JP2017025798A - Control device of internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suitably restore a NOx occlusion capacity without impairing an exhaust characteristic even if rich control is required during the acceleration of an internal combustion engine, related to a control device of the on-vehicle internal combustion engine which performs lean burn operation.SOLUTION: An internal combustion engine 10 having an NSR catalyst 42 and an in-cylinder injection valve 22 is controlled. Stoichiometric control for operating the internal combustion engine 10 in a stoichiometric air-fuel ratio region, lean control for operating the internal combustion engine 10 in a lean air-fuel ratio region, and rich control for discharging the NOx of the NSR catalyst 42 are selectively performed. When performing the rich control by interrupting the stoichiometric control during a rise of an engine rotational speed, at least one of the following items is performed out of (1) the enhancement of fuel pressure, (2) the acceleration of a fuel injection start crank angle, and (3) the further enrichment of an air-fuel ratio at a start of the rich control compared with the case that the rich control is performed while interrupting the lean control.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a control device for an internal combustion engine, and more particularly to a control device for an internal combustion engine that is suitable as a control device for a vehicle-mounted internal combustion engine that performs lean burn operation.

  Patent Document 1 discloses an internal combustion engine that performs lean burn operation. This internal combustion engine includes a NOx storage reduction catalyst in the exhaust passage. From the internal combustion engine during lean burn operation, exhaust gas containing NOx at a high concentration is discharged. The NOx occlusion reduction catalyst can occlude NOx in the exhaust gas within the range of the occlusion capacity. For this reason, the internal combustion engine described above can prevent NOx from being released into the atmosphere until the NOx storage reduction catalyst reaches a state where the NOx is saturated.

The internal combustion engine performs rich control to make the air-fuel ratio of the air-fuel mixture rich when the NOx occlusion amount in the NOx occlusion reduction catalyst reaches a preset threshold value. The NOx storage reduction catalyst releases stored NOx when the air-fuel ratio of the exhaust gas becomes rich. The released NOx is reduced by a reducing agent such as CO and HC contained in rich exhaust gas and purified to N ₂ , CO _{2 and the} like. Therefore, according to the internal combustion engine described above, the NOx storage capacity of the NOx storage reduction catalyst can be appropriately regenerated without deteriorating the exhaust characteristics.

  In a vehicle in which the above internal combustion engine is mounted together with an automatic transmission, there may occur a situation in which gear shifting and rich control are performed at the same timing. Before and after the start of the rich control, a change in the torque of the internal combustion engine due to a change in the air-fuel ratio is likely to occur. For this reason, when they are overlapped, the shock at the time of shifting is easily increased.

  Patent Document 1 discloses a technique for executing rich control before a gear shift when a gear shift is predicted in order to prevent such an increase in shock. According to this technique, it is possible to reliably avoid an increase in shift shock due to execution of rich control.

JP 09-184438 A JP-A-10-131786 JP 2002-013428 A Japanese Patent Laid-Open No. 2007-327412 International Publication No. 2004/097200

  By the way, in the internal combustion engine described above, the rich control may be performed while the engine speed is increasing. The moving speed of the piston increases as the engine speed increases. On the other hand, the time required to inject the same amount of fuel is constant regardless of the engine speed if the injection conditions are the same. Since the internal combustion engine always injects fuel under the same conditions, the crank angle at the end of injection changes to the retard side as the engine speed increases. During rich control requiring a large amount of fuel injection, the injection end crank angle exceeds the engine-specific threshold in the high speed region, and the fuel injected into the cylinder reaches the cylinder inner wall without being blocked by the piston. May lead to a condition.

  The fuel adhering to the wall surface may generate particulate matter (PM) without completely burning in the cylinder. For this reason, in the above-described internal combustion engine, when rich control is executed in a high rotation range, a situation may occur in which exhaust characteristics of the internal combustion engine deteriorate.

  The present invention has been made to solve the above-described problems, and even when rich control is required during acceleration of an internal combustion engine, the NOx occlusion ability can be appropriately restored without impairing exhaust characteristics. An object of the present invention is to provide a control device for an internal combustion engine that can perform the above-described operation.

In order to achieve the above object, a first invention is an apparatus for controlling an internal combustion engine comprising a NOx storage reduction catalyst disposed in an exhaust passage and an in-cylinder injection valve that injects fuel directly into the cylinder. ,
Stoichiometric control for operating the internal combustion engine in a stoichiometric air-fuel ratio region including a stoichiometric air-fuel ratio;
Lean control for operating the internal combustion engine in an air-fuel ratio range leaner than the stoichiometric air-fuel ratio range;
And rich control for selectively passing exhaust gas belonging to an air-fuel ratio region richer than the stoichiometric air-fuel ratio region to the NOx storage-reduction catalyst in order to release NOx stored in the NOx storage-reduction catalyst. Run,
When the rich control is executed by interrupting the stoichiometric control while the engine speed is increasing, compared to the case where the rich control is executed by interrupting the lean control,
(1) Increasing the pressure of fuel supplied to the in-cylinder injection valve;
(2) accelerating the fuel injection start crank angle of the in-cylinder injection valve, and (3) further enriching the air-fuel ratio at the start of the rich control,
And executing at least one of the following.

  In the first invention, (1) when the fuel pressure to the in-cylinder injection valve is increased, the time required for fuel injection is shortened. In this case, the end crank angle of fuel injection is accelerated in the advance direction as compared with the case where the fuel pressure is not changed. Also, (2) when the start crank angle of fuel injection is increased, the end crank angle of fuel injection is also increased in the advance direction. (3) When the air-fuel ratio at the start of the rich control is further enriched, the NOx release speed is increased as compared with the case where the rich control is not performed, and the execution time of the rich control is shortened.

  In an internal combustion engine, if the crank angle at the end of injection exceeds a threshold value specific to the engine, a situation occurs in which the fuel injected from the in-cylinder injection valve reaches the in-cylinder wall surface without being blocked by the piston. If the injection end crank angle is advanced by the effect (1) or (2), it is easy to end the fuel injection before the threshold value even in the high speed region. Further, when the execution time of the rich control is shortened due to the effect (3), it becomes easy to complete the rich control before the engine speed greatly increases with acceleration. If the rich control is completed before the engine speed increases, it is possible to avoid the crank angle at the end of injection from exceeding the threshold value. Therefore, according to the present invention, even when rich control is required in the process in which the engine rotational speed increases with acceleration, the NOx storage capacity can be properly recovered without impairing the exhaust characteristics of the internal combustion engine. it can.

It is a figure which shows the structure of Embodiment 1 of this invention. It is a figure which shows the driving | operation area | region (lean driving | operation area | region) in which lean control is performed in Embodiment 1 of this invention. It is a figure which shows the relationship between the cylinder injection start crank angle of fuel, and the number of PM particle | grains in exhaust gas. It is a figure which shows the relationship between the engine speed NE and the cylinder injection end crank angle. FIG. 6 is a diagram showing the relationship between in-cylinder injection time τ and in-cylinder injection amount q according to fuel pressure. It is a timing chart for demonstrating the subject at the time of performing rich control during acceleration. 3 is a timing chart for explaining the operation of the first embodiment of the present invention. It is a flowchart of the routine performed in Embodiment 1 of the present invention. It is a flowchart of the routine performed in Embodiment 2 of this invention. FIG. 5 is a diagram showing the relationship between the in-cylinder injection start crank angle of fuel and the number of PM particles in exhaust gas according to engine speed. It is a figure which shows the relationship between the engine speed NE, the fuel pressure, and the possible advance amount of the starting crank angle of fuel injection. It is a figure which shows the relationship between the engine speed NE, the cooling water temperature of an internal combustion engine, and the possible advance amount of the starting crank angle of fuel injection. It is a flowchart of the routine performed in Embodiment 3 of the present invention. It is a timing chart for demonstrating operation | movement of Embodiment 3 of this invention. It is a timing chart for demonstrating operation | movement of Embodiment 4 of this invention. It is a flowchart of the routine performed in Embodiment 4 of this invention. It is a flowchart of the routine performed in Embodiment 5 of this invention. It is a timing chart for demonstrating operation | movement of Embodiment 6 of this invention. It is a flowchart of the routine performed in Embodiment 6 of this invention.

Embodiment 1 FIG.
[Configuration of Embodiment 1]
FIG. 1 shows the configuration of Embodiment 1 of the present invention. The present embodiment includes an internal combustion engine 10. The internal combustion engine 10 includes variable valve mechanisms 12 and 14 on both the intake side and the exhaust side. The variable valve mechanisms 12 and 14 can change the operation timing of the intake valve 16 and the exhaust valve 18.

  The internal combustion engine 10 includes a piston 19, a spark plug 20, an in-cylinder injection valve 22, and a port injection valve 24 for each cylinder. Fuel is supplied to the in-cylinder injection valve 22 via a fuel pressure variable mechanism 26. The fuel pressure variable mechanism 26 can change the pressure of the fuel supplied to the in-cylinder injection valve 22. The internal combustion engine 10 is also equipped with a water temperature sensor 28 and a crank angle sensor 30. According to the crank angle sensor 30, the rotational speed NE of the internal combustion engine 10 can be detected.

  The exhaust passage 32 of the internal combustion engine 10 communicates with the turbine 36 of the turbocharger 34. The exhaust passage 32 is provided with a bypass passage 38 that bypasses the turbine 36, and a wastegate valve 40 is incorporated therein. A three-way catalyst 41, a NOx storage reduction catalyst (NSR catalyst) 42, and a NOx selective reduction catalyst (SCR catalyst) 43 are disposed downstream of the turbine 36 and the bypass passage 38.

  An air-fuel ratio sensor 44 is disposed upstream of the three-way catalyst 41. The air-fuel ratio sensor 44 generates a signal that is linear with respect to the air-fuel ratio. Oxygen sensors 45 and 46 are arranged downstream of the three-way catalyst 41 and downstream of the NSR catalyst 42. The oxygen sensors 45 and 46 output signals that change stepwise between the lean side and the rich side with the theoretical air-fuel ratio as a boundary. A NOx sensor 47 that emits a signal corresponding to the NOx concentration is installed downstream of the SCR catalyst 43.

  The intake passage 50 of the internal combustion engine 10 includes an air flow meter 54 downstream of the air filter 52. A compressor 56 of the turbocharger 34 is disposed downstream of the air flow meter 54. Further, an intercooler 58, a supercharging pressure sensor 60, and an electronic throttle valve 62 are disposed downstream of the compressor 56. In the vicinity of the throttle valve 62, a throttle opening sensor 64 is provided. A surge tank 66 communicates with the downstream side of the throttle valve 62. An intake pressure sensor 68 is disposed in the surge tank 66.

  The present embodiment includes an electronic control unit (hereinafter referred to as “ECU”) 70. In addition to the various sensors and actuators described above, an accelerator opening sensor 72 is electrically connected to the ECU 70. The ECU 70 can detect a request from the vehicle driver based on the output of the accelerator opening sensor 72.

[Basic operation of the first embodiment]
(Lean burn operation)
FIG. 2 shows an operation region where the lean burn operation is performed in the first embodiment of the present invention. Hereinafter, this region is referred to as a “lean operation region”, and control for operating the internal combustion engine in that region is referred to as “lean control”. In FIG. 2, a solid line denoted by reference numeral 74 indicates the fully open characteristic of the internal combustion engine 10. Moreover, the boundary line shown with the code | symbol 76 has shown the outline of the lean operation area | region. As shown in FIG. 2, in the present embodiment, NE1 and TRQ1 are the upper limit values of the engine rotational speed NE and the torque TRQ in the lean operation region.

The internal combustion engine 10 operates in an air-fuel ratio region (hereinafter referred to as “stoichiometric air-fuel ratio region”) including stoichiometry in a high-rotation and high-load region (hereinafter referred to as “stoichiometric operation region”) exceeding the lean operation region. Do. Hereinafter, this operation is referred to as “stoichiometric operation”, and control for causing the internal combustion engine 10 to perform stoichiometric operation is referred to as “stoichiometric control”. During stoichiometric operation, exhaust gas containing CO, HC and NOx is exhausted in response to excess or deficiency of oxygen. The three-way catalyst 41 stores NOx and O ₂ to purify exhaust gas under an excessive oxygen condition, and purifies CO and HC by releasing oxygen under an oxygen deficient condition. For this reason, during the stoichiometric operation, impurities in the exhaust gas can be appropriately purified mainly by the function of the three-way catalyst 41.

The lean burn operation is performed using an air-fuel ratio region that is leaner than the stoichiometric air-fuel ratio region. During lean burn operation, CO and HC are not generated, and exhaust gas containing NOx at a high concentration is discharged. In this case, the exhaust gas containing NOx blows through the three-way catalyst 41 and flows into the NSR catalyst 42. The NSR catalyst 42 can occlude NOx within the range of its occlusion capacity, and releases NOx during occlusion when it is supplied with exhaust gas having a rich air-fuel ratio. The NOx released at this time is reduced by HC and CO contained in the rich exhaust gas and purified to N ₂ and CO ₂ .

(Rich control)
In this embodiment, the NOx occlusion amount is calculated by integrating the NOx amount flowing into the NSR catalyst 42 during the lean burn operation. When the NOx occlusion amount reaches a determination value corresponding to NOx saturation, the air-fuel ratio is changed to a richer value than the stoichiometric air-fuel ratio region. Hereinafter, this control is referred to as “rich control”.

  When the rich control is started, NOx stored in the NSR catalyst 42 starts to be released. When the release of NOx is completed, the exhaust gas rich in the air-fuel ratio is blown out downstream of the NSR catalyst 42, and the output of the oxygen sensor 46 is inverted. The ECU 70 receives the reversal and ends the rich control, and again starts the control for the lean burn operation. NOx generated during the lean burn operation is appropriately purified mainly by the NSR 42 by repeating the above processing.

(Relationship between fuel injection timing and number of PM particles)
FIG. 3 shows the relationship between the in-cylinder injection start crank angle used during rich control and the number of PM particles. In FIG. 3, the horizontal axis represents the crank angle before top dead center (BTDC) with the compression top dead center as the base point (0 degree). That is, “330” shown on the horizontal axis represents the crank angle advanced by 30 degrees from the exhaust top dead center (360 degrees BTDC). “180” means an intake bottom dead center.

  In-cylinder injection of the internal combustion engine 10 is performed in a process in which the piston 19 passes through the exhaust top dead center (360 degrees BTDC) toward the intake bottom dead center (180 degrees BTDC). While the piston 19 is too close to the exhaust top dead center, the fuel injected from the in-cylinder injection valve 22 adheres to the vicinity of the top surface of the piston 19 and generates droplets. Such droplets are difficult to burn completely in the cylinder and tend to generate PM. For this reason, as shown in FIG. 3, when the in-cylinder injection start crank angle becomes earlier than a certain threshold value (here, about 310 degrees BTDC), the number of PM particles shows a rapid increasing tendency.

  On the other hand, when the piston 19 is too close to the bottom dead center, the fuel injected from the in-cylinder injection valve 22 reaches the in-cylinder wall surface without being blocked by the piston 19. In this case, coupled with the fact that the flow rate of the intake air has started to settle, droplets are likely to be generated on the inner wall surface of the cylinder. Such droplets are also difficult to burn completely in the cylinder, and tend to generate PM. For this reason, as shown in FIG. 3, when the in-cylinder injection start crank angle becomes slower than a certain threshold value (here, about 230 degrees BTDC), the number of PM particles shows a rapid increasing tendency.

(Relationship between in-cylinder injection end crank angle and engine speed)
FIG. 4 shows the relationship between the crank angle at which in-cylinder injection started under the same conditions ends and the engine speed NE. The vertical axis shown in FIG. 4 indicates the BTDC crank angle with the compression top dead center as the base point (0 degree), similarly to the horizontal axis shown in FIG. The moving speed of the piston 19 increases as the engine speed NE increases. On the other hand, the time required for injecting the same amount of fuel is constant even if the engine speed changes as long as the fuel injection conditions are the same. For this reason, the crank angle at which the fuel injection started under the same conditions ends becomes a value closer to the intake bottom dead center as the engine speed NE increases as shown in FIG.

  A boundary line denoted by reference numeral 78 in FIG. 4 indicates a limit crank angle (hereinafter referred to as “limit crank angle”) at which droplets are not generated on the inner wall surface of the cylinder. When the crank angle is larger than the limit crank angle 78, the in-cylinder injected fuel is blocked by the piston 19 and is not sprayed directly on the in-cylinder wall surface. On the other hand, when the crank angle is smaller than the limit crank angle 78, the injected fuel directly reaches the inner wall surface of the cylinder and generates droplets there. In the internal combustion engine 10 of the present embodiment, when the injection condition is constant, 204 ° BTDC with the compression top dead center as a base point becomes the limit crank angle 78. When the engine rotational speed NE exceeds 5000 rpm, the crank angle at the end of injection becomes smaller than the limit crank angle 78, and wall surface droplets are generated.

  As described above, in the present embodiment, rich control is started when the NOx storage amount reaches the NOx saturation determination value. Since the NOx occlusion amount increases only during lean burn operation, the start of rich control is required in principle during lean burn operation. Since the lean burn operation is performed in a region where the rotation speed is low, the injection end crank angle does not fall below the limit crank angle 78 when the rich control is started under the operation.

  However, at the time of acceleration of the internal combustion engine 10, after the transition from the lean operation region to the stoichiometric operation region, a rich control start request may occur, for example, at point A shown in FIG. If the engine speed NE further increases thereafter, a situation may occur in which the injection end crank angle enters the wall droplet generation region before the rich control is completed.

[Features of Embodiment 1]
(Higher fuel pressure)
FIG. 5 shows the relationship between the fuel injection time (in-cylinder injection time τ) by the in-cylinder injection valve 22 and the in-cylinder injection amount q in two cases with different fuel pressures. As shown in FIG. 5, the time τ required to inject the same amount of fuel q is shorter when the fuel pressure is higher than when the pressure is lower. Therefore, if the fuel pressure applied to the in-cylinder injection valve 22 is increased, the time required for fuel injection is shortened, and it becomes easy to prevent the injection end crank angle from falling below the limit crank angle in the high rotation region. Therefore, in the present embodiment, when the execution of rich control is required in the stoichiometric operation region during acceleration of the internal combustion engine 10, the fuel pressure is increased compared to the case where rich control is required in the lean operation region. Rich control was performed.

(Details of rich control)
Next, the details of the rich control executed by the present embodiment and the effects achieved by the present embodiment will be described with reference to FIGS.
FIG. 6 is a timing chart for explaining the operation when the rich control is performed with the reference fuel pressure at the time of acceleration of the internal combustion engine 10. However, the “reference fuel pressure” is a fuel pressure normally used in the internal combustion engine 10. When rich control is requested during execution of lean control, the rich control is performed using the reference fuel pressure.

  In this embodiment, the NSR 42 releases NOx when supplied with rich exhaust gas as described above. Here, in the present embodiment, the three-way catalyst 41 is disposed upstream of the NSR catalyst 42. For this reason, after the rich control is started, the enriched exhaust gas first flows into the three-way catalyst 41.

  Before the start of rich control, lean burn operation is performed. For this reason, the three-way catalyst 41 is in an oxygen saturated state at the start of the rich control. The three-way catalyst 41 in this state purifies the rich exhaust gas, and sends the exhaust gas without excess or deficiency of oxygen downstream. For this reason, the exhaust gas rich in the air-fuel ratio does not flow into the NSR catalyst 42 until the stored oxygen of the three-way catalyst 41 is consumed after the rich control is started. In other words, after the rich control is started, the NOx release of the NSR catalyst 42 is not started until the stored oxygen of the three-way catalyst 41 is consumed.

  In order to improve the fuel consumption characteristics of the internal combustion engine 10, it is desirable that the rich control period be short. Therefore, it is desirable that the period until the NSR catalyst 42 starts to release NOx after the start of the rich control is also short.

  A waveform 80 shown in the uppermost column of FIG. 6 shows a change in the air-fuel ratio that occurs when rich control is executed in the present embodiment. As shown by the waveform 80, in the present embodiment, after the rich control is started, the air-fuel ratio of the air-fuel mixture is first greatly enriched (for example, about 12.5). The oxygen of the three-way catalyst 41 is consumed faster as the exhaust gas flowing into the three-way catalyst 41 becomes richer. For this reason, if the air-fuel ratio is greatly enriched at the start of the rich control, the period until the NSR catalyst 42 starts to release NOx can be shortened.

  When the oxygen of the three-way catalyst 41 is consumed after the start of the rich control, the output of the oxygen sensor 45 disposed downstream thereof is inverted to the rich side. When detecting this inversion, the ECU 70 weakens the enrichment of the air-fuel mixture and continues the rich control with a rich air-fuel ratio of about 14.0. As a result, the NOx occluded in the NSR catalyst 42 is properly released, and the NOx occlusion capacity is recovered.

  The timing chart shown in FIG. 6 shows an operation in a situation where the fuel pressure is constant before and after the start of the rich control and the engine speed NE is continuously increasing (waveforms 82 and 84). reference). The end-crank angle of in-cylinder injection indicated by the waveform 86 becomes a retarded value as the air-fuel ratio becomes richer, and changes to the retarded side as the engine speed NE increases. In the operation shown in FIG. 6, the injection end crank angle 86 enters the region below the limit crank angle 78 in the second half of the first stage enrichment and in the second half of the second stage enrichment. In this case, exhaust gas containing a large number of PM particles is discharged in the second half of the first stage and the second half of the second stage.

  FIG. 7 is a timing chart of the operation shown in the present embodiment when execution of rich control is requested during acceleration of the internal combustion engine 10. A waveform indicated by reference numeral 88 in FIG. 7 represents the pressure of the fuel supplied to the in-cylinder injection valve 22. As shown in FIG. 7, in the present embodiment, when execution of rich control is requested during acceleration of the internal combustion engine 10, the fuel pressure 88 is increased prior to the start of rich control.

  When the fuel pressure 88 is increased, the in-cylinder injection time τ is shortened, and the injection end crank angle 90 of each cycle is advanced as compared with the case where the reference fuel pressure is used. Therefore, according to the present embodiment, the rich control can be ended without causing a situation where the injection end crank angle 90 is less than the limit crank angle 78 even when the internal combustion engine 10 is accelerated. Therefore, according to the present embodiment, even when the internal combustion engine 10 is accelerated, the NOx storage capacity of the NSR catalyst 42 can be properly recovered without impairing the exhaust characteristics.

(Routine process of the first embodiment)
FIG. 8 shows a flowchart of a routine executed by the ECU 70 in the present embodiment. The routine shown in FIG. 8 is started repeatedly at a predetermined cycle after the internal combustion engine 10 is started. When this routine is started, it is first determined whether or not a rich control request due to out-of-region has occurred (step 100). Specifically, it is determined whether the following three conditions are both satisfied.
(Condition 1) Has the NOx storage amount of the NSR catalyst 42 reached the NOx saturation judgment value?
(Condition 2) Is the current operating range a stoichiometric operating range?
(Condition 3) Is the engine speed increasing?

  When at least one of the above three conditions is not satisfied, it can be determined that the rich control can be completed without causing droplets on the cylinder inner wall surface even if the reference fuel pressure is maintained. Therefore, in the routine shown in FIG. 8, if the determination in step 100 is negative, the current routine is terminated without performing any special processing.

  On the other hand, when all of the above three conditions are satisfied, it can be determined that there is a possibility that droplets are generated on the cylinder inner wall surface when rich control is performed at the reference fuel pressure. For this reason, in the routine shown in FIG. 8, if the determination in step 100 is affirmative, processing for increasing the fuel pressure and performing rich control is performed thereafter.

  Specifically, first, the in-cylinder injection amount q required in the current operation cycle is calculated (step 102). The ECU 70 calculates the air-fuel ratio of the air-fuel mixture in a routine different from the routine shown in FIG. For example, after the rich control is started, the air-fuel ratio is calculated to be 12.5 during the period of the first stage of enrichment. Further, the air-fuel ratio is calculated to be 14.0 during the period of the second stage enrichment. In this step 102, in addition to the air-fuel ratio calculated in this way, the in-cylinder injection amount q is calculated based on the intake air amount Ga detected by the air flow meter 54, the injection ratio of port injection and in-cylinder injection, and the like. The

Next, the in-cylinder injection time τ when the reference fuel pressure is used is calculated (step 104). The ECU 70 stores the injection amount per unit time of the in-cylinder injection valve 22 under the reference fuel pressure (hereinafter referred to as “unit time injection amount”). Here, the in-cylinder injection time τ is calculated by applying the unit time injection amount to the following equation.
(In-cylinder injection time τ) = (In-cylinder injection amount q) / (Unit time injection amount) (1)

Next, the end crank angle of in-cylinder injection when the reference fuel pressure is used is calculated (step 106). Here, specifically, the following processing is sequentially performed.
1. Reading of the start crank angle of in-cylinder injection.
2. Based on the engine speed NE and in-cylinder injection time τ, the crank angle width that changes during in-cylinder injection is calculated.
3. The injection end crank angle is calculated by adding the crank angle width of 2 to the start crank angle of 1.

  When the above processing is completed, it is then determined whether or not the in-cylinder injection end crank angle is below the limit crank angle (step 108). If the determination is negative, it can be determined that the fuel injection ends before reaching the state where the fuel is directly sprayed onto the cylinder inner wall surface. In this case, since it is not necessary to increase the fuel pressure, the current routine is immediately terminated thereafter.

  On the other hand, if the determination in step 108 is affirmative, it can be determined that the injection end crank angle enters the region where droplets are generated on the cylinder inner wall surface with the reference fuel pressure maintained. In this case, a command for raising the fuel pressure to the maximum pressure that can be used in the internal combustion engine 10 is issued to the variable fuel pressure mechanism 26 (step 110). If the fuel pressure is increased, the time τ required for fuel injection is shortened, and the injection end crank angle shifts to the advance side. For this reason, according to said process, when rich control is performed during the acceleration of the internal combustion engine 10, it can prevent effectively that an injection end crank angle falls below a limit crank angle. Therefore, according to the present embodiment, even when the internal combustion engine 10 is accelerated, the NOx occlusion ability of the NSR catalyst 42 can be properly recovered while minimizing the deterioration of the exhaust characteristics.

Embodiment 2. FIG.
Next, a second embodiment of the present invention will be described with reference to FIG. The present embodiment can be realized by causing the ECU 70 to execute the routine shown in FIG. 9 instead of the routine shown in FIG. 8 using the configuration shown in FIG.

[Features of Embodiment 2]
(Fuel pressure increasing method)
In the first embodiment described above, when the generation of droplets cannot be prevented with the reference fuel pressure, the fuel pressure is always changed to the maximum pressure. However, in order to increase the fuel pressure, it is necessary to make a booster mechanism such as a pump work. For this reason, in order to raise the fuel pressure to the maximum pressure, the boosting mechanism is forced to do a great deal of work.

  On the other hand, the fuel pressure at the time of rich control is sufficient as long as the crank angle at which the injection ends is not less than the limit crank angle, and the maximum pressure is required in all situations where droplet generation cannot be prevented at the reference fuel pressure. It is not something to do. Therefore, in the present embodiment, the fuel pressure is increased so that the injection end crank angle coincides with the limit crank angle in a situation where the generation of droplets cannot be prevented with the reference fuel pressure.

(Routine process of the second embodiment)
FIG. 9 is a flowchart of a routine executed by the ECU 70 in the present embodiment. This routine is repeatedly started at a predetermined cycle after the internal combustion engine 10 is started, similarly to the routine shown in FIG. Further, steps 100 to 108 of this routine are the same as the corresponding steps shown in FIG. In the following, description will be made with respect to a part specific to the routine shown in FIG.

In the routine shown in FIG. 9, when it is determined in step 108 that the injection end crank angle is less than the limit crank angle, the target injection period is then calculated (step 112). Here, specifically, the following processing is sequentially performed.
1. Reading of the start crank angle of in-cylinder injection.
2. Reading of the limit crank angle.
3. Calculate the target injection period according to the following equation (Target injection period) = (Start crank angle)-(Limit crank angle) (2)
According to the processing in step 108, the injection period for ending the fuel injection at the limit crank angle can be calculated as the target injection period.

Next, a target injection time corresponding to the target injection period is calculated (step 114). If the engine speed NE (rpm) is known, the time required for the crank angle to change by the width of the target injection period can be calculated. In step 114, specifically, the time is calculated as the target injection time according to the relationship of the following equation.
(Target injection time) = (Target injection period) × 60 / (NE × 360) (3)

Next, the fuel pressure for injecting the in-cylinder injection amount q within the target injection time is calculated as the target fuel pressure (step 116).
Here, the relationship of the following equation is established between the injection amount from the in-cylinder injection valve 22, the fuel pressure, and the injection time.
(Injection amount) =
(Unit time injection amount) × √ {(actual fuel pressure) / (reference fuel pressure)} × injection time
... (4)

Applying “in-cylinder injection amount q” to (injection amount), “target fuel pressure” to (actual fuel pressure), and “target injection time” to (injection time), the target fuel pressure is Can be expressed as
(Target fuel pressure) =
{(In-cylinder injection amount q) / (unit time injection amount × target injection time)} ² × (reference fuel pressure)
... (5)

  In step 116, the target fuel pressure is calculated according to the relationship of the above equation (5). The target fuel pressure calculated in this way is a pressure for ending the fuel injection at the limit crank angle. Therefore, if the fuel pressure is changed to the target fuel pressure, the injection end crank angle can be made equal to the limit crank angle.

  In the routine shown in FIG. 9, it is then determined whether the target fuel pressure is equal to or lower than the maximum fuel pressure that can be used in the internal combustion engine 10 (step 118). As a result, when it is determined that the target fuel pressure is equal to or lower than the maximum fuel pressure, the fuel pressure is changed to the target fuel pressure (step 120). On the other hand, if the determination in step 118 is negative, the target fuel pressure is changed to the maximum fuel pressure (step 122), and then the process of step 120 is performed.

  According to the above processing, when the generation of droplets cannot be prevented at the reference fuel pressure, the fuel pressure is increased so that the injection end crank angle is aligned with the limit crank angle within a range not exceeding the maximum fuel pressure. Can do. According to this process, it is possible to efficiently prevent the generation of droplets accompanying the rich control during acceleration without unnecessarily increasing the fuel pressure. For this reason, according to the present embodiment, the NOx occlusion ability of the NSR catalyst 42 can be appropriately recovered even during acceleration without impairing good exhaust characteristics.

Embodiment 3 FIG.
Next, a third embodiment of the present invention will be described with reference to FIGS. The present embodiment can be realized by causing the ECU 70 to execute the routine shown in FIG. 13 instead of the routine shown in FIG. 8 using the configuration shown in FIG.

[Features of Embodiment 3]
(Advance of crank angle for starting injection)
FIG. 10 is a diagram showing the relationship between the in-cylinder injection start crank angle and the number of PM particles under two reference fuel pressures in two cases where the engine speed NE is different. In FIG. 10, the broken line connecting the ◯ plots is the relationship when NE is 2000 rpm, and the broken line connecting the □ plots is the relationship when NE is 4000 rpm.

  As described with reference to FIG. 3, the number of PM particles was started at the timing when the injected fuel directly adheres to the top surface of the piston 19 in addition to the case where the injection end crank angle was less than the limit crank angle. Also increases in case. That is, a large amount of PM particles is generated even when the injection start crank angle is too close to the top dead center. Hereinafter, when the injection start crank angle is gradually brought closer to the top dead center, the crank angle that becomes the inflection point at which the number of PM particles rapidly increases is referred to as “start side limit crank angle”.

  Near the starting limit crank angle, the higher the engine speed NE, the faster the piston 19 moves away from the in-cylinder injection valve 22. As the speed at which the piston 19 moves away is higher, the fuel injected from the in-cylinder injection valve 22 is less likely to adhere to the top surface of the piston 19. In other words, the fuel injected in the vicinity of the start side limit crank angle becomes more difficult to generate PM particles as the engine rotational speed NE is higher. For this reason, as shown in FIG. 10, the limit crank angle on the start side becomes a larger value (a value closer to top dead center) as NE is higher.

  In other words, the start crank angle of in-cylinder injection can be changed to the advance side as the engine rotational speed NE is higher without causing a rapid increase in the number of PM particles. If the injection start crank angle is advanced, the injection end crank angle is naturally advanced. For this reason, if the injection start crank angle during the rich control is advanced in accordance with the increase in the engine rotational speed NE, it is easy to end the injection before generating droplets on the cylinder inner wall surface in the high rotation region. Can create a state. Therefore, in this embodiment, the start crank angle of in-cylinder injection is advanced according to the engine speed NE as necessary during execution of rich control.

(Calculation of the advance amount of the crank angle at the start of injection)
In the present embodiment, in conjunction with the process of advancing the injection start crank angle as described above, the process of increasing the fuel pressure is performed as necessary, as in the case of the second embodiment. Here, in the vicinity of the limit crank angle on the start side, the higher the fuel pressure, the easier the fuel adheres to the piston top surface. Therefore, the advanceable amount of the injection start crank angle varies depending on the fuel pressure. Therefore, in the present embodiment, the fuel pressure is taken into consideration when determining the advance amount of the injection start crank angle.

  Further, whether or not the fuel injected near the limit crank angle on the start side generates PM particles is greatly influenced by the fuel vaporization. The fuel vaporization is greatly influenced by the in-cylinder temperature of the internal combustion engine 10. For this reason, in this embodiment, when determining the advance amount of the injection start crank angle, the cooling water temperature having a correlation with the in-cylinder temperature is also taken into consideration.

  As described above, in this embodiment, the advance amount of the starting crank angle of fuel injection is determined based on the engine speed NE, the fuel pressure of in-cylinder injection, and the coolant temperature of the internal combustion engine 10. The ECU 70 stores a map that determines the advanceable amount of the injection start crank angle in relation to these three parameters.

  FIG. 11 shows the relationship read from the above map by specifying the cooling water temperature. FIG. 12 shows the relationship read from the map by specifying the fuel pressure. Numerical values such as 0, 5, 10, 15, and 20 shown in the two-dimensional coordinates of FIGS. 11 and 12 represent the advance amount that can be given to the reference injection start crank angle at the respective coordinates. . The ECU 70 can appropriately calculate the advance amount of the injection start crank angle by referring to these relationships. However, the “reference injection start crank angle” is an injection start crank angle normally used in the internal combustion engine 10. When rich control is requested during execution of lean control, the rich control is performed using the reference injection start crank angle.

(Routine process of the third embodiment)
FIG. 13 is a flowchart of a routine executed by the ECU 70 in the present embodiment in order to realize the above function. The routine shown in FIG. 13 is the same as the routine shown in FIG. 9 except that steps 130 to 134 are inserted after step 122. In the following, detailed description of parts common to both will be omitted, and parts specific to the routine shown in FIG. 13 will be mainly described.

  The routine shown in FIG. 13 is repeatedly started at a predetermined cycle after the internal combustion engine is started. According to this routine, when the execution of the rich control is requested during the acceleration process, the target fuel pressure for adjusting the injection end crank angle to the limit crank angle is calculated by the processing of steps 100 to 116. If the target fuel pressure is equal to or lower than the maximum fuel pressure, rich control is performed with the fuel pressure as the value (see steps 118 and 120).

  On the other hand, if the target fuel pressure exceeds the maximum fuel pressure, the target fuel pressure is replaced with the maximum fuel pressure (see step 122). In this case, if rich control is performed as it is, the injection end crank angle is less than the limit crank angle, and PM particles can be generated slightly.

  In the routine shown in FIG. 13, following the process of step 122, the engine speed NE and the coolant temperature ethw are read (step 130). Next, the advanceable amount corresponding to NE, ethw, and the maximum fuel pressure is calculated with reference to the map described above (step 132). Next, the injection start crank angle is changed to the advance side within the range of the advanceable amount (step 134). Thereafter, rich control fuel injection is performed through the processing of step 120.

  According to the above processing, when an increase in the amount of PM particles accompanying the execution of rich control is predicted, first, the advancement of the injection end crank angle is achieved by the increase in fuel pressure. If the injection end crank angle is retarded from the limit crank angle even when the maximum fuel pressure is used, the injection start crank angle is advanced to further increase the injection end crank angle. Figured. For this reason, according to the present embodiment, it is possible to avoid an increase in the number of PM particles accompanying the execution of the rich control in a wider operation region than in the case of the first or second embodiment.

[Modification of Embodiment 3]
(Modification 1)
In the third embodiment described above, the method of advancing the injection start crank angle in the high rotation region is combined with the method of the second embodiment (steps 112 to 116 in FIG. 9). It is not limited to this. That is, the method for advancing the injection start crank angle may be combined with the method of the first embodiment (step 100 in FIG. 8).

(Modification 2)
Furthermore, the technique for preventing the increase in the number of PM particles by advancing the injection start crank angle may be used independently from the technique of the first or second embodiment. That is, when steps 112 to 122 are removed from the routine shown in FIG. 13 and the injection end crank angle is less than the limit crank angle (step 108), the process of always advancing the injection start crank angle (steps 130 to 134). ) May be executed.

  The timing chart shown in FIG. 14 represents the operation of the modification 2 described above. In FIG. 14, the injection start crank angle 91 is advanced as the engine rotational speed NE increases after the rich control is started. As a result, the injection end crank angle 92 maintains a value larger than the limit crank angle 78 over the entire execution period of the rich control. As shown in this example, according to Modification 2 described above, as in the case of Embodiments 1 to 3, the rich control can be appropriately completed in the acceleration environment without increasing the number of PM particles. it can.

Embodiment 4 FIG.
Next, a fourth embodiment of the present invention will be described with reference to FIGS. The present embodiment can be realized by causing the ECU 70 to execute the routine shown in FIG. 16 instead of the routine shown in FIG. 8 using the configuration shown in FIG. However, in this embodiment, since the fuel pressure is not changed, the fuel pressure variable mechanism 26 can be omitted.

[Features of Embodiment 4]
(Rich air / fuel ratio at the start of rich control)
FIG. 15 is a timing chart for explaining features of rich control performed in the present embodiment. In FIG. 15, a dashed-dotted waveform 80 shown in the uppermost column represents a change in the air-fuel ratio used in rich control in the first to third embodiments. This waveform 80 represents a state in which the air-fuel ratio is enriched to about 12.5 with the start of the rich control, and then the enrichment is weakened and the air-fuel ratio is changed to 14.0. Also in the present embodiment, when the rich control is requested during the execution of the lean control, the change along the waveform 80 is given to the air-fuel ratio.

  A waveform 93 indicated by a solid line in FIG. 15 represents a change in the air-fuel ratio used when rich control is required in the acceleration process in the present embodiment. This waveform 93 represents a state in which the air-fuel ratio is greatly enriched to about 11.5 to 12.0 at the start of rich control and then changed to about 14.0. The waveform 93 also shows that the rich control end time is earlier than in the first to third embodiments because the air-fuel ratio is greatly enriched at the start of the rich control.

  When rich control is started while the engine speed NE is increasing, the NE immediately after the start is not so fast. Therefore, immediately after the start of the rich control, a relatively long time from the start of fuel injection until the crank angle reaches the limit crank angle is secured. Under such circumstances, even if the air-fuel ratio is greatly enriched, the injection end crank angle is difficult to fall below the limit crank angle.

  In the present embodiment, the air-fuel ratio (11.5 to 12.0) set at the start of the rich control consumes the stored oxygen of the three-way catalyst 41, that is, the rich exhaust gas downstream of the three-way catalyst 41. Is maintained until it blows through. During this time, the engine rotational speed NE continues to increase, so that the injection end crank angle gradually approaches the limit crank angle.

  However, the air-fuel ratio of 11.5 to 12.0 indicates that the injection end crank angle is the limit crank regardless of the increase of NE until the stored oxygen of the three-way catalyst 41 is consumed after the start of the rich control. It is a value set as a value that does not fall below the corner. Therefore, in the present embodiment, as shown in the lowermost column of FIG. 15, the injection end crank angle 94 is limited while the air-fuel ratio is maintained at 11.5 to 12.0 after the rich control is started. It can be kept in a region larger than the crank angle 78.

  The stored oxygen of the three-way catalyst 41 is consumed by reacting with a reducing agent such as HC and CO contained in rich exhaust gas. The concentration of the reducing agent in the exhaust gas increases as the air-fuel ratio of the exhaust gas becomes richer. For this reason, the time required to consume the stored oxygen of the three-way catalyst 41 becomes shorter as the exhaust gas is richer.

  In the present embodiment, in the rich control in the acceleration process, the air-fuel ratio at the start of the rich control is enriched compared to the rich control that interrupts the lean control. For this reason, in the rich control in the acceleration process, the stored oxygen of the three-way catalyst 41 can be consumed in a shorter time than in the case of the rich control that interrupts the lean control. Since this time can be shortened, the time required for the rich control in the acceleration process is naturally shorter than the time required for the rich control that interrupts the lean control.

  When the stored oxygen of the three-way catalyst 41 is consumed and the air-fuel ratio is changed to about 14.0, the in-cylinder injection time τ becomes shorter at that time. For this reason, the injection end crank angle 94 is temporarily separated above the limit crank angle 78. Thereafter, when the engine speed NE continues to increase, the injection end crank angle 94 approaches the limit crank angle 78 again. If this state continues for a long period of time, the injection end crank angle 94 is eventually delayed to the limit crank angle 78.

  However, in this embodiment, the time required for rich control in the acceleration process is shortened as described above. For this reason, according to the present embodiment, when rich control is started in the course of acceleration, the control can be terminated before the engine speed NE increases significantly. As a result, according to the present embodiment, it is possible to effectively prevent the injection end crank angle from falling below the limit crank angle 78 during execution of rich control accompanying acceleration.

(Routine process of the fourth embodiment)
FIG. 16 is a flowchart of a routine executed by the ECU 70 in the present embodiment. This routine is repeatedly started at a predetermined cycle after the internal combustion engine 10 is started, similarly to the routine shown in FIG. In FIG. 16, the same steps as those shown in FIG. 9 are denoted by the same reference numerals, and detailed description thereof is omitted.

  In the routine shown in FIG. 16, first, in step 100, it is determined whether or not a rich control request due to out-of-region has occurred. If this determination is affirmative, next, the engine speed NE and the coolant temperature ethw of the internal combustion engine 10 are read (step 140).

  Next, the target air-fuel ratio at the start of rich control is set (step 142). Specifically, first, an appropriate value within the range of 11.5 to 12.0 is calculated as the rich-side guard value based on the engine speed NE and the coolant temperature ethw. However, the rich-side guard value is a rich-side limit air-fuel ratio at which purification in the internal combustion engine 10 can be guaranteed. Next, the rich-side guard value calculated in this way is set as the target air-fuel ratio at the start of rich control.

  When the above process is completed, the in-cylinder injection amount q corresponding to the set target air-fuel ratio is calculated by the process of step 102. Next, in step 104, the in-cylinder injection time τ is calculated. Further, at step 106, an injection end crank angle corresponding to the above τ is calculated. Then, in step 108, it is determined whether or not the above-mentioned injection end crank angle is below the limit crank angle. These processes are performed in the same manner as the processes in steps 102 to 108 shown in FIG.

  In the routine shown in FIG. 16, when the condition of step 108 is satisfied, it can be determined that the fuel injection does not end up to the limit crank angle if the rich side guard value is set to the target air-fuel ratio. In this case, in step 112, a target injection period for ending the fuel injection by the limit crank angle is calculated. Next, the target injection time corresponding to the target injection period is calculated by the process of step 114. These processes are performed in the same manner as the processes in steps 112 and 114 shown in FIG.

  When the above processing is completed, the target injection time calculated in step 114 is then replaced with the in-cylinder injection time τ (step 146). As a result, the in-cylinder injection time τ calculated in step 104, that is, the in-cylinder injection time τ corresponding to the rich side guard value is corrected to the in-cylinder injection time τ for completing the injection at the limit crank angle. .

  When it is determined in step 108 in the routine shown in FIG. 16 that the injection end crank angle is not less than the limit crank angle, fuel injection is performed by the limit crank angle even if the rich side guard value is used as the target air-fuel ratio. It can be determined that it will end. In this case, since it is not necessary to correct the in-cylinder injection time τ, the processing of steps 112, 114 and 144 is jumped.

  When the above processing ends, next, the start of rich control is commanded (step 146). As a result, the first stage rich control for consuming the stored oxygen of the three-way catalyst 41 is started.

  Next, in this routine, it is determined whether or not the stored oxygen of the three-way catalyst 41 has been consumed (step 148). Specifically, it is determined here whether or not the output of the oxygen sensor 45 located downstream of the three-way catalyst 41 is richly inverted. As a result, when it is determined that the rich inversion has not been detected yet, it is determined that the stored oxygen of the three-way catalyst 41 has not been consumed yet, and the current process is terminated. Then, until the rich inversion of the oxygen sensor 45 is detected, the processing after step 100 is repeated, and the first-stage rich control is continued.

  If it is determined in step 148 that the stored oxygen has been consumed, the first rich control is terminated and a command is issued to start the second rich control (step 150). Thereafter, the ECU 70 performs the second stage rich control using an air-fuel ratio of about 14.0 to release NOx of the NSR catalyst 42 by another routine.

  As described above, according to the present embodiment, the rich control requested in the acceleration process can be completed in a shorter time than the rich control executed by interrupting the lean control. For this reason, according to the present embodiment, fuel injection can be completed up to the limit crank angle even during rich control in the acceleration process, and an increase in the number of PM particles can be effectively prevented.

Embodiment 5. FIG.
Next, a fifth embodiment of the present invention will be described with reference to FIG. The present embodiment can be realized by causing the ECU 70 to execute the routine shown in FIG. 17 instead of the routine shown in FIG. 8 using the configuration shown in FIG. However, since the fuel pressure is not changed in this embodiment, the fuel pressure variable mechanism 26 can be omitted.

[Features of Embodiment 5]
(Method for determining the air-fuel ratio at the start of rich control)
In the above-described fourth embodiment, when rich control is requested in the acceleration process, the target air-fuel ratio (in-cylinder injection amount) at the start of rich control is determined by the following procedure.
1. First, the target air-fuel ratio is set to the rich side guard value.
2. If the target air-fuel ratio is set to the rich-side guard value, if the fuel injection does not end before the limit crank angle, the in-cylinder injection time τ is shortened so that the fuel injection ends at the limit crank angle.

On the other hand, this embodiment is characterized in that the target air-fuel ratio (in-cylinder injection amount) at the start is determined by the following procedure when rich control is performed in the acceleration process.
1. Set the target injection period to finish fuel injection at the limit crank angle.
2. In the target injection period, when the air-fuel ratio of the air-fuel mixture becomes richer than the rich-side guard value, the target air-fuel ratio is set as the rich-side guard value.

(Routine processing of the fifth embodiment)
FIG. 17 is a flowchart of a routine executed by the ECU 70 in the present embodiment. This routine is repeatedly started at a predetermined cycle after the internal combustion engine 10 is started, similarly to the routine shown in FIG. In FIG. 17, the same steps as those shown in FIG. 16 are denoted by the same reference numerals, and detailed description thereof is omitted.

  In the routine shown in FIG. 17, when it is determined in step 100 that the rich control in the acceleration process is required, the process of step 112 is performed through step 140. Here, specifically, the injection period for ending injection at the limit crank angle is calculated as the target injection period.

Next, the air-fuel ratio corresponding to the target injection period is calculated as the target air-fuel ratio (step 152). Here, specifically, the following processes are sequentially executed.
1. The target injection period is converted into in-cylinder injection time τ based on the engine speed NE.
2. The in-cylinder injection amount q corresponding to the in-cylinder injection time τ is calculated.
3. The air-fuel ratio is calculated from the in-cylinder injection amount q and the intake air amount Ga detected by the air flow meter 54.
4). Set the air / fuel ratio to the target air / fuel ratio.

  When the above processing is completed, it is next determined whether or not the target air-fuel ratio is smaller than the rich-side guard value (step 154). When this condition is satisfied, it can be determined that the target air-fuel ratio has entered a rich region where purification cannot be guaranteed. In this case, since the target air-fuel ratio cannot be used as it is, the target air-fuel ratio is replaced with the rich guard value (step 156).

  On the other hand, if the condition of step 154 is not satisfied, it can be determined that the target air-fuel ratio is a purifiable air-fuel ratio. In this case, since the target air-fuel ratio can be used as it is, step 156 is jumped.

  After the above processing is completed, the in-cylinder injection time τ corresponding to the target air-fuel ratio is calculated by the processing of steps 102 and 104, and the first rich control is performed by the processing of steps 146 and 148. When the stored oxygen of the three-way catalyst 41 is consumed, the second rich control is started by the processing in steps 148 and 150.

  According to the above processing, the air-fuel ratio at the start of rich control can be enriched without increasing the number of PM particles, as in the case of the fourth embodiment. For this reason, according to the present embodiment, the same effect as in the fourth embodiment can be achieved.

Embodiment 6 FIG.
Next, a sixth embodiment of the present invention will be described with reference to FIGS. The present embodiment can be realized by causing the ECU 70 to execute the routine shown in FIG. 19 instead of the routine shown in FIG. 8 using the configuration shown in FIG. However, since the fuel pressure is not changed in this embodiment, the fuel pressure variable mechanism 26 can be omitted.

[Features of Embodiment 6]
(Air-fuel ratio of the second stage rich control)
In the fifth embodiment described above, when the rich control is required in the acceleration process, the time required for the rich control is shortened by greatly enriching the air-fuel ratio used in the first stage rich control. From the viewpoint of shortening the time required for the rich control, it is desirable to enrich the air-fuel ratio as much as possible not only in the first stage but also in the second stage rich control. However, the air-fuel ratio during the rich control affects the fuel consumption characteristics of the internal combustion engine 10. For this reason, enriching the air-fuel ratio more than necessary is not desirable from the viewpoint of obtaining good fuel efficiency characteristics.

  Under circumstances where the engine speed NE increases due to normal acceleration, the fuel injection can be terminated before the limit crank angle over the entire rich control by greatly enriching the air-fuel ratio of the first rich control. it can. Therefore, under such circumstances, it is not necessary to further enrich the air-fuel ratio used in the second stage rich control.

  However, when the internal combustion engine 10 is mounted on a vehicle together with an automatic transmission, kickdown may occur during acceleration, and the engine speed NE may increase rapidly. In order to terminate the injection to the limit crank angle over the entire range of the rich control under such a sudden rise, in addition to the first stage rich control, the second stage rich control also sets the air-fuel ratio. It is effective to make it greatly rich. Therefore, in the present embodiment, when rich control is requested in the acceleration process, it is determined whether or not automatic transmission (kick down) of the transmission is predicted. When automatic shift is predicted, the air-fuel ratio is greatly enriched not only in the first stage rich control but also in the second stage rich control.

(Operation of Embodiment 6)
FIG. 18 is a timing chart showing an operation when rich control is requested in the acceleration process and occurrence of kickdown is predicted during execution of rich control. In FIG. 18, an alternate long and short dash line waveform 80 shown in the uppermost column indicates an air-fuel ratio change pattern used when rich control is requested during execution of lean control in the present embodiment.

  In the example of FIG. 18, rich control in the acceleration process is started at time t1. In the present embodiment, as in the case of the fifth embodiment, in the rich control in the acceleration process, the air-fuel ratio is greatly enriched unconditionally at the start (about 11.5 to 12.0). As a result, in the example shown in FIG. 18, the injection end crank angle 98 substantially coincides with the limit crank angle 78 after time t1.

  The example illustrated in FIG. 18 illustrates a state where the required torque gradient 96 reaches the threshold value x in the execution process of the rich control. In this case, the ECU 70 predicts the occurrence of kickdown, and maintains the state in which the air-fuel ratio is greatly enriched even after time t2 when the second-stage rich control is to be performed. Therefore, in FIG. 18, the state where the injection end crank angle 98 substantially coincides with the limit crank angle 78 is maintained after time t2.

  If kickdown occurs at time t3, then the engine speed NE suddenly increases (see waveform 97). When the engine rotational speed NE shows such a rapid increase, the amount of fuel that can be injected up to the limit crank angle 78 decreases rapidly. And if the amount of fuel decreases rapidly, the air-fuel ratio will increase rapidly. For this reason, the air-fuel ratio during rich control shown in FIG. 18 shows a rapid increase after time t3 (see waveform 95).

  When the air-fuel ratio changes like the waveform 95 in the process of rich control, a large amount of reducing agent can be sent to the NSR catalyst 42 between times t2 and t3. That is, in this case, the NOx release of the NSR catalyst 42 can be rapidly advanced between the times t2 and t3. For this reason, even after the time t3, even if the air-fuel ratio suddenly goes to stoichiometric, a sufficient amount of NOx can be released by the time the rich control is completed. Therefore, according to the present embodiment, even when rich control is required in the process of acceleration and kickdown of the automatic transmission occurs in the process of rich control, without greatly increasing the number of PM particles, The NOx storage capacity of the NSR catalyst 42 can be properly recovered.

(Routine processing of the sixth embodiment)
FIG. 19 is a flowchart of a routine executed by the ECU 70 in the present embodiment. This routine is repeatedly started at a predetermined cycle after the internal combustion engine 10 is started, similarly to the routine shown in FIG. The routine shown in FIG. 19 is the same as the routine shown in FIG. 16 except that step 160 is inserted after step 146. In the following, steps common to both are denoted by common reference numerals and detailed description thereof is omitted.

  In the routine shown in FIG. 19, it is determined whether the shift of the automatic transmission is predicted or, more specifically, the kickdown associated with acceleration is predicted after the rich control of the first stage is started in step 146. (Step 160). Since the technology for predicting the kickdown of the automatic transmission is well known in the field of vehicle control, detailed description thereof is omitted here. In step 160, the ECU 70 determines whether or not a kick down is predicted by the known method.

  If it is determined that the occurrence of kickdown is not predicted, the processing from step 148 onward is performed thereafter. In this case, the system according to the present embodiment operates in the same manner as in the fifth embodiment.

  On the other hand, if the occurrence of kick-down is predicted in step 160, the routine is terminated without performing steps 148 and 150 thereafter. In this case, the rich control at the first stage and the second stage are not distinguished from each other until the end of the rich control is determined, that is, until it is determined at step 100 that the request for the rich control has not occurred. Steps 100 to 146 are repeatedly executed. According to the above processing, until the oxygen sensor 46 downstream of the NSR catalyst 42 is richly inverted, the rich control is continued so that the injection end crank angle is aligned with the limit crank angle. As a result, the NOx release of the NSR catalyst 42 proceeds at a high speed, and the NOx occlusion capacity is recovered within a short period without significantly deteriorating the exhaust characteristics.

(Modification of Embodiment 6)
By the way, in the above-described sixth embodiment, a method of maintaining the state in which the air-fuel ratio is greatly enriched even after the rich control for consuming the stored oxygen of the three-way catalyst 41 is completed is described in the fifth embodiment (FIG. 17). Steps 100 to 146) are combined with the above method. However, the present invention is not limited to this. That is, the method of maintaining the air-fuel ratio in a greatly rich state may be combined with the method of the fourth embodiment (steps 100 to 146 in FIG. 16).

DESCRIPTION OF SYMBOLS 10 Internal combustion engine 19 Piston 22 In-cylinder injection valve 26 Fuel pressure variable mechanism 28 Water temperature sensor 30 Crank angle sensor 41 Three-way catalyst 42 NOx storage reduction catalyst (NSR catalyst)
45, 46 Oxygen sensor 54 Air flow meter

Claims (1)

A control device for an internal combustion engine, comprising: a NOx occlusion reduction catalyst arranged in an exhaust passage;
Stoichiometric control for operating the internal combustion engine in a stoichiometric air-fuel ratio region including a stoichiometric air-fuel ratio;
Lean control for operating the internal combustion engine in an air-fuel ratio range leaner than the stoichiometric air-fuel ratio range;
And rich control for selectively passing exhaust gas belonging to an air-fuel ratio region richer than the stoichiometric air-fuel ratio region to the NOx storage-reduction catalyst in order to release NOx stored in the NOx storage-reduction catalyst. Run,
When the rich control is executed by interrupting the stoichiometric control while the engine speed is increasing, compared to the case where the rich control is executed by interrupting the lean control,
(1) Increasing the pressure of fuel supplied to the in-cylinder injection valve;
(2) accelerating the fuel injection start crank angle of the in-cylinder injection valve, and (3) further enriching the air-fuel ratio at the start of the rich control,
A control device for an internal combustion engine that executes at least one of the following.

Images