the selective reduction catalyst 23.
More specifically, the temperature correction amount Dt_no2, when continued to lower the temperature of the oxidation catalyst and CSF by decreasing towards the lower limit from the upper limit value "0" (Tdoc_L-Tdoc_scr_opt) , the NO ₂ sensor The catalyst deterioration determination value DET_SCR_AGD is determined based on the timing when the output value Vno2 falls below the deterioration determination threshold value Vno2_JD_th. In this way, the algorithm for determining the catalyst deterioration determination value DET_SCR_AGD while continuously changing the temperature correction amount Dt_no2 is obtained by changing the parameter related to the EGR correction coefficient Kegr_no2 in the equations (9) to (12) in the first embodiment with the temperature It can be constructed by replacing with a correction amount Dt_no2.
Conversely, when the temperature correction amount Dt_no2 is increased from the lower limit value (Tdoc_L-Tdoc_scr_opt) toward the upper limit value “0” to continue to increase the temperature of the oxidation catalyst and CSF, the output of the NO ₂ sensor The catalyst deterioration determination value DET_SCR_AGD can also be determined based on the timing when the value Vno2 exceeds the deterioration determination threshold value Vno2_JD_th. This algorithm can be constructed by replacing the parameter relating to the EGR correction coefficient Kegr_no2 in the equations (13) to (16) in the first embodiment with the temperature correction amount Dt_no2.

<NO ₂ generation priority mode>
In the NO ₂ generation priority mode, the NO ₂ -NOx ratio controller 31B generates a large amount of NO ₂ in the oxidation catalyst and the CSF, and increases the amount of NO ₂ in the exhaust gas flowing into the selective reduction catalyst (36) ), The temperature correction amount Dt_no2 is set to “0”.

As described above, the NO ₂ -NOx ratio controller can be operated in three different control modes of the NO ₂ sensor feedback mode, the catalyst deterioration determination mode, and the NO ₂ generation priority mode. The preferred time for executing each mode is the same as in the first embodiment.
That is a catalyst degradation determination value DET_SCR_AGD is "1", if it can be determined that the selective reduction catalyst has not deteriorated prohibits execution of the NO ₂ sensor feedback mode, than during execution of NO ₂ sensor feedback mode It is preferable to improve fuel consumption. In addition, when the catalyst deterioration determination value DET_SCR_AGD is “2” or “3”, that is, when it can be determined that the deterioration of the selective reduction catalyst has progressed to some extent, it is preferable to permit execution of the NO ₂ sensor feedback mode.
Further, when the deterioration degree of the oxidation catalyst and the CSF is determined and it is determined that the deterioration degree is small, the execution of the NO ₂ sensor feedback mode is permitted. When it is determined that the deterioration degree is large, the NO ₂ sensor It is preferable to prohibit the execution of the feedback mode and execute the NO ₂ generation priority mode, for example.

FIG. 28 is a time chart illustrating an example of a mode switching procedure in the NO ₂ -NOx ratio controller. In the example shown in FIG. 28, the NO ₂ -NOx ratio controller is operated in the order of the NO ₂ generation priority mode, the catalyst deterioration determination mode, and the NO ₂ sensor feedback mode after starting the engine at time “0”. Show.
As shown in FIG. 28, execution of the NO ₂ sensor feedback mode is prohibited during a period until the predetermined time has elapsed after the start of the engine (during warm-up), and instead, the NO ₂ generation priority mode is set. Run. That is, during the warm-up, the NO ₂ generation priority mode is executed, and the NO ₂ generation efficiency is quickly raised together with the HC and CO purification rates. After a predetermined time has elapsed since the start of the engine (after warming up), the execution of the NO ₂ sensor feedback mode is permitted, and the NO ₂ -NOx ratio is maintained at an optimum value in the vicinity of 0.5. In addition, the NOx purification rate is kept high together with the HC and CO purification rates.
As described above, the NO ₂ sensor feedback mode is similarly prohibited and the NO ₂ generation priority mode is executed when the temperature of the oxidation catalyst is lower than the activation temperature in addition to the warming up immediately after the engine is started. Then, when the temperature of the oxidation catalyst is equal to or higher than its activation temperature, execution of the NO ₂ sensor feedback mode may be permitted.
Accordingly, as described with reference to FIG. 13 in the first embodiment, HC, CO, and NOx can all be efficiently purified during warm-up immediately after engine startup and after warm-up.

[Modification 1 of the third embodiment]
Next, Modification 1 of the third embodiment will be described.
In this modification, the degree of deterioration is determined based on the temperature correction amount Dt_no2 during operation in the NO ₂ sensor feedback mode without separately setting the catalyst deterioration determination mode as in the third embodiment.
More specifically, while varying the temperature correction amount Dt_no2 with NO ₂ feedback mode, with respect to temperature correction amount Dt_no2 at the time when the output value Vno2 of NO ₂ sensor is below the degradation determination threshold value Vno2_JD_th, the equation (19 )), The deterioration determination parameter J_SCR that is inversely proportional to the progress of the deterioration degree of the selective reduction catalyst is calculated, and this parameter is compared with the threshold value in the same manner as the above equation (20). A catalyst deterioration determination value DET_SCR_AGD is determined.

[Modification 2 of the third embodiment]
Next, Modification 2 of the third embodiment will be described.
As shown in FIG. 23, between the lower limit value Tdoc_L and the optimum temperature Tdoc_scr_opt, when the temperature of the oxidation catalyst decreases, the NO ₂ generation efficiency decreases and the NO ₂ -NOx ratio also decreases. In the third embodiment, the temperature region lowering the temperature NO ₂ generation efficiency decreases the oxidation catalyst [Tdoc_L, Tdoc_scr_opt] by changing the target temperature Tdoc_cmd _within, the optimum value of NO 2 -NOx ratio Control near. That in the third _embodiment, when reducing the NO 2 -NOx ratio, to lower the temperature of the oxidation catalyst.

By the way, as shown in FIG. 23, the optimum temperature Tdoc_dcr_opt is substantially equal to the temperature at which the NO oxidation efficiency in the oxidation catalyst and CSF becomes maximum. In this case, between the optimum temperature Tdoc_scr_opt and the predetermined upper limit value Tdoc_H, when the temperature of the oxidation catalyst increases, the NO oxidation efficiency decreases and the NO ₂ -NOx ratio also decreases. In this modification, thus the temperature region Raising the temperature NO ₂ generation efficiency decreases the oxidation catalyst [Tdoc_scr_opt, Tdoc_H] by changing the target temperature Tdoc_cmd _in, NO 2 -NOx ratio the optimal value Control near to. That is, in the present modification, the target temperature Tdoc_cmd is set in the region [Tdoc_scr_opt, Tdoc_H], and the target temperature Tdoc_cmd is corrected to increase in this region in the opposite direction to that in the third embodiment, whereby NO ₂ -Reduce the NOx ratio.

Therefore, in the NO ₂ sensor feedback mode in the present modification, the temperature correction amount Dt_no2 is changed in the direction opposite to that in the third embodiment. Specifically, when the output value Vno2 of NO ₂ sensor is below NO ₂ detection threshold Vno2_th by reducing the temperature correction amount Dt_no2, the target temperature Tdoc_cmd, reduce the temperature of the thus oxidation catalyst and CSF. Further, when the output value Vno2 of the NO ₂ sensor is larger than the NO ₂ detection threshold value, the temperature correction amount Dt_no2 is increased to increase the target temperature Tdoc_cmd, and thus the temperature of the oxidation catalyst and CSF. As described above, the arithmetic expression for determining the temperature correction amount Dt_no2 is, for example, that the signs of the initial subtraction amount DDt_DEC, the return amount DDt_INC, and the feedback gain Ki_no2 in the above equations (33) and (34) are inverted. Can be configured. Further, as the target temperature Tdoc_cmd setting region is changed from [Tdoc_L, Tdoc_scr_opt] to [Tdoc_scr_opt, Tdoc_H], the upper limit value of the temperature correction amount Dt_no2 in the equation (35) is changed from “0” to Tdoc_H−Tdoc_sc, The lower limit value is changed from Tdoc_L-Tdoc_scr_opt to “0”.

Even in the catalyst deterioration determination mode, the temperature correction amount Dt_no2 is changed in the direction opposite to that in the third embodiment. Specifically, when the temperature correction amount Dt_no2 is increased from the lower limit value “0” toward the upper limit value (Tdoc_H−Tdoc_scr_opt), the temperature of the oxidation catalyst and the CSF is continuously increased, and the output of the NO ₂ sensor The catalyst deterioration determination value DET_SCR_AGD is determined based on the timing when the value Vno2 falls below the deterioration determination threshold value Vno2_JD_th. Alternatively, when the temperature of the oxidation catalyst and the CSF is continuously decreased by decreasing the temperature correction amount Dt_no2 from the upper limit value (Tdoc_H-Tdoc_scr_opt) toward the lower limit value “0”, the output value Vno2 of the NO ₂ sensor is The catalyst deterioration determination value DET_SCR_AGD is determined based on the timing at which the deterioration determination threshold Vno2_JD_th is exceeded.

By the way, when the temperature of the oxidation catalyst and CSF is raised from the optimum temperature Tdoc_scr_opt to optimize the NO ₂ -NOx ratio, the temperature of the selective reduction catalyst on the downstream side also rises and the NOx purification rate decreases. Therefore, if the temperature is raised too much, there is a risk that NOx emission to the outside of the system will worsen. Therefore, the upper limit value Tdoc_H of the target temperature Tdoc_cmd is set so that the NOx purification rate in the selective reduction catalyst is reliably increased when the temperature of the oxidation catalyst and the CSF is increased to reduce the NO ₂ -NOx ratio. When the temperature is increased from Tdoc_scr_opt to Tdoc_H, the effect of increasing the NOx purification rate by reducing and optimizing the NO ₂ -NOx ratio rather than the effect of reducing the NOx purification rate by increasing the temperature of the selective reduction catalyst It is preferable to set so as to be larger.

  Further, when the temperature of the oxidation catalyst and CSF is increased as in this modification, the amounts of HC and CO discharged from the engine also tend to increase. However, the HC and CO oxidation efficiency in the oxidation catalyst is different from the NO oxidation efficiency characteristic that shows upward convex characteristics as shown in FIG. Therefore, even if the amount of HC and CO discharged from the engine to increase the temperature of the oxidation catalyst and CSF increases, the discharge of HC and CO to the outside of the system does not deteriorate greatly.

As described above, in this modification, when the NO ₂ -NOx ratio is reduced, the temperatures of the oxidation catalyst and the CSF are raised, so that the average temperature of the exhaust system is higher than that in the third embodiment. Further, since the gasoline engine has a higher exhaust system average temperature than the diesel engine, it is relatively difficult to control the temperature of the oxidation catalyst and the CSF as in the third embodiment. On the other hand, since the exhaust temperature can be raised relatively easily by, for example, retarding the ignition timing, the present modification that raises the temperature of the oxidation catalyst and the CSF, as in this modification, It is particularly suitable for gasoline engine exhaust purification systems.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described with reference to the drawings. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

In the first to third embodiments described above, when the output value Vno2 of the NO ₂ sensor provided on the downstream side of the selective reduction catalyst 23 is larger than the threshold value Vno2_th, that is, the exhaust gas flowing into the selective reduction catalyst is in a state of excessive NO _2. In this case, the NO ₂ purification rate in the selective reduction catalyst was maintained near the maximum by reducing the NO ₂ -NOx ratio. On the other hand, in this embodiment, a NO sensor that detects NO is provided on the downstream side of the selective reduction catalyst, and when the output value Vno is larger than the threshold value Vno_th, that is, the exhaust gas flowing into the selective reduction catalyst is in a state of excessive NO. In some cases, by increasing the NO ₂ -NOx ratio in reverse, the NOx purification rate in the selective reduction catalyst is maintained near the maximum.
In the first embodiment, the NO ₂ -NOx ratio is adjusted by increasing or decreasing the EGR amount. Similar to the first embodiment, the present embodiment also adjusts the NO ₂ -NOx ratio by increasing or decreasing the EGR amount.

  FIG. 29 is a block diagram showing the configuration of the exhaust purification system 2C of the engine 1C according to the present embodiment including the NO sensor 43 and its ECU 3C.

In order to detect the state of the exhaust purification system 2A, a NO sensor 43C is connected to the ECU 3C. This NO sensor 43C detects the amount or concentration of NO in the exhaust downstream of the selective reduction catalyst 23 in the exhaust pipe 11, and supplies a signal Vno substantially proportional to the detected value to the ECU 3C.
By the way, there is no existing sensor suitable for in-vehicle use as a NO sensor sensitive only to NO in exhaust gas. However, NOx in the exhaust gas, since it may be regarded as being almost all composed of only NO and NO _2, NO sensors, such as described above can be constructed by combining a NOx sensor and NO ₂ sensor . In addition, since existing NOx sensors are sensitive not only to NOx but also to NH ₃ , it is preferable to combine an NH ₃ sensor in addition to the NOx sensor and the NO ₂ sensor. That is, a desired output value proportional to the NO concentration or amount can be obtained by subtracting the output value of the NO ₂ sensor and the output value of the NH ₃ sensor from the output value of the NOx sensor.

As shown in FIG. 29, the control block relating to the determination of the EGR valve command value Legr_cmd includes a NO ₂ -NOx ratio controller 31C, a reference EGR amount map value calculation unit 32C, and an EGR controller 33. .
Only the differences between the first embodiment and this embodiment will be described below.

First, in this embodiment, since the NO excessive state is detected by the NO sensor 43C, the map in the reference EGR amount map value calculation unit 32C indicates that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst 23 is the first. Contrary to 1st Embodiment, it is preferable to set so that it may become NO excessive.

In the NO sensor feedback mode in the NO ₂ -NOx ratio controller 31C, when the output value Vno of the NO sensor 43C is larger than a predetermined deterioration determination threshold value Vno_th, that is, when the NO is excessive, the EGR amount is increased and the feed is increased. The NO ₂ -NOx ratio is increased by decreasing the NO amount. That is, the direction in which the EGR amount, the feed NO amount, and the NO ₂ -NOx ratio change in the NO sensor feedback mode is opposite to that in the first embodiment.
Therefore, the arithmetic expression in the NO sensor feedback mode of the NO ₂ -NOx ratio controller 31C is configured by inverting the signs of the initial subtraction amount Dkegr_DEC, the return amount Dkegr_INC, and the feedback gain Ki_no2 in the equations (6) to (8). can do.

FIG. 30 is a time chart showing changes in the NO sensor output value Vno, the EGR correction coefficient Kegr_no, and the target EGR amount Gegr_cmd when the NO ₂ -NOx ratio controller is operated in the NO sensor feedback mode configured as described above. It is a chart.

  From time t1 to t2, the output value Vno of the NO sensor is equal to or less than the NO detection threshold value Vno_th. In this case, the EGR correction coefficient Kegr_no gradually decreases toward the lower limit value. As a result, the target EGR amount Gegr_cmd gradually decreases away from the map value Gegr_map, and as a result, the NO amount discharged from the engine gradually increases as compared with the case where the EGR correction coefficient Kegr_no2 is not decreased. .

  Next, at time t2, the output value Vno of the NO sensor exceeds the NO detection threshold value Vno2_th. At this moment, the EGR correction coefficient Kegr_no decreases by the initial subtraction amount. Thereby, the target EGR amount Gegr_cmd is instantaneously changed to a larger value so as to approach the map value Gegr_map. Thereafter, from time t2 to time t3 when the output value Vno of the NO sensor falls below the NO detection threshold value Vno_th, the EGR correction coefficient Kegr_no increases by an amount proportional to the output deviation. As a result, the target EGR amount Gegr_cmd becomes larger so as to approach the map value Gegr_map, and as a result, the NO amount discharged from the engine gradually increases as compared with the case where the EGR correction coefficient Kegr_no2 is not decreased. .

Similarly, the arithmetic expression in the catalyst deterioration determination mode of the NO ₂ -NOx ratio controller 31C is configured by inverting the sign of the decrease amount Dkegr_JD_DEC or the increase amount Dkegr_JD_INC to reverse the direction in which the EGR correction coefficient Kegr_no2 changes. it can.

[Fifth Embodiment]
Next, a fifth embodiment of the present invention will be described.
In the present embodiment, as in the fourth embodiment, a NO sensor is provided on the downstream side of the selective reduction catalyst, and when the output value Vno is larger than the threshold value Vno_th, that is, the exhaust gas flowing into the selective reduction catalyst is in a state of excessive NO. In this case, the NO ₂ purification rate in the selective reduction catalyst is maintained in the vicinity of the maximum by increasing the NO ₂ -NOx ratio.

In the second embodiment, the NO ₂ -NOx ratio is adjusted by changing the air-fuel ratio of the air-fuel mixture and increasing / decreasing the oxygen concentration of the exhaust gas. Similar to the second embodiment, this embodiment also adjusts the NO ₂ -NOx ratio by changing the air-fuel ratio of the air-fuel mixture and increasing or decreasing the oxygen concentration of the exhaust gas.
Hereinafter, differences between the second embodiment and the present embodiment will be described.

First, in the present embodiment, since a NO excessive state is detected by the NO sensor, the map in the reference air-fuel ratio map value calculation unit shows that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is the second embodiment. On the contrary, it is preferable that the setting is made so as to make the NO excessive.

Further, in the NO sensor feedback mode in the NO ₂ -NOx ratio controller, when the output value Vno of the NO sensor is larger than the deterioration determination threshold value Vno_th, that is, when the NO excessive state, the fuel injection parameter, the boost pressure, and The NO ₂ -NOx ratio is increased by changing the air-fuel ratio of the air-fuel mixture to a leaner side according to at least one of the EGR amounts and increasing the oxygen concentration of the exhaust gas. That is, the direction in which the air-fuel ratio of the air-fuel mixture, the oxygen concentration of the exhaust gas, and the NO ₂ -NOx ratio change in the NO sensor feedback mode is opposite to that in the second embodiment.
Therefore, the arithmetic expression in the NO sensor feedback mode of the NO ₂ -NOx ratio controller can be configured by inverting the signs of the initial subtraction amount DDaf_DEC, the return amount DDaf_INC, and the feedback gain Ki_af_no2 in the equations (23) to (25). .
Similarly, the arithmetic expression in the catalyst deterioration determination mode of the NO ₂ -NOx ratio controller can be configured by reversing the direction in which the air-fuel ratio correction coefficient changes from that in the second embodiment.

[Sixth Embodiment]
Next, a sixth embodiment of the present invention will be described.
In the present embodiment, as in the fourth embodiment, a NO sensor is provided on the downstream side of the selective reduction catalyst, and when the output value Vno is larger than the threshold value Vno_th, that is, the exhaust gas flowing into the selective reduction catalyst is in a state of excessive NO. In this case, the NO ₂ purification rate in the selective reduction catalyst is maintained in the vicinity of the maximum by increasing the NO ₂ -NOx ratio.

In the third embodiment, the NO ₂ -NOx ratio is adjusted by increasing or decreasing the temperature of the oxidation catalyst. Similar to the third embodiment, this embodiment also adjusts the NO ₂ -NOx ratio by increasing or decreasing the temperature of the oxidation catalyst.
Hereinafter, differences between the third embodiment and this embodiment will be described.

First, in the present embodiment, since the NO excessive state is detected by the NO sensor, the map in the reference post-injection amount map value calculation unit has the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst in the third embodiment. On the contrary, it is preferable that the setting is made so as to make the NO excessive.

Further, in the NO sensor feedback mode in the NO ₂ -NOx ratio controller, when the output value Vno of the NO sensor is larger than the deterioration determination threshold value Vno_th, that is, when the NO is excessive, at least one of the after injection amount and the post injection amount. The NO ₂ -NOx ratio is increased by raising the temperature of the oxidation catalyst. That is, the after-injection amount, the post-injection amount, the temperature of the oxidation catalyst, and the direction in which the NO ₂ -NOx ratio changes in the NO sensor feedback mode are opposite to those in the third embodiment.
Therefore, the arithmetic expression in the NO sensor feedback mode of the NO ₂ -NOx ratio controller can be configured by inverting the signs of the initial subtraction amount DDt_DEC, the return amount DDt_INC, and the feedback gain Ki_no2 in the equations (33) to (35). .
Similarly, the arithmetic expression in the catalyst deterioration determination mode of the NO ₂ -NOx ratio controller can be configured by reversing the direction in which the temperature correction amount changes from that in the third embodiment.

[Seventh Embodiment]
Next, a seventh embodiment of the present invention will be described with reference to the drawings. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.
FIG. 31 is a diagram showing a configuration of a control block related to determination of the EGR valve command value Legr_cmd among the control blocks configured in the ECU 3D of the exhaust purification system according to the present embodiment.

As shown in FIG. 31, the control block for determining the EGR valve command value Legr_cmd includes a feedforward NO ₂ -NOx ratio controller 31D, a reference EGR amount map value calculation unit 32, an EGR controller 33, and a feed NOx estimator. comprises a _34D, and NO 2 -NOx ratio estimator 35D, model modifier and 36D, the.

According to this control block, the target EGR amount Gegr_cmd multiplies the reference EGR amount Gegr_map calculated by the reference EGR amount map value calculation unit 32 by the EGR correction coefficient Kegr_scr calculated by the feedforward NO ₂ -NOx ratio controller 31D. (See the following formula (37)).
The target EGR amount Gegr_cmd is defined not only by multiplying the reference EGR amount Gegr_map by the EGR correction coefficient Kegr_scr, but also by adding the EGR correction coefficient Kegr_scr to the reference EGR amount Gegr_map as shown in the equation (37). May be.

The feed NOx estimator 34D calculates an estimated value NOx_eng_hat of the feed NOx amount based on parameters related to engine combustion, such as the engine speed, fresh air amount, fuel injection amount, fuel injection pattern, EGR rate, and intake air temperature. . Specifically, the feed NOx amount estimated value NOx_eng_hat is calculated by multiplying the reference value NOx_eng_bs determined by searching a predetermined map and the correction coefficient Knox_eng as shown in the following equation (38). Is done.

FIG. 32 is a diagram showing a specific example of a map for determining the reference value NOx_eng_bs for the feed NOx amount estimated value NOx_eng_hat based on the engine speed NE and the fuel injection amount Gfuel among the parameters. According to the example shown in FIG. 32, the reference value NOx_eng_hat is determined to be larger as the engine speed NE is higher and the fuel injection amount Gfuel is larger.
FIG. 33 is a diagram showing a specific example of a map for determining the correction coefficient Knox_eng for the reference value NOx_eng_bs based on the EGR rate Regr among the parameters. According to the example shown in FIG. 33, the correction coefficient Knox_eng is determined to be smaller as the EGR rate Regr is higher.

Returning to FIG. 31, the NO ₂ -NOx ratio estimator 35D determines the NO _{2 of} the exhaust gas flowing into the selective reduction catalyst based on the estimated feed NOx amount NOx_eng_hat and the parameters related to the NO oxidation efficiency of the oxidation catalyst and CSF. An amount estimation value NO2_csf_hat, an NO amount estimation value NO_csf_hat, and an NO ₂ -NOx ratio estimation value Rscr_no_nox are calculated.

Specifically, the estimated value NO2_csf_hat of the NO ₂ amount flowing into the selective reduction catalyst is calculated based on the following formula (39). Here, the parameter Rox_no_no2_bs is a reference value for the NO oxidation efficiency in the oxidation catalyst and CSF, and is determined by searching a predetermined map based on the estimated value Gex of the exhaust volume and the estimated value of feed NOx amount NOx_eng_hat. The The parameter Kno_no2_tdoc is a correction coefficient for the NO oxidation efficiency Rox_no_no2_bs, and is determined by searching a predetermined map based on the oxidation catalyst temperature Tdoc. In the following equation (39), the parameter Kmod_no2 is a correction coefficient calculated by a model corrector 36D described later.

  FIG. 34 is a diagram showing a specific example of a map for determining the reference value Rox_no_no2_bs based on the estimated value Gex of the exhaust volume and the estimated value of feed NOx amount NOx_eng_hat. As described with reference to FIG. 3, the oxidation efficiency of the oxidation catalyst and the CSF tends to decrease as the passing amount of the substance to be oxidized per unit time increases and decreases as the exhaust volume increases. In view of this, according to the example shown in FIG. 34, the reference value Rox_no_no2_bs is determined to be smaller as the estimated value Gex of the exhaust volume becomes larger and the estimated value of feed NOx amount NOx_eng_hat becomes larger.

  FIG. 35 is a diagram showing a specific example of a map for determining the correction coefficient Kno_no2_tdoc based on the oxidation catalyst temperature Tdoc. This correction coefficient Kno_no2_tdoc is a coefficient for reflecting the characteristics of the NO oxidation efficiency in the oxidation catalyst and CSF as shown in FIG. 2 with respect to temperature, and is set between 0 and 1. According to the example shown in FIG. 35, the correction coefficient Kno_no2_tdoc is set to approximately 1 between about 200 to 350 ° C. at which the NO oxidation efficiency increases, and outside this range, as the oxidation catalyst temperature Tdoc decreases or increases. It is set to approach zero.

Further, the estimated value NO_csf_hat of the NO amount flowing into the selective reduction catalyst is calculated by subtracting the NO ₂ amount estimated value NO2_csf_hat from the feed NOx amount estimated value NOx_eng_hat, as shown in the following equation (40).

Further, the estimated value Rscr_no_nox of the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is obtained by dividing the NO ₂ amount estimated value NO2_csf_hat by the feed NOx amount estimated value NOx_eng_hat, as shown in the following equation (41). Calculated.

Returning to FIG. 31, the configuration of the feedforward NO ₂ -NOx ratio controller 31D will be described. The feedforward NO ₂ -NOx ratio controller 31D operates in any of the following four control modes, and calculates the EGR correction coefficient Kegr_scr based on different algorithms in each control mode.
1. 1. Feed forward control mode 2. Catalyst deterioration judgment mode 3. Fuel economy priority mode NO ₂ Generation Priority Mode Hereinafter, the procedure for calculating the EGR correction coefficient Kegr_scr using these four control modes will be described in order.

<Feed forward control mode>
In the feedforward control mode, the controller 31D controls the NO ₂ -NOx ratio of the exhaust flowing into the selective reduction catalyst toward an optimum value that maximizes the NOx purification rate. More specifically, the estimated value Rscr_no_nox of the _NO 2 -NOx ratio converges to the vicinity of the target _NO 2 -NOx ratio set to the optimum value for maximizing the NOx purification rate Rscr_no_nox_cmd (e.g., 0.5) Thus, the EGR correction coefficient Kegr_scr is determined.

First, in this feedforward control mode, the NO ₂ -NOx ratio error E_no2_no is defined based on any of the following formulas (42-1) to (42-3).

Equation (42-1) defines the error E_no2_no as a deviation between the estimated value Rscr_no_nox and a predetermined target NO ₂ -NOx ratio Rscr_no_nox_cmd, and is the simplest method.

Equation (42-2) defines the error E_no2_no as a deviation between the effective NO ₂ -NOx ratio Eff_no_nox defined by the following equation (43) and the target NO ₂ -NOx ratio Rscr_no_nox_cmd. The selective reduction catalyst, since there is a storage effect of NO and NO ₂ as described above, effective _NO 2 -NOx ratio in the actual selective reduction catalyst, a delay occurs with respect to _NO 2 -NOx ratio of the exhaust flowing It becomes like. Therefore, in the following formula (43), the effective NO ₂ -NOx ratio Eff_no_nox is defined with a value that causes a delay with respect to the estimated value Rscr_no_nox in order to consider the storage effect in the selective reduction catalyst. Here, in the following equation (43), the coefficient Klag is a delay coefficient and is set between −1 and 0.

Equation (42-3) defines the error E_no2_no by the surplus NO ₂ amount estimated value Ex_no2 defined by the following equation (44). The surplus NO ₂ amount estimated value Ex_no2 corresponds to an integral value of a deviation between the NO ₂ amount NO2_csf_hat flowing into the selective reduction catalyst and the target value of the NO ₂ amount (Rscr_no_nox_cmd · NOx_eng_hat). Since the selective reduction catalyst has a storage effect as described above, the error is defined by the equation (42-1), and the estimated value Rscr_no_nox is not always controlled so as to coincide with the target NO ₂ -NOx ratio Rscr_no_nox_cmd. Even if control is performed such that the error is defined by (42-3) and the surplus NO ₂ amount with respect to the target value becomes 0, it may be sufficiently effective.

  In the feedforward control mode, for example, the following equations (45) to (50) are set so that the error E_no2_no defined by any one of the equations (42-1) to (42-3) becomes zero. An EGR correction amount Kegr_scr is calculated based on the response assignment control algorithm. As the feedback algorithm, a conventionally known algorithm such as a PID control algorithm or an optimum regulator may be used.

First, as shown in the following formula (45), the switching function setting parameter VPOLE (k) determined between −1 and 0 by searching a predetermined table and the previous value Ex_no2 (k−1) of the error. ) And the current error value Ex_no2 (k) is calculated and defined as a switching function σ (k).

  For example, when a phase plane is defined with the horizontal axis representing the previous error value E_no2_no (k−1) and the vertical axis representing the current error value E_no2_no (k), the switching function σ (k) defined by Expression (45) The combination of the errors E_no2_no (k) and E_no2_no (k-1) where becomes 0 becomes a switching straight line with a slope of -VPOLE. On this switching straight line, by setting -VPOLE to a value smaller than 1 and larger than 0, E_no2_no (k-1)> E_no2_no (k), so that the error E_no2_no converges to 0. . That is, the switching function setting parameter VPOLE is a parameter that specifies the convergence of the error E_no2_no.

FIG. 36 is a diagram showing a setting table for the switching function setting parameter VPOLE.
As shown in FIG. 36, the switching function setting parameter VPOLE has an estimated value Rscr_no_nox (when an error is defined by the equation (42-1)), an effective NO ₂ -NOx ratio Eff_no_nox (an error is defined by the equation (42-2)). Or a surplus NO ₂ amount estimated value Ex_no2 (when an error is defined by the equation (42-3)) and a low convergence value (for example, −0.35) and a smaller convergence value (for example, -0.95).

More specifically, when the estimated value Rscr_no_nox or the effective NO ₂ -NOx ratio Eff_no_nox is in a range [R_L, R_H] including the target NO ₂ -NOx ratio Rscr_no_nox_cmd, the estimated value Rscr_no_nox or the effective NO ₂ -NOx ratio The value of the parameter VPOLE is set to a low convergence value so that Eff_no_nox is controlled to drift within this range [R_L, R_H]. When the estimated value Rscr_no_nox or the effective NO ₂ -NOx ratio Eff_no_nox is outside the range [R_L, R_H], the value of the parameter VPOLE is set so that it quickly matches the high convergence value as the distance from the range [R_L, R_H] increases. The estimated value Rscr_no_nox or the effective NO ₂ -NOx ratio Eff_no_nox is quickly converged to the target NO ₂ -NOx ratio Rscr_no_nox_cmd.
Similarly, when the surplus NO ₂ amount estimated value Ex_no2 is within a range [R_Ex_L, R_Ex_H] including 0, the value of the parameter VPOLE is set to a low convergence value, and when it is outside the range [R_Ex_L, R_Ex_H], The value of the parameter VPOLE is set so that it quickly matches the high convergence value as the distance from the range [R_Ex_L, R_Ex_H] increases.

Here, as shown in FIG. 36, the lower limit value R_L (or R_Ex_L) of the range in which the parameter VPOLE is set to the low convergence value is closer to the target value Rscr_no_nox_xmd (or 0) than the upper limit value R_H (or R_Ex_H). It is preferable to set to. This is because, as described above, the reduction in the NOx purification rate in the selective reduction catalyst is larger than when the NO ₂ -NOx ratio is smaller than the optimum value.

Next, as shown in the following formula (46), the provisional value Kegr_scr_temp ′ of the EGR correction amount is calculated by the sum of the reaching law input Urch and the adaptive law input Uadp.

In the above equation (46), the reaching law input Urch is an input for placing the deviation state quantity on the switching line, and as shown in the following equation (47), the switching function σ is multiplied by a predetermined feedback gain Krch. It is calculated by.

Further, in the above equation (46), the adaptive law input Uadp is an input for suppressing the influence of modeling error and disturbance and placing the deviation state quantity on the switching straight line. As shown in the following equation (48), It is calculated by the sum of the previous value Uadp (k−1) of the adaptive law input and the product of the switching function σ and the predetermined feedback gain Kadp.

Next, the provisional value Kegr_scr_temp ′ of the EGR correction amount calculated by the above equation (46) is limited between the upper limit value 1 and the lower limit value 0 as shown in the following equation (49).

Then, the EGR correction amount Kegr_scr is determined by the limited provisional value Kegr_scr_temp as shown in the following formula (50). Note that, when the exhaust volume estimation value Gex is at a high exhaust volume greater than the predetermined threshold Gex_H, it is considered that the NO ₂ generation efficiency in the oxidation catalyst and the CSF is lowered without any particular control. As in the NO ₂ generation priority mode described in, the EGR correction coefficient Kegr_scr is forcibly set to “1”.

<Catalyst deterioration judgment mode>
Returning to FIG. 31, in the catalyst deterioration determination mode, the controller 31D calculates a catalyst deterioration determination value DET_SCR_AGD indicating the degree of deterioration of the selective reduction catalyst. That is, when the continued decrease or increase the _NO 2 -NOx ratio of the exhaust flowing into the selective reduction catalyst, based on the timing at which the output value Vno2 is that exceeds or falls below a predetermined deterioration determination threshold Vno2_th of NO ₂ sensor, The catalyst deterioration determination value DET_SCR_AGD is calculated.
<Fuel consumption priority mode>
In the fuel efficiency priority mode, the controller 31D sets the EGR correction coefficient Kegr_scr to a predetermined fuel efficiency priority EGR correction coefficient Kegr_no2_opt, similarly to the equation (17).
<NO ₂ generation priority mode>
In the NO ₂ generation priority mode, the controller 31D sets the EGR correction coefficient Kegr_scr to “1” similarly to the above equation (18).
Note that the specific procedure of these modes and the preferred time for executing each mode are the same as those in the first embodiment, and thus detailed description thereof will be omitted.

The model corrector 36D corrects the estimated values NO2_csf_hat, NO_csf_hat, Rscr_no_nox calculated by the NO ₂ -NOx ratio estimator 35D based on the output value Vno2 of the NO ₂ sensor provided downstream of the selective reduction catalyst. Therefore, the correction coefficient Kmod_no2 is calculated.

The correction coefficient Kmod_no2 is multiplied by an uncorrected map value (Kno_no2_tdoc · Rox_no_no2_bs · NOx_eng_hat) as shown on the right side of the above equation (39), thereby calculating the estimated value NO2_csf_hat and thus the estimated value Rscr_no. Therefore, the estimated value Rscr_no_nox is corrected to a large value (NO ₂ excessive side) when the correction coefficient Kmod_no2 becomes large, and is corrected to a small value (NO excessive side) when the correction coefficient Kmod_no2 becomes small.

When the controller 31D is operated in the feedforward control mode as described above, the estimated value Rscr_no_nox is ideally maintained at the target value Rscr_no_nox_cmd, and the NOx purification rate in the selective reduction catalyst is maintained at the maximum state. The amount of NO ₂ discharged to the downstream side of the reduction catalyst is very small.

Therefore, when the output value Vno2 of the NO ₂ sensor provided on the downstream side of the selective reduction catalyst exceeds the predetermined threshold value Vno2_th, that is, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is in a state where the NO _{2 is} excessive. If it can be determined, this is because the estimated value Rscr_no_nox is estimated to be larger than the actual NO ₂ -NOx ratio, or the target value Rscr_no_nox_cmd is set to a value larger than the actual optimum value. It is done.
Conversely, when the output value Vno2 of the NO ₂ sensor is equal to or less than the threshold value Vno2_th, that is, when it can be determined that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is excessive, this is the estimated value Rscr_no_nox Is estimated to be smaller than the actual NO ₂ -NOx ratio, or the target value Rscr_no_nox_cmd is set to a value larger than the actual optimum value.
In addition, as a factor by which the estimated value Rscr_no_nox is estimated to a value different from the actual NO ₂ -NOx ratio, for example, deterioration of the oxidation catalyst and CSF can be cited. Further, factors that cause the target value Rscr_no_nox_cmd to be set to a value different from the actual optimum value include adhesion and inflow of HC to the selective reduction catalyst.

Therefore, the model corrector 36D increases the value of the correction coefficient Kmod_no2 when the output value Vno2 of the NO ₂ sensor exceeds the threshold value Vno2_th in order to eliminate the shift in the feedforward control mode as described above. When the estimated value Rscr_no_nox of the NO ₂ -NOx ratio is corrected to the increasing side, and the output value Vno 2 is less than or equal to the threshold value Vno 2_th, the estimated value Rscr_no_nox of the NO ₂ -NOx ratio is decreased by decreasing the value of the correction coefficient Kmod_no2. To decrease.

More specifically, the correction coefficient Kmod_no2 is determined based on the following equations (51) to (53).
First, the output deviation DVno2 is defined by subtracting a predetermined threshold value Vno2_th from the output value Vno2 of the NO ₂ sensor by the following equation (51). Since the meaning of the output deviation DVno2 has been described with reference to FIG. 6 in the first embodiment, detailed description thereof is omitted here.

Next, as shown in the following equation (52), the update value Dkmod of the correction coefficient Kmod_no2 is determined, and as shown in the following equation (53), the correction coefficient Kmod_no2 is updated by the determined update value Dkmod, and this is corrected. The provisional value of the coefficient is Kmod_no2_temp. Furthermore, the correction coefficient Kmod_no2 is determined by limiting the provisional value Kmod_no2_temp between the upper limit value 1 and the lower limit value 0 as shown in the following equation (54).

As described above, when the output value Vno2 is equal to or less than the threshold value Vno2_th, the correction coefficient Kmod_no2 is decreased. Therefore, in the above formula (52), the NO ₂ non-detection update value Dkmod_ng is set to a negative value. Further, when the output value Vno2 is larger than the threshold Vno2_th Since increases the correction factor Kmod_no2, in the above formula (52), NO ₂ detected when updating value Dkmod_po_L is set to a positive value.
Further, in the above formula (52), the first value Dkmod_po_H of NO ₂ detected during an update value that is employed only when the output value Vno2 exceeds the threshold Vno2_th is the NO ₂ detecting when large enough positive than the update value Dkmod_po_L Set to a value. This is to prevent the NO ₂ purification rate from temporarily deteriorating when the correction coefficient Kmod_no2 is changed as described with reference to FIG. 9 in the first embodiment.

FIG. 37 is a time chart showing changes in the correction coefficient Kmod_no2 determined by the model corrector as described above.
From time t1 to t2, the output value Vno2 of the NO ₂ sensor is equal to or less than the threshold value Vno2_th. In this case, the correction coefficient Kmod_no2 decreases toward the lower limit value 0 by the negative update value Dkmod_ng defined by the equation (52). As a result, the estimated value NO2_csf_hat is corrected to the decreasing side so as to be away from the map value Kno_no2_tdoc · Rox_no_no2_bs · NOx_eng_hat (see Expression (39)) before the correction by the correction coefficient Kmod_no2.

Next, at time t2, the output value Vno2 of the NO ₂ sensor exceeds the threshold value Vno2_th. At this moment, the correction coefficient Kmod_no2 increases by the initial value Dkmod_po_H of the update value set by the equation (52). Thereby, the estimated value NO2_csf_hat is instantaneously changed to a large value so as to approach the map value before correction. Thereafter, the correction coefficient Kmod_no2 increases by a positive update value Dkmod_po_L from time t2 to time t3 when the output value Vno2 of the NO ₂ sensor falls below the threshold value Vno2_th. Thus, the estimated value NO2_csf_hat is corrected to the increasing side so as to approach the map value before correction.

Next, the effects of the feedforward control mode will be examined with reference to the simulation results shown in FIGS.
FIG. 38 shows simulation results when the oxidation catalyst and CSF are new, and the feedforward control mode and the model corrector are deactivated. Here, deactivating the feedforward control mode refers more specifically to the case where the EGR correction coefficient Kegr_scr is continuously set to “1”, and deactivating the model corrector is a correction coefficient. This is a case where Kmod_no2 is forcibly continuously set to “1”.
FIG. 39 shows simulation results when the feed forward control mode is activated while the oxidation catalyst and CSF are new and the model corrector is deactivated.
FIG. 40 shows a simulation result when the oxidation catalyst and the CSF are new, and the feedforward control mode and the model corrector are operated together.
FIG. 41 shows simulation results when the oxidation catalyst and CSF are deteriorated products, and the feedforward control mode and the model corrector are deactivated.
FIG. 42 shows a simulation result when the feed forward control mode is activated while the oxidation catalyst and the CSF are deteriorated products and the model corrector is deactivated.
FIG. 43 shows a simulation result when the oxidation catalyst and the CSF are deteriorated products, and the feedforward control mode and the model corrector are operated.

38 to 43, the value indicated by the thin line in the column of NO ₂ -NOx ratio is the effective NO ₂ -NOx in the selective reduction catalyst obtained in consideration of the storage effect of NO and NO ₂ in the selective reduction catalyst. It is a calculated value obtained from the estimated value Rscr_no_nox of the NO ₂ -NOx ratio.

If the oxidation catalyst and the CSF are new, the NO ₂ generation efficiency by these catalysts is high, and as shown in FIG. 38, the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst becomes NO ₂ excessive. For this reason, the NO purification rate is maintained high, but the NO ₂ purification rate becomes low, and the amount of NO ₂ discharged to the downstream side of the selective reduction catalyst increases.

When only the feedforward control mode is operated from the result shown in FIG. 38, as shown in FIG. 39, the target EGR amount Gegr_cmd is set to a value smaller than the map value Gegr_map, whereby the feed NOx amount is set in the feedforward control mode. Although the NO ₂ -NOx ratio is controlled to an appropriate value from the NO ₂ excessive state, the NO purification rate and the NO ₂ purification rate in the selective reduction catalyst are both kept high as a result, although the NO ₂ -NOx ratio is increased as compared with the non-operating state. It will be.

When the model corrector is further operated from the result shown in FIG. 39, the NO ₂ purification rate slightly deteriorates as shown in FIG. This is because the model corrector periodically creates a state of excessive NO ₂ by continuously decreasing the correction coefficient Kmod_no2 until the output value Vno2 of the NO ₂ sensor exceeds the threshold value Vno2_th. Thus, when the model corrector is operated in a state where the oxidation catalyst and CSF are not deteriorated, the NO ₂ purification rate will be lower than when the model corrector is not operated, but compared with the result of FIG. Since it is maintained at a sufficiently large value, it can be said that it is within the allowable range.

Next, when the oxidation catalyst and the CSF are deteriorated products, the NO ₂ production efficiency by these catalysts is reduced as compared with the example shown in FIG. 38, the NO ₂ -NOx ratio is maintained at an appropriate value, and as a result, NO ₂ The purification rate will be kept relatively high.

When only the feedforward control mode is operated from the result shown in FIG. 41, the NO ₂ -NOx ratio is controlled to a stable value with small fluttering as shown in FIG. 42, but the model corrector is not operated. Feedback it oxidation catalyst and CSF is a degraded state Sarezu estimate of resulting _NO 2 -NOx ratio becomes a value smaller than the actual _NO 2 -NOx ratio. As a result, exhaust gas containing excessive NO flows into the selective reduction catalyst, and the NO purification rate is significantly reduced.

When the model corrector is further operated from the result shown in FIG. 41, the value of the correction coefficient Kmod_no2 changes from the initial value of 1 to the vicinity of 0 as shown in FIG. 43, according to the deterioration degree of the oxidation catalyst and CSF. Since the estimated value Rscr_no_nox is corrected to an appropriate value, the NO ₂ -NOx ratio is maintained at an appropriate value, and as a result, both the NO purification rate and the NO ₂ purification rate are maintained high.
From the above, the effectiveness of operating the feedforward control mode and the model corrector was verified.

As described above, in the present embodiment, as in the first and fourth embodiments, the EGR correction coefficient that is a correction value for the map value Gegr_map of the target EGR amount Gegr_cmd as a parameter for increasing or decreasing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst. Although Kegr_scr is employed and the EGR correction coefficient Kegr_scr is determined by the feedforward NO ₂ -NOx ratio controller 31D, the present invention is not limited to this.
For example, as in the second and fifth embodiments, as a parameter for increasing or decreasing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the target value AF_cmd of the air-fuel ratio of the engine air-fuel mixture is corrected from the reference value AF_map. The value Daf_no2 may be adopted and the correction value Daf_no2 may be determined based on the same algorithm as in the present embodiment.
Further, for example, as in the third and sixth embodiments, as a parameter for increasing / decreasing the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the target value Tdoc_cmd of the temperature of the oxidation catalyst and CSF from the reference value Tdoc_scr_opt is used. The correction value Dt_no2 may be adopted and the correction value Dt_no2 may be determined based on the same algorithm as in the present embodiment.

  Further, in the present embodiment, as shown in Expression (39), the estimated values NO2_csf_hat, NO_csf_hat, and Rscr_no_nox are corrected by multiplying the map value before correction by the correction coefficient Kmod_no2, and thereby the shift in the feedforward control mode is corrected. However, it is not limited to this. As described above, in addition to the estimated value Rscr_no_nox, the target value Rscr_no_nox_cmd can also change from the actual optimum value. Therefore, even if the target value Rscr_no_nox_cmd is corrected with the correction coefficient Kmod_no2 instead of correcting the estimated value NO2_csf_hat as described above, the same effect can be obtained.

[Modification 1 of the seventh embodiment]
Next, Modification 1 of the seventh embodiment will be described.
In the seventh embodiment, the feed NOx estimator and the NO ₂ -NOx ratio estimator search for predetermined maps, respectively, to thereby determine the feed NOx amount estimated value NOx_eng_hat, NO ₂ -NOx ratio estimated value Rscr_no_nox, NO _2. The amount estimation value NO2_csf_hat and the NO amount estimation value NO_csf_hat were calculated.

In the feed NOx estimator and the NO ₂ -NOx ratio estimator in this modification, these estimated values are calculated using a neural network. Hereinafter, the neural networks constructed in the feed NOx estimator and the NO ₂ -NOx ratio estimator in this modification will be described in order.

FIG. 44 is a diagram showing a neural network structure of the feed NOx estimator.
This neural network is configured by connecting a plurality of neurons that output in accordance with a predetermined function, and outputs a value Y (k) according to an m-component input vector U (k). As shown in FIG. 44, this neural network includes an input layer composed of m neurons W _1j (j = 1 to m) and m × (n−1) neurons W _ij (i = 2 to 2). n, j = 1 to m), and a hierarchical type including three layers, that is, an intermediate layer composed of n, j = 1 to m) and an output layer composed of one neuron Y.
Input layer: W _1j (j = 1, 2,..., M)
Intermediate layer: W _ij (i = 2, 3,..., N, j = 1, 2,..., M)
Output layer: Y

The operation of m neurons W _1j (j = _{1 to} m) in the input layer will be described.
A signal T _1j (k) is input to the neuron W _1j in the input layer. As this input signal T _1j (k), the j-th component U _j (k) of the input vector U (k) is used as shown in the following equation (55).

The neuron W _{1j in the} input layer is coupled to the m neurons W _2j (j = 1 to m) in the intermediate layer with a predetermined weight, and the signal V _1j (k) is connected to the m neurons W _{2j that} are coupled. Is output. That is, the neuron W _{1j generates} m signals V _1j (k) corresponding to the input signal T _1j (k) according to the sigmoid function f (x) as shown in the following equations (56) and (57). Output to neuron W _2j .

FIG. 45 is a diagram showing the sigmoid function f (x). FIG. 45 shows the case where ε = 0 and β = 0.5, 1.0, 2.0, 3.0 in the above equation (57).
The range of the sigmoid function f (x) is [ε, ε + 1]. Further, as shown in FIG. 45, the sigmoid function f (x) approaches a step function centered on x = 0 as β is increased.

  In the above equation (57), the coefficient β represents the slope gain of the sigmoid function f (x), and the coefficient ε represents the offset value of the sigmoid function f (x). The slope gain β is set by learning of a neural network described later. The offset value ε is set by learning of a neural network described later or set to a predetermined value.

Next, the operation of (n−1) × m neurons W _ij (i = 2 to n, j = 1 to m) in the intermediate layer will be described.
An intermediate layer neuron W _ij (i = 2 to n, j = ₁ to m) has a predetermined value for each of m signals V _{i−1, j} (j = ₁ to m) output from the neurons to be coupled. The sum of signals multiplied by the weights ω _{i−1, j} (j = _{1 to m} ) is input. Therefore, a signal T _ij (k) as shown in the following equation (58) is input to the neuron W _ij in the intermediate layer.

Among the neurons in the intermediate layer, the neurons excluding m coupled to the output layer, that is, (n−2) × m neurons W _ij (i = 2 to n−1, j = 1 to m) The layer is connected to m neurons W _{i + 1, j} (j = _{1 to} m) with a weight ω _ij , and a signal V _ij (k) is output to the connected neurons W _{i + 1, j} . That is, this neuron W _ij (i = 2 to n−1, j = 1 to m) is in accordance with the input signal T _ij (k) according to the sigmoid function f (x) as shown in the following equation (59). The signal V _ij (k) is output to m neurons W _{i + 1, j} .

Further, m neurons W _nj (j = 1 to m) in the intermediate layer are connected to the neuron Y in the output layer with a weight ω _nj , and a signal V _nj (k) is output to the neuron Y in the output layer. To do. That is, these neurons W _nj (j = 1 to m) _generate a signal V _nj (k) corresponding to the input signal T _nj (k) according to the sigmoid function f (x) as shown in the following equation (60). Output to neuron Y.

Next, the operation of the neuron Y in the output layer will be described.
The neuron Y of the output layer is multiplied by a predetermined weight ω _{n, j} (j = ₁ to m) by m signals V _{n, j} (j = ₁ to m) output from the neuron of the intermediate layer to be coupled. The sum of the received signals is input. Therefore, a signal T (k) as shown in the following formula (61) is input to the neuron Y in the output layer.

The neuron Y in the output layer outputs a signal Y (k) corresponding to the input signal T (k) according to the sigmoid function g (x) as shown in the following equations (62) and (63).

  The sigmoid function g (x) behaves qualitatively as the function f (x) shown in FIG. 45 described above, but differs from the sigmoid function f (x) in that the range is [δ, δ + α]. . In the above equation (63), the coefficient γ represents the slope gain of the sigmoid function g (x), and the coefficient δ represents the offset value of the sigmoid function g (x). The coefficient α indicates an output gain for setting the degree of freedom that the output of the neural network can take. The slope gain γ and the output gain α are set by learning with a neural network described later. The offset value δ is set by learning of a neural network described later, or set to a predetermined value.

The component of the input vector U (k) for the neural network in the feed NOx estimator configured as described above is defined as shown in the following equation (64). Thus, the components of the input vector U (k) include a plurality of physical quantities (engine speed NE, cylinder intake fresh air amount Gair, cylinder intake EGR amount Gegr, total fuel required for estimating the feed NOx amount. Injection amount Gfuel, post injection amount Gpost, pilot injection amount Gpilot, post injection timing θpost, main injection timing θmain, and mixture gas temperature Tair). In addition, the component of the input vector U (k) includes data on physical quantities of different types necessary for estimating the feed NOx amount in this way, and also includes data on physical quantities at different times, The reproducibility of the dynamic behavior of the estimated value during transient operation can be further improved.

  Here, as shown in the above equation (64), in this modification, the time k is different from the time k-3 with respect to the two physical quantities of the engine speed NE and the cylinder intake fresh air amount Gair. Data were included. In this way, when data at different times for a specific physical quantity is included, data at times that are far enough to say that they are clearly different values are included in one input vector, so that the input data becomes transient. The neural network can recognize that the data is related to the state. Furthermore, the calculation load can be reduced by thinning out the data at successive times of time k-1 and time k-2 as in the above equation (64). In addition, when data at consecutive different times with a small difference are included, the learning accuracy may be reduced. However, by using data thinned out at an appropriate time interval as in the above equation (64), Such a decrease in learning accuracy can be prevented.

In order to set the output Y (k) of the neural network for such an input vector U (k) as an estimated value NOx_eng_hat of the feed NOx amount as shown in the following equation (65), the neural network includes the following: Learning as shown is performed.

  First, by operating the actually prepared engine and its exhaust purification system, the value of the component of the input vector U in the above equation (64) and the measured value NOx_eng_mg of the feed NOx amount at that time are recorded, and the learning data Prepare. Such learning data preferably includes not only steady data but also transient data when the engine is changed transiently.

Next, neural network learning is performed based on the acquired learning data. That is, error propagation is performed so that the estimated error Enn (= NOx_eng_hat (k) −NOx_eng_mg (k)) defined by the deviation between the output NOx_eng_hat of the neural network when learning data is input and the measured value NOx_eng_mg is minimized. Various gains (α, β, γ, δ, ε) of the neuron functions f (x), g (x), and each of the gains of the functions f (x), g (x) of the neuron, A weight ω _ij (i = 1 to n, j = 1 to m) indicating the strength of connection of neurons is set.

In the above description, the configuration of the neural network that calculates only the estimated value NOx_eng_hat of the feed NOx amount without distinguishing between NO and NO ₂ has been described. However, the present invention is not limited to this. For example, _two neural networks that independently calculate the feed NO amount and the feed NO ₂ amount may be constructed to distinguish NO from NO ₂ .

A neural network in the NO ₂ -NOx ratio estimator, that is, a neural network for calculating an estimated value of the amount of NO ₂ flowing into the selective reduction catalyst can be constructed in the same manner as the feed NOx estimator. In this case, the component of the input vector U (k) includes a plurality of physical quantities (exhaust volume estimated value Gex) necessary for estimating the amount of NO ₂ flowing into the selective reduction catalyst, as shown in the following equation (66). , Cylinder intake fresh air amount Gair, cylinder intake EGR amount Gegr, total fuel injection amount Gfuel, post injection amount Gpost, pilot injection amount Gpilot, post injection timing θpost, main injection timing θmain, oxidation catalyst temperature Tdoc, soot accumulation on CSF Amount Ms). Further, similarly to the above equation (64), in order to improve the reproducibility of the estimated value during the transient operation, it is preferable to include data at different times for the specific physical quantity. The soot accumulation amount Ms on the CSF is calculated by searching a soot emission amount map from the elapsed time from the previous CSF regeneration process, the travel distance, the integrated value of the exhaust volume, the engine speed, and the engine load. It is estimated from the integrated value of the estimated value and the integrated value of the traveling energy.

Since the oxygen concentration of the exhaust gas greatly affects the NO ₂ production efficiency in the oxidation catalyst and CSF, a neural network for calculating the estimated value of the NO ₂ amount flowing into the selective reduction catalyst, and hence the estimated value of the NO ₂ -NOx ratio. The input to the network preferably includes parameters correlated with the oxygen concentration of the exhaust gas, such as the cylinder intake fresh air amount Gair, the cylinder intake EGR amount Gegr, and the total fuel injection amount Gfuel. Further, since the amount of unburned HC flowing into the oxidation catalyst and CSF and the amount of soot accumulation in CSF greatly affect the NO ₂ generation efficiency in the oxidation catalyst and CSF, the unburned HC is input to the neural network. Includes parameters related to the unburned HC amount of exhaust, such as total fuel injection amount Gfuel, post injection amount Gpost, pilot injection amount Gpilot, post injection timing θpost, main injection timing θmain, and soot accumulation amount Ms. It is preferable that

Similarly to the above equation (41), the NO ₂ amount estimated value NO2_csf_hat calculated using the neural network as described above is divided by the feed NOx amount estimated value NOx_eng_hat to flow into the selective reduction catalyst. An estimated value Rscr_no_nox of the NO ₂ -NOx ratio of the exhaust is calculated.

[Modification 2 of the seventh embodiment]
Next, Modification 2 of the seventh embodiment will be described.
In the seventh embodiment, the correction coefficient Kmod_no2 for correcting the estimated values NO2_csf_hat, NO_csf_hat, Rscr_no_nox is calculated based on the output value Vno2 of the NO ₂ sensor provided on the downstream side of the selective reduction catalyst. In the model corrector, the deviation between the estimated value Rscr_no_nox or the target value Rscr_no_nox_cmd and the actual value is determined by the output value Vno2 of the NO ₂ sensor, and the correction coefficient Kmod_no2 is increased or decreased to eliminate this deviation.

  The model corrector in the present modification is based on the detection value Vno of the NO sensor provided on the downstream side of the selective reduction catalyst (the same as the NO sensor 43 in the fourth embodiment), and the correction coefficient Kmod_no2 in the above equation (39). Similarly, the correction coefficient Kmod_no for correcting the estimated value NO2_csf_hat is calculated by multiplying the map value before correction.

If the output value Vno the NO sensor exceeds a threshold Vno_th, that is, when _{the NO} 2 -NOx ratio of the exhaust flowing into the selective reduction catalyst can be determined that the state of the NO excessive, this estimate Rscr_no_nox actual _NO 2 -NOx It is considered that the estimated value is smaller than the ratio or the target value Rscr_no_nox is set to a value smaller than the actual optimum value. Therefore, in this case, the model corrector corrects the estimated value Rscr_no_nox of the NO ₂ -NOx ratio to the decreasing side by decreasing the value of the correction coefficient Kmod_no.

Further, when the output value Vno the NO sensor is less than or equal to the threshold Vno_th, that is, when _{the NO} 2 -NOx ratio of the exhaust flowing into the selective reduction catalyst can be determined that the state of the NO ₂ excessive, this estimate Rscr_no_nox actual NO It is considered that the estimated value is larger than the _2- NOx ratio, or the target value Rscr_no_nox_cmd is set to a value larger than the actual optimum value. Therefore, in this case, the model corrector corrects the estimated value Rscr_no_nox of the NO ₂ -NOx ratio to the increase side by increasing the value of the correction coefficient Kmod_no.

  An arithmetic expression for determining the correction coefficient Kmod_no showing the above behavior is obtained by defining the output deviation DVno similar to the above expression (51) with respect to the output value Vno and the threshold value Vno_th of the NO sensor, In (52) to (54), it can be configured by inverting the sign of the gain (Dkmod_po_H, Dkmod_po_L, Dkmod_ng) for updating the correction coefficient Kmod_no.

The present invention is not limited to the above-described embodiment, and various modifications can be made.
For example, in order to adjust the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst, the EGR amount is increased or decreased in the first and fourth embodiments (EGR method), and in the second and fifth embodiments, the mixture is mixed In the third and sixth embodiments, the temperature of the oxidation catalyst was increased or decreased (temperature adjustment method).
In the present invention, these EGR method, AF method, and temperature adjustment method may be combined as well as executed independently.
In the case of a gasoline engine, it is preferable to apply the AF method because the air-fuel ratio of the air-fuel mixture is easily changed compared to a diesel engine, and in the case of a diesel engine, the temperature adjustment method or the EGR method may be applied. preferable.
The temperature adjustment method adjusts the NO ₂ -NOx ratio by increasing / decreasing the temperature of the oxidation catalyst. However, since it takes time to change the temperature of the oxidation catalyst in the exhaust pipe, other EGR methods and AF methods are used. Compared with NO, the change in the NO ₂ -NOx ratio is slow. Therefore, the temperature adjustment method is preferably executed in combination with other EGR methods or AF methods.

  In the above embodiment, the oxidation catalyst 21 is provided directly under the engine 1, and the CSF 22 having both the soot collection function and the oxidation function of CO, HC, NO, etc. is provided downstream thereof. Not limited to this. Instead of such CSF 22, a filter having only a soot collecting function and not an oxidizing function, an oxidation catalyst having only an oxidizing function and not a soot collecting function may be used.

1, 1A, 1B, 1C ... engine (internal combustion engine)
2, 2A, 2B, 2C ... Exhaust gas purification system (exhaust gas purification system)
3, 3A, 3B, 3C, 3D ... ECU (control means)
31D ... feed forward NO ₂ -NOx ratio controller (NO ₂ -NOx ratio controller)
34D: Feed NOx estimator (estimating means)
35D _... NO 2 -NOx ratio estimator (estimating means)
36D ... Model corrector (correction means)
11 ... Exhaust pipe (exhaust passage)
21 ... Oxidation catalyst (oxidation catalyst)
22 ... CSF (oxidation catalyst)
23 ... selective reduction catalyst (selective reduction catalyst)
43 ... NO ₂ sensor (NO ₂ detecting means)
43C ... NO sensor (NO detection means)

Claims (31)

An oxidation catalyst provided in the exhaust passage of the internal combustion engine;
A selective reduction catalyst that is provided downstream of the oxidation catalyst in the exhaust passage and selectively reduces NOx in the exhaust;
An exhaust gas purification system for an internal combustion engine , comprising: an EGR device that recirculates a part of the exhaust gas flowing through the exhaust passage to the intake passage of the internal combustion engine,
Reference value calculating means for calculating a reference value of an EGR amount corresponding to an amount of exhaust gas recirculated by the EGR device based on an operating state of the internal combustion engine;
EGR amount control means for controlling the EGR device so that the EGR amount becomes a target value determined based on the reference value;
The NO ₂ generation priority control using the reference value as a target value, the NO ₂ -NOx ratio corresponding to the ratio of NO _{2 to} NOx in the exhaust gas flowing into the selective reduction catalyst, and the NOx purification rate in the selective reduction catalyst And a control means capable of executing NO ₂ -NOx ratio optimization control for controlling toward an optimum value to be maximized,
In the NO ₂ -NOx ratio optimization control, the target value is set within a region below the reference value so that the state in which the NO ₂ concentration in the exhaust downstream of the selective reduction catalyst is 0 or in the vicinity thereof is maintained. Change
The reference value is a value such that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature is greater than the optimum value,
The control means includes
Until the predetermined time elapses after starting the internal combustion engine, or when the temperature of the oxidation catalyst is lower than the predetermined temperature, the execution of the NO ₂ -NOx ratio optimization control is prohibited and the NO ₂ Perform generation priority control ,
The NO ₂ -NOx ratio optimization control is allowed to be executed after a predetermined time has elapsed since the start of the internal combustion engine or when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature. An exhaust purification system for an internal combustion engine. NO ₂ detection means for detecting NO _{2 in} the exhaust downstream of the selective reduction catalyst in the exhaust passage ;
Correction value calculating means for calculating a correction value of the EGR amount based on the detection value of the NO ₂ detecting means;
A target value calculating means for calculating a target value of the EGR amount based on the reference value and the correction value;
In the NO ₂ -NOx ratio optimization control, the correction value calculation means is configured such that the target value of the EGR amount decreases away from the reference value when the value detected by the NO ₂ detection means is larger than a predetermined value. The exhaust gas purification system for an internal combustion engine according to claim 1 , wherein the NO ₂ -NOx ratio is reduced toward the optimum value by calculating a correction value . In the NO ₂ -NOx ratio optimization control, the NO value of the exhaust gas flowing into the selective reduction catalyst is increased by decreasing the target value and increasing the NO amount flowing into the oxidation catalyst by reducing the EGR amount. The exhaust gas purification system for an internal combustion engine according to claim 2, wherein the _2- NOx ratio is reduced toward the optimum value. The ratio of the amount of NO ₂ flowing out from the oxidation catalyst to the amount of NO flowing into the oxidation catalyst is defined as NO ₂ production efficiency,
In the NO ₂ -NOx ratio optimization control , when reducing the NO ₂ -NOx ratio toward the optimum value ,
Than the effect of increasing the NO ₂ production efficiency due to the decrease in the amount of HC and CO discharged from the internal combustion engine,
The combustion parameter correlated with the combustion state of the internal combustion engine is set so that the effect of lowering the NO ₂ production efficiency due to the increase in the amount of NO flowing into the oxidation catalyst becomes larger. Item 3. An exhaust purification system for an internal combustion engine according to Item 2. In the NO ₂ -NOx ratio optimization control, the correction value calculation means is configured so that the target value of the EGR amount is within the reference value or less when the value detected by the NO ₂ detection means is equal to or less than a predetermined value. by calculating the correction value so as to increase the exhaust gas purification system for an internal combustion engine according to claims 2 to any of 4, wherein to reduce the amount of NO flowing into the oxidation catalyst. The control means includes
Determining the degree of deterioration of the selective reduction catalyst;
When it is determined that the degree of deterioration is small, execution of the NO ₂ -NOx ratio optimization control is prohibited, and the EGR amount is set so that the fuel consumption of the internal combustion engine is improved,
If it is determined that the degree of degradation is large, the exhaust gas purification system for an internal combustion engine according to claim 2, characterized in that to allow the execution of the NO ₂ -NOx ratio optimization control. It said control means, exhaust gas purification system for an internal combustion engine according to claim 6, characterized in that to determine the degree of degradation of the selective reduction catalyst based on the correction value. An oxidation catalyst provided in the exhaust passage of the internal combustion engine;
An exhaust gas purification system for an internal combustion engine, comprising: a selective reduction catalyst that is provided downstream of the oxidation catalyst in the exhaust passage and selectively reduces NOx in the exhaust;
Reference value calculating means for calculating a reference value of the air-fuel ratio of the air-fuel mixture of the internal combustion engine based on the operating state of the internal combustion engine;
Air-fuel ratio control means for controlling at least one of a fuel injection amount, a supercharging pressure, and an EGR amount so that the air-fuel ratio becomes a target value determined based on the reference value;
The NO ₂ generation priority control using the reference value as a target value, the NO ₂ -NOx ratio corresponding to the ratio of NO _{2 to} NOx in the exhaust gas flowing into the selective reduction catalyst, and the NOx purification rate in the selective reduction catalyst And a control means capable of executing NO ₂ -NOx ratio optimization control for controlling toward an optimum value to be maximized,
In the NO ₂ -NOx ratio optimization control, the target value is set within a region below the reference value so that the state in which the NO ₂ concentration in the exhaust downstream of the selective reduction catalyst is 0 or in the vicinity thereof is maintained. Change
The reference value is a value such that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature is greater than the optimum value,
The control means includes
Until the predetermined time elapses after starting the internal combustion engine, or when the temperature of the oxidation catalyst is lower than the predetermined temperature, the execution of the NO ₂ -NOx ratio optimization control is prohibited and the NO ₂ Perform generation priority control ,
The NO ₂ -NOx ratio optimization control is allowed to be executed after a predetermined time has elapsed since the start of the internal combustion engine or when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature. An exhaust purification system for an internal combustion engine. NO in the exhaust gas downstream of the selective reduction catalyst in the exhaust passage. ₂ NO to detect ₂ Detection means;
NO ₂ Correction value calculation means for calculating the correction value of the air-fuel ratio based on the detection value of the detection means;
Target value calculation means for calculating a target value of the air-fuel ratio based on the reference value and the correction value;
The correction value calculation means is the NO ₂ In the NOx ratio optimization control, the NO ₂ When the detected value by the detecting means is larger than the predetermined value, the correction value is calculated so that the target value of the air-fuel ratio is deviated from the reference value to the rich side, whereby the NO ₂ The exhaust gas purification system for an internal combustion engine according to claim 8, wherein the NOx ratio is reduced toward the optimum value. The ratio of the amount of NO ₂ flowing out from the oxidation catalyst to the amount of NO flowing into the oxidation catalyst is defined as NO ₂ production efficiency,
In reducing the NO ₂ -NOx ratio toward the optimum value in the NO ₂ -NOx ratio optimization control,
Than the effect of increasing the NO ₂ production efficiency due to the decrease in the amount of NO flowing into the oxidation catalyst ,
It is correlated with the combustion state of the internal combustion engine so that the effect of lowering the NO ₂ generation efficiency due to the decrease in the oxygen concentration of the exhaust gas and the increase in the amount of HC and CO discharged from the internal combustion engine becomes larger. The exhaust gas purification system for an internal combustion engine according to claim 9 , wherein a certain combustion parameter is set. In the NO ₂ -NOx ratio optimization control, the correction value calculation means is configured so that the target value of the air-fuel ratio is within the range of the reference value or less when the value detected by the NO ₂ detection means is less than a predetermined value. The exhaust gas purification system for an internal combustion engine according to claim 9 or 10, wherein the correction value is calculated so as to change to a lean side .   The control means changes the air-fuel ratio of the air-fuel mixture according to at least one of a fuel injection parameter, a supercharging pressure, and an EGR amount corresponding to the amount of exhaust gas recirculated by the EGR device. The exhaust gas purification system for an internal combustion engine according to any one of 11. The control means includes
Determining the degree of deterioration of the selective reduction catalyst;
When it is determined that the degree of deterioration is small, the execution of the NO ₂ -NOx ratio optimization control is prohibited, and the air-fuel ratio of the mixture is set so that the fuel efficiency of the internal combustion engine is improved,
The exhaust purification system for an internal combustion engine according to claim 9, wherein when it is determined that the degree of deterioration is large, execution of the NO ₂ -NOx ratio optimization control is permitted. It said control means, exhaust gas purification system for an internal combustion engine according to claim 13, characterized in that to determine the degree of degradation of the selective reduction catalyst based on the correction value. An oxidation catalyst provided in the exhaust passage of the internal combustion engine;
A selective reduction catalyst that is provided downstream of the oxidation catalyst in the exhaust passage and selectively reduces NOx in the exhaust;
The temperature of the oxidation catalyst is controlled to a predetermined target temperature by changing a command value of a fuel injection parameter that defines a fuel injection mode of the internal combustion engine from a reference value determined based on an operating state of the internal combustion engine. An exhaust gas purification system for an internal combustion engine comprising an oxidation catalyst temperature control means ,
The NO ₂ generation priority control using a predetermined optimum temperature as the target temperature and the NO ₂ -NOx ratio corresponding to the ratio of NO _{2 to} NOx in the exhaust gas flowing into the selective reduction catalyst are determined as the NOx purification rate in the selective reduction catalyst. And a control means capable of executing NO ₂ -NOx ratio optimization control for controlling toward an optimum value that maximizes
In the NO ₂ -NOx ratio optimization control, the target temperature of the oxidation catalyst is set to be equal to or lower than the optimum temperature so that the state in which the NO ₂ concentration in the exhaust downstream of the selective reduction catalyst is 0 or in the vicinity thereof is maintained. Within the domain of
The NO ₂ production efficiency in the oxidation catalyst is maximized at the optimum temperature with respect to the temperature, and exhibits a convex characteristic.
The reference value is a value that maintains the temperature of the oxidation catalyst at the optimum temperature and that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is larger than the optimum value,
The control means includes
Until the predetermined time elapses after starting the internal combustion engine, or when the temperature of the oxidation catalyst is lower than the predetermined temperature, the execution of the NO ₂ -NOx ratio optimization control is prohibited and the NO ₂ Perform generation priority control ,
The NO ₂ -NOx ratio optimization control is allowed to be executed after a predetermined time has elapsed since the start of the internal combustion engine or when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature. An exhaust purification system for an internal combustion engine. The fuel injection parameters are a main injection amount, an after injection amount, a post injection amount, a main injection timing, and an after injection timing of the internal combustion engine,
The oxidation catalyst temperature control means decreases at least one of the command value of the main injection amount, the after injection amount, and the post injection amount from the respective reference values, or advances the main injection timing and the after injection timing from the respective reference values. 16. The exhaust gas purification system for an internal combustion engine according to claim 15 , wherein the temperature of the oxidation catalyst is lowered by angling. In the NO ₂ -NOx ratio optimization control, the combustion parameters correlated with the combustion state of the internal combustion engine are the same even if the CO and HC exhausted from the internal combustion engine are oxidation catalysts whose oxidation ability has decreased with a decrease in temperature. 17. The exhaust gas purification system for an internal combustion engine according to claim 16, wherein the exhaust gas purification system is set to be equal to or less than an amount that can be processed. NO in the exhaust gas downstream of the selective reduction catalyst in the exhaust passage. ₂ NO to detect ₂ Detection means;
NO ₂ Correction value calculating means for calculating a correction value of the temperature of the oxidation catalyst based on the detection value of the detecting means;
A target temperature setting means for setting a target temperature of the oxidation catalyst based on the optimum temperature and the correction value;
The correction value calculation means is the NO ₂ In the NOx ratio optimization control, the NO ₂ When the detection value detected by the detection means is larger than a predetermined value, the correction value is calculated so that the target temperature decreases away from the optimum temperature, whereby the NO ₂ The exhaust gas purification system for an internal combustion engine according to any one of claims 15 to 17, wherein a -NOx ratio is reduced toward the optimum value. It said correction value calculating means, in the NO ₂ -NOx ratio optimization control, when the detection value by the NO ₂ detecting means is equal to or less than the predetermined value is increased in a range wherein the target temperature is below the optimum temperature The exhaust gas purification system for an internal combustion engine according to claim 18 , wherein the correction value is calculated as follows . The lower limit temperature of the region with respect to the temperature of the oxidation catalyst is the selective reduction when the temperature of the oxidation catalyst is lowered from the temperature at which NO oxidation efficiency is maximized to the lower limit temperature in the NO ₂ -NOx ratio optimization control. The effect of improving the NOx purification rate by reducing the NO ₂ -NOx ratio toward its optimum value is set to be larger than the effect of reducing the NOx purification rate by reducing the temperature of the catalyst. An exhaust purification system for an internal combustion engine according to any one of claims 15 to 19 , characterized in that: The control means includes
Determining the degree of deterioration of the selective reduction catalyst;
When it is determined that the degree of deterioration is small, the execution of the NO ₂ -NOx ratio optimization control is prohibited,
If it is determined that the degree of degradation is large, the exhaust gas purification system for an internal combustion engine according to claim 18 or 19 and permits the execution of the NO ₂ -NOx ratio optimization control. It said control means, exhaust gas purification system for an internal combustion engine according to claim 21, characterized in that to determine the degree of degradation of the selective reduction catalyst based on the correction value. An oxidation catalyst provided in the exhaust passage of the internal combustion engine;
A selective reduction catalyst that is provided downstream of the oxidation catalyst in the exhaust passage and selectively reduces NOx in the exhaust;
The temperature of the oxidation catalyst is controlled to a predetermined target temperature by changing a command value of a fuel injection parameter that defines a fuel injection mode of the internal combustion engine from a reference value determined based on an operating state of the internal combustion engine. An exhaust gas purification system for an internal combustion engine comprising an oxidation catalyst temperature control means ,
The NO ₂ generation priority control using a predetermined optimum temperature as the target temperature and the NO ₂ -NOx ratio corresponding to the ratio of NO _{2 to} NOx in the exhaust gas flowing into the selective reduction catalyst are determined as the NOx purification rate in the selective reduction catalyst. And a control means capable of executing NO ₂ -NOx ratio optimization control for controlling toward an optimum value that maximizes
In the NO ₂ -NOx ratio optimization control, the target temperature of the oxidation catalyst is set to be equal to or higher than the optimum temperature so that the state in which the NO ₂ concentration in the exhaust downstream of the selective reduction catalyst is 0 or close thereto is maintained. Within the domain of
The NO ₂ production efficiency in the oxidation catalyst is maximized at the optimum temperature with respect to the temperature, and exhibits a convex characteristic.
The reference value is a value that maintains the temperature of the oxidation catalyst at the optimum temperature and that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst is larger than the optimum value,
The control means includes
Until the predetermined time elapses after starting the internal combustion engine, or when the temperature of the oxidation catalyst is lower than the predetermined temperature, the execution of the NO ₂ -NOx ratio optimization control is prohibited and the NO ₂ Perform generation priority control ,
The NO ₂ -NOx ratio optimization control is allowed to be executed after a predetermined time has elapsed since the start of the internal combustion engine or when the temperature of the oxidation catalyst is equal to or higher than a predetermined temperature. An exhaust purification system for an internal combustion engine. The fuel injection parameters are a main injection amount, an after injection amount, a post injection amount, a main injection timing, and an after injection timing of the internal combustion engine,
The oxidation catalyst temperature control means increases the command value of at least one of the main injection amount, the after injection amount, and the post injection amount from the respective reference values, or delays the main injection timing and the after injection timing from the respective reference values. 24. The exhaust gas purification system for an internal combustion engine according to claim 23 , wherein the temperature of the oxidation catalyst is raised by making an angle . NO in the exhaust gas downstream of the selective reduction catalyst in the exhaust passage. ₂ NO to detect ₂ Detection means;
NO ₂ Correction value calculating means for calculating a correction value of the temperature of the oxidation catalyst based on the detection value of the detecting means;
A target temperature setting means for setting a target temperature of the oxidation catalyst based on the optimum temperature and the correction value;
The correction value calculation means is the NO ₂ In the NOx ratio optimization control, the NO ₂ When the detection value detected by the detection means is larger than a predetermined value, the correction value is calculated so that the target temperature rises away from the optimum temperature, whereby the NO ₂ 25. The exhaust gas purification system for an internal combustion engine according to claim 23, wherein the NOx ratio is reduced toward the optimum value. Said correction value calculating means, in the NO ₂ -NOx ratio optimization control, when the detection value by the NO ₂ detecting means is equal to or less than the predetermined value, lowering the target temperature is within the range of more than the optimum temperature 26. The exhaust gas purification system for an internal combustion engine according to claim 25 , wherein the correction value is calculated as follows . The upper limit temperature of the region relative to the temperature of the oxidation catalyst is the selective reduction when the temperature of the oxidation catalyst is increased from the temperature at which the NO oxidation efficiency is maximized to the upper limit temperature in the NO ₂ -NOx ratio optimization control. The effect of improving the NOx purification rate by reducing the NO ₂ -NOx ratio toward its optimum value is set to be larger than the effect of reducing the NOx purification rate by increasing the temperature of the catalyst. 27. The exhaust gas purification system for an internal combustion engine according to any one of claims 23 to 26 . The control means includes
Determining the degree of deterioration of the selective reduction catalyst;
When it is determined that the degree of deterioration is small, the execution of the NO ₂ -NOx ratio optimization control is prohibited,
If it is determined that the degree of degradation is large, the exhaust gas purification system for an internal combustion engine according to claim 25 or 26 and permits the execution of the NO ₂ -NOx ratio optimization control. When the control means continues to change the correction value so that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst decreases, the detection value by the NO ₂ detection means falls below a deterioration determination threshold value. 29. The exhaust gas purification system for an internal combustion engine according to any one of claims 6 , 13, 21 and 28, wherein deterioration determination control for determining the degree of deterioration of the selective reduction catalyst based on timing can be further executed . When the control means continues to change the correction value so that the NO ₂ -NOx ratio of the exhaust gas flowing into the selective reduction catalyst increases, the detection value by the NO ₂ detection means exceeds the deterioration determination threshold value. 29. The exhaust gas purification system for an internal combustion engine according to any one of claims 6 , 13, 21 and 28, wherein deterioration determination control for determining the degree of deterioration of the selective reduction catalyst based on timing can be further executed . The control means includes
Determining the degree of deterioration of the oxidation catalyst;
When it is determined that the degree of deterioration is small, execution of the NO ₂ -NOx ratio optimization control is permitted, and when it is determined that the degree of deterioration is large, execution of the NO ₂ -NOx ratio optimization control is prohibited. An exhaust purification system for an internal combustion engine according to any one of claims 1 to 30, wherein

Images