JP7095317B2 - Engine exhaust purification control device - Google Patents

Description

The technique disclosed herein relates to an engine exhaust purification control device.

In an engine provided with a so-called NOx catalyst such as a NO storage catalyst, it is known to control the fuel injection amount (so-called DeSOx control) in order to remove SOx adsorbed on the NOx catalyst.

For example, Patent Document 1 discloses that, as an example of DeSOx control, a rich step for enriching the air-fuel ratio of exhaust gas and a lean step for making the air-fuel ratio leaner are alternately repeated.

When the lean step disclosed in Patent Document 1 is carried out, the exhaust gas contains more oxygen than when the rich step is carried out. In this case, oxygen in the exhaust gas will flow out to the downstream side of the NOx catalyst.

On the other hand, various exhaust gas treatment devices such as a DPF for a diesel engine may be provided in the exhaust passage on the downstream side of the NOx catalyst. Therefore, in Patent Document 1, oxygen is supplied to the downstream side of the NOx catalyst by carrying out the above-mentioned lean step, and the exhaust gas treatment device (DPF) is provided by utilizing the oxygen supplied in this way. Purification is disclosed.

Japanese Patent No. 4241032

By the way, as described in Patent Document 1, when the rich step and the lean step are alternately performed, a large amount of oxygen is secured in the lean step, but oxygen is insufficient in the rich step. There was a possibility. This is inconvenient for purifying the exhaust gas treatment device by the method described in the same document. On the other hand, in order to efficiently purify the NOx catalyst, it is required to surely perform the rich step.

As described above, there is room for improvement in efficiently purifying both the NOx catalyst and the exhaust gas treatment device, not one of them.

The technique disclosed here was made in view of this point, and the purpose thereof is to efficiently purify both the NOx catalyst and the exhaust gas treatment device in the exhaust gas purification control device of the engine. be.

The inventors of the present application do not use a common NOx catalyst for all cylinders, but if a plurality of NOx catalysts are used, the rich step and the lean step are alternately performed and supplied to the exhaust gas treatment device. Focusing on the ability to secure the desired oxygen, we have found the technology disclosed here.

Specifically, the techniques disclosed herein include an exhaust passage connected to each of a plurality of cylinders, a NOx catalyst provided in the exhaust passage, and an exhaust provided in an exhaust passage downstream of the NOx catalyst. The present invention relates to an exhaust gas purification control device for an engine equipped with a gas treatment device. The exhaust gas purification control device includes an injector provided in each of the plurality of cylinders and a controller connected to each of the plurality of injectors. The exhaust gas treatment device is a particle filter configured to collect particulate matter in the exhaust gas.

The exhaust passage includes a first exhaust passage connected to a part of the plurality of cylinders, a second exhaust passage connected to another part of the plurality of cylinders, the first exhaust passage, and the above. The second exhaust passage is formed by merging on the downstream side of each, and has a third exhaust passage provided with the exhaust gas treatment device, and the NOx catalyst is provided in the first exhaust passage. It has a first NOx catalyst and a second NOx catalyst provided in the second exhaust passage.

The controller alternately performs a NOx catalyst regeneration control in which the rich step of making the air-fuel ratio near the theoretical air-fuel ratio or richer than the theoretical air-fuel ratio and the lean step of making the air-fuel ratio leaner than the rich step are performed alternately. A control signal is output to each of the plurality of injectors so as to perform the above.

When the controller performs the rich step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control, the controller performs the lean step in the cylinder leading to the second exhaust passage. A control signal is output to each of the plurality of injectors, and the controller is configured to remove the sulfur component from the first and second NOx catalysts by executing the NOx catalyst regeneration control over a plurality of cycles. Further, when the lean step is performed, the controller further injects fuel into the injector at an injection timing that does not burn in the cylinder after the fuel injection corresponding to the operating state of the engine. And output the control signal .

Here, the term "near the theoretical air-fuel ratio" is used to mean that the air-fuel ratio is within the range of 14 to 16.

According to this configuration, when the NOx catalyst regeneration control is performed, the controller alternately performs the rich step and the lean step. By executing such NOx catalyst regeneration control, the NOx catalyst can be purified as in the so-called DeSOx control.

However, simply alternating between the rich step and the lean step may result in a shortage of oxygen in the rich step, as described above.

On the other hand, according to the above configuration, instead of using a common NOx catalyst for all cylinders, a first NOx catalyst provided in the first exhaust passage and a second NOx catalyst provided in the second exhaust passage are used. ing.

Therefore, in the controller, when the air-fuel ratio of the exhaust gas flowing through the first NOx catalyst becomes rich, that is, when the rich step is performed in the cylinder leading to the first exhaust passage, the air-fuel ratio of the exhaust gas flowing through the second NOx catalyst is lean. A lean step is carried out in the cylinder leading to the second exhaust passage so as to be.

According to this configuration, the exhaust gas passing through the first NOx catalyst may be deficient in oxygen, but the exhaust gas passing through the second NOx catalyst contains a large amount of oxygen, and therefore downstream of them. Exhaust gas containing more oxygen will flow through the third exhaust passage provided in the above, as compared with the case where the rich step is performed in all the cylinders.

This makes it possible to stably supply oxygen to the exhaust gas treatment device and, by extension, efficiently purify the exhaust gas treatment device. On the other hand, in the cylinder unit, since the rich step and the lean step are alternately performed, not only the exhaust gas treatment device but also the NOx catalyst can be efficiently purified.

As described above, according to the above configuration, both the exhaust gas treatment device and the NOx catalyst can be efficiently purified .

Further, the "sulfur component" includes so-called sulfur oxide (SOx).

Further, the exhaust gas treatment device may be a so-called GPF or DPF.

Further, according to the above configuration, DeSOx control for removing the sulfur component from the NOx catalyst and regeneration control of the filter for burning and removing particulate matter are simultaneously performed, and thus each control is made efficient. It is possible to complete the process.

In particular, as described above, even when the rich step is performed in some cylinders, the lean step is performed in other cylinders, so that the exhaust gas treatment device is used. It is possible to supply sufficient oxygen and, by extension, realize efficient regeneration control.

In addition, the controller performs additional injections, such as so-called post injections, when performing lean steps. Since the fuel injected by this additional injection does not burn in the cylinder, it is discharged to the exhaust passage as unburned fuel (HC). Such HC can be used as a reducing agent.

Therefore, if the exhaust gas treatment device has an oxidizing function, a large amount of oxygen is supplied, and the exhaust gas treatment device generates heat of reaction more frequently than before. In order to burn the particulate matter (PM) collected in the exhaust gas treatment device, it was necessary to supply relatively high temperature exhaust gas to the exhaust gas treatment device. Exhaust gas can be made colder than before. Therefore, it is possible to reduce the injection amount of fuel such as the above-mentioned additional injection, and at the same time, it is possible to keep the NOx catalyst at a lower temperature than before by the amount of lowering the exhaust gas.

In general, it is known that when the temperature of the NOx catalyst rises excessively, SOx adsorbed on the NOx catalyst aggregates, making it difficult to remove the SOx from the NOx catalyst. This is not preferable in that it causes a decrease in the efficiency of DeSOx control.

On the other hand, the above configuration is effective from the viewpoint of suppressing a decrease in the efficiency of DeSOx control because the NOx catalyst can be kept at a lower temperature than the conventional one.

Further, the controller determines the magnitude relationship between the sulfur poisoning amount in the first NOx catalyst and the sulfur poisoning amount in the second NOx catalyst based on the injection amount of fuel to the plurality of cylinders, and the controller determines the magnitude relationship. When the amount of sulfur poisoning in the first NOx catalyst is larger than the amount of sulfur poisoning in the second NOx catalyst when the NOx catalyst regeneration control is performed, the cylinder leading to the first exhaust passage is compared with the case where the amount is smaller. The time for performing the rich step may be lengthened.

For example, as the vehicle equipped with the engine repeatedly accelerates and decelerates, the fuel injection amount becomes biased, which may cause a difference in the sulfur poisoning amount.

Further, in the rich step, the exhaust becomes a reducing atmosphere. On the other hand, SOx adsorbed on the NOx catalyst is released in such a reducing atmosphere.

Therefore, in a situation where the amount of sulfur poisoning in the first NOx catalyst is large and it is considered that a large amount of SOx is adsorbed on the catalyst, the time for performing the rich step is lengthened to obtain a larger amount of SOx in the first NOx. It can be released from the catalyst.

Further, when the controller performs the lean step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control, the controller performs the rich step in the cylinder leading to the second exhaust passage. As such, the control signal may be output to each of the plurality of injectors.

According to this configuration, the first NOx catalyst and the second NOx catalyst can be purified.

Further, when a predetermined condition is satisfied before the start of the NOx catalyst regeneration control, the controller additionally injects fuel at an injection timing such that it burns in the cylinder after the fuel injection corresponding to the operating state of the engine. By performing the additional injection of the fuel, a control signal is output to the injector so as to perform NOx catalyst rich purge control that makes the air-fuel ratio richer than before the predetermined condition is satisfied. May be good.

When the NOx catalyst regeneration control is performed to purify the exhaust gas treatment device, the NOx catalyst may be desorbed as a result of the excessive temperature rise of the NOx catalyst.

On the other hand, according to the above configuration, when a predetermined condition is satisfied, the controller executes NOx catalyst rich purge control by performing additional injection such as so-called post injection. This makes it possible to remove NOx from the NOx catalyst.

That is, by removing NOx from the NOx catalyst before starting the NOx catalyst regeneration control, it is possible to suppress the desorption of NOx caused by the NOx catalyst regeneration control.

Further, the exhaust gas treatment device may have an oxidizing function.

As described above, when the exhaust gas treatment device has an oxidation function, unburned fuel such as HC can be oxidized in this exhaust gas treatment device even during NOx catalyst regeneration control. As a result, the emission performance of the engine can be improved.

Further, another technique disclosed herein includes an exhaust passage connected to each of a plurality of cylinders, a NOx catalyst provided in the exhaust passage, and an exhaust provided in an exhaust passage downstream of the NOx catalyst. The present invention relates to an exhaust gas purification control device for an engine equipped with a gas treatment device. This exhaust gas purification control device is provided in an injector provided in each of the plurality of cylinders, a controller connected to each of the plurality of injectors, and the third exhaust passage on the downstream side of the exhaust gas treatment device. The SCR catalyst is provided, and NOx sensors provided on the upstream side and the downstream side of the SCR catalyst are provided.

The exhaust passage includes a first exhaust passage connected to a part of the plurality of cylinders, a second exhaust passage connected to another part of the plurality of cylinders, the first exhaust passage, and the above. The second exhaust passage is formed by merging on the downstream side of each, and has a third exhaust passage provided with the exhaust gas treatment device, and the NOx catalyst is provided in the first exhaust passage. It has a first NOx catalyst and a second NOx catalyst provided in the second exhaust passage.

The controller alternately performs a NOx catalyst regeneration control in which the air-fuel ratio is near the theoretical air-fuel ratio or a rich step that makes the air-fuel ratio richer than the theoretical air-fuel ratio and a lean step that makes the air-fuel ratio leaner than the rich step. When the control signal is output to each of the plurality of injectors, and the controller performs the rich step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control. , A control signal is output to each of the plurality of injectors so that the lean step is performed in the cylinder leading to the second exhaust passage.

Then, when the rich step is performed, the injector is such that after the fuel injection corresponding to the operating state of the engine, the additional injection of fuel is executed at the injection timing such that the fuel is burned in the cylinder. The controller outputs a control signal to the engine, estimates the amount of ammonia adsorbed by the SCR catalyst based on the detection signal of the NOx sensor, and when the adsorbed amount is large, in the rich step than when it is small. Delay the timing to execute the additional injection.

In general , ammonia may be generated by the binding of NOx, a reducing agent, or the like. The generation of ammonia is not desirable in order to suppress odors and the like.

On the other hand, when the timing for executing the additional injection is retarded as in the above configuration, the combustion temperature when the fuel injected by the additional injection burns in each cylinder is reduced. As a result, the temperature of the exhaust gas discharged from each cylinder is lowered. Since a predetermined heat energy is required for the production of ammonia, the above configuration is effective in suppressing the generation of ammonia.

Further, the controller may reduce the amount of fuel additionally injected in the rich step when the adsorption amount of the ammonia is large as compared with the case where the adsorption amount is small.

By reducing the amount of additional fuel injected as in this configuration, the thermal energy generated by the combustion of the fuel can be suppressed. This is advantageous in suppressing the generation of ammonia.

As described above, the technique disclosed herein can efficiently purify both the NOx catalyst and the exhaust gas treatment device.

FIG. 1 is a schematic diagram illustrating an engine configuration. FIG. 2 is a schematic diagram illustrating the layout of the exhaust system by omitting a part thereof. FIG. 3 is a schematic diagram illustrating the configuration of the DPF. FIG. 4 is a block diagram illustrating the configuration of the exhaust gas purification control device of the engine. FIG. 5 is a diagram illustrating an operating area of the engine. The upper figure of FIG. 6 is a diagram illustrating a map for determining a post injection timing, and the lower figure of FIG. 5 is a diagram illustrating a map for determining a post injection amount. FIG. 7 is a diagram illustrating a map for determining the ratio between the basic lean time and the basic rich time. FIG. 8 is a flowchart illustrating the specific contents of DeSOx control. FIG. 9 is a time chart showing a specific example of DeSOx control.

Hereinafter, embodiments of the exhaust gas purification control device for the engine will be described in detail with reference to the drawings. The following description is an example of an engine exhaust gas purification control device. FIG. 1 is a schematic diagram illustrating the configuration of the engine 1, and FIG. 2 is a schematic diagram illustrating the layout of the exhaust system by omitting a part thereof. Further, FIG. 3 is a schematic diagram illustrating the configuration of the DPF 44, and FIG. 4 is a block diagram illustrating the configuration of the exhaust gas purification control device of the engine 1.

The engine 1 is a diesel engine to which fuel mainly composed of light oil is supplied, and is configured as a 4-stroke engine equipped with four cylinders (cylinders) and is mounted on a four-wheeled automobile (vehicle). ing. The crankshaft 7, which is the output shaft of the engine 1, is connected to the drive wheels of an automobile via a transmission (not shown), and when the engine 1 operates, the output is transmitted to the drive wheels to drive the automobile. It is designed to do.

The engine 1 is a two-stage turbocharged engine. That is, as shown in FIG. 1, the first turbocharger 51 configured to supercharge the gas introduced into the combustion chamber 6 and the exhaust passage 40 leading to the combustion chamber 6 of the engine 1 A second turbocharger 52 is provided. The exhaust passage 40 is provided with a NOx catalyst 41, which will be described later, and a DPF 44 as an exhaust gas treatment device, in order from the upstream side.

Hereinafter, the overall configuration of the engine 1 will be described in detail.

(1) Overall configuration The engine 1 includes a cylinder block 3 provided with a plurality of cylinders 2 (only one is shown in FIG. 1), a cylinder head 4 provided on the upper surface of the cylinder block 3, and each cylinder. It has a piston 5 inserted in 2. The piston 5 is connected to the crankshaft 7 via a connecting rod.

Further, a cavity is formed on the top surface of each piston 5. The combustion chamber 6 is partitioned for each cylinder 2 by the cavity, the inner wall surface of the cylinder 2, and the cylinder head 4.

The cylinder head 4 is formed with an intake port 16 for introducing intake air into the combustion chamber 6 and an exhaust port 17 for deriving exhaust air from the combustion chamber 6 for each cylinder 2. The intake port 16 is open to the combustion chamber 6, and an intake valve 12 for opening and closing the opening is provided. Similarly, the exhaust port 17 is also open to the combustion chamber 6, and an exhaust valve 13 for opening and closing the opening is arranged.

The cylinder head 4 is also provided with an injector 10 for injecting fuel into the combustion chamber 6 and a set of glow plugs 11 for raising the temperature of the gas in each cylinder 2 for each cylinder 2. ..

In the example shown in FIG. 1, the tip of the injector 10 is arranged so as to face the combustion chamber 6 from the ceiling surface of the combustion chamber 6 (specifically, the surface partitioned by the cylinder head 4). A plurality of injection ports are provided at the tip of the injector 10, and are configured to be able to control the opening degree of each injection port.

The ECU 100, which will be described later, inputs a pulse signal (control signal) to the injector 10 in order to control the injection mode of the fuel through the injector 10. The fuel injection mode can be controlled through the pulse width of the pulse signal, the input timing, the number of inputs, and the like.

Specifically, the injector 10 can carry out a main injection mainly for obtaining engine torque and a post injection in which the combustion energy hardly contributes to the engine torque. Here, the main injection is to inject the fuel before or near the compression top dead center so that the injected fuel starts to burn from the vicinity of the compression top dead center. On the other hand, the post injection is to inject the fuel at the timing on the retard side of the main injection (specifically, the timing during the expansion stroke).

Further, the glow plug 11 has a heat generating portion at the tip, which generates heat according to the energized voltage when the glow plug 11 is energized. Although not shown, the heat generating portion faces the inside of the combustion chamber 6 and is arranged so as to be located near the tip portion of the injector 10. For example, the heat generating portion of the glow plug 11 is located between the sprays ejected from each injection port of the injector 10 so as not to come into direct contact with the sprays.

An intake passage 20 is connected to one side surface of the engine 1, while an exhaust passage 40 is connected to the other side surface. Here, the intake passage 20 communicates with the intake port 16 of each cylinder 2 and introduces fresh air into each combustion chamber 6. On the other hand, the exhaust passage 40 communicates with the exhaust port 17 of each cylinder 2 and discharges the burned gas (exhaust gas) from each combustion chamber 6. The first turbocharger 51 and the second turbocharger 52 described above are arranged in the intake passage 20 and the exhaust passage 40.

In the intake passage 20, in order from the upstream side, the air cleaner 21, the compressor 51a of the first turbocharger 51 (hereinafter, appropriately referred to as the first compressor 51a), and the compressor 52a of the second turbocharger 52 (hereinafter, appropriately referred to as appropriate). , The second compressor 52a), the intercooler 22, the intake shutter valve 23, and the surge tank 24 are provided. The intake shutter valve 23 is basically in a fully open state, but when the engine 1 is stopped, for example, it is in a fully closed state so as not to cause a shock.

The intake passage 20 is also provided with an intake bypass passage 25 that bypasses the second compressor 52a and an intake bypass valve 26 that opens and closes the intake bypass passage 25. The intake bypass valve 26 is switched between a fully closed state and a fully open state.

As shown in FIG. 2, the exhaust passage 40 has a first exhaust passage 40A connected to a part of four cylinders 2 (specifically, two cylinders 2 located at both ends in the cylinder row direction). The second exhaust passage 40B, the first exhaust passage 40A, and the second exhaust passage connected to the other part of the four cylinders 2 (specifically, the two cylinders 2 located in the center in the cylinder row direction). It has a third exhaust passage 40C, which is formed by merging 40B on the downstream side of each.

In this configuration example, the first and second exhaust passages 40A and 40B are provided with separate catalysts 43 and 43 configured to exhibit substantially the same function, respectively. Each of the two catalysts 43 comprises the NOx catalyst 41 described above.

The catalyst 43 includes a NOx catalyst 41 for purifying NOx and an oxidation catalyst (DOC: Diesel Oxidation Catalyst) 42. Here, the oxidation catalyst 42 may be provided integrally with the NOx catalyst 41 or in the exhaust passage 40 on the upstream side of the NOx catalyst 41. In this configuration example, the first catalyst 43 is configured by coating the surface of the catalyst material forming the NOx catalyst 41 with the catalyst material forming the oxidation catalyst 42.

The NOx catalyst 41 stores NOx in the exhaust in a lean state where the air-fuel ratio of the exhaust is larger than the stoichiometric air-fuel ratio (state in which the excess air ratio λ is λ> 1), and the stored NOx is used as the air-fuel ratio of the exhaust. Is near the stoichiometric air-fuel ratio (λ≈1) or richer than the stoichiometric air-fuel ratio (λ <1), that is, under a reducing atmosphere in which the exhaust passing through the NOx catalyst 41 contains a large amount of unburned HC. It is a NOx storage-reducing catalyst (NSC: NOx Storage Catalyst) that reduces in.

In this configuration example, the air-fuel ratio of the exhaust gas corresponds to the air-fuel ratio of the air-fuel mixture in the combustion chamber 6. That is, when one is rich, the other is rich, when one is near the stoichiometric air-fuel ratio, the other is near the stoichiometric air-fuel ratio, and when one is lean, the other is lean. Here, the term "near the theoretical air-fuel ratio" means that the air-fuel ratio of the exhaust gas or the air-fuel mixture is within the range of 14 to 16.

Specifically, the NOx catalyst 41 according to this configuration example carries platinum as an oxidation-reducing agent and barium as an occlusion agent, and when the air-fuel ratio of the exhaust is lean, NOx in the exhaust is carried. Is oxidized by platinum and occluded in barium. On the other hand, when the air-fuel ratio of the exhaust is richer than this, NOx is released from barium and reduced by using HC or the like as a reducing agent.

Further, in this NOx catalyst 41, when the air-fuel ratio of the exhaust is in a lean state, sulfur oxides (SOx) in the exhaust are also oxidized by platinum and occluded in barium. SOx occluded in the NOx catalyst 41 is released and reduced in a state where the air-fuel ratio of the exhaust gas is rich, similar to NOx. SOx is an example of a "sulfur component".

Further, the NOx catalyst 41 is adapted to generate and release ammonia (NH3) when reducing the stored NOx. Specifically, at the time of NOx reduction, nitrogen in NOx stored in the NOx catalyst 41, NOx passing through the NOx catalyst 41, hydrogen, which is a reducing agent introduced into the NOx catalyst 41, and the like are combined. , NH3 is generated.

The oxidation catalyst 42 has a function of adsorbing a hydrocarbon (HC), and is configured to purify the HC adsorbed by the function. Specifically, the oxidation catalyst 42 uses oxygen in the exhaust gas to oxidize HC, that is, unburned fuel, carbon monoxide (CO), and the like to change them into water and carbon dioxide. Here, this oxidation reaction that occurs in the oxidation catalyst 42 is an exothermic reaction, and when the oxidation reaction occurs in the oxidation catalyst 42, the temperature of the exhaust is increased.

As described above, the oxidation catalyst 42 temporarily adsorbs HC when the oxidation catalyst 42 is not activated and the HC cannot be sufficiently purified, such as during a cold start, and after the oxidation catalyst 42 is activated. It has a function to release and purify the adsorbed HC.

In the following description, the catalyst 43 and the NOx catalyst 41 provided in the first exhaust passage 40A are referred to as "first catalyst 43A" and "first NOx catalyst 41A", respectively. Similarly, in the following description, the catalyst 43 and the NOx catalyst 41 provided in the second exhaust passage 40B are referred to as "second catalyst 43B" and "second NOx catalyst 41B", respectively. The same designation may be used for the oxidation catalyst 42.

On the other hand, in the third exhaust passage 40C, in order from the upstream side, the turbine 52b of the second turbo supercharger 52 (hereinafter, appropriately referred to as the second turbine 52b) and the turbine 51b of the first turbo supercharger 51 (hereinafter, hereinafter). , As appropriate, the first turbine 51b), the DPF (Diesel Particulate Filter) 44, the urea injector 45 that injects urea into the exhaust passage 40 on the downstream side of the DPF 44, and the urea injected from the urea injector 45. An SCR (Selective Catalytic Reduction) catalyst 46 that purifies NOx using the SCR catalyst 46 and a slip catalyst 47 that oxidizes and purifies unreacted ammonia discharged from the SCR catalyst 46 are provided.

Here, the DPF 44 is located downstream of the two catalysts 43 in the exhaust passage 40, and is configured as a particle filter for collecting particulate matter (PM: Particulate Matter) in the exhaust gas.

Specifically, in the DPF 44, a large number of passages are formed in a grid pattern with porous ceramics or the like. As shown in FIG. 3, these many passages include a passage 44a in which the exhaust upstream side is open and the exhaust downstream side is closed, and a passage 44b in which the exhaust upstream side is closed and the exhaust downstream side is open. These two types of passages 44a and 44b are alternately arranged in a staggered manner. The exhaust gas that has flowed into the passage 44a passes through the partition wall 44c that separates the passage 44a and the passage 44b, and escapes to the passage 44b. The partition wall 44c functions as a filter for preventing PM from coming out from the passage 44a to the passage 44b. That is, PM is collected by this partition wall 44c.

Further, the DPF 44 according to this configuration example also has an oxidizing function in addition to the PM collecting function. Specifically, the partition wall 44c of the DPF 44 is coated with the oxidation catalyst layer 44d. Hydrocarbons (HC), that is, unburned fuel, carbon monoxide (CO), and the like are adsorbed on the oxidation catalyst layer 44d, and these are oxidized to water and carbon dioxide by oxygen in the exhaust. It has become like. This oxidation reaction that occurs in the oxidation catalyst layer 44d is an exothermic reaction, and when the oxidation reaction occurs in the oxidation catalyst layer 44d, the temperature of the exhaust is increased.

The PM collected by the DPF44 burns when exposed to high temperatures and is supplied with oxygen and is removed from the DPF44. The temperature at which PM is burned and removed is about 650 ° C, which is a relatively high temperature. Therefore, in order to burn PM and remove it from DPF44, it is necessary to raise the temperature of DPF44 to a relatively high temperature. Generally, the temperature required to remove PM from the DPF 44 (about 650 ° C.) is higher than the temperature required to remove SOx from the NOx catalyst 41 (about 600 ° C.).

Further, the urea injector 45 injects urea into the exhaust passage 40 on the downstream side of the DPF 44. The urea injected from the urea injector 45 is introduced into the SCR catalyst 46.

The SCR catalyst 46 hydrolyzes the urea injected from the urea injector 45 to generate NH3, and reacts (reduces) this NH3 with NOx in the exhaust to purify it. As described above, in this configuration example, the urea injected from the urea injector 45 functions as a reducing agent for the SCR catalyst 46.

Specifically, in the SCR catalyst 46, the introduced NH3 is adsorbed, and NOx is reduced by the reaction between the adsorbed NH3 and NOx. Further, as described above, when NOx is reduced in the NOx catalyst 41, NH3 is also released from the NOx catalyst 41, and the SCR catalyst 46 releases NH3 released from the NOx catalyst in the exhaust gas. NOx is also purified by reducing it with NOx.

Both the SCR catalyst 46 and the NOx catalyst 41 can purify NOx, but they have different temperatures at which the NOx purification rate (NOx storage rate) becomes high. The NOx purification rate (NOx storage rate) of the SCR catalyst 46 becomes high when the temperature of the exhaust gas is relatively high, and the NOx purification rate (NOx storage rate) of the NOx catalyst 41 is such that the temperature of the exhaust gas is relatively low. It becomes high at the time.

In this configuration example, NOx is purified using both the NOx catalyst 41 and the SCR catalyst 46. Specifically, when the temperature of the SCR catalyst 46 is lower than the first temperature and the NOx purification rate by the SCR catalyst 46 is low, NOx purification is performed only by the NOx catalyst 41, and the temperature of the SCR catalyst 46 is equal to or higher than the second temperature. (The second temperature is higher than the first temperature), and when the NOx purification rate by the SCR catalyst 46 is high, NOx purification is performed only by the SCR catalyst 46. Then, when the temperature of the SCR catalyst 46 is between the first temperature and the second temperature, NOx purification is performed by both the NOx catalyst 41 and the SCR catalyst 46. Further, even when the exhaust gas flow rate is large and the NOx purification rate by the SCR catalyst 46 is low, NOx purification is performed by both the NOx catalyst 41 and the SCR catalyst 46.

The exhaust passage 40 (specifically, the third exhaust passage 40C) also has an exhaust bypass passage 48 that bypasses the second turbine 52b, an exhaust bypass valve 49 that opens and closes the exhaust bypass passage 49, and a waste that bypasses the first turbine 51b. A gate passage 53 and a wastegate valve 54 for opening and closing the gate passage 53 are provided. The exhaust bypass valve 49 and the wastegate valve 54 are switched between a fully closed state and a fully open state, respectively, and are changed to an arbitrary opening degree between these states.

The engine 1 further includes an EGR device 55 that recirculates a portion of the exhaust gas to the intake air. The EGR device 55 includes a portion of the exhaust passage 40 (specifically, the third exhaust passage 40C) upstream of the upstream end of the exhaust bypass passage 48, and the intake shutter valve 23 and the surge tank 24 of the intake passage 20. It has an EGR passage 56 that connects the portion between them, a first EGR valve 57 that opens and closes the EGR passage 56, and an EGR cooler 58 that cools the exhaust gas passing through the EGR passage 56. Further, the EGR device 55 has an EGR cooler bypass passage 59 that bypasses the EGR cooler 58, and a second EGR valve 60 that opens and closes the EGR cooler bypass passage 59.

Next, the control system of the engine 1 will be described in detail.

(2) Control system The exhaust gas purification control device for an engine includes an ECU (Engine Control Unit) 100 for operating the engine 1. The ECU 100 is a controller based on a well-known microcomputer. The ECU 100 is composed of a central processing unit (CPU) for executing a program, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory) for storing the program and data, and an electric signal. It is equipped with an input / output bus for input / output. The ECU 100 is an example of a controller.

As shown in FIG. 3, various sensors SW1 to SW10 are connected to the ECU 100. The sensors SW1 to SW10 output the detection signal to the ECU 100. Such sensors include:

That is, the water temperature sensor SW1 attached to the engine 1 and detecting the temperature of the cooling water, and the airflow sensor SW2 for detecting the flow rate of fresh air flowing through the intake passage 20 arranged downstream of the air cleaner 21 in the intake passage 20. In addition, the intake air temperature sensor SW3 that detects the temperature of fresh air, the crank angle sensor SW4 that is attached to the engine 1 and detects the rotation angle of the crankshaft 7, and the accelerator pedal mechanism (not shown) are attached to the accelerator pedal. Accelerator opening sensor SW5 that detects the accelerator opening corresponding to the amount of operation, O2 sensor SW6 that is provided in the exhaust passage 40 and detects the oxygen concentration in the exhaust gas, and vehicle speed sensor SW7 that detects the speed (vehicle speed) of the vehicle. , The boost pressure sensor SW8 attached to the surge tank 24 and detecting the pressure of the air introduced into the combustion chamber 6, and the SCR between the DPF 44 and the SCR catalyst 46 in the exhaust passage 40 and in the same passage. The NOx sensor SW9 is provided between the catalyst 46 and the slip catalyst 47, respectively, and detects the NOx concentration in the exhaust gas. Further, an exhaust temperature sensor SW10 for detecting the temperature of the exhaust gas flowing through the exhaust passage 40 is attached to the exhaust passage 40.

The ECU 100 determines the operating state of the engine 1 and the vehicle based on these detection signals, and calculates the control amount of each device. The ECU 100 inputs the control signal related to the calculated control amount to the injector 10, the glow plug 11, the intake shutter valve 23, the intake bypass valve 26, the exhaust bypass valve 49, the wastegate valve 54, the first EGR valve 57, and the second EGR. Output to valve 60.

For example, the ECU 100 acquires the actual boost pressure (hereinafter, also referred to as “actual boost pressure”) at the time of detection based on the detection signal by the boost pressure sensor SW8. At the same time, the ECU 100 calculates a target value of the boost pressure (hereinafter, also referred to as “target boost pressure”) based on the detection signals from other sensors. Then, the ECU 100 adjusts the opening degree of the intake bypass valve 26, the exhaust bypass valve 49, the wastegate valve 54, etc. so that the actual boost pressure becomes the target boost pressure.

Then, the ECU 100 controls the operation of the first turbocharger 51 and the second turbocharger 52 by adjusting the opening degree of the intake bypass valve 26, the exhaust bypass valve 49, the wastegate valve 54, and the like.

In a general engine, in order to purify and regenerate various devices provided in the exhaust passage 40 such as the first catalyst 43 and the DPF 44, DeNOx control for removing NOx from the NOx catalyst 41 and SOx from the NOx catalyst 41 It is conceivable to execute DeSOx control (NOx catalyst regeneration control) for removing PM and DPF regeneration control for removing PM from DPF44 one by one in order.

On the other hand, in this engine 1, both are started at the same time in order to efficiently complete the DeSOx control and the DPF regeneration control. Therefore, in the following description, the DeSOx control and the DPF regeneration control may be collectively referred to as "DeSOx + DPF regeneration control".

Hereinafter, the basic contents of the fuel injection control will be described, and then the details of the DeNOx control, the DeSOx control, the DPF regeneration control, and the like will be described in order.

-Fuel injection control-
First, the fuel injection control in this configuration example will be described. This fuel injection control is started when the ignition of the vehicle is turned on and the power is turned on to the ECU 100, and is repeatedly executed at a predetermined cycle.

First, the ECU 100 determines the driving state of the vehicle. Specifically, the ECU 100 is currently set at least in the accelerator opening detected by the accelerator opening sensor SW5, the vehicle speed detected by the vehicle speed sensor SW7, the crank angle detected by the crank angle sensor SW4, and the transmission of the vehicle. Acquire the gear stage.

Next, the ECU 100 determines the target acceleration of the vehicle based on the determined driving state of the vehicle. Specifically, in the memory of the ECU 100 or the like, acceleration characteristic maps defined for various vehicle speeds and various gear stages are stored in advance. The ECU 100 selects an acceleration characteristic map corresponding to the current vehicle speed and gear stage from such an acceleration characteristic map, and determines a target acceleration corresponding to the current accelerator opening by referring to the selected map. ..

Next, the ECU 100 determines the target torque of the engine 1 for realizing the determined target acceleration. In this case, the ECU 100 determines the target torque within the range of the torque that can be output by the engine 1 based on the current vehicle speed, gear stage, road surface gradient, road surface μ, and the like.

Next, the ECU 100 calculates the fuel injection amount to be injected from the injector 10 based on the target torque and the engine rotation speed in order to output the determined target torque from the engine 1. This fuel injection amount is the fuel injection amount (main injection amount) to be injected in the main injection.

On the other hand, in parallel with the above-mentioned processing, the ECU 100 sets the fuel injection pattern according to the operating state of the engine 1. Specifically, when the above-mentioned DeNOx control or the like is performed, the ECU 100 sets a fuel injection pattern such that at least post injection is performed in addition to the main injection. In this case, the ECU 100 also determines the fuel injection amount (post injection amount) to be injected in the post injection, the injection timing (post injection timing), and the like.

After that, the ECU 100 outputs a control signal to the injector 10 so as to realize the calculated main injection amount and the set injection pattern. Here, when performing post-injection, a control signal that realizes the above-mentioned post-injection amount and post-injection timing is output. That is, the ECU 100 controls the injector 10 so that a desired amount of fuel is injected in a desired injection pattern.

(2-1) Normal Control First, the control (normal control) performed during normal steady operation without performing DeNOx control, DeSOx control, DPF regeneration control, etc. will be described.

In this flow control, the air-fuel ratio of the air-fuel mixture in the combustion chamber 6 is set to a leaner state (λ> 1) than the stoichiometric air-fuel ratio in order to improve fuel efficiency. For example, in normal control, the excess air ratio λ of the air-fuel mixture is about λ = 1.7. Further, in this normal control, the post injection is limited and only the main injection is carried out. Further, in normal control, the operation of the glow plug 11 is stopped. Further, in normal control, the first EGR valve 57, the second EGR valve 60, the intake bypass valve 26, the exhaust bypass valve 49, and the wastegate valve 54 are set to the operating state of the engine 1, for example, the engine speed and the engine load, respectively. Accordingly, the EGR rate is controlled to an appropriate value. As described above, the boost pressure is also controlled so as to realize the target boost pressure according to the operating state of the engine 1.

(2-2) DeNOx Control Next, DeNOx control, which is a control for reducing NOx stored in the NOx catalyst 41 (hereinafter, also referred to as “occluded NOx”) and releasing (leaving) it from the NOx catalyst 41, will be described. do.

As described above, in the NOx catalyst 41, the occluded NOx is reduced in a state where the air-fuel ratio of the air-fuel mixture is close to the stoichiometric air-fuel ratio (λ≈1) or in a rich state (λ <1) smaller than the stoichiometric air-fuel ratio. .. Therefore, in order to reduce the occluded NOx, it is necessary to reduce the air-fuel ratio of the air-fuel mixture as compared with the normal operation (when the normal control described later is performed).

Therefore, when the predetermined condition is satisfied (in this configuration example, when the DPF regeneration permission flag is set to "1"), the ECU 100 is after the fuel injection (main injection) corresponding to the operating state of the engine 1. By performing additional injection of fuel, a control signal is output to the injector 10 so as to execute DeNOx control that makes the air-fuel ratio richer than before the predetermined condition is satisfied.

Specifically, in this configuration example, the ECU 100 reduces the occluded NOx by performing post-injection as additional injection to reduce the air-fuel ratio of the air-fuel mixture. That is, the ECU 100 causes the injector 10 to execute the post injection in addition to the main injection. In such DeNOx control, for example, the excess air ratio λ of the air-fuel mixture is set to about λ = 0.94 to 1.06. By doing so, the NOx occluded in the NOx catalyst 41 will be reduced.

Next, DeNOx control in this configuration example will be described.

The ECU 100 performs active DeNOx control (NOx catalyst rich purge control) as DeNOx control when the engine load is in the medium load range (see the second region R12 in FIG. 5), and when the engine load is near the upper limit (NOx catalyst rich purge control). The passive DeNOx control performed in the first region R11 of FIG. 5) can be used properly.

For example, in the ECU 100, as active DeNOx control, when the NOx storage amount in the NOx catalyst 41 (hereinafter, also simply referred to as “NOx storage amount”) is a predetermined amount or more (typically, the NOx storage amount is near the limit). ), The post injection from the injector 10 is continuously executed at the injection timing such that the fuel injected by the post injection burns in the combustion chamber 6 (inside the cylinder 2). By doing so, a large amount of occluded NOx can be forcibly reduced, and eventually the NOx purification performance of the NOx catalyst 41 can be ensured.

The injection timing (post injection timing) of the post injection in the active DeNOx control is set in advance, for example, in the first half of the expansion stroke and between 30 and 70 ° CA after the compression top dead center. .. By doing so, the fuel injected by the post-injection is discharged as unburned fuel (that is, HC) as it is, and the oil dilution by the post-injected fuel is suppressed.

Further, in this configuration example, in order to promote the combustion of the fuel injected by the post injection, the glow plug 11 is energized to heat the air-fuel mixture during the active DeNOx control.

On the other hand, in the ECU 100, as passive DeNOx control, even when the NOx storage amount is less than a predetermined amount, when the air-fuel ratio changes to the rich side due to the acceleration of the vehicle, the fuel injected by the post injection is injected into the combustion chamber 6. Post-injection from the injector 10 is executed at an injection timing that does not burn inside (inside the cylinder 2). In the passive DeNOx control, the post injection is executed by taking advantage of the situation where the main injection amount increases and the air-fuel ratio of the air-fuel mixture decreases, such as when accelerating the vehicle, so the situation such as when not accelerating. The amount of post injection for achieving the target air-fuel ratio is relatively small as compared with the case where post injection is executed by multiplying by. This makes it possible to suppress deterioration of fuel consumption due to post-injection while forcibly reducing the storage NOx.

The injection timing of the post injection in the passive DeNOx control (post injection timing) is set to be at least on the retard side of the injection timing of the post injection in the active DeNOx control. For example, the injection timing of the post injection in the passive DeNOx control can be set in the latter half of the expansion stroke and in the vicinity of 110 ° CA after the compression top dead center. By doing so, it is possible to suppress the generation of smoke (soot) caused by the combustion of the fuel injected by the post injection.

Here, with reference to FIG. 5, the operating region of the engine 1 that executes each of the passive DeNOx control and the active DeNOx control will be described. The horizontal axis of FIG. 5 shows the engine speed, and the vertical axis of FIG. 5 shows the engine load. Further, in FIG. 5, the curve L1 shows the maximum torque line of the engine 1.

In this configuration example, the ECU 100 executes active DeNOx control when the engine load and the engine speed are included in the second region R12 shown in FIG. Here, in the second region R12, the engine load is the first reference load Lo1 or more and less than the second reference load Lo2 (> the first reference load Lo1), and the engine speed is the first reference speed N1 or more. It is an operating area less than the second reference rotation speed (> first reference rotation speed N1).

On the other hand, the ECU 100 executes passive DeNOx control when the engine load and the engine speed are included in the first region R11 shown in FIG. Here, the first region R11 is a region where the engine load is higher than the second region R12, and the engine load is a predetermined third reference load (> second reference load Lo2) or more.

As described above, the content of DeNOx control is used properly in the first region R11 and the second region R12 for the following reasons.

The exhaust temperature is low in the operating region where the engine load is low or the engine load is relatively high but the engine speed is low. Along with this, the temperature of the NOx catalyst 41 tends to be lower than the temperature at which the stored NOx can be reduced. Therefore, in this configuration example, DeNOx control is restricted in such an operating region.

Further, as described above, post-injection is performed in DeNOx control, but when the post-injected fuel is discharged to the exhaust passage 40 as it is without burning, the EGR cooler 58 or the like is generated by the deposit caused by this unburned fuel. There is a risk of blockage. Therefore, it is preferable that the fuel injected by the post injection is burned in the combustion chamber 6. However, in a region where the engine load is high or the engine load is relatively low but the engine speed is high, combustion occurs due to the high temperature in the combustion chamber 6 or the short time per crank angle. It may be difficult to sufficiently mix the fuel injected by the post-injection with the air before the gas in the chamber 6 is discharged. In that case, there is a possibility that the fuel injected by the post injection cannot be sufficiently burned in the combustion chamber 6. Furthermore, soot may increase due to insufficient mixing of fuel and air. Therefore, in such an operating region, DeNOx control is basically stopped.

However, in the first region R11 where the engine load is very high, the air-fuel ratio of the air-fuel mixture can be kept small even during normal operation due to the large amount of main injection. Therefore, in the first region R11, the post injection amount required for reducing the occluded NOx can be reduced, and the influence caused by the unburned fuel being discharged to the exhaust passage 40 can be suppressed to be small.

Therefore, the ECU 100 implements active DeNOx control for burning the post-injected fuel in the combustion chamber 6 in the second region R12 where neither the engine load nor the engine speed is too low and not too high. On the other hand, the ECU 100 implements passive DeNOx control in the first region R11 so that the post-injected fuel is not burned in the combustion chamber 6. The first region R11 is a region where the exhaust temperature is sufficiently high and the oxidation catalyst 42 is sufficiently activated. Therefore, the unburned fuel discharged to the exhaust passage 40 is purified by the oxidation catalyst 42.

Further, among the operating regions in which DeNOx control is to be stopped, SCR is particularly high in the region on the higher load side than the second region R12 or on the high rotation side (region R13 on the lower load side than the first region R11). Since the NOx purification performance by the catalyst 46 is sufficiently ensured, it is possible to reliably prevent the emission of NOx from the vehicle without executing DeNOx control.

In the first region R11 on the higher load side than the region R13 in which the NOx is purified by the SCR catalyst 46, the amount of exhaust gas becomes large, and although the SCR catalyst 46 cannot completely purify the NOx, passive DeNOx control is executed as described above. By doing so, NOx can be purified by the NOx catalyst 41.

Here, the control content when the operating state of the engine 1 changes as shown by the arrow in FIG. 5 will be specifically described. First, when the operating state of the engine 1 enters the second region R12 (see reference numeral A12), the ECU 100 executes active DeNOx control. Then, when the operating state of the engine 1 deviates from the second region R12 (see reference numeral A13), the ECU 100 temporarily stops the active DeNOx control. At this time, the SCR catalyst 46 purifies NOx. Then, when the operating state of the engine 1 reenters the second region R12 (see reference numeral A14), the ECU 100 restarts the active DeNOx control. By doing so, the active DeNOx control is not terminated until the NOx occluded in the NOx catalyst 41 drops to substantially zero.

Next, a temperature range in which passive DeNOx control or active DeNOx control is performed in this configuration example will be described. As described above, the NOx catalyst 41 exhibits NOx purification performance in a relatively low temperature range, while the SCR catalyst 46 exhibits a NOx purification performance in a relatively high temperature range, specifically, the NOx catalyst 41 exhibits NOx purification performance. Demonstrates NOx purification performance in a temperature range higher than the range.

In this configuration example, even when the operating state of the engine 1 is in the first region R1, when the temperature of the SCR catalyst 46 is raised to a temperature at which NOx can be purified, the SCR catalyst 46 is used. Since NOx can be purified, active DeNOx control is not executed. By doing so, it is possible to suppress the deterioration of fuel efficiency caused by the execution of DeNOx control.

Specifically, the ECU 100 allows execution of active DeNOx control only when the temperature of the SCR catalyst 46 (hereinafter referred to as “SCR temperature”) is lower than a predetermined SCR determination temperature, and the SCR temperature is SCR. When the temperature is above the determination temperature, the execution of active DeNOx control is prohibited. Here, the "SCR determination temperature" refers to a determination value set near the lower limit of the temperature range in which NOx can be purified (the temperature range of the SCR catalyst 46). The SCR determination temperature is equal to the above-mentioned second temperature.

(2-3) DeSOx + DPF regeneration control DeSOx control is a control for reducing and removing SOx (an example of a sulfur component, hereinafter appropriately referred to as occluded SOx) stored in the NOx catalyst 41.

As described above, the occluded SOx is reduced in a state where the air-fuel ratio is close to the stoichiometric air-fuel ratio (λ≈1) or in a rich state (λ <1) smaller than the stoichiometric air-fuel ratio. Along with this, even with DeSOx control, in addition to the main injection, in order to make the air-fuel ratio of the air-fuel mixture close to the stoichiometric air-fuel ratio (λ≈1) or richer than the stoichiometric air-fuel ratio (λ <1). Perform post injection.

However, since SOx has a stronger binding force than NOx, in order to reduce the occluded SOx, the temperature of the NOx catalyst 41 and the temperature of the exhaust gas passing through the NOx catalyst 41 are set to a higher temperature (about 600 ° C.) than during DeNOx control. There is a need to. On the other hand, as described above, if the unburned fuel is oxidized in the oxidation catalyst 42, the temperature of the exhaust gas passing through the first catalyst 43 and the NOx catalyst 41 can be increased.

Therefore, in this configuration example, as DeSOx control, post-injection is performed in the same manner as in DeNOx control to make the air-fuel ratio of the exhaust richer than in normal operation and to be near or smaller than the theoretical air-fuel ratio (hereinafter, appropriately, as appropriate). The rich step (simply making it rich) and the post-injection while making the air-fuel ratio of the exhaust leaner than the theoretical air-fuel ratio (hereinafter, as appropriate, simply making it lean) are performed to make the oxidation catalyst 42 air and unburned fuel. A control signal is output to the injector 10 of each of the four injectors 10 so as to alternately carry out a lean step of supplying the fuel and oxidizing them with the oxidation catalyst 42.

The ECU 100 is configured to remove sulfur components from the first and second NOx catalysts 41A and 41B as NOx catalysts 41 by executing such DeSOx control over a plurality of cycles.

On the other hand, the DPF regeneration control is a control for regenerating the purification capacity of the DPF 44 by removing the PM collected in the DPF 44. As described above, the PM collected in the DPF 44 can be burned off at a high temperature. In order to burn PM, it is required to supply oxygen to DPF44.

On the other hand, if HC or the like, that is, unburned fuel is oxidized in the oxidation catalyst 42 contained in the first catalyst 43 provided on the upstream side of the DPF 44, the temperature of the exhaust flowing into the DPF 44 and thus the temperature of the DPF 44 are increased. be able to. Furthermore, the temperature of the DPF 44 can be raised by allowing the unburned fuel to undergo an oxidation reaction also in the oxidation catalyst layer 44d of the DPF 44. In order to promote such an oxidation reaction, it is required to supply oxygen to DPF44.

Here, focusing on the lean step that forms the above-mentioned DeSOx control, in this lean step, a larger amount of air than the stoichiometric is supplied to the exhaust passage 40. Since such air passes through the NOx catalyst 41 and reaches DPF44, it seems that this lean step can be used to regenerate DPF44.

However, in DeSOx control, rich steps and lean steps are alternately performed. Therefore, although a large amount of oxygen is secured in the lean step, there is a possibility that oxygen will be insufficient in the rich step. This is inconvenient for purifying DPF44. Therefore, it is conceivable to lengthen the time for carrying out the lean step with respect to the time for carrying out the rich step, but in order to efficiently purify the NOx catalyst, not only the lean step but also the rich step is carried out. It is also required.

The inventors of the present application do not use a common NOx catalyst for all cylinders 2, but as shown in FIG. 2, if the first NOx catalyst 41A and the second NOx catalyst 41B are used, the rich step and the lean step are alternately performed. While doing so, it was found that oxygen to be supplied to DPF44 could be secured.

That is, in this configuration example, the DeSOx control also serves as the DPF regeneration control. In that sense, as described above, the control shown below may be referred to as "DeSOx + DPF reproduction control".

Specifically, when the ECU 100 performs the rich step in the cylinder 2 leading to the first exhaust passage 40A (hereinafter, also referred to as the first cylinder group 2A) when performing DeSOx control, the ECU 100 goes to the second exhaust passage 40B. A control signal is output to each injector 10 so that the lean step is performed in the communicating cylinder 2 (hereinafter, also referred to as the second cylinder group 2B).

On the other hand, when the ECU 100 performs the DeSOx control, when the lean step is performed in the first cylinder group 2A, the ECU 100 sends a control signal to each injector 10 so as to perform the rich step in the second cylinder group 2B. Output.

In this way, by performing the lean step and the rich step alternately and simultaneously, SOx is removed from the first NOx catalyst 41A and the second NOx catalyst 41B, and oxygen is sufficiently supplied to the DPF 44 arranged downstream thereof. , DPF44 can be regenerated with the oxygen.

Here, when the rich step is performed, the ECU 100 executes additional fuel injection (post injection) as in the active DeNOx control. Specifically, the ECU 100 has an injection timing (the first half of the expansion stroke, for example, compression top dead center) such that the post-injected fuel burns in the cylinder 2 after the fuel injection corresponding to the operating state of the engine 1. A control signal is output to the injector 10 so that the post injection is executed at 30 to 70 ° CA) after the point. Then, in the rich step, the air-fuel ratio of the air-fuel mixture and the exhaust is set to be close to the stoichiometric air-fuel ratio by setting the excess air ratio λ of the air-fuel mixture and the exhaust to about 1.0. For example, in the rich step, the excess air ratio λ of the air-fuel mixture and the exhaust gas is set to about λ = 0.94 to 1.06. As a result, a reducing agent such as CO can be supplied to the NOx catalyst 41, and SOx adsorbed on the NOx catalyst 41 can be reduced and removed.

Further, in the rich step, as in the case of active DeNOx control, the first EGR valve 57 is fully closed while the second EGR valve is fully closed in order to improve the stability of this combustion while suppressing the soot associated with the combustion of the post-injected fuel. The valve 60 is opened and the opening degree of the second EGR valve 60 is made smaller than that during normal operation. Further, the ECU 100 controls the intake shutter valve 23, the exhaust bypass valve 49, and the wastegate valve 54 so that the air-fuel ratio of the air-fuel mixture is low, so that the intake air amount is smaller than that in the normal operation.

On the other hand, when the lean step is performed, the ECU 100 has an injection timing (expansion) such that the post-injected fuel does not burn in the cylinder 2 after the fuel injection corresponding to the operating state of the engine 1 as in the passive DeNOx control. In the latter half of the stroke, for example, a control signal is output to the injector 10 so that the post injection is executed at 110 ° CA) after the compression top dead center. Then, in the lean step, the air-fuel ratio of the air-fuel mixture is made leaner than the stoichiometric air-fuel ratio by setting the air excess ratio λ of the air-fuel mixture to 1 or more. For example, in the lean step, the excess air ratio λ of the air-fuel mixture is set to about λ = 1.2 to 1.4.

Further, in the lean step, the first EGR valve 57 and the second EGR valve 60 are fully closed in order to prevent the EGR cooler and the like from being blocked by the deposit caused by the unburned fuel.

As described above, in DeSOx control, it is necessary to make the air-fuel ratio of the air-fuel mixture in the combustion chamber 6 lean. Therefore, it is difficult to perform DeSOx control in the first region R11 where the engine load is high and the air-fuel ratio of the air-fuel mixture cannot be sufficiently lean. On the other hand, as described above, the control for burning the post injection is preferably performed in the second region R12. Therefore, in this configuration example, DeSOx control is performed only when the engine 1 is being operated in the second region R12.

As described above, if the air-fuel ratio of the air-fuel mixture in the combustion chamber 6 is made leaner than the theoretical air-fuel ratio and the post injection is performed without burning the fuel, PM can be burned and removed. Therefore, the lean step. It is possible to remove the combustion of PM at the time of implementation.

-Details of rich step and lean step-
Hereinafter, a method for determining the control amount in the rich step and the lean step will be described.

As described above, the ECU 100 determines the main injection amount based on the required torque of the engine 1. On the other hand, the post-injection amount in the rich step is determined based on the adsorption amount of NH3 in the SCR catalyst 46 (hereinafter, also simply referred to as the NH3 adsorption amount). Although details are omitted, the NH3 adsorption amount is estimated by the ECU 100 based on the detection signal of the NOx sensor SW9.

Specifically, when the NH3 adsorption amount is large, the ECU 100 outputs a control signal to the injector 10 so as to reduce the post injection amount in the rich step as compared with the case when the NH3 adsorption amount is small (see the figure below in FIG. 6). reference).

More specifically, the ECU 100 monotonically reduces the post injection amount as the rich step is repeated a plurality of times, and when the number of repetitions (DeSOx cycle number) exceeds a predetermined number, the post is posted as shown by the broken line in FIG. The injection amount is made constant.

That is, the amount of the sulfur component desorbed from the NOx catalyst 41 by the DeSOx control can be expected to be larger as the amount of the sulfur component adsorbed on the NOx catalyst 41 increases. Therefore, as the rich step is repeated, the sulfur component can be sufficiently desorbed even if the post injection amount is reduced. Fuel efficiency can be improved by reducing the amount of post injection.

In this configuration example, the post injection timing in the rich step is also determined based on the NH3 adsorption amount as well as the post injection amount.

Specifically, when the NH3 adsorption amount is large, the ECU 100 outputs a control signal to the injector 10 so as to retard the post injection timing more than when the NH3 adsorption amount is small (see the upper figure of FIG. 6). ).

More specifically, the ECU 100 gradually retards the post injection timing as the rich step is repeated a plurality of times, and when the number of repetitions (DeSOx cycle number) exceeds a predetermined number, as shown by the broken line in FIG. The post injection timing is made constant.

Although the details will be described later, by adopting such a configuration, the combustion temperature caused by the fuel injected by the post injection can be lowered, and thereby the generation of NH3 in the NOx catalyst 41 can be suppressed.

Further, the rich step and the lean step are not replaced each time the combustion stroke is performed once in each cylinder 2, but are switched after the rich step or the lean step is performed for a predetermined time, respectively. Hereinafter, the execution time per step when the rich step is continuously performed is referred to as a rich duration. Similarly, the execution time per step when the lean step is continuously performed is referred to as a lean duration.

The basic rich time, which is the target value of the rich duration, and the basic lean time, which is the target value of the lean duration, are determined based on the sulfur poisoning amount (hereinafter, also referred to as S poisoning amount) in the NOx catalyst 41, respectively. It is supposed to be done.

Specifically, the ECU 100 receives S in the first NOx catalyst 41A based on the fuel injection amount (total injection amount obtained by adding the main injection amount and the post injection amount) in each of the first and second NOx catalysts 41A and 41B. The poisonous amount (hereinafter, also referred to as the first poisoned amount) and the S poisoned amount in the second NOx catalyst 41B (hereinafter, also referred to as the second poisoned amount) are calculated. Subsequently, the ECU 100 determines the magnitude relationship between the calculated first poisoning amount and the second poisoning amount.

Then, when the ECU 100 performs DeSOx control, when the first poisoning amount is larger than the second poisoning amount, the time for performing the rich step in the first cylinder group 2A is longer than when it is small. (That is, set a longer basic rich time).

More specifically, the ECU 100 has a ratio (corresponding to the vertical axis of FIG. 7) obtained by dividing the basic lean time by the basic rich time as the difference obtained by subtracting the second poisoned amount from the first poisoned amount increases. ) Is increased. Then, when the difference becomes equal to or more than a predetermined value, the ECU 100 keeps this ratio constant at the predetermined value as shown in FIG. 7. This predetermined value is selected so that the excess air ratio for all the cylinders 2 is 1.2 to 1.3.

The difference formed by subtracting the second poisoning amount from the first poisoning amount is calculated from, for example, the difference between the fuel injection amount in the first cylinder group 11A and the fuel injection amount in the second cylinder group 11B. You can also do it.

Although the details will be described later, by adopting such a configuration, it is possible to secure the reproduction efficiency of the DPF 44 while increasing the efficiency of the DeSOx control.

(2-4) Specific contents of the control process Next, the specific contents of the active DeNOx control and the DeSOx + DPF reproduction control will be described with reference to the flowchart of FIG. The flow shown in FIG. 8 is repeatedly executed by the ECU 100 at a predetermined cycle, and is also executed in parallel with the above-mentioned fuel injection control and the like.

In step S1, the ECU 100 determines whether or not the DPF regeneration permission flag is 1. If this determination is NO, the process proceeds to step S20, the ECU 100 performs normal control, and then ends the process (returns to step S1). On the other hand, if the determination in step S1 is YES, the process proceeds to step S2. The DPF regeneration permission flag is a flag that becomes 1 when the regeneration of the DPF 44 is permitted and becomes 0 when the regeneration of the DPF is prohibited. In the present embodiment, the DPF regeneration permission flag is set to 1 when the PM accumulation amount of the DPF44 is equal to or greater than the regeneration start accumulation amount, and is set to 0 when the PM accumulation amount is equal to or less than the regeneration end accumulation amount. The regeneration end deposit amount is a PM deposit amount reduced to such an extent that it is no longer necessary to remove PM from the DPF44, and is set to a value near 0, for example.

First, in step S1, the ECU 100 determines whether or not the value of the DPF regeneration permission flag is “1”. If the value of the flag is "0" (step S1: NO), the control process proceeds to step S20. In this case, the ECU 100 ends the process (returns to step S1) after performing the above-mentioned normal control. On the other hand, if the value of the DPF regeneration permission flag is "1" (step S1: YES), the control process proceeds to step S2.

The DPF regeneration permission flag is a flag that becomes "1" when the regeneration of the DPF 44 is permitted and becomes "0" when the reproduction of the DPF 44 is prohibited. In this configuration example, the DPF regeneration permission flag is set to "1" when the PM accumulation amount of the DPF44 is equal to or more than the predetermined regeneration start amount, and is set to "0" when the PM accumulation amount is equal to or less than the regeneration end amount.

Here, the PM deposit amount is, for example, the front-rear differential pressure of the DPF 44 calculated from the pressure sensors provided on the upstream side and the downstream side of the DPF 44 (the difference between the pressure on the upstream side and the pressure on the downstream side of the DPF 44). It is calculated from the above. Further, the regeneration start amount is set to a value smaller than the PM accumulation amount that can be collected by the DPF 44. On the other hand, the regeneration end amount is a PM accumulation amount reduced to such an extent that it is no longer necessary to remove PM from the DPF 44, and is set to a value near 0, for example.

In the following step S2, the ECU 100 determines whether or not the operating state of the engine 1 is in the second region R12. If this determination is NO, the control process proceeds to step S20 to perform normal control, and then ends the process (returns to step S1), as in the case of prohibiting the reproduction of the DPF 44. On the other hand, if the determination in step S2 is YES, the process proceeds to step S3.

In the following step S3, the ECU 100 implements active DeNOx control.

In step S4 following step S3, the ECU 100 determines whether or not the storage NOx amount is equal to or less than the preset DeNOx end determination value, that is, until the storage NOx amount is equal to or less than the DeNOx end determination value with the implementation of the active DeNOx control. Determine if it has decreased. If this determination is NO, the process returns to step S2. On the other hand, if this determination is YES, the process proceeds to step S5. That is, the ECU 100 continues the active DeNOx control until the storage NOx amount becomes equal to or less than the DeNOx end determination value and the determination in step S4 becomes YES.

Here, the amount of stored NOx indicates the amount (flow rate) of RawNOx discharged from the engine 1, and is estimated from the exhaust flow rate, the excess air ratio λ of the air-fuel mixture, and the like. Further, the DeNOx end determination value is set to a value near 0, for example.

During the DeNOx control, the ECU 100 estimates the amount of NH3 released from the NOx catalyst 41, and controls the urea injector 45 so that the larger the estimated amount, the smaller the urea injection amount.

Then, when the determination in step S4 is YES, the ECU 100 proceeds to step S5. In step S5, the ECU 100 stops the active DeNOx control and starts the DeSOx + DPF reproduction control. Specifically, the ECU 100 starts the rich step in the first cylinder group 2A, while starting the lean step in the second cylinder group 2B. The lean step may be started in the first cylinder group 2A, while the rich step may be started in the second cylinder group 2B. In this case, the following description may be appropriately read as the first cylinder group 2A and the second cylinder group 2B alternately.

In step S6 following step S5, the ECU 100 determines whether or not the operating state of the engine 1 is in the second region R12. If this determination is NO, the process proceeds to step S20, the ECU 100 performs normal control, and then ends the process (returns to step S1).

On the other hand, if the determination in step S6 is YES, the control process proceeds to step S7. That is, as soon as the operating state of the engine 1 deviates from the second region R12, the DeSOx + DPF regeneration control is stopped and the normal control is returned. In this case, by sequentially processing steps S1 and the like, the DeSOx + DPF regeneration control is restarted as soon as the operating state of the engine 1 is included in the second region R12.

Then, in step S7, the ECU 100 determines whether or not the rich duration in the first cylinder group 2A is equal to or longer than the above-mentioned basic rich time. If this determination is NO, the process proceeds to step S8 to maintain the current step, while if this determination is YES, the process proceeds to step S9 to execute step switching.

Specifically, in step S8, the ECU 100 continuously performs the rich step in one of the first cylinder group 2A and the second cylinder group 2B, and continuously performs the lean step in the other. For example, if the rich step is started in the first cylinder group 2A and the lean step is started in the second cylinder group 2B, the process proceeds to step S8 without performing the process shown in step S9 even once (that is,). (When the number of repetitions is "1"), the rich step is continued in the first cylinder group 2A, while the lean step is continued in the second cylinder group 2B.

On the other hand, in step S9, the ECU 100 switches from the rich step to the lean step in one of the first cylinder group 2A and the second cylinder group 2B, and switches from the lean step to the rich step in the other.

In step S8 following step S8 or step S9, the amount of adsorbed SOx adsorbed on the first NOx catalyst 41A and the amount of adsorbed SOx adsorbed on the second NOx catalyst 41B are both equal to or less than the predetermined DeSOx end determination value. Determine if it has become. If this determination is NO, the process returns to step S6. On the other hand, if this determination is YES, the process proceeds to step S11 and the DeSOx + DPF reproduction control is terminated. That is, the ECU 100 continues the DeSOx + DPF regeneration control until the amount of adsorption SOx becomes equal to or less than the DeSOx end determination value and the determination in step S10 becomes YES.

Here, the adsorbed SOx amount is calculated by subtracting the amount of SOx desorbed by DeSOx control (SOx desorption amount) from the cumulative value of the first detoxification amount (second detoxification amount) described above. .. The SOx detachment amount is calculated based on the exhaust gas temperature detected by the exhaust temperature sensor SW10 and the air-fuel ratio. It is also possible to estimate the temperature of the NOx catalyst 41 using various state quantities and use this estimated value instead of the exhaust gas temperature. On the other hand, the DeSO end value is set to a value near 0, for example.

After finishing the DeSOx + DPF reproduction control in step S11, the control process proceeds to step S12. In this step S12, the ECU 100 starts a control specialized for regeneration of the DPF 44 (hereinafter, also simply referred to as “DPF regeneration control”).

In this configuration example, the ECU 100 performs post-injection while making the air-fuel ratio of the air-fuel mixture leaner than the stoichiometric air-fuel ratio as DPF regeneration control, and causes air and unburned fuel to flow into the oxidation catalyst 42 to oxidize them. Control to oxidize with the catalyst 42 is carried out. Specifically, in the DPF regeneration control, the post-injection is performed at the timing when the post-injected fuel does not burn in the combustion chamber 6 (in the latter half of the expansion stroke, for example, 110 ° CA after the compression top dead center). .. Unlike the DeSOx + DPF regeneration control, in this DPF control, a lean air-fuel mixture is burned in all the cylinders 2. Further, in the DPF regeneration control, since it is not necessary to burn the post injection, the energization to the glow plug 11 is stopped.

In step S13 following step S12, the ECU 100 determines whether or not the PM accumulation amount of the DPF 44 is equal to or less than the above-mentioned regeneration end amount. If this determination is NO, the process returns to step S12. On the other hand, if this determination is YES, the process proceeds to step S20, normal control is performed, and a return is performed (returning to step S1). That is, the ECU 100 continues the DPF regeneration control until the PM accumulation amount becomes equal to or less than the regeneration end amount and the determination in step S13 becomes YES.

(2-5) Control Example Next, a specific example of control by the ECU 100 (particularly, DeSOx + DPF reproduction control) will be described with reference to FIG. FIG. 9 is a diagram illustrating changes in various parameters when DeSOx + DPF reproduction control is executed in this configuration example.

In each figure of FIG. 9, the solid line shows the configuration example of the present disclosure, while the broken line shows the comparative example. In the comparative example, the conventional DeSOx control is implemented. That is, in the comparative example, the lean step and the rich step are alternately performed as in the configuration example of the present disclosure, but unlike the configuration example of the present disclosure, in the first cylinder group 2A. When the rich step is carried out, the rich step is also carried out in the second cylinder group 2B. The same applies to lean steps.

Further, in the comparative example, when the conventional DeSOx control is completed, the DPF regeneration control similar to the process shown in step S12 of FIG. 8 is started.

Further, in FIG. 8, “deSOx_Flag” indicates whether to perform a rich step or a lean step in the corresponding cylinder group. Specifically, when the value of this flag is "1", the rich step is executed, and when the value of the flag is "0", the lean step is executed.

In FIG. 9, when the DPF regeneration permission flag changes from 0 to 1 at time t1, active DeNOx control is performed. Specifically, the air-fuel ratio of the exhaust is made richer than the stoichiometric air-fuel ratio, and post-injection is performed. At this time, the injection timing of the post injection is relatively advanced (first half of the expansion stroke) so that the post-injected fuel burns in each combustion chamber 6 (cylinder 2). Further, the first EGR valve 57 is fully closed, and the opening degree of the second EGR valve 60 is smaller than the opening degree during normal operation, that is, immediately before time t1 (to the closed side), but to the open side rather than the fully closed side. Will be done.

With the implementation of active DeNOx control, the NOx storage amount gradually decreases after time t1 (not shown). Further, as the post-injected fuel burns in the combustion chamber 6 and the temperature of the exhaust gas increases, the temperature of the DPF 44 gradually increases after the time t1. Specifically, after time t1, both the temperature before DPF44 and the temperature inside DPF44 increase. In FIG. 9, the “temperature before the DPF” indicates the exhaust temperature in the third exhaust passage 40C on the immediate upstream side of the DPF 44. On the other hand, the "temperature inside the DPF" indicates the temperature inside the DPF 44.

Although not shown, the temperature of the oxidation catalyst 42 gradually increases after time t1.

When the amount of occluded NOx becomes equal to or less than the DeNOx end determination value at time t2, the active DeN
Ox control is stopped. Subsequently, in the case of the comparative example, the usual DeSOx control is carried out, while in the case of the configuration example of the present disclosure, the above-mentioned DeSOx + DPF regeneration control is carried out.

In the case of the comparative example, the lean step and the rich step are not performed differently in the first cylinder group 2A and the second cylinder group 2B, but one of the rich step and the lean step is alternately performed in all the cylinders 2. Repeat to execute.

When the comparative example is carried out, in the lean step, the O2 concentration in front of DPF44 (the third exhaust passage 40C on the immediate upstream side of DPF44) is secured, so that the temperature of the exhaust gas rises due to the oxidation reaction in the oxidation catalyst layer 44d. However, both the temperature before DPF44 and the temperature inside DPF44 rise. This makes it possible to burn the PM collected in the DPF 44.

On the other hand, in the rich step of the comparative example, although the SOx adsorbed on the NOx catalyst is desorbed (not shown), the O2 concentration in the DPF44 is insufficient as compared with the case where the lean step is performed (Fig. Example). Then, it becomes substantially 0), and the heat generation due to the oxidation reaction becomes insufficient. Therefore, although the temperature before DPF44 is kept substantially constant, the temperature inside DPF44 drops, which is disadvantageous in burning PM.

Therefore, when the comparative example is carried out, the PM (soot deposition amount in the figure) decreases in the lean step, but the soot deposition amount increases in the rich step. As a result, within the range from time t2 to less than t5, the soot deposition amount does not substantially decrease, and by performing normal DPF regeneration control after time t5, normal DPF regeneration control is performed from time t5 to time t8. By doing so, it becomes possible to reduce the amount of soot deposited. Then, at time t8, this DPF regeneration control is completed.

Further, in the configuration of the comparative example, there is also a problem that CO and total hydrocarbon (THC) functioning as a reducing agent temporarily increase due to insufficient O2 concentration during the rich step. As can be seen from the comparison between the "CO and THC amounts before DPF" and the "CO and THC amounts after DPF" in FIG. 9, when the configuration of the comparative example is used, the DPF 44 substantially contains CO and THC. Do not remove. The "CO and THC amount after DPF" indicates the CO and THC amount in the third exhaust passage 40C on the direct downstream side of the DPF 44 and on the direct upstream side of the SCR catalyst 46.

On the other hand, when the configuration example of the present disclosure is implemented, the rich step is carried out in the first cylinder group 2A and the lean step is carried out in the second cylinder group 2B within the range from time t2 to less than t3. To.

More specifically, at time t2, a rich step was carried out in the first cylinder group 2A, and the injection timing was set to be relatively advanced and the injected fuel was set to burn in the combustion chamber 6. As the post-injection is performed, the air-fuel ratio of the exhaust is made close to or richer than the stoichiometric air-fuel ratio. At the same time, a lean step is carried out in the second cylinder group 2B, and post-injection is carried out in which the injection timing is relatively retarded and the injected fuel is set so as not to burn in the combustion chamber 6. At the same time, the air-fuel ratio of the exhaust is made leaner than the theoretical air-fuel ratio.

Further, within the range from time t3 to less than t4, the lean step is carried out in the first cylinder group 2A, while the rich step is carried out in the second cylinder group 2B.

In this way, the lean step and the rich step are repeatedly executed alternately and alternately in the first cylinder group 2A and the second cylinder group 2B.

When the configuration example of the present disclosure is carried out, when the rich step is carried out in one cylinder group, the lean step is carried out in the other cylinder group. Therefore, since the O2 concentration before DPF44 is kept substantially constant, sufficient heat generation due to the oxidation reaction is sufficiently secured. Therefore, in this configuration example, the amount of soot deposited is monotonously reduced. At the same time, when the rich step is performed in one of the cylinder groups, SOx can be detached in the NOx catalyst 41 leading to that cylinder group. As a result, as shown in FIG. 9, it is possible to reduce the amount of soot deposited and at the same time reduce the amount of SOx adsorbed.

Further, in the configuration example of the present disclosure, since the O2 concentration is sufficiently secured, the oxidation reaction in the oxidation catalyst layer 44d occurs more frequently, and the calorific value caused by the oxidation reaction increases. As the calorific value caused by the oxidation reaction increases, the temperature before DPF44 can be made lower than before (see FIG. 9).

Further, as can be seen from the comparison between the "CO and THC amounts before DPF" and the "CO and THC amounts after DPF" in FIG. 9, the DPF44 can generate CO and THC by O2 supplied through DeSOx + DPF regeneration control. It can be oxidized and removed. As described above, the DPF44 according to this configuration example can remove not only PM such as soot but also CO and THC due to the combination of DeSOx + DPF regeneration control and the fact that DPF44 has an oxidizing function. .. This is particularly effective in improving the emission performance of the engine 1.

In the configuration example of the present disclosure, after time t6, the SOx adsorption amount is sufficiently reduced in both the first NOx catalyst 41A and the second NOx catalyst 41B, and the process shifts to the normal DPF regeneration control. Normal DPF regeneration control is equivalent to continuously performing a lean step in all cylinders 2. This will further reduce the amount of soot deposited.

In the configuration example of the present disclosure, the soot deposition amount monotonically decreases during the DeSOx + DPF regeneration control. Therefore, the DPF regeneration control can be completed at time t7, which is earlier than the comparative example.

(3) Summary As described above, when the DeSOx control is performed, the ECU 100 alternately performs the rich step and the lean step as shown in the time chart of FIG. As a result, SOx adsorbed on the NOx catalyst 41 can be separated.

However, if the rich step and the lean step are performed alternately, the O2 concentration may be insufficient in the rich step as shown in the comparative example shown in FIG.

Therefore, the engine 1 shown as a configuration example does not use the NOx catalyst 41 common to all the cylinders 2, but has the first NOx catalyst 41A provided in the first exhaust passage 40A and the second NOx catalyst 41 as shown in FIG. The configuration focuses on proper use with the second NOx catalyst 41B provided in the exhaust passage 40B.

That is, as shown in steps S6 to S10 of FIG. 8, when the air-fuel ratio of the exhaust gas flowing through the first NOx catalyst 41A becomes rich, that is, when the rich step is performed in the first cylinder group 2A, the ECU 100 performs the second NOx. A lean step is performed in the second cylinder group 2B so that the air-fuel of the exhaust gas flowing through the catalyst 41B becomes lean.

According to this, although the exhaust gas passing through the first NOx catalyst 41A may be deficient in oxygen, the exhaust gas passing through the second NOx catalyst 41B contains a large amount of oxygen. Exhaust gas containing a large amount of oxygen flows through the third exhaust passage 40C provided downstream as compared with the case where the rich step is performed in all the cylinders 2.

As a result, as shown in FIG. 9, oxygen can be stably supplied to the DPF 44, and the DPF 44 can be efficiently purified. On the other hand, since the rich step and the lean step are alternately performed in the cylinder 2 unit, not only the DPF 44 but also the NOx catalyst 41 can be efficiently purified.

In this way, both the DPF 44 and the NOx catalyst 41 can be efficiently purified.

Further, the ECU 100 executes post injection when the lean step is performed. Since the fuel injected by this post injection does not burn in the cylinder 2, it is discharged to the exhaust passage 40 as unburned fuel (HC). Such HC can be used as a reducing agent.

Therefore, as shown in FIG. 3, when the DPF 44 has an oxidizing function, the heat of reaction is generated more frequently in the DPF 44 as a large amount of oxygen is supplied. In order to burn the PM collected in the DPF44, it was necessary to supply a relatively high temperature exhaust gas to the DPF44, but the exhaust gas can be made colder than before by the amount of such reaction heat. can. Therefore, it is possible to reduce the injection amount of fuel such as the above-mentioned additional injection, and at the same time, it is possible to keep the NOx catalyst 41 at a lower temperature than before by the amount of lowering the exhaust gas.

In general, it is known that when the temperature of the NOx catalyst 41 rises excessively, the SOx adsorbed on the NOx catalyst 41 aggregates, making it difficult to desorb the NOx catalyst 41 from the NOx catalyst 41. This is not preferable in that it causes a decrease in the efficiency of DeSOx control.

On the other hand, since the NOx catalyst 41 can be kept at a lower temperature than the conventional one, the engine 1 is also effective from the viewpoint of suppressing a decrease in the efficiency of DeSOx control.

Further, since the DPF 44 has an oxidizing function, unburned fuel such as HC can be oxidized in the DPF 44 even during the DeSOx + DPF regeneration control. As a result, the emission performance of the engine 1 can be improved.

By the way, as the vehicle equipped with the engine 1 repeatedly accelerates and decelerates, the fuel injection amount becomes biased, and as a result, the first cylinder group 11A and the second cylinder group 11B are covered with S. There may be a difference in the amount of poison.

Further, in the rich step, the exhaust becomes a reducing atmosphere. On the other hand, the SOx adsorbed on the NOx catalyst 41 is released in such a reducing atmosphere.

Therefore, in a situation where the amount of S poisoning in the first NOx catalyst 41A is large and it is considered that a large amount of SOx is adsorbed on the catalyst 41A, the basic rich time is lengthened as shown in FIG. As a result, a larger amount of SOx can be released from the first NOx catalyst 41A.

On the other hand, if the basic rich time is excessively long, the DPF44 may not be sufficiently purified in the lean step, and its regeneration efficiency may decrease.

Therefore, when the amount of S poisoning in the first NOx catalyst 41A is sufficiently large, it is set to a constant predetermined value as shown in FIG. 7 without lengthening the basic rich time. As a result, the reproduction efficiency of the DPF 44 can be ensured.

Further, when DeSOx + DPF regeneration control is performed, the NOx catalyst 41 may be excessively heated, and as a result, the NOx occluded in the catalyst 41 may be desorbed.

On the other hand, in this configuration example, as shown in steps S3 to S5 of FIG. 8, active DeNOx control is executed before the start of DeSOx + DPF reproduction control. According to this, NOx can be removed from the NOx catalyst 41 prior to starting the DeSOx + DPF regeneration control.

Further, in general, ammonia may be generated due to the binding of NOx, a reducing agent, or the like. The generation of ammonia is not desirable in order to suppress odors and the like.

On the other hand, as shown in the upper figure of FIG. 6, when the adsorption amount of NH3 is large, the post injection timing is retarded more than when it is small. Then, the combustion temperature when the fuel injected by the post injection burns in each cylinder 2 is reduced. As a result, the temperature of the exhaust gas discharged from each cylinder 2 is lowered. Since the production of ammonia requires a predetermined heat energy, such a configuration is effective in suppressing the generation of ammonia. Further, by suppressing the generation of ammonia, the burden on the slip catalyst 47 can be reduced.

Further, as shown in the lower figure of FIG. 6, when the adsorption amount of NH3 is large, the post injection amount is reduced as compared with the case where it is small, so that the thermal energy generated by the combustion of the fuel injected by the post injection is suppressed. be able to. This is advantageous in suppressing the generation of ammonia. Further, by suppressing the generation of ammonia, the burden on the SCR catalyst 46 can be reduced. Further, by suppressing the generation of ammonia, the burden on the slip catalyst 47 can be reduced << other embodiments >>.
In the above embodiment, when the PM deposit amount of DPF44 becomes equal to or more than the regeneration start deposit amount, the active DeNOx control and the DeSOx + DPF regeneration control are sequentially performed, but the configuration is not limited to this. .. For example, the active DeNOx control and the DeSOx + DPF regeneration control may be sequentially performed by triggering that the NOx storage amount becomes a predetermined amount or more.

1 engine 2 cylinder (cylinder)
2A 1st cylinder group (cylinder leading to the 1st exhaust passage)
2B 2nd cylinder group (cylinder leading to the 2nd exhaust passage)
10 Injector 40 Exhaust passage 40A 1st exhaust passage 40B 2nd exhaust passage 40C 3rd exhaust passage 41 NOx catalyst 41A 1st NOx catalyst 41B 2nd NOx catalyst 44 DPF (exhaust gas treatment device, particle filter)
46 SCR catalyst 100 ECU (controller)
SW9 NOx sensor

Claims (7)

Exhaust gas of an engine including an exhaust passage connected to each of a plurality of cylinders, a NOx catalyst provided in the exhaust passage, and an exhaust gas treatment device provided in an exhaust passage downstream of the NOx catalyst. It is a purification control device,
Injectors provided in each of the plurality of cylinders and
A controller connected to each of the plurality of injectors.
The exhaust gas treatment device is a particle filter configured to collect particulate matter in the exhaust gas.
The exhaust passage includes a first exhaust passage connected to a part of the plurality of cylinders, a second exhaust passage connected to another part of the plurality of cylinders, the first exhaust passage, and the above. It has a third exhaust passage, which is formed by merging the second exhaust passage on the downstream side of each and is provided with the exhaust gas treatment device.
The NOx catalyst has a first NOx catalyst provided in the first exhaust passage and a second NOx catalyst provided in the second exhaust passage.
The controller alternately performs a NOx catalyst regeneration control in which the rich step of making the air-fuel ratio near the theoretical air-fuel ratio or richer than the theoretical air-fuel ratio and the lean step of making the air-fuel ratio leaner than the rich step are performed alternately. A control signal is output to each of the plurality of injectors so as to perform the above.
When the controller performs the rich step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control, the controller performs the lean step in the cylinder leading to the second exhaust passage. A control signal is output to each of the plurality of injectors, and the control signal is output.
The controller is configured to remove sulfur components from the first and second NOx catalysts by executing the NOx catalyst regeneration control over a plurality of cycles.
Further, when the lean step is performed, the controller further injects fuel into the injector at an injection timing that does not burn in the cylinder after the fuel injection corresponding to the operating state of the engine. And output the control signal
An engine exhaust purification control device that is characterized by this. In the exhaust gas purification control device for the engine according to claim 1 ,
The controller determines the magnitude relationship between the sulfur poisoning amount in the first NOx catalyst and the sulfur poisoning amount in the second NOx catalyst based on the fuel injection amount to the plurality of cylinders.
When the controller performs the NOx catalyst regeneration control, when the amount of sulfur poisoning in the first NOx catalyst is larger than the amount of sulfur poisoning in the second NOx catalyst, the controller goes to the first exhaust passage as compared with the case where the amount is smaller. An engine exhaust purification control device characterized by prolonging the time for performing the rich step in a communicating cylinder. In the exhaust gas purification control device for an engine according to claim 1 or 2 .
When the controller performs the lean step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control, the controller performs the rich step in the cylinder leading to the second exhaust passage. An engine exhaust purification control device characterized by outputting a control signal to each of the plurality of injectors. In the exhaust gas purification control device for an engine according to any one of claims 1 to 3 .
When a predetermined condition is satisfied before the start of the NOx catalyst regeneration control, the controller performs additional fuel injection at an injection timing such that combustion occurs in the cylinder after fuel injection corresponding to the operating state of the engine. An engine exhaust purification control device characterized by outputting a control signal to the injector so as to perform NOx catalyst rich purge control that makes the air-fuel ratio richer than before the predetermined condition is satisfied. In the exhaust gas purification control device for an engine according to any one of claims 1 to 4 .
The exhaust gas treatment device is an engine exhaust gas purification control device characterized by having an oxidation function. Exhaust gas of an engine including an exhaust passage connected to each of a plurality of cylinders, a NOx catalyst provided in the exhaust passage, and an exhaust gas treatment device provided in an exhaust passage downstream of the NOx catalyst. It is a purification control device,
Injectors provided in each of the plurality of cylinders and
A controller connected to each of the plurality of injectors,
An SCR catalyst provided in the third exhaust passage on the downstream side of the exhaust gas treatment device, and
A NOx sensor provided on the upstream side and the downstream side of the SCR catalyst is provided.
The exhaust passage includes a first exhaust passage connected to a part of the plurality of cylinders, a second exhaust passage connected to another part of the plurality of cylinders, the first exhaust passage, and the above. It has a third exhaust passage, which is formed by merging the second exhaust passage on the downstream side of each and is provided with the exhaust gas treatment device.
The NOx catalyst has a first NOx catalyst provided in the first exhaust passage and a second NOx catalyst provided in the second exhaust passage.
The controller alternately performs a NOx catalyst regeneration control in which the rich step of making the air-fuel ratio near the theoretical air-fuel ratio or richer than the theoretical air-fuel ratio and the lean step of making the air-fuel ratio leaner than the rich step are performed alternately. A control signal is output to each of the plurality of injectors so as to perform the above.
When the controller performs the rich step in the cylinder leading to the first exhaust passage when performing the NOx catalyst regeneration control, the controller performs the lean step in the cylinder leading to the second exhaust passage. A control signal is output to each of the plurality of injectors, and the control signal is output.
When the rich step is performed, the controller sends the fuel injection corresponding to the operating state of the engine to the injector so as to execute an additional injection of fuel at an injection timing such that the fuel is burned in the cylinder. Output the control signal,
The controller estimates the adsorption amount of ammonia in the SCR catalyst based on the detection signal of the NOx sensor, and when the adsorption amount is large, the timing of executing the additional injection in the rich step is later than when the adsorption amount is small. Make a corner
An engine exhaust purification control device that is characterized by this. In the exhaust gas purification control device for an engine according to claim 6 .
The controller is an engine exhaust gas purification control device, characterized in that when the amount of adsorbed ammonia is large, the amount of fuel additionally injected in the rich step is reduced as compared with the case where the amount of ammonia adsorbed is small.

Images