JPH10238415A - Air-fuel ratio control system of internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce NOx and to restrict a deterioration in derivability by varying an EGR amount, taking into consideration of variation of real air-fuel ratio by fed back correction of air-fuel ratio. SOLUTION: While a part of an exhaust gas exhausted to an exhaust passage 23 of an engine 11 is recirculated in an intake air passage 22 side, detected data from an O2 sensor 29 is monitored to carry out a feedback control such that air fuel ratio of air-fuel mixture supplied to the engine 11 corresponds to target air-fuel ratio. An ECU 51 controls a recirculation amount of the exhaust gas based on an operating state of the engine 11, and makes the recirculation amount of the exhaust gas in the case where a feedback correction factor is set to a rich side with respect to the target air-fuel ratio greater than that of the exhaust gas in the case where the feedback correction factor is set to a lean side with respect to the target air-fuel ratio.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an air-fuel ratio control device for an internal combustion engine, and more particularly, to an exhaust gas recirculation (EGR).
The present invention relates to an air-fuel ratio control device for an internal combustion engine that performs feedback control of the air-fuel ratio of the internal combustion engine in consideration of control.

[0002]

2. Description of the Related Art Conventionally, exhaust gas recirculation (hereinafter referred to as EGR) control for returning a part of exhaust gas to an intake side has been performed for the purpose of reducing NOx in the exhaust gas and improving the fuel consumption rate. For example, in the technology described in Japanese Patent Application Laid-Open No. 60-173361, a change in the air-fuel ratio is monitored by a lean sensor, and when the base air-fuel ratio changes, the EGR amount is also corrected according to the change. . EGR amount is N
It is better to increase as much as possible in terms of reduction of Ox and improvement of the fuel consumption rate. However, if it is too large, a misfire will occur. Therefore, as shown in FIG. In this conventional example, misfire is prevented and drivability is improved based on the relationship between the base air-fuel ratio and the EGR amount.

[0003]

Incidentally, in the air-fuel ratio control of the internal combustion engine, generally, the actual air-fuel ratio is feedback-controlled so as to coincide with the target air-fuel ratio corresponding to the operation state of the internal combustion engine. Under the feedback control, the actual air-fuel ratio is not actually constant because the base air-fuel ratio itself always repeats lean / rich.

However, in the above-described conventional air-fuel ratio control device,
Although the EGR amount is simply corrected based on the change in the base air-fuel ratio, it is effective in preventing misfire. However, it has not been possible to make a fine correction of the EGR amount in accordance with the operating state of the engine.

That is, in the air-fuel ratio feedback control, usually, a correction value is given to make the actual air-fuel ratio coincide with the target air-fuel ratio (base air-fuel ratio) and the feedback correction is performed. Fluctuate between Here, when the EGR amount is increased, when the actual air-fuel ratio leans to the lean side, a surge due to misfire easily occurs as described above. Therefore, the EGR amount must be set in consideration of the feedback correction at the time of the maximum lean. No. Conversely, when the actual air-fuel ratio leans to the rich side,
It means that the amount of R can be increased. However, at present, the amount of EGR is limited to the extent that no surge occurs, so that a fine E
The GR amount cannot be controlled, and the effect of the EGR cannot be fully utilized.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned circumstances, and has as its object to reduce NOx and improve drivability by changing the EGR amount in consideration of a change in actual air-fuel ratio due to air-fuel ratio feedback correction. It is an object of the present invention to provide an air-fuel ratio control device for an internal combustion engine which can suppress deterioration.

[0007]

In order to achieve the above object, according to the first aspect of the present invention, a part of exhaust gas discharged to an exhaust passage of an internal combustion engine is recirculated to an intake passage side of the engine. Meanwhile, in an air-fuel ratio control device for an internal combustion engine that monitors detection data from an air-fuel ratio detection unit and feedback-controls an air-fuel ratio of an air-fuel mixture supplied to the internal combustion engine to a target air-fuel ratio, based on an operation state of the engine. Recirculation amount control means for controlling the amount of recirculation of the exhaust gas; and a recirculation amount of the exhaust gas when the feedback correction coefficient is set to the rich side with respect to the target air-fuel ratio. Exhaust gas recirculation amount correction means for correcting the controlled exhaust gas recirculation amount so as to be larger than the exhaust gas recirculation amount when the correction coefficient is set to the lean side. It has as its gist that there was Unishi.

According to the second aspect of the present invention, in addition to the configuration of the first aspect, the exhaust gas recirculation amount correcting means is variably set to a rich or lean side with respect to the target air-fuel ratio. The gist of the invention is to provide a transition estimating means for estimating the transition of the feedback correction coefficient, and to correct the exhaust gas recirculation amount prior to the transition of the estimated feedback correction coefficient.

According to a third aspect of the present invention, in addition to the configuration of the second aspect, the exhaust gas recirculation amount correcting means changes the feedback correction coefficient estimated by the changing means. The gist is that when the actual feedback correction coefficient fluctuates earlier than before, the exhaust gas recirculation amount correction is performed based on the actual feedback correction coefficient fluctuation.

In such a configuration, the recirculation amount control means controls the recirculation amount of the exhaust gas based on the operating state of the internal combustion engine, and sets a feedback correction coefficient on the rich side with respect to the target air-fuel ratio in the feedback control. The recirculation amount of the exhaust gas is controlled so as to be larger than the recirculation amount of the exhaust gas when the feedback correction coefficient is set to the lean side with respect to the target air-fuel ratio.

According to the second aspect of the invention, in addition to the operation of the first aspect, the exhaust gas recirculation correction is executed prior to the feedback correction coefficient estimated by the transition estimating means. I have. This takes into account the time lag until the recirculated exhaust gas circulates again into the combustion chamber, pre-reads the state of the actual air-fuel ratio and executes control of the exhaust gas recirculation means earlier by the delay. is there.

According to the third aspect of the present invention, in addition to the effect of the second aspect of the present invention, when an actual change occurs due to the estimated change of the feedback correction coefficient, the change of the feedback correction value On the basis of this, the recirculation correction of the recirculated exhaust gas is executed.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which an air-fuel ratio control device for an internal combustion engine according to the present invention is embodied in a gasoline engine will be described in detail with reference to FIGS.

FIG. 1 is a schematic configuration diagram showing an air-fuel ratio control device of a four-cylinder engine mounted on a vehicle in this embodiment. As shown in FIG. 1, an engine 11 as an internal combustion engine takes in outside air from an air cleaner 13 through an intake passage 12. Air cleaner 13
An air flow meter 14 is provided on the downstream side of
The intake air amount Q is detected. An intake air temperature sensor 15 is provided between the air cleaner 13 and the air flow meter 14, and the intake air temperature sensor 15 detects the intake air temperature THA. Also, the engine 11 takes in the outside air and
The fuel injected from the injector 16 provided for each cylinder is taken in the vicinity of a. Then, a mixture of the taken-in fuel and outside air is introduced into the combustion chamber 11a through the intake valve 18 provided for each cylinder, and exploded and burned in the combustion chamber 11a to obtain a driving force. The exhaust gas after explosion and combustion is led out of the combustion chamber 11a to an exhaust passage 23 where exhaust manifolds for respective cylinders are gathered via an exhaust valve 22, and is discharged to the outside.

In the middle of the intake passage 12, there is provided a throttle valve 25 which is opened and closed by operation of an accelerator pedal 41. By opening and closing the throttle valve 25, the amount of intake air to the intake passage 12 is adjusted. A surge tank 26 for smoothing the pulsation of the intake air is provided downstream of the throttle valve 25. The surge tank 26
Is provided with an intake air pressure sensor 28 that detects the intake air pressure (intake pressure) PiM. In the intake passage 12, near the throttle valve 25, a throttle sensor 27 for detecting the opening degree (throttle opening degree) TA is provided.

On the other hand, in the middle of the exhaust passage 23, an oxygen sensor (air-fuel ratio sensor) 2 for detecting the oxygen concentration OX in the exhaust gas is provided.
9 are provided. The engine 11 has a water temperature sensor 3 for detecting the temperature (cooling water temperature) THW of the cooling water.
0 is provided.

An ignition signal distributed by a distributor 32 is applied to an ignition plug 31 provided for each cylinder of the engine 11. The distributor 32 distributes the high voltage output from the igniter 35 to each of the ignition plugs 31 in synchronization with the crank angle of the engine 11, and the ignition timing of each of the ignition plugs 31 is the high voltage output timing from the igniter 35. Is determined by

The distributor 32 is provided with a rotation speed sensor 36 for detecting the rotation speed (engine rotation speed) NE of the engine 1 from the rotation of a rotor (not shown) built in the distributor 32. The distributor 32 is also provided with a crank angle sensor 37 that detects a change in the crank angle of the engine 11 at a predetermined rate according to the rotation of the rotor.

The automatic transmission 39 drivingly connected to the engine 11 is provided with a vehicle speed sensor 40. The vehicle speed sensor 40 detects the vehicle speed (vehicle speed) SP at that time.
D can be detected and a signal indicating the value can be output. The accelerator pedal 41 is provided with an accelerator sensor 42 for detecting the accelerator opening ACCP.

Further, in this embodiment, a known exhaust gas circulation (EGR) device 43 is provided. This EG
The R device 43 includes an EGR passage 44 and an EGR valve 45 provided in the middle of the EGR passage 44. The EGR passage 44 is connected to the intake passage 1 downstream of the throttle valve 25.
2 and the exhaust passage 23. When the EGR valve 45 is opened, part of the exhaust gas discharged to the exhaust passage 23 is
Flows to The exhaust gas flows to the intake passage 12 via the EGR valve 45. That is, part of the exhaust gas is E
It is recirculated into the intake air-fuel mixture by the GR device 43. At this time, the recirculation amount of the exhaust gas is adjusted by adjusting the opening degree of the EGR valve 45.

Each of the sensors 14, 15, 27-30, 3
The operating state of the engine 11 and the like are appropriately detected by 6, 37, 40, 42, and the like, and these constitute an operating state detecting means.

Next, the electrical configuration of the present embodiment will be described with reference to the block diagram of FIG. ECU 51
Denotes a central processing unit (CPU) 52, a read-only memory (ROM) 5 in which predetermined control programs, maps, and the like are stored in advance.
3. Random access memory (RAM) 54 for temporarily storing the calculation results and the like of CPU 52, backup RAM for storing pre-stored data even after power is turned off
55 and the like. The ECU 51 also connects these components to the external input circuit 56, the external output circuit 57, etc.
Are configured as logical operation circuits connected with each other.

The external input circuit 56 includes the air flow meter 14, the intake air temperature sensor 15, the throttle sensor 2
7, intake pressure sensor 28, oxygen sensor 29, water temperature sensor 3
0, a rotation speed sensor 36, a crank angle sensor 37, a vehicle speed sensor 40, an accelerator sensor 42, and the like are connected to each other. Then, the CPU 52 transmits each of the sensors 14, 15, 27 to 30, 36, 37, 4 via the external input circuit 56.
The output signals from 0 and 42 are read as input values. CP
U52 is connected to the injector 16, the igniter 35 and the EGR connected to the external output circuit 57 based on these input values.
The valve 45 and the like are suitably controlled. The learning values and flags in this embodiment are stored in the RAM 54 or the backup RAM 55 described above.

Next, among the processes executed by the ECU 51, the "air-fuel ratio A / F feedback routine" for matching the stoichiometric air-fuel ratio with the actual air-fuel ratio is shown in FIGS. 3, 4 and 11 (a). A description will be given based on this. This routine is executed by interruption every predetermined crank angle.

First, in step 101, the ECU 41
Reads a detection signal (for example, cooling water temperature THW, throttle opening TA, engine speed NE, oxygen concentration OX, etc.) output from each of the sensors in step 101. Then, in step 102, it is determined whether or not the feedback processing condition of the air-fuel ratio A / F is satisfied based on the signals from these sensors. The feedback processing conditions refer to basic conditions such as, for example, whether a predetermined time has elapsed after the engine is started, and whether the coolant temperature is equal to or higher than a predetermined temperature. If the feedback condition is not satisfied, a feedback correction coefficient FAF is determined in step 103.
Is set to "1.0", the skip counter Cskip is cleared in step 104, and the process is temporarily terminated.

On the other hand, when it is determined that the feedback condition is satisfied, in the next step 105, the current air-fuel ratio A is determined based on the signal VO2 indicating the oxygen concentration OX read this time.
It is determined whether / F is lean with respect to the stoichiometric air-fuel ratio. When it is determined that the current air-fuel ratio A / F is on the lean side, it is necessary to control the feedback correction coefficient FAF so as to be on the rich side. So first,
At step 106, it is determined whether or not the count value of the delay counter CDLY is larger than zero.

If it is determined that the count value is larger than 0, the process proceeds to step 107 where the delay counter CDL is determined.
Y is cleared once, and the count value is decremented by one in step 108. If it is determined in step 106 that the count value is 0 or less, the process immediately proceeds to step 10.
Move to 8. That is, if the count value of the delay counter CDLY is larger than 0, 0 is set to an initial value and then decremented. If the count value as a result of the previous routine is smaller than 0, the previous count value is further decremented.

Next, at step 109, it is determined whether or not the count value of the delay counter CDLY is smaller than a predetermined lean count value TDL. If it is determined that the count value of the delay counter CDLY is smaller than the predetermined lean count value TDL, the count value of the delay counter CDLY is guarded by the predetermined lean count value TDL in step 110 (delay). The count value of the counter CDLY is assumed to be the lean count value TDL), the rich flag F1 is set to “0” to indicate that the vehicle is in the lean state in step 111, and then the process proceeds to step 112. On the other hand, if it is determined in step 109 that the count value of the delay counter CDLY is equal to or greater than the predetermined lean count value TDL, the process directly proceeds to step 112 without setting the rich flag F1 to “0”.

In step 105, when it is determined that the current air-fuel ratio A / F is not on the lean side, that is, when the air-fuel ratio A / F is on the rich side, the feedback correction coefficient FAF is set to the lean side. You need to control. In that case, first, at step 113, it is determined whether or not the count value of the delay counter CDLY is smaller than 0. Here, when it is determined that the count value is smaller than 0, the delay counter C is determined in step 114.
DLY is cleared once, and the count value is incremented by one in step 114. If it is determined in step 113 that the count value is 0 or more, the process immediately proceeds to step 115. That is, the delay counter C
If the count value of DLY is smaller than 0, 0 is incremented as an initial value, and if the count value as a result of the previous routine is larger than 0, further increment is performed.

Next, at step 116, it is determined whether or not the count value of the delay counter CDLY is larger than a predetermined rich side count value TDR. If it is determined that the count value of the delay counter CDLY is larger than the predetermined rich side count value, the process proceeds to step 11.
7, the count value of the delay counter CDLY is guarded by the rich count value TDR (the count value of the delay counter CDLY is changed to the rich count value TDR).
), The rich flag F1 is set to “1” to indicate the rich state in step 118, and then the process proceeds to step 112. On the other hand, if it is determined in step 116 that the count value of the delay counter CDLY is equal to or greater than the predetermined rich-side count value TDR, the rich flag F
1 is not set to “1”, and the process immediately proceeds to the processing of step 112.

In step 112, it is determined whether the rich flag F1 has been inverted. Here, the rich flag F1
Is determined to be inverted, the feedback correction coefficient FAF is inverted from the rich side to the lean side in the current process or from the rich side to the reverse side. If it is determined that the rich flag F1 has been inverted, step 119
It is determined whether or not the currently set rich flag F1 is "0". On the other hand, if the rich flag F1 is not inverted at step 112, it is determined that the air-fuel ratio A / F is on the same side as the previous time, and the routine proceeds to step 120, where the currently set rich flag F1 is set to "0". Is determined.

If it is determined in step 119 that the currently set rich flag F1 is "0", it is determined that the air-fuel ratio A / F has been inverted from the rich side to the lean side in this process, and the process proceeds to step 119. It moves to 121. In step 121, a predetermined rich side skip value RSR
Is added to the feedback correction coefficient FAF to set a new feedback correction coefficient FAF (rich-side skip). That is, since the air-fuel ratio A / F is on the lean side, it is necessary to bring it to the direction of the stoichiometric air-fuel ratio, that is, the rich side. As shown in FIG. 11A, the behavior of the feedback correction coefficient FAF at this time appears as a rich skip of the skip value RSR at the p position.

Next, in step 122, a flag (hereinafter referred to as a rich skip flag) FRskip indicating that rich skip has been performed is set to "1". And
In step 123, the skip counter Cskip is incremented by one. Thereafter, in step 124, the feedback correction coefficient FAF is guarded with a predetermined value so as not to go too far to the rich side or lean side, and in step 125, the skip counter Cskip is guarded so as to be within the predetermined count value, and the process is temporarily terminated.

If it is determined in step 119 that the currently set rich flag F1 is "1", it is determined that the air-fuel ratio A / F has been inverted from the lean side to the rich direction in this process, and the process proceeds to step 119. Go to 126. In step 126, a predetermined lean side skip value RS
L is subtracted from the feedback correction coefficient FAF to set a new feedback correction coefficient FAF (lean skip). That is, since the air-fuel ratio A / F is on the rich side, it is necessary to bring this to the direction of the stoichiometric air-fuel ratio, that is, the lean side. As shown in FIG. 11A, the behavior of the feedback correction coefficient FAF appears as a lean skip of the skip value RSL at the q position. Then, in step 127, a flag indicating that the lean-side skip has been performed (hereinafter, referred to as a lean skip flag) FLskip is set to “1”. Next, the process proceeds to step 123.

On the other hand, if it is determined in step 120 that the currently set rich flag F1 is "0", the air-fuel ratio A / F is determined to be in the lean state this time as in the previous case, and the process proceeds to step 128. Transition. Step 12
In step 8, a predetermined rich correction value KiR is added to the feedback correction coefficient FAF, and a new feedback correction coefficient F
Set AF. Then, in a step 129, the rich skip flag FRskip is set to “0”. Next, the process proceeds to step 123. As long as the air-fuel ratio A / F is not inverted to the rich state, the rich correction value KiR is integrated, so that the feedback correction coefficient FAF gradually increases.

If it is determined in step 120 that the currently set rich flag F1 is "1", the air-fuel ratio A / F is assumed to be in the rich state this time as in the previous case, and the process proceeds to step 130. Transition. Step 13
At 0, a predetermined lean correction value KiL is subtracted from the feedback correction coefficient FAF to set a new feedback correction coefficient FAF. Then, in step 131, the lean skip flag FLskip is set to “0”. Next, the process proceeds to step 123. As long as the air-fuel ratio A / F does not reverse to the lean state, the lean correction value KiL is integrated, so that the feedback correction coefficient FAF gradually decreases.

As described above, when the air-fuel ratio A / F is in a rich state, the feedback correction coefficient FAF is corrected to the lean side in order to make it close to the stoichiometric air-fuel ratio. Is done. When the air-fuel ratio A / F is inverted from the rich state to the lean state, the rich-side skip value RSR is added to the feedback correction coefficient FAF. Conversely, when the air-fuel ratio A / F is inverted from the lean state to the rich state, the lean-side skip is added. Subtract the value RSL.

Next, in the processing executed by the ECU 51, the average value of the time from the rich skip to the lean skip and the average time from the lean skip to the rich skip of the feedback correction coefficient FAF is calculated. FIG. 5 shows the calculation routine.
It will be described based on FIG. This routine is executed by interruption every predetermined crank angle.

In step 201, the ECU 51 determines whether or not the skip counter Cskip is 0. Here, if the skip counter Cskip is 0, it is determined that the skip has not yet been performed, and the area used for the feedback correction coefficient FAF cycle calculation in the RAM 54 is cleared in step 202. In step 201, the skip counter Csk
If it is determined that ip is not 0, it is determined in step 203 whether the lean skip flag FLskip has been set to “1”, that is, whether lean skip has been performed.

If it is determined that the lean skip has not been performed, it is determined in step 204 whether the rich skip flag FRskip has been set to "1", that is, whether the rich skip has been performed. You. If it is determined that the rich side skip has not been executed, the process is temporarily terminated.

On the other hand, if the rich skip is executed in step 204, the time at that time is set as the rich skip time TRt in step 205. Next, at step 206, it is determined whether or not the count value of the skip counter Cskip is one. Here, if the count value of the skip counter Cskip is 1, it is the first skip, and the process is temporarily terminated, assuming that there is not enough data to obtain the inter-skip period.

In step 206, the skip counter Csk
If the count value of ip is not 1, the rich skip was performed this time, and the previous time was lean skip. Therefore,
In step 207, the current rich skip time TR
The previous lean skip time TLt is subtracted from t, and the value is set as the lean integration time TL (see FIG. 11A). Next, in step 208, it is determined whether the count value of the skip counter Cskip is 3 or less.
When the count value is 3 or less, data of the lean integration time TL is not obtained twice or more, and no average value is obtained. Therefore, when the count value of the skip counter Cskip is 3 or less, the first lean integration time TL is directly used in step 209.
1 is set as the value of the average lean cycle time AVETL.

On the other hand, if the count value of the skip counter Cskip is 4 or more, in step 210, 1/2 of the sum of the previous average lean cycle time AVETL and the current lean integration time TL is calculated as the new average lean cycle time. AVE
Set as TL.

If it is determined in step 203 that the lean skip flag FLskip is set to "1", that is, if it is determined that the lean skip has been executed, then in step 211 the time at that time is set to the lean skip time TLt.
Is set as Next, at step 212, it is determined whether the count value of the skip counter Cskip is one. Here, if the count value of the skip counter Cskip is 1, it is the first skip, and the process is temporarily terminated, assuming that there is not enough data to obtain the inter-skip period. If the count value of the skip counter Cskip is not 1, the lean skip was performed this time, and the previous time was a rich skip. Therefore, in step 213, the previous rich skip time TRt is subtracted from the current lean skip time TLt, and the value is subtracted from the rich integration time Tt.
It is set as R (see FIG. 11A). Next, in step 214, it is determined whether the count value of the skip counter Cskip is 3 or less. The determination is made based on whether the count value is 3 or less for the same reason as described above.
Here, the count value of the skip counter Cskip is 3
If not, in step 215, the first rich integration time TR1 is set as the average rich cycle time AVETR.

On the other hand, if the count value of the skip counter Cskip is 4 or more, in step 216, the half of the sum of the previous average rich cycle time AVETR and the current rich integration time TR is added to the new average rich cycle time. AVE
Set as TR.

By the above processing, the time from the rich skip to the lean skip of the feedback correction coefficient FAF and the average rich (lean) cycle time A as the average value from the lean skip to the rich skip are obtained.
VETR and AVETL can be calculated. The average rich (lean) cycle times AVETR and AVETL are used to calculate the expected timing of the next skip of the feedback correction coefficient FAF described later.

Next, among the processes executed by the ECU 51, a "correction duty value ADEGR skip timing calculation routine" for calculating a correction duty value ADEGR skip timing obtained by correcting the EGR duty value DEGR is shown in FIGS. This will be described based on 11 (b). This routine is executed by interruption every predetermined crank angle. In the present embodiment, the EGR according to the operation state
Is calculated, the duty value DEGR is set as a reference value 1.0, and a correction coefficient is given to the reference value to calculate a correction duty value ADEGR. Since the correction duty value ADEGR follows the lean / rich skip behavior of the feedback correction coefficient FAF, the skip timing Tsk of the correction duty value ADEGR itself is used.
It is necessary to calculate ip.

Here, as a premise of the description of the "correction duty value ADEGR skip timing calculation routine", first, a "base duty value DEGR calculation routine" for calculating the base duty value DEGR will be described with reference to FIG. This routine is executed by interruption every predetermined crank angle.

Base duty value D in EGR control
EGR is the air flow meter 14 and the crank angle sensor 3
The rotational speed NE is determined based on an intake air amount QN per one rotation of the crankshaft calculated based on an output signal from the rotation speed sensor 7 and an output signal from the rotation speed sensor 36.

First, the ECU 51 reads the current intake air amount QN and the rotational speed NE in step 301,
02 is a map (not shown) for calculating the duty value of the EGR amount stored in the RAM 54 in advance.
Search for. Then, at step 303, the intake air amount QN
And a base duty value DEGR based on the duty value corresponding to the rotation speed NE.

In this embodiment, the ECU 51 executes the "correction duty value skip timing calculation routine" with the base duty value DEGR of the EGR calculated in such a routine as the reference value 1.0.

First, at step 401, it is determined whether or not feedback control of the air-fuel ratio A / F is being performed.
If not, the process ends once. If the feedback control is being performed, it is determined in step 402 whether the count value of the skip counter Cskip is 2 or less. When the count value is 2 or less, the process is temporarily terminated because the rich integration time TR and the lean integration time TL have not been determined yet. That is, FIG.
As shown in (b), until the skip of the feedback correction coefficient FAF is counted three times, the skip timing of the EGR cannot be predicted, so that the skip is not skipped and a constant value (a duty of 0.9 against the base value of 1.0) is used. Value DEG
R, which will be described later).

On the other hand, if the count value is not less than 2 in step 402, that is, if the count value is 3 or more, first in step 403 the rich skip flag FR
It is determined whether skip is set to "1", that is, whether rich skip has been performed. If the rich skip has been performed, the DEGR skip timing Tskip is calculated in step 404 by adding the average rich cycle time AVETR up to the previous time to the current time and further reducing the predetermined delay time Tdlay.

Here, the delay time Tdlay is a time from when the EGR valve 45 is operated to when the EGR gas is introduced into the combustion chamber 11a. That is, although the EGR gas is returned to the surge tank 26, it takes time for the EGR gas to reach the combustion chamber 11a from the surge tank 26.
The skip timing is set earlier in consideration of the delay time. This delay time Tdlay is experimentally obtained. In this way, the skip timing Tskip of the correction duty value ADEGR can be predicted.

The duty value skip timing Tskip will be described in more detail with reference to FIG.
For example, when the duty value skip timing Tskip1 is the third skip (the count value of the skip counter Cskip is 3) in the calculation of the feedback correction coefficient FAF, the average lean cycle time AVE relative to the current time timer1 at that timing.
It is obtained by adding the lean integration time TL2 corresponding to TL and further reducing the delay time Tdlay. Therefore, the duty value skip timing Tskip1 is skipped earlier by the delay time Tdlay than the expected fourth skip timing of the feedback correction coefficient FAF.

On the other hand, if the rich skip flag FRskip has not been set in step 403, it is determined in step 405 whether the lean skip flag FLskip has been set, that is, whether or not lean skip has been performed. Here, the lean skip flag F
If Lskip is not set, the process is temporarily terminated. On the other hand, if the lean skip has been performed, at step 406 the lean time AV
The duty value skip timing Tskip is calculated by adding the ETL and further reducing the predetermined delay time Tdlay.

Next, of the processing executed by the ECU 41, a "correction duty value ADEGR calculation routine" for calculating a correction duty value ADEGR by giving a correction coefficient to the base duty value DEGR is described with reference to FIGS. Will be described. This routine is executed by interruption every predetermined crank angle.

In step 501, it is determined whether or not the air-fuel ratio A / F has been feedback-corrected.
The duty value is set to 0, that is, the process is temporarily terminated without performing the EGR control. If the feedback correction coefficient FAF has been set in the subsequent step 503,
Count value of the skip counter Cskip is 2
It is determined whether: When the count value is 2 or less, the correction duty value ADEGR is set to 0.9 with respect to the reference value 1.0, and the process is temporarily terminated. Here, 0.
The value of 9 is a limit value that does not cause misfire at the time of the maximum lean of the air-fuel ratio A / F.

If the count value is not less than 2, it is determined in step 505 whether the count value is 3.
If the count value is not 3, it is determined that the count value is 4 or more, and it is determined in step 506 whether the rich skip flag FRskip has been set, that is, whether the rich skip of the feedback correction coefficient FAF has been performed. Is done. When it is determined that the rich skip has been performed, the lean flag F2 is set to "0" in step 507.

Then, in step 508, a value obtained by multiplying the delay coefficient DLAY by the rich correction value KiR is added to the feedback correction coefficient FAF, and this value is used as the correction duty value ADEGR at that time. As shown in FIG. 12, when the abscissa represents the delay coefficient DLAY corresponding to the delay time and the rich side correction value KiR is the coefficient of the slope, the ordinate represents the delay coefficient DL.
It is the product of AY and the rich side correction value KiR.

On the other hand, if it is determined in step 506 that the rich skip flag FRskip has not been set, then in step 509 the lean skip flag FLski
It is determined whether or not p has been set, that is, whether or not the feedback correction coefficient FAF has performed lean skip. Here, when it is determined that the lean skip has been performed, the lean flag F2 is set to “1” in step 510. Then, in step 511, the value obtained by multiplying the feedback correction coefficient FAF by the coefficient DLAY by the lean correction value KiL is subtracted, and the resulting value is used as the correction duty value A at that time.
DEGR. As shown in FIG. 13, the horizontal axis represents the delay coefficient DLAY corresponding to the delay time, and the lean correction value KiL
Is the coefficient of the slope, the vertical axis is the product of the delay coefficient DLAY and the lean correction value KiL.

If it is determined in step 509 that the lean skip flag FLskip has not been set, it is determined in step 512 whether the current time is earlier than the correction duty value skip timing Tskip. The determination in step 512 is to calculate the correction duty value ADEGR in the case where four or more skips have been made in the feedback correction of the air-fuel ratio A / F and the current time is in the integrated state. Here, if it is determined that the current time is before the correction duty value skip timing Tskip, it is determined in step 513 whether the lean flag F2 is “0”. Here, if it is determined that the lean flag F2 is “0”, step 514 is executed.
Then, the rich correction value KiR is added to the previous correction duty value ADEGR, and the correction value is integrated to the rich side. on the other hand,
When it is determined in step 513 that the lean flag F2 is not "0" but "1", in step 515, the lean correction value KiL is subtracted from the previous correction duty value ADEGR to integrate the correction value toward the lean side.

If it is determined in step 512 that the current time is not before the corrected duty value skip timing Tskip, it is determined in step 516 whether the current time is the corrected duty value skip timing Tskip. Here, if it is determined that the current time is the correction duty value skip timing Tskip, it is determined in step 517 whether the rich flag F1 is “0”, that is, whether the air-fuel ratio A / F is on the lean side. You. If it is determined that the rich flag F1 is "0", the lean flag F2 is set to "0" in step 518, and the correction duty value ADEGR is set in step 519.
Is added to the rich side skip value RSR. That is, the correction duty value ADEGR performs a rich skip by the skip value RSR.

On the other hand, at step 517, the rich flag F1
Is not "0", that is, if it is determined whether the air-fuel ratio A / F is on the rich side, the lean flag F2 is set to "1" in step 520, and the lean side skip is performed from the correction duty value ADEGR in step 521. Subtract the value RSL. That is, the correction duty value ADEGR skips to the lean side by the skip value RSL.

If it is determined in step 516 that the current time is not the corrected duty value skip timing Tskip, then in step 522 the current time is delayed by the corrected duty value skip timing Tskip by a delay time Tdlay.
Is determined before the time (the next predicted FAF skip timing). If it is determined that the timing is before the predicted FAF skip timing, the process proceeds to step 513. That is, normal integration processing (step 514 or step 515) is performed.

Further, at step 522, the current time is delayed by the correction duty value skip timing Tskip by the delay time Tskip.
If it is determined that the current time is not before the time added with the delay, that is, if it is determined that the current time is the next predicted FAF skip timing or later, in step 523, the current time is corrected duty value skip timing Tskip. Is determined to be a time obtained by adding the delay time Tdlay to the time. . If the determination is "NO", the process proceeds to step 513. This is for the case where the current time does not skip even after the expected FAF skip timing, and the integration is continued by resetting. On the other hand, if “YES”, then step 52
In step 4, it is determined whether the lean flag F2 is set to "0". If it is "0", the process proceeds to step 510. In the case of “1”, the process proceeds to step 507. Here, the reason for shifting to step 510 or step 507 is as follows. In other words, the fact that the processing reaches step 523 is the expected FA
That is, the feedback correction coefficient FAF is not skipping even though it is the F skip timing. If the feedback correction coefficient FAF is not skipped, only the correction duty value ADEGR is inverted to continue integration, and the feedback correction coefficient FAF
And the correction duty value ADEGR are too large. In consideration of such a possibility, the correction duty value ADEGR is changed to the current feedback correction coefficient FA.
Reset is performed based on F.

If it is determined in step 505 that the count value of the skip counter Cskip is 3, it is determined in step 524 whether the current time is earlier than the correction duty value skip timing Tskip. The determination in step 524 is to calculate the correction duty value ADEGR in the case where the air-fuel ratio A / F is skipped three times in the feedback correction and the current time is in the integrated state. If it is determined that the current time is not earlier than the correction duty value skip timing Tskip, the process proceeds to step 5
Move to 16. On the other hand, if it is determined that the current time is before the correction duty value skip timing Tskip, the rich flag F1 is set to “0” in step 525.
It is determined whether the air-fuel ratio A / F is lean, and the process proceeds to step 508 because the air-fuel ratio A / F is lean and the feedback correction coefficient FAF is rich. If the value is “1”, the air-fuel ratio A / F is in a rich state, and the feedback correction coefficient FAF is on the lean side.

By such processing, the EGR correction duty value ADEGR can be calculated. Figure 1 below
The correction duty value ADEGR calculated based on the above routine based on 1 (a) and (b) will be described. When the feedback control of the air-fuel ratio A / F is started, the EGR
Correction duty value ADE in duty value correction control
GR is corrected to 0.9 with respect to the reference value of 1.0. Feedback correction coefficient FAF in feedback control
Until there is a third skip, the value is controlled by that value, and in the third skip (lean skip in the present embodiment), the correction value (delay coefficient DLAY) is synchronized with the skip in accordance with the feedback correction coefficient FAF at that time. × lean correction coefficient KiL) to reduce the initial correction duty value AD
Calculate EGR.

The lean correction coefficient KiL is integrated from the initial correction duty value ADEGR set as described above (the graph becomes lower right). Correction duty value ADE
GR reaches the first correction duty value skip timing Tskip1 in which the fourth skip is predicted before the fourth skip. Here, the skip timing T
In skip1, a rich skip value RSR is given to the correction duty value ADEGR to perform a rich skip. And
The second correction duty value skip timing Tski
The rich correction coefficient KiR is integrated toward p2 (the graph rises to the right). The skip timing Tskip2 that predicts the subsequent fifth skip is calculated in the same manner.

Here, the third lean integration time TL3 is shorter than the preceding and following integration times. Therefore, the average lean cycle time AV calculated from the past two lean integration times TL
The actual skip (the sixth time) comes earlier than the next correction duty value skip timing Tskip3 based on the ETL. In this case, as described in the above routine, the process proceeds to step 507, and the initial correction duty value ADEGR is newly set based on the feedback correction coefficient FAF at this time. That is, 6
The correction duty value ADEGR is also skipped in synchronization with the skipping of the feedback correction coefficient FAF for the second time. At the same time, the lean integration time TL3 is taken in and the next (fourth)
The correction duty value skip timing Tskip4 is set (see FIG. 6).

As described above, the correction duty value ADEGR
Is calculated so as to follow the lean-rich behavior of the feedback correction coefficient FAF. Then, E is calculated according to the correction duty value ADEGR thus calculated.
The CU 51 controls the EGR valve 45.

Next, of the processing executed by the ECU 51, the correction duty value ADE calculated as described above
An "actual duty value ADEGRF control routine" for actually controlling the EGR valve 45 for the GR to make the actual duty value DEGRF coincide with the correction duty value ADEGR will be described with reference to FIG. This routine is executed by interruption every predetermined crank angle.

First, at step 601, it is determined whether or not the engine is idling. EG if idling
Since the R control is unnecessary, the correction duty value ADEGR is set to 0 in step 602, and the process is once ended. If it is determined that the vehicle is not in the idling state, it is determined in step 603 whether or not the condition of the EGR control is satisfied based on a signal from each sensor.
If the condition is not satisfied, the flow shifts to step 602 to end the process once. Here, the condition of the EGR control is a basic condition based on a case where a predetermined time has elapsed from the start of the engine or a load state.

In step 604, the EGR counter C
It is determined whether or not EGR is 0, that is, whether or not this is the first process. If the count value of the EGR counter CEGR is 0, it is determined that the processing is the first processing, and the initial actual duty value ADEGRF is set to 0.9 in step 605.
And Then, in step 606, an averaging execution flag FSUM indicating whether or not averaging is being performed is set to "1". Here, “smoothing” is the correction duty value ADG
This means smoothing when the actual duty value ADEFRF is synchronized with R. Next, at step 607, the EGR
The count value of the counter CEGR is incremented by one, and the process ends once.

At step 604, the EGR counter C
If the EGR count value is not 0, that is, if it is the second or subsequent routine, the process proceeds to step 608, where it is determined whether the smoothing execution flag FSUM is “0”. In the case of the smoothed state, the current actual duty value ADEFRF is calculated in step 609 by the calculated correction duty value A.
It is determined whether it is smaller than DEGR. The actual duty value ADEFRF is the calculated correction duty value ADEF
If it is determined that it is smaller than R, the rich-side correction value K that is twice the actual duty value ADEFRF in step 610
Add iR. The process of step 610 is a smoothing process, and as shown by the one-dot chain line in FIG. 14, a surge may occur if the correction duty value ADEGR is suddenly overtaken. In order to catch up with the correction duty value ADEGR.

Next, the actual duty value ADEFRF corrected in step 611 is changed to the calculated duty value ADEF.
It is determined whether it is larger than R, and if not, the process is temporarily terminated and steps 601 to 611 are repeated. On the other hand, if it is determined that the actual duty value ADEFRF is larger than the calculated correction duty value ADEGR, it is not necessary to perform smoothing control any more.
To guard the actual duty value ADEGR so as to be the same as the calculated correction duty value ADEGR. At this time, the actual duty value ADEGRF is equal to the correction duty value AD.
You have caught up to EGR. And step 61
At step 3, the smoothing execution flag FSUM is set to "0", and the process is temporarily terminated.

At step 609, the current actual duty value ADEFRF is calculated to be the calculated correction duty value ADEF.
If it is determined that it is not smaller than R, step 616
It is determined whether the current actual duty value DEGRF is larger than the calculated correction duty value ADEGR.
Here, if it is determined that the size is not large, the process proceeds to step 618. On the other hand, the actual duty value ADEGR
If it is determined that F is larger than the calculated duty value ADEGR, at step 615 the actual duty value AD
The lean correction value KiL, which is twice as large as EGRF, is subtracted.
This is because when the initial actual duty value ADEGRF is set to 0.9 in step 605 as shown in FIG.
This assumes a case where the EGRF is large, and is a case where the EGRF moves toward the lean side.

Next, the actual duty value ADEFRF corrected in step 616 is used as the calculated correction duty value AD.
It is determined whether the value is smaller than the EGR. If the value is not smaller than the EGR, the process is temporarily terminated and steps 601 to 616 are repeated. On the other hand, if it is determined that the actual duty value ADEFRF is smaller than the calculated correction duty value ADEGR, there is no need to perform smoothing control any more, so the process proceeds to step 612.

As described above, the “actual duty value ADEGR
In the "F control routine", the actual duty value ADEGR
F immediately matches the correction duty value ADEGR, and the actual duty value ADEFRF is changed to the correction duty value ADEF.
It is controlled to synchronize with the behavior of R. Then, while the EGR amount based on the actual duty value ADEFRF is supplied, feedback control of the air-fuel ratio A / F based on the feedback correction coefficient FAF is performed, and the feedback control is applied to the fuel injection control of the injector 16.

The feedback control and the EGR control of the air-fuel ratio A / F are executed in the manner described above. According to the present embodiment, the following excellent effects can be obtained. -When the feedback correction coefficient FAF is in a lean state, the EGR amount is reduced, and when the feedback correction coefficient FAF is in a rich state, the EGR amount is increased. Therefore, E according to the driving situation
The GR amount can be recirculated without waste. As shown in FIG. 16, when the feedback correction coefficient FAF shifts to the rich side with respect to the EGR amount which has been determined in consideration of the maximum lean in the conventional feedback control, E is correspondingly increased.
The GR amount can be increased. Therefore, NOx can be reduced and drivability can be improved more than before, and the fuel consumption rate can be further reduced.

In the EGR control, it usually takes time until the EGR valve 45 is operated and EGR gas is introduced into the combustion chamber 11a. Therefore, in the feedback correction of the air-fuel ratio A / F, the feedback correction coefficient FA
F changes, and even if the EGR valve 45 is operated in response to the change, the feedback correction coefficient FAF
The EGR control is not performed according to the fluctuation of. In the present embodiment, the EGR control is performed by estimating the feedback correction coefficient FAF after a predetermined time obtained by adding the delay time to the current time in consideration of the delay time.
, Drivability can be improved, and the fuel consumption rate can be further reduced.

In this embodiment, the rich integration time TR or the lean integration time TL, which is the elapsed time between the skip timings, is obtained, and the correction duty value ADEGR is fed back based on the average rich cycle time AVETR or the average lean cycle time AVETL. The next skip timing of the correction coefficient FAF is expected. Therefore, for example, even if the lean integration time TL is extremely short once only due to a signal input defect, the influence on the prediction of the next skip timing is suppressed as much as possible.

In the feedback control at the beginning of the feedback correction of the air-fuel ratio A / F from the average rich cycle time AVETR, a predetermined skip value RSR is used when going from the maximum lean to the rich side, and conversely, when going from the maximum rich to the lean side. , RSL to skip the feedback correction coefficient FAF. Here, the EGR control is executed by estimating the fluctuation of the feedback correction coefficient FAF. The delay time Tdlay is equal to the EGR valve 45.
Is operated until the EGR gas is introduced into the combustion chamber 11a. That is, although the EGR gas is returned to the surge tank 26, it takes a long time for the EGR gas to reach the combustion chamber 11a from the surge tank 26. Therefore, the skip timing is set earlier in consideration of the delay time. is there. This delay time Tdlay is experimentally obtained. Thus, the correction duty value ADE
The GR skip timing Tskip can be estimated.

In the above embodiment, when the skip timing next to the feedback correction coefficient FAF comes before the correction duty value skip timing Tskip, the correction duty value ADEGR is skipped simultaneously with the skip timing. Is controlled. In this case, the initial correction duty value ADEGR is newly set based on the current feedback correction coefficient FAF.
Is set to be controlled. Therefore, the control amount of the correction duty value ADEGR does not greatly deviate from the expected control amount of the correction coefficient FAF.

The calculated correction duty value ADEGR
In order to catch up with the actual duty value ADEFRF, first, control is performed with a correction value of 0.9 with respect to the reference value 1.0, and then smoothing control is performed so as to reach the correction duty value ADEGR while performing a smoothing process. I have. Therefore, there is no sudden change in the EGR amount, so that the possibility of occurrence of a surge is reduced.

The embodiments of the present invention are not limited to the above, and may be carried out in the following manner by appropriately changing a part of the configuration without departing from the spirit of the invention. .

In the above embodiment, the correction duty value A
The DEGR is controlled by a constant correction value of 0.9 with respect to the reference value of 1.0 until the third skip timing comes. However, this correction value can be changed according to the correction amount of the feedback correction coefficient FAF. Also, the time for controlling with the constant correction value does not necessarily have to be until the third skip timing.

In the above-described embodiment, in the feedback control, when skipping from the maximum lean to the rich side, and conversely, when shifting from the maximum rich to the lean side, predetermined skip values RSR and RSL are given to provide the feedback correction coefficient FA.
F is skipped. However, it is not always necessary to skip.

In the above embodiment, the gasoline engine 1 is used as the internal combustion engine. However, a vehicle equipped with a diesel engine can be used. Other technical ideas grasped by the above embodiments will be described below.

(1) The air-fuel ratio control device for an internal combustion engine according to claim 3, wherein the actual fluctuation of the feedback correction coefficient is rich / lean inversion. (2) The internal combustion engine according to (3) or (1), wherein the exhaust gas recirculation amount correction means corrects the exhaust gas recirculation amount in synchronization with a change in the actual feedback correction coefficient. Engine air-fuel ratio control device.

[0091]

As described above in detail, according to the first aspect of the present invention, finer EGR control is possible, so that NOx can be reduced and drivability can be improved more than before, and further, fuel consumption can be improved. It has an excellent effect that the rate can be reduced.

According to the second aspect of the present invention, in addition to the effect of the first aspect of the invention, a time delay until the EGR gas reaches the combustion chamber can be eliminated, so that more accurate EGR control according to the actual rich / lean state can be performed. It becomes possible.

According to the third aspect of the present invention, when the actual air-fuel ratio goes rich, the amount of exhaust gas also increases, and when the actual air-fuel ratio goes lean, the amount of exhaust gas decreases. Therefore, NOx can be reduced more than ever, drivability can be improved, and the fuel consumption rate can be further reduced.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing an air-fuel ratio control device for an engine according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an electrical configuration of an ECU.

FIG. 3 is a flowchart showing an “air-fuel ratio A / F feedback routine” executed by the ECU.

FIG. 4 is a flowchart showing an “air-fuel ratio A / F feedback routine” executed by the ECU.

FIG. 5 is a flowchart showing a “feedback correction coefficient FAF cycle calculation routine” executed by the ECU.

FIG. 6 is a flowchart showing a “correction duty value A” executed by the ECU;
The flowchart showing the "DEGR skip timing calculation routine."

FIG. 7 is a flowchart showing a “base duty value DEGR calculation routine” executed by the ECU.

FIG. 8 is a flowchart showing a “correction duty value A” executed by the ECU;
The flowchart showing the "DEGR calculation routine."

FIG. 9 is a flowchart showing a “correction duty value A” executed by the ECU;
The flowchart showing the "DEGR calculation routine."

FIG. 10 is a flow chart showing an operation of the “actual duty value A” executed by the ECU
9 is a flowchart showing a DEGRF control routine ".

FIG. 11 is a timing chart showing skip timing of a feedback correction coefficient FAF and the like.

FIG. 12 is a partially enlarged view of a timing chart for explaining means for calculating an initial correction duty value ADEGR from a feedback correction coefficient FAF.

FIG. 13 is a partially enlarged view of a timing chart for explaining means for calculating an initial correction duty value ADEGR from a feedback correction coefficient FAF.

FIG. 14 is a partially enlarged view of a timing chart for explaining a process for catching an actual duty value ADEFRF to a correction duty value ADEGR.

FIG. 15 is a partially enlarged view of a timing chart for explaining a process for catching an actual duty value ADEFRF to a correction duty value ADEGR.

FIG. 16 is a timing chart showing a relationship between a feedback correction coefficient FAF and an EGR amount.

FIG. 17 is a graph illustrating a relationship between an EGR rate and an air-fuel ratio A / F in conventional EGR control.

[Explanation of symbols]

11: engine as internal combustion engine, 11a: combustion chamber, 2
2 ... intake passage, 23 ... exhaust passage, 29 ... oxygen sensor, 4
3: EGR device, 51: ECU constituting exhaust gas recirculation amount correction means and exhaust gas recirculation control means.

Claims (3)

[Claims] An exhaust gas discharged to an exhaust passage of an internal combustion engine is recirculated to an intake passage side of the engine while monitoring data supplied from an air-fuel ratio detecting means and supplied to the internal combustion engine. An air-fuel ratio control device for an internal combustion engine that performs feedback control of an air-fuel ratio of an air-fuel mixture to a target air-fuel ratio, comprising: a recirculation amount control unit that controls a recirculation amount of the exhaust gas based on an operation state of the engine; The recirculation amount of the exhaust gas when the feedback correction coefficient is set to the rich side with respect to the fuel ratio is larger than the recirculation amount of the exhaust gas when the feedback correction coefficient is set to the lean side with respect to the target air-fuel ratio. An exhaust gas recirculation amount correction means for correcting the amount of exhaust gas recirculation to be controlled so as to increase. 2. The exhaust gas recirculation amount correcting means includes a transition estimating means for estimating a transition of a feedback correction coefficient variably set to a rich side or a lean side with respect to the target air-fuel ratio. 2. The air-fuel ratio control device for an internal combustion engine according to claim 1, wherein the correction of the recirculation amount of the exhaust gas is performed prior to the change of the correction coefficient. 3. The exhaust gas recirculation amount correction means, when an actual change occurs in the feedback correction coefficient prior to a transition of the feedback correction coefficient estimated by the transition means, the actual feedback correction coefficient 3. The air-fuel ratio control device for an internal combustion engine according to claim 2, wherein the correction of the recirculation amount of the exhaust gas is performed based on the variation of the exhaust gas.