JP2011231682A - Control device for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress deterioration of emission characteristics even if asymmetric abnormality occurs in an exhaust gas sensor in an internal combustion engine using alcohol-containing fuel.SOLUTION: A control device for the internal combustion engine includes: a means for calculating a degree of asymmetry between the output characteristic of an A/F sensor 12 when an air-fuel ratio of exhaust gas is changed from a rich side to a lean side and the output characteristic of the A/F sensor 12 when the air-fuel ratio of exhaust gas is changed from the lean side to the rich side; a means for detecting a concentration of alcohol; a means for learning the amount of correction Kr reflected in the amount of fuel injected based on the integrated value ΔVf of an output deviation between the output value Vr of an Osensor 14 and a reference value Vrefr corresponding to a theoretical air-fuel ratio; a means for restricting the amount of correction Kr by a predetermined guard width; and a means for variably setting the guard width according to a change in the amount of correction Kr corresponding to the concentration of alcohol and the degree of asymmetry. Preferably, a means is further provided for variably setting the update rate of the amount of correction Kr according to a change in the amount of correction Kr corresponding to the concentration of alcohol and the degree of asymmetry.

Description

  The present invention relates to a control device for an internal combustion engine, and more particularly to a control device for an internal combustion engine that can use a fuel containing alcohol.

  Asymmetry abnormality is known as one form of abnormality that occurs in the air-fuel ratio sensor. The asymmetric abnormality is an abnormality that is asymmetrical when the output characteristic of the air-fuel ratio sensor changes from lean to rich and when it changes from rich to lean. FIG. 8 is a diagram for explaining an asymmetric abnormality of the air-fuel ratio sensor. In the figure, (A) shows the output characteristics of a normal air-fuel ratio sensor, and (B) shows the output characteristics when an asymmetric abnormality occurs. As shown in (A) in this figure, in a normal air-fuel ratio sensor, when the air-fuel ratio changes from lean to rich as opposed to the output characteristics when the air-fuel ratio changes from rich to lean across the stoichiometric air-fuel ratio (stoichiometric), Contrast with output characteristics. On the other hand, as shown in (B) in the figure, in the air-fuel ratio sensor in which the asymmetrical abnormality occurs, the output characteristics when changing from rich to lean are different from the output characteristics when changing from lean to rich. ing. When such an asymmetrical abnormality occurs, a steady deviation of the air-fuel ratio occurs, and the emission characteristics deteriorate.

  As a technique related to the asymmetric abnormality of the air-fuel ratio sensor as described above, one disclosed in Japanese Patent Application Laid-Open No. 2009-299545 is known. More specifically, in this prior art, when the fuel rich state changes to the fuel lean state, the change value to the fuel lean state estimated based on the injected fuel amount and the intake air amount and the air-fuel ratio sensor are used. The difference from the change value to the fuel lean state determined by the signal is calculated. Conversely, when the fuel lean state changes to the fuel rich state, the difference is estimated based on the injected fuel amount and the intake air amount. The difference between the change value to the fuel rich state and the change value to the fuel rich state determined by the signal from the air-fuel ratio sensor is calculated. When the difference between the absolute values of the differences calculated in each process is equal to or greater than the abnormality determination threshold value, it is determined that the air-fuel ratio sensor is asymmetric abnormality.

JP 2009-299545 A JP 2000-54895 A JP 2009-74388 A

  By the way, in the air-fuel ratio control of the internal combustion engine, feedback control based on the output value of the air-fuel ratio sensor arranged upstream of the catalyst (so-called main feedback control) and feedback control based on the output value of the oxygen sensor arranged downstream of the catalyst ( So-called sub-feedback control). In the sub-feedback control, an output deviation between the output value of the oxygen sensor and the reference value corresponding to the theoretical air-fuel ratio is obtained, and the feedback value calculated by the PID control (or PI control) is reflected in the fuel injection amount. .

  In the sub-feedback control, learning using a smoothed integral term related to PID control as a learning value is also performed. This process is called sub-feedback learning. The integral term reflects the vibration of the output signal of the oxygen sensor. By smoothing the integral term, the vibration component can be attenuated and only the steady component included in the integral term can be extracted. . This steady component corresponds to a steady deviation between the center value of the air-fuel ratio of the exhaust gas flowing into the catalyst and the stoichiometric air-fuel ratio. If the main feedback control is performed, this deviation corresponds to a constant signal error of the air-fuel ratio sensor. Therefore, by storing the above steady component as a learning value, the constant signal error of the air-fuel ratio sensor is compensated even immediately after the start of the sub-feedback control or when the sub-feedback control is stopped. Thus, the air-fuel ratio of the exhaust gas flowing into the catalyst can be brought close to the stoichiometric air-fuel ratio.

  Therefore, by performing the sub-feedback learning, it is possible to suppress the deterioration of emissions even when an asymmetric abnormality occurs in the air-fuel ratio sensor. However, a predetermined guard is set for the learning value for the purpose of preventing erroneous correction. For this reason, when the degree of asymmetric abnormality is large, it is assumed that the learning value is limited to the guard width. In particular, when the fuel used includes alcohol fuel, the amount of hydrogen discharged increases, and the steady-state deviation of the air-fuel ratio becomes larger. For this reason, in the conventional technique, the learning value is limited by a predetermined guard width, and there is a possibility that desired sub-feedback learning cannot be performed.

  The present invention has been made to solve the above-described problems, and suppresses deterioration of emission characteristics even in a case where an asymmetric abnormality occurs in an exhaust gas sensor in an internal combustion engine using a fuel containing alcohol. It is an object of the present invention to provide a control device for an internal combustion engine that can be used.

In order to achieve the above object, a first invention is a control device for an internal combustion engine,
A control device for an internal combustion engine capable of using a fuel containing alcohol,
A catalyst disposed in the exhaust passage;
A main exhaust gas sensor disposed upstream of the catalyst and emitting an output corresponding to the air-fuel ratio of the exhaust gas upstream of the catalyst;
An asymmetry calculating means for calculating an asymmetry between the output characteristic of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from rich to lean and that when changing from lean to rich;
A sub exhaust gas sensor disposed downstream of the catalyst and emitting an output corresponding to the air-fuel ratio of the exhaust gas downstream of the catalyst;
Sub feedback learning means for learning a correction amount to be reflected in the fuel injection amount based on an integrated value of an output deviation between the output value of the sub exhaust gas sensor and a reference value corresponding to the theoretical air-fuel ratio;
Guard means for limiting the correction amount by a predetermined guard width;
Fuel property detecting means for detecting the alcohol concentration of the fuel;
Guard width control means for variably setting the guard width in response to a change in the correction amount according to the alcohol concentration and the degree of asymmetry;
It is characterized by providing.

According to a second invention, in the first invention,
The asymmetry calculation means includes:
Means for detecting a correlation value (hereinafter referred to as a first correlation value) of output responsiveness of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from a predetermined rich air-fuel ratio to a lean air-fuel ratio;
Means for detecting a correlation value (hereinafter referred to as a second correlation value) of output responsiveness of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from the lean air-fuel ratio to the rich air-fuel ratio;
Means for calculating the degree of asymmetry by dividing the first correlation value by the second correlation value;
It is characterized by including.
The second invention is the first or second invention,
The guard width control means increases the guard width as the alcohol concentration increases.

A fourth invention is any one of the first to third inventions,
The guard width control means increases the guard width as the degree of asymmetry increases.

According to a fifth invention, in any one of the first to fourth inventions,
The sub-feedback learning means is
Updating means for learning the correction amount at a predetermined update speed;
An update rate control means for variably setting the update rate in response to a change in the correction amount according to the alcohol concentration and the degree of asymmetry;
It is characterized by including.

According to a sixth invention, in the fifth invention,
The update speed control means increases the update speed as the alcohol concentration increases.

A seventh invention is the fifth or sixth invention, wherein
The update speed control means increases the update speed as the degree of asymmetry increases.

According to an eighth invention, in any one of the fifth to seventh inventions,
The sub-feedback learning means includes means for calculating a value obtained by multiplying an integrated value of the output deviation by a predetermined gain as the correction amount,
The update speed control means includes
And a means for increasing the gain in response to the increase in the correction amount.

According to a ninth invention, in any one of the fifth to eighth inventions,
The update speed control means includes
The method further comprises means for narrowing the learning interval of the correction amount corresponding to the increase of the correction amount.

  When the output characteristics of the main exhaust gas sensor become asymmetric, a steady deviation of the air-fuel ratio occurs. In addition, when a fuel with a high alcohol concentration is used, steady deviation increases due to an increase in the amount of hydrogen discharged. In the sub-feedback learning using the sub-exhaust gas sensor, the steady-state deviation of these air-fuel ratios is corrected by reflecting the correction amount learned based on the integrated value of the output deviation in the fuel injection amount. According to the first aspect of the invention, the guard width of the guard provided in the correction amount is variably controlled in response to the change in the correction amount according to the alcohol concentration of the fuel and the asymmetry degree of the output characteristics. For this reason, according to the present invention, it is possible to effectively prevent erroneous correction while suppressing unnecessary restrictions on the correction amount of sub-feedback learning.

  According to the second aspect of the invention, the correlation value (first correlation value) of the output response when the air-fuel ratio of the exhaust gas is changed from the predetermined rich air-fuel ratio to the lean air-fuel ratio, and conversely, the rich from the lean air-fuel ratio. A correlation value (second correlation value) of output responsiveness when the air-fuel ratio is changed is detected, and a value obtained by dividing the first correlation value by the second correlation value is calculated as the degree of asymmetry. Therefore, according to the present invention, the asymmetry degree of the output characteristic of the main exhaust gas sensor can be calculated effectively.

  According to the third aspect of the invention, the higher the alcohol concentration of the fuel, that is, the greater the steady-state deviation of the air-fuel ratio, the greater the guard width of the sub feedback learning correction amount. For this reason, according to the present invention, it is possible to effectively guard against erroneous correction and to effectively prevent an unnecessary restriction from being imposed on the correction amount corresponding to the steady-state deviation of the air-fuel ratio caused by the alcohol concentration. .

  According to the fourth aspect of the invention, the greater the asymmetry of the output characteristics of the main exhaust gas sensor, that is, the greater the steady deviation of the air-fuel ratio, the greater the guard width of the correction amount for sub feedback learning. For this reason, according to the present invention, erroneous correction is effectively guarded, and unnecessary restriction is effectively prevented from being imposed on the correction amount corresponding to the steady deviation of the air-fuel ratio caused by the asymmetry of the output characteristics. be able to.

  According to the fifth aspect, the update rate of the correction amount is variably controlled in response to the change in the correction amount according to the alcohol concentration of the fuel and the asymmetry degree of the output characteristics. For this reason, according to the present invention, since the update rate of the correction amount can be controlled in accordance with the magnitude of the steady deviation of the air-fuel ratio, it is possible to achieve both suppression of emission deterioration and suppression of variation.

  According to the sixth invention, the higher the alcohol concentration of the fuel, that is, the larger the correction amount of the sub feedback learning, the faster the update rate of the correction amount. For this reason, according to the present invention, when the steady-state deviation of the air-fuel ratio is large, the effect of the correction is obtained at an early stage to quickly suppress the deterioration of the emission, and when the steady-state deviation of the air-fuel ratio is large, it is reliably corrected. To effectively suppress variations in emissions.

  According to the seventh invention, the update rate of the correction amount is increased as the asymmetry degree of the output characteristic of the main exhaust gas sensor is larger, that is, as the correction amount of the sub feedback learning is larger. For this reason, according to the present invention, when the steady-state deviation of the air-fuel ratio is large, the effect of the correction is obtained at an early stage to quickly suppress the deterioration of the emission, and when the steady-state deviation of the air-fuel ratio is large, it is reliably corrected. To effectively suppress variations in emissions.

  According to the eighth aspect, it is possible to effectively increase the update speed of the correction amount by increasing the gain multiplied by the correction amount.

  According to the ninth aspect, the correction amount update speed can be effectively increased by narrowing the correction amount learning interval.

1 is a diagram illustrating an overall configuration of an internal combustion engine to which an air-fuel ratio control apparatus according to an embodiment of the present invention is applied. It is a flowchart which shows the routine of the main feedback control performed in the system of this Embodiment. It is a flowchart which shows the routine of the sub feedback control performed in the system of this Embodiment. It is a calculation map of sub F / B correction amount Kr. It is a map which prescribes | regulates the guard value of the sub F / B correction amount according to an asymmetry degree and alcohol concentration. It is a map which prescribes | regulates the gain Pr of the sub F / B correction value according to an asymmetry degree and alcohol concentration. It is a flowchart which shows the routine performed in Embodiment 1 of this invention. It is a figure for demonstrating the asymmetrical abnormality of an air fuel ratio sensor.

  Embodiments of the present invention will be described below with reference to the drawings. In addition, the same code | symbol is attached | subjected to the element which is common in each figure, and the overlapping description is abbreviate | omitted. The present invention is not limited to the following embodiments.

Embodiment 1 FIG.
[Configuration of Embodiment 1]
FIG. 1 is a diagram showing an overall configuration of an internal combustion engine to which an air-fuel ratio control apparatus according to an embodiment of the present invention is applied. This internal combustion engine is an internal combustion engine mounted on an FFV (Flexible Fuel Vehicle) and can be operated with gasoline as a fuel, and can also be operated with a fuel obtained by mixing alcohol such as ethanol or methanol and gasoline. Is.

  As shown in this figure, an exhaust passage 4 is connected to the engine body 2. In the exhaust passage 4, three-way catalysts 6, 8 for purifying harmful components (NOx, CO, HC) in the exhaust gas are arranged in two stages. The upstream three-way catalyst 6 is disposed close to the exhaust manifold, and the downstream three-way catalyst 8 is disposed under the floor of the vehicle.

An A / F sensor (entire air / fuel ratio sensor) 12 is attached upstream of the three-way catalyst 6, and an O ₂ sensor (oxygen sensor) 14 is attached downstream of the three-way catalyst 6. The A / F sensor 12 is a sensor that shows output characteristics linear with respect to the air-fuel ratio. The O ₂ sensor 14 is a sensor that outputs a signal corresponding to the oxygen concentration in the gas, and has an output characteristic that an output value is inverted with respect to the air-fuel ratio with respect to the stoichiometric air-fuel ratio. Specifically, when the exhaust gas downstream of the three-way catalyst 6 is rich with respect to the stoichiometric air-fuel ratio, a rich output (for example, 0.8 V) is emitted, and when the exhaust gas is lean, A lean output (for example, 0.2V) is generated.

  In addition, the system of the present embodiment includes a fuel property sensor 16 that can detect the alcohol concentration of the fuel. As the fuel property sensor 16, for example, a sensor that detects the alcohol concentration by measuring the dielectric constant, refractive index, etc. of the fuel can be used.

The system engine of the present embodiment is provided with an ECU (Electronic Control Unit) 10 as a control device that comprehensively controls the operation of the entire system. The A / F sensor 12, the O ₂ sensor 14, and the fuel property sensor 16 described above are connected to the ECU 10. The ECU 10 feedback-controls the fuel injection amount based on the output values of the A / F sensor 12 and the O ₂ sensor 14 so that the air-fuel ratio of the exhaust gas flowing into the three-way catalyst 6 becomes the stoichiometric air-fuel ratio.

(Air-fuel ratio control of the present embodiment)
The air-fuel ratio control executed by the ECU 10 includes main feedback control and sub feedback control. In the main feedback control, a feedback value (main F / B correction amount) to be reflected in the calculation of the fuel injection amount is calculated based on the deviation between the output signal of the A / F sensor 12 and the theoretical air-fuel ratio. In the sub feedback control, a feedback value (sub F / B correction amount) to be reflected in the calculation of the fuel injection amount is calculated based on the deviation between the output signal of the O ₂ sensor 14 and the reference signal. The contents of feedback control and sub-feedback control according to the present embodiment will be specifically described below.

(Main feedback control)
First, the content of the main feedback control will be specifically described. FIG. 2 is a flowchart showing a main feedback control routine executed in the system of the present embodiment. The routine shown in this figure is periodically executed at predetermined processing cycles such as every fixed crank.

  In the routine shown in FIG. 2, first, a basic injection amount Qb is calculated such that the air-fuel ratio of the air-fuel mixture in the combustion chamber is stoichiometric (step 100). Next, the output Vf of the A / F sensor 12 is detected (step 102). Next, an output difference ΔVf (= Vf−Vref) between the output Vf detected in step 102 and the sensor output Vref corresponding to the stoichiometric air-fuel ratio (stoichiometric) is calculated (step 104).

Next, the main F / B correction amount Kf (= Pf × ΔVf) is calculated by multiplying the calculated output difference ΔVf by a predetermined gain Pf (step 106). Next, the basic injection amount Qb calculated in step 100, the main F / B correction amount Kf calculated in step 106, and the sub F / B correction amount Kr calculated in sub-feedback control to be described later are expressed by the following equations. By substituting for (1), the final fuel injection amount Qfnl is calculated (step 108).
Qfnl = Kf × Qb + Kr (1)

  According to the above-described main feedback control, the larger the sensor output Vf is larger than the sensor output Vref corresponding to the theoretical air-fuel ratio (ΔVf> 0), that is, the greater the actual pre-catalyst air-fuel ratio moves away from the stoichiometric side, the larger the correction. Since the amount Kf is calculated, the basic injection amount Qb is corrected to increase. Conversely, the smaller the sensor output Vf is smaller than the sensor output Vref (ΔVf <0), that is, the smaller the actual pre-catalyst air-fuel ratio is far from the stoichiometric side, the smaller the correction amount Kf is calculated. Qb is corrected to decrease. Thus, in the main feedback control, the fuel injection amount is increased or decreased so that the air-fuel ratio detected by the A / F sensor 12 matches the stoichiometric air-fuel ratio.

(Sub feedback control)
Next, the content of the sub feedback control will be specifically described. FIG. 3 is a flowchart showing a sub-feedback control routine executed in the system of the present embodiment. The routine shown in this figure is periodically executed at predetermined processing cycles such as every fixed crank.

In the routine shown in FIG. 3, first, counting of a timer provided in the ECU 10 is started (step 200). Next, the sensor output Vr of the O ₂ sensor 14 is detected (step 202). Next, an output difference ΔVr (= Vrefr−Vr) between the sensor output Vr and the sensor output Vrefr corresponding to the theoretical air-fuel ratio is calculated, and this output difference ΔVr is integrated with the previous integrated value (step 204).

  Next, it is determined whether or not the timer value exceeds a predetermined value ts (step 206). As a result, when the timer value does not exceed the predetermined value ts, this routine is immediately terminated. On the other hand, when the timer value exceeds the predetermined value ts in this step 206, the process proceeds to the next step, and the output difference integrated value ΣΔVr at this time is updated and stored as the learned value ΔVrg (step 208). ).

  Next, the sub-F / B correction amount Kr (= Pr × ΔVrg) is updated by multiplying the stored learning value ΔVrg by a predetermined gain Pr (step 210). Next, the output difference integrated value ΣΔVr and the timer are reset (step 212). The updated sub F / B correction amount Kr is substituted into the above equation (1) in step 108. Thereby, the sub F / B correction amount Kr can be reflected in the final fuel injection amount Qfnl.

  The reason why the sensor output difference ΔVr is accumulated for a predetermined time ts is to detect a steady deviation amount (steady deviation) of the sensor output Vr with respect to the sensor output Vrefr. The predetermined value ts that defines the integration time is set to a time much longer than one engine cycle. Therefore, the learning value ΔVrg and the sub F / B correction amount Kr are updated in a period much longer than one engine cycle.

  The sub F / B correction amount Kr may be calculated from the map shown in FIG. FIG. 4 is a calculation map of the sub F / B correction amount Kr. As shown in this map, the sensor output Vr becomes zero as the sensor output Vrr is smaller than the sensor output Vrefr corresponding to the theoretical air-fuel ratio on a time average (ΔVrg> 0), that is, as the actual post-catalyst air-fuel ratio moves away from the stoichiometry. On the other hand, a larger sub F / B correction amount Kr is obtained, and the basic injection amount Qb is increased and corrected when the final injection amount is calculated. On the contrary, as the sensor output Vr is larger than the sensor output Vrefr in terms of time average (ΔVrg <0), that is, as the actual post-catalyst air-fuel ratio moves away from the stoichiometric side, a smaller correction amount Kr with respect to 0 is obtained. The basic injection amount Qb is corrected to decrease.

(Features of this embodiment)
Next, features of the present embodiment will be described. In the sub-feedback control of the present embodiment, the control amount is limited within a predetermined guard range in order to prevent erroneous correction from being executed. More specifically, as shown in FIG. 4, the sub F / B correction amount Kr has a mechanism that can only take values within the upper and lower guard value ranges. For this reason, for example, even when the calculated sub F / B correction amount Kr is equal to or greater than the upper limit guard value, the control sub F / B correction amount Kr is fixed to the upper limit guard value. As the guard range, a preset value is used as a range of learning values for restricting execution of erroneous correction.

  Here, when an asymmetrical abnormality as shown in FIG. 8 occurs in the output of the A / F sensor 12, the sensor output Vref corresponding to the theoretical air-fuel ratio deviates from the true value. For this reason, even if the sensor output Vf converges in the vicinity of the sensor output Vref in the main feedback control described above, a steady deviation occurs in the actual air-fuel ratio.

  In the sub-feedback control, as described above, the steady shift of the sensor output Vr with respect to the sensor output Vrefr can be detected and corrected. Therefore, the main feedback control is executed using the A / F sensor 12 in which the asymmetric abnormality has occurred, and even when the steady deviation occurs, the steady deviation is effectively corrected by the sub feedback control. Can do.

  However, as the degree of asymmetry of the sensor output of the A / F sensor 12 increases, the above-described steady deviation increases. For this reason, depending on the degree of asymmetry, the calculated sub F / B correction amount Kr is outside the range of the upper and lower guard values, and it is assumed that sufficient sub feedback correction cannot be performed due to the guard effect. Is done.

  In particular, in the engine of the present embodiment, a fuel containing alcohol is used. If the alcohol concentration is high, the amount of hydrogen in the exhaust gas increases, so the output of the A / F sensor 12 shifts to the rich side. For this reason, in the feedback control of the present embodiment that also corrects the influence of hydrogen, it is sufficiently assumed that the calculated sub F / B correction amount Kr exceeds the upper and lower guard values. On the other hand, since the guard value is provided to prevent erroneous correction, it is not wise to loosen the guard in the dark clouds.

  Therefore, in the system according to the present embodiment, the range of the guard value provided in the sub-feedback is variably controlled according to the alcohol concentration of the fuel and the degree of asymmetry of the A / F sensor 12. FIG. 5 is a map that defines the guard value for the sub F / B correction amount according to the degree of asymmetry and the alcohol concentration. As shown in this figure, the guard value of the sub F / B correction value is set to a larger value as the alcohol concentration is higher and the sensor output asymmetry degree of the A / F sensor 12 is larger. As a result, the guard value of the sub F / B correction value can be set to a larger value as the steady state deviation increases, so that a situation in which the sub feedback control is unnecessarily limited can be effectively suppressed.

  As described above, in the sub-feedback control, the sensor output difference ΔVr is integrated for a predetermined time ts in order to detect a steady deviation amount (steady deviation) of the sensor output Vr with respect to the sensor output Vrefr. For this reason, the sub F / B correction value Kr is updated every predetermined time ts. However, when an asymmetrical abnormality has occurred in the A / F sensor 12, it is preferable to increase the update rate of the sub feedback control to quickly suppress the emission deterioration. However, if the update speed of the sub feedback control is increased when the sensor output is normal, it becomes a factor that increases the variation in emissions.

  Therefore, in the system of the present embodiment, the update speed of the sub feedback control is variably set according to the asymmetry degree of the A / F sensor 12 and the alcohol concentration of the fuel. As a specific method for increasing the update speed, for example, a method of narrowing the integration time ts of the sensor output difference ΔVr, that is, a learning interval, a method of increasing the gain Pr of the sub F / B correction value Kr, and the like are considered. It is done. FIG. 6 is a map that defines the gain Pr of the sub F / B correction value according to the degree of asymmetry and the alcohol concentration. As shown in the figure, the gain Pr of the sub F / B correction value is set to a larger value as the alcohol concentration is higher and asymmetry of the sensor output of the A / F sensor 12 is larger. Accordingly, the gain Pr of the sub F / B correction value Kr can be increased as the steady-state deviation of the sensor output of the A / F sensor 12 is increased, and thus the increase in variation when the sensor output is normal can be prevented. It is possible to achieve both a quick reflection of sub-feedback when an asymmetric abnormality occurs in the output.

  In the map shown in FIG. 6, the gain Pr of the sub F / B correction value corresponding to the degree of asymmetry and the alcohol concentration is defined. However, the learning interval described above may be defined instead of the gain Pr. Good. In this case, it is possible to effectively control the update rate by defining the learning interval to be narrower as the alcohol concentration is higher and the sensor output asymmetry degree of the A / F sensor 12 is larger.

[Specific Processing in First Embodiment]
Next, specific processing for realizing the above-described control will be described with reference to FIG. First, FIG. 7 is a flowchart showing a routine executed by ECU 10 in the first embodiment of the present invention. The routine shown in FIG. 7 is repeatedly executed during engine operation.

  In the routine shown in FIG. 7, first, it is determined whether or not a feedback condition is satisfied (step 300). Specifically, it is determined whether or not the A / F sensor 12 and the O2 sensor 14 are activated. As a result, when it is determined that the feedback condition is not satisfied, this routine is immediately terminated. On the other hand, when it is determined in step 100 that the feedback condition is satisfied, the process proceeds to the next step, and the asymmetry of the A / F sensor 12 is acquired (step 302). Specifically, the asymmetry calculated in steps 308 to 318 described later and stored as a learning value is read.

  Next, the alcohol concentration of the fuel is detected based on the detection signal of the fuel property sensor 16 (step 304). Next, the control amount of the fuel injection control is changed (step 306). Here, specifically, first, using the map shown in FIG. 5, the sub feedback correction amount corresponding to the asymmetry degree of the A / F sensor 12 and the alcohol concentration of the fuel acquired in the above-described steps 302 and 304 described above. The guard value is calculated. In step 306, the gain Pr of the sub-feedback control corresponding to the asymmetry of the A / F sensor 12 and the alcohol concentration of the fuel obtained in steps 302 and 304 described above is obtained using the map shown in FIG. Calculated.

  Next, it is determined whether or not an execution condition for active control is satisfied (step 308). As an execution condition for the active control, for example, whether or not the intake air amount has reached a predetermined amount can be mentioned. As a result, when it is determined that the active control execution condition is not satisfied, this routine is immediately terminated. On the other hand, if it is determined in step 308 that the execution condition for active control is satisfied, the process proceeds to the next step, and the control target air-fuel ratio is changed from the rich side (or lean side) to the stoichiometric side. Active control of the injection amount that is forcibly changed to the side (or rich side) is executed (step 310).

Next, it is determined whether or not a flag prohibiting the estimation of the degree of asymmetry is issued from other controls, and whether or not the active control has been executed once continuously (step 312). As a result, when this establishment is not recognized, this routine is immediately terminated. On the other hand, if the determination in step 312 is confirmed, the process proceeds to the next step, and a determination value is calculated (step 314). Here, specifically, first, the response time of the A / F sensor 12 when the control target air-fuel ratio is changed from the predetermined rich air-fuel ratio to the lean air-fuel ratio, and conversely, from the predetermined lean air-fuel ratio to the rich air-fuel ratio. The response time of the A / F sensor 12 when changing to the fuel ratio is calculated. Then, the determination value of the asymmetry degree is calculated by substituting the calculated response time into the following equation (2).
Asymmetry determination value = | Rich → Lean response time / Lean → Rich response time | (2)

  Next, it is determined whether or not the number of executions of active control is a predetermined number (for example, 5 times) or more (step 316). As a result, when the number of executions of active control has not yet reached the predetermined number, this routine is immediately terminated. On the other hand, if it is determined in step 316 that the determination is satisfied, the process proceeds to the next step, and asymmetry calculation and learning processing is performed (step 318). Specifically, the average value of the past five asymmetry determination values calculated in step 314 is updated as the latest asymmetry. The updated asymmetry is used in step 302 above.

  As described above, according to the system of the first embodiment, the sub F / B correction value Kr corresponding to the asymmetry of the A / F sensor 12 and the alcohol concentration of the fuel corresponds to the sub F / B. The guard value of the B correction value can be set variably. As a result, it is possible to prevent the correction of the steady deviation caused by the asymmetric abnormality of the sensor output from being restricted unnecessarily, and it is possible to effectively suppress the deterioration of the emission.

  Further, according to the system of the first embodiment, the sub F / B correction value corresponding to the change of the sub F / B correction value Kr according to the asymmetry degree of the A / F sensor 12 and the alcohol concentration of the fuel. The update speed can be variably set. As a result, it is possible to quickly correct the steady deviation caused by the asymmetric abnormality of the sensor output.

  Incidentally, in the system of the first embodiment described above, the guard value of the sub F / B correction value is variably set according to the map shown in FIG. 5, but the alcohol concentration of the fuel and the asymmetry of the A / F sensor 12 are set. It is also possible to define a guard value function according to the above and set the guard value based on the function. The same applies to the update speed of the sub F / B correction value, and the function of the learning interval or the gain Pr according to the alcohol concentration of the fuel and the asymmetry degree of the A / F sensor 12 is not limited to the map shown in FIG. Then, the update speed may be set based on the function.

  In the system of the first embodiment described above, the response time of the A / F sensor 12 when the predetermined rich air-fuel ratio is changed to the lean air-fuel ratio is used as a value indicating the response characteristic of the A / F sensor 12. However, for example, correlation values of other response characteristics represented by delay time, sensor trajectory length, or area may be used.

In the first embodiment described above, the three-way catalyst 6 is the “catalyst” in the first invention, the A / F sensor 12 is the “main exhaust gas sensor” in the first invention, and the O ₂ sensor 14. Is the “sub exhaust gas sensor” in the first invention, Vrefr is the “reference value” in the first invention, ΔVr is the “output deviation” in the first invention, and ΣΔVr is in the first invention. Kr corresponds to the “integrated value” and the “correction amount” in the first invention. Further, when the ECU 10 executes the process of step 302, the “asymmetry calculation means” in the first invention executes the processes of steps 200 to 212, so that “ The “sub-feedback learning means” executes the process of step 304 above, and the “fuel property detection means” in the first invention executes the process of step 306 so that “ "Guard width control means" is realized respectively.

  In the first embodiment described above, the rich → lean response time is changed to the “first correlation value” in the second invention, and the lean → rich response time is changed to the “second correlation value” in the second invention. , Respectively.

  In the first embodiment described above, the ECU 10 executes the process of step 208, so that the “updating means” in the fifth aspect of the invention executes the process of step 306. The “update speed control means” in the first aspect of the invention is realized.

  In the first embodiment described above, the gain Pr corresponds to the “gain” in the eighth invention.

2 Engine (Internal combustion engine)
4 Exhaust passage 6 Three-way catalyst (S / C)
8 Three-way catalyst (U / F)
10 ECU (Electronic Control Unit)
12 A / F sensor 14 O ₂ sensor 16 Fuel property sensor

Claims (9)

A control device for an internal combustion engine capable of using a fuel containing alcohol,
A catalyst disposed in the exhaust passage;
A main exhaust gas sensor disposed upstream of the catalyst and emitting an output corresponding to the air-fuel ratio of the exhaust gas upstream of the catalyst;
An asymmetry calculating means for calculating an asymmetry between the output characteristic of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from rich to lean and that when changing from lean to rich;
A sub exhaust gas sensor disposed downstream of the catalyst and emitting an output corresponding to the air-fuel ratio of the exhaust gas downstream of the catalyst;
Sub feedback learning means for learning a correction amount to be reflected in the fuel injection amount based on an integrated value of an output deviation between the output value of the sub exhaust gas sensor and a reference value corresponding to the theoretical air-fuel ratio;
Guard means for limiting the correction amount by a predetermined guard width;
Fuel property detecting means for detecting the alcohol concentration of the fuel;
Guard width control means for variably setting the guard width in response to a change in the correction amount according to the alcohol concentration and the degree of asymmetry;
A control device for an internal combustion engine, comprising: The asymmetry calculation means includes:
Means for detecting a correlation value (hereinafter referred to as a first correlation value) of output responsiveness of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from a predetermined rich air-fuel ratio to a lean air-fuel ratio;
Means for detecting a correlation value (hereinafter referred to as a second correlation value) of output responsiveness of the main exhaust gas sensor when the air-fuel ratio of the exhaust gas is changed from the lean air-fuel ratio to the rich air-fuel ratio;
Means for calculating the degree of asymmetry by dividing the first correlation value by the second correlation value;
The control apparatus for an internal combustion engine according to claim 1, comprising:   3. The control apparatus for an internal combustion engine according to claim 1, wherein the guard width control means increases the guard width as the alcohol concentration increases.   4. The control device for an internal combustion engine according to claim 1, wherein the guard width control means increases the guard width as the degree of asymmetry increases. 5. The sub-feedback learning means is
Updating means for learning the correction amount at a predetermined update speed;
An update rate control means for variably setting the update rate in response to a change in the correction amount according to the alcohol concentration and the degree of asymmetry;
The control apparatus for an internal combustion engine according to any one of claims 1 to 4, further comprising:   6. The control apparatus for an internal combustion engine according to claim 5, wherein the update speed control means increases the update speed as the alcohol concentration increases.   The control apparatus for an internal combustion engine according to claim 5 or 6, wherein the update speed control means increases the update speed as the degree of asymmetry increases. The sub-feedback learning means includes means for calculating a value obtained by multiplying an integrated value of the output deviation by a predetermined gain as the correction amount,
The update speed control means includes
8. The control device for an internal combustion engine according to claim 5, further comprising means for increasing the gain in response to the increase in the correction amount. The update speed control means includes
9. The control apparatus for an internal combustion engine according to claim 5, further comprising means for narrowing a learning interval of the correction amount corresponding to the increase of the correction amount.

Images