JPH0792011B2 - Fuel injection amount learning control device for internal combustion engine - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection amount learning control device for an internal combustion engine, and in particular, learns and controls the fuel injection amount during acceleration / deceleration by correcting acceleration increase amount and deceleration increase amount. The present invention relates to a fuel injection amount learning control device for an internal combustion engine.

[Conventional technology]

Conventionally, the basic fuel injection time TP, which is determined according to the engine load (intake air amount or intake pipe pressure) and the engine speed, and the residual oxygen concentration in the exhaust gas are detected, and a signal inverted at the theoretical air-fuel ratio is output. There is known a fuel injection amount control device that controls the fuel injection amount in a feedback manner so that the air-fuel ratio becomes the stoichiometric air-fuel ratio by using the air-fuel ratio feedback correction coefficient FAF obtained from the output of the O ₂ sensor. In such a fuel injection amount control device, it is determined whether the vehicle is accelerating or decelerating, the fuel injection amount is increased when it is determined that the vehicle is accelerating, and the fuel injection amount is reduced when it is determined that the vehicle is decelerating. There is.

However, when the mileage of the vehicle becomes long, deposits adhere to the intake valve and intake port due to changes over time, and due to the effects of this deposit, the air-fuel ratio shows a lean tendency during acceleration and the air-fuel ratio tends to reach a rich tendency during deceleration. Will be shown. That is, during acceleration, the throttle opening becomes large, so the intake pipe pressure is high and the amount of fuel evaporation is small, so the injected fuel is absorbed in the deposit, and as a result, the fuel supplied to the combustion chamber becomes insufficient and empty. It shows a lean lean fuel ratio. On the other hand, during deceleration, since the throttle opening is small and the intake pipe pressure is low, the fuel absorbed in the deposit evaporates, and since this evaporated fuel is supplied to the combustion chamber, much fuel is supplied to the combustion chamber. Therefore, the air-fuel ratio has a tendency to reach. FIG. 2 shows the difference between the air-fuel ratio feedback correction coefficient and the air-fuel ratio of the internal combustion engine (new internal combustion engine) in the initial state during acceleration and the internal combustion engine with deposits. As can be understood from the figure, since the internal combustion engine in the initial state has no deposit, the air-fuel ratio feedback correction coefficient changes at a constant cycle as shown by the broken line, and the air-fuel ratio also becomes a value near the theoretical air-fuel ratio. Controlled. On the other hand, in the internal combustion engine with deposits, the air-fuel ratio becomes lean during acceleration, and as a result, the air-fuel ratio feedback correction coefficient greatly changes as shown by the solid line. Further, FIG. 3 shows the relationship between the deviation ΔPM of the intake pipe pressure every 360 ° CA and the change frequency of the air-fuel ratio. In the figure, the solid line NEW shows the change frequency of the air-fuel ratio in the internal combustion engine in the initial state, and the broken line OLD shows the change frequency of the air-fuel ratio of the internal combustion engine with deposits. As can be seen from the figure, in an internal combustion engine with deposits, the acceleration (ΔPM
(When is positive), the frequency of the air-fuel ratio becomes lean is high, and at the time of deceleration (when ΔPM is negative), the frequency of air-fuel ratio becomes rich.

For this reason, conventionally, as disclosed in Japanese Patent Laid-Open No. 59-203829, the acceleration increase coefficient K _ACC is reduced and the air-fuel ratio is reduced during acceleration and when the air-fuel ratio A / F is greater than the target air-fuel ratio.
When the A / F is leaner than the target air-fuel ratio, the acceleration increase coefficient K _ACC is increased, and when the air-fuel ratio A / F is less than the target air-fuel ratio during deceleration, the deceleration reduction coefficient K _DCL is increased and the air-fuel ratio A / F is increased. Deceleration reduction coefficient K when F is leaner than the target air-fuel ratio
Learning control is performed by reducing _DCL .

Incidentally, as a technique related to the present invention, Japanese Patent Application Laid-Open No. 59-203829
There is a technique described in Japanese Patent Laid-Open No. 60-204937.

[Problems to be Solved by the Invention]

However, in the conventional learning control, the acceleration increase coefficient and the deceleration decrease coefficient are learned immediately when the air-fuel ratio does not become equal to the target air-fuel ratio during acceleration / deceleration. However, there is a problem in that erroneous learning occurs due to the change in the air-fuel ratio and the exhaust emission and driver viability deteriorate. That is, in the conventional learning control, since the learning is performed without considering the deviation of the air-fuel ratio, there is a problem that the learning according to the deviation of the air-fuel ratio due to the change over time cannot be performed.

The present invention has been made to solve the above-mentioned problems, and it is possible to correct the deviation of the air-fuel ratio due to a change over time by determining the deviation tendency of the air-fuel ratio during transition such as acceleration / deceleration, and the fuel of the internal combustion engine. It is an object to provide an injection amount learning control device.

[Means for Solving the Problems]

In order to achieve the above object, the present invention is a transient state determination means for determining a transient state, a fuel injection means for increasing or decreasing the amount of fuel determined in accordance with the correction coefficient at the time of the transient state, and fuel injection means in the exhaust gas. Oxygen concentration sensor to detect the residual oxygen concentration, the time when the oxygen concentration sensor output shows a Rich state from the target air-fuel ratio and the time when the oxygen concentration sensor output shows a lean state from the target air-fuel ratio Deviation calculation means for calculating a deviation between the deviation and a correction coefficient for correcting the correction coefficient during the transition so that the deviation becomes a value within the predetermined range when the deviation becomes a value outside the predetermined range in a transient state. And means.

[Action]

According to the present invention, the transient state determination means determines whether the operating state is the transient state. The fuel injection means increases or decreases the amount of fuel determined according to the correction coefficient at the time of transition when it is determined to be in the transient state, and injects the fuel. Further, the deviation calculation means is
The deviation between the time when the output of the oxygen concentration sensor that detects the residual oxygen concentration in the exhaust gas shows a rich state from the target air-fuel ratio and the time when the output of the oxygen concentration sensor shows a lean state from the target air-fuel ratio Calculate Here, if the time when the oxygen concentration sensor output shows a rich state from the target air-fuel ratio is longer than the time when the oxygen concentration sensor output shows a lean state from the target air-fuel ratio, the air-fuel ratio rich state Is continued longer than the lean state of the air-fuel ratio, which means that the air-fuel ratio is biased to the latch side.
That is, the air-fuel ratio has a tendency to reach.
Therefore, it is possible to determine from the above deviation whether the air-fuel ratio in a certain period shows a rich tendency or a lean tendency. Then, the correcting means sets the deviation to a value within the predetermined range when the deviation becomes a value outside the predetermined range in the transient state, that is, when the air-fuel ratio shows a rich tendency and a lean tendency. Correct the correction coefficient during transition. In this way, when the air-fuel ratio shows a rich tendency and a lean tendency, by correcting the correction coefficient during the transition so that the deviation becomes a value within a predetermined range, the rich tendency and the lean tendency of the air-fuel ratio are corrected. The fuel injection amount can be controlled so that the fuel ratio is controlled near the target air-fuel ratio. In this way, by correcting the correction coefficient at the time of transition and learning-controlling the fuel injection amount, the change over time of the air-fuel ratio due to deposit adhesion or the like is corrected.

〔The invention's effect〕

As described above, according to the present invention, the air-fuel ratio is biased to the latch side from the deviation between the time when the oxygen concentration sensor output shows the air-fuel ratio lean state and the time when the air-fuel ratio lean state is shown. The air-fuel ratio is deviated to the lean side, and the correction coefficient at the time of transition is learned, so that the deviation of the air-fuel ratio due to aging can be corrected without mis-learning to correct the exhaust emission and driver. It is possible to obtain an effect that it is possible to prevent the deterioration of the vitality.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.
In the following, numerical values that do not hinder the present invention will be used for description, but the present invention is not limited to these numerical values. FIG. 4 schematically shows a four-cylinder, four-cycle spark ignition engine (engine) equipped with a fuel injection amount learning control device according to an embodiment of the present invention.

This engine is controlled by a microcomputer 44 as a control circuit. A throttle valve 8 is arranged on the downstream side of the air cleaner 2, and a surge tank 12 is provided on the downstream side of the throttle valve 8. An intake air temperature sensor 4 for detecting the intake air temperature is provided near the air cleaner 2.
Is attached to the throttle valve 8, and an idle switch 10 that is turned on when the throttle valve is fully closed is attached to the throttle valve 8. A diaphragm type or semiconductor type pressure sensor 6 is attached to the surge tank 12. The output signal from the pressure sensor 6 has a small time constant for removing the pulsating component of the intake pipe pressure (for example, 3 to 5 msec) and a filter 7 composed of a CR filter having a good response (see FIG. 5). Process). Also, the throttle valve 8
A bypass passage 14 is provided so as to bypass the throttle valve and connect the upstream side of the throttle valve and the downstream side of the throttle valve. An ISC (idle speed control) valve 16 whose opening is adjusted by a solenoid is attached to the bypass passage 14, and the duty ratio of the current flowing through the solenoid is controlled to control the amount of air flowing through the bypass passage 14. The rotation speed during idling is controlled to the target value. The surge tank 12 is in communication with the combustion chamber of the engine 20 via the intake manifold 18 and the intake port 22. A fuel injection valve 24 is attached to each cylinder so as to project into the intake manifold 18.

The combustion chamber of the engine 20 includes a catalyst device 27 filled with a three-way catalyst through an exhaust port 26 and an exhaust manifold 28.
Is in communication with. An O ₂ sensor 30 that detects the oxygen concentration in the exhaust gas and outputs a signal inverted at the stoichiometric air-fuel ratio is attached to the exhaust manifold 28. A cooling water temperature sensor 34 is attached to the engine block 32 so as to penetrate the engine block 32 and project into the water jacket. The cooling water temperature sensor 34 detects the engine cooling water temperature and outputs a water temperature signal, and the water temperature signal represents the engine temperature. The engine oil temperature may be detected to represent the engine temperature.

A spark plug 38 is attached to each cylinder so as to penetrate the cylinder head of the engine 20 and project into the combustion chamber. The ignition plug 38 is connected to a microcomputer 44 via a distributor 40 and an igniter 42 having an ignition coil. This Distributor
A rotation angle sensor 48 composed of a signal rotor fixed to the distributor shaft and a pick-up fixed to the distributor housing is mounted inside 40. The rotation angle sensor 48 outputs an engine rotation speed signal every 30 ° CA, for example.

As shown in FIG. 5, the micro computer 44 is a micro processing unit (MPU) 60, read only
Memory (ROM) 62, random access memory (RAM)
64, Back-up Plum (BU-RAM) 66, I / O port 6
8, an input port 70, output ports 72, 74, 76, and a bus 75 such as a data bus or a control bus that connects them. The BU-RAM 66 stores an acceleration increase coefficient and a deceleration decrease coefficient as a correction coefficient at the time of transition described below. An A / D converter 78 and a multiplexer 80 are sequentially connected to the input / output port 68. To the multiplexer 80, the pressure sensor 6 is connected via a filter 7 and a buffer 82 formed of a resistor R and a capacitor C, the cooling water temperature sensor 34 is connected via a buffer 84, and the cooling water temperature sensor 34 is connected via a buffer 85. The temperature sensor 4 is connected. The MPU 60 controls the multiplexer 80 and the A / D converter 78 to sequentially convert the pressure sensor 6 output, the cooling water temperature sensor 34 output, and the intake air temperature sensor 4 output, which are input via the filter 7, into a digital signal and RAM64. To memorize. Therefore, the multiplexer 80, the A / D converter 78, the MPU 60 and the like act as sampling means for sampling the pressure sensor output and the like at predetermined time intervals. The O ₂ sensor 30 is connected to the input port 70 via the comparator 88 and the buffer 86, and the rotation angle sensor 48 is connected to the input port 70 via the waveform shaping circuit 90. Further, the idle switch 10 is connected to the input port 70 via a buffer (not shown). The output port 72 is connected to the igniter 42 via a drive circuit 92, and the output port 74 is a drive circuit 94 having a down counter.
Connected to the fuel injector 24 via, and output port 76
Is connected to the solenoid of the ISC valve 16 via a drive circuit 96. Note that 98 is a clock and 99 is a counter. Programs and the like for the control routines described below are stored in advance in the ROM 62.

Next, a control routine of the first embodiment in which the present invention is applied to the above engine will be described.

FIG. 7 shows a routine for calculating the air-fuel ratio feedback correction coefficient FAF for controlling the air-fuel ratio feedback control. In step 180, it is judged whether the air-fuel ratio feedback back-up condition is satisfied. Whether or not the air-fuel ratio feedback condition is satisfied is judged according to the operating condition. For example, the engine cooling water temperature is not lower than the predetermined value (for example, 40 ° C) and the fuel is not in the fuel cut. , It is judged that the air-fuel ratio feedback condition is satisfied when the fuel amount is not increasing and the air-fuel ratio lean control is not being executed. Any one of the above conditions in step 180
If it is determined that the feed back condition is not satisfied, that is, if the feed back condition is not met, the air-fuel ratio feed back correction coefficient FAF is set to 1.0 in step 182, and the post air-fuel ratio feed back correction coefficient FAF is stored in a predetermined area of the RAM.

On the other hand, when it is determined in step 180 that all of the above conditions are satisfied and the air-fuel ratio feedback condition is satisfied, in step 183, the O ₂ sensor output is acquired and then in step 184 the O ₂ sensor output changes the air-fuel ratio latch. It is determined whether or not it is shown. When it is determined that the output of the O ₂ sensor indicates the air-fuel ratio limit, the flag CAFL is reset in step 186, and then it is determined in step 188 whether the flag CAFR is reset. flag
When CAFR is reset, it is the time when the output of the O ₂ sensor reverses from lean to rich, so the step is stopped.
At 190, the proportional constant R _S is subtracted from the air-fuel ratio feedback correction coefficient FAF, and then at step 191, the flag CAFR is set. When it is determined in step 188 that the flag CAFR has been set, that is, after the output of the O ₂ sensor is inverted from lean to rich, in step 189, the integration constant K is calculated from the air-fuel ratio feedback correction coefficient FAF.
Subtract _i .

When it is determined in step 184 that the O ₂ sensor output indicates the lean air-fuel ratio, in step 192 the flag CAFR is reset, and then in step 194 it is determined whether the flag CAFL is reset. flag
When CAFR is reset, it is the time when the output of the O ₂ sensor reverses from the air-fuel ratio switch to lean, so the step is stopped.
At 196, the proportional constant R _S is added to the air-fuel ratio feedback correction coefficient FAF, and then the flag CAFL is set at step 197. On the other hand, when it is determined in step 194 that the flag CAFL has been set, that is, after the output of the O ₂ sensor is inverted from lean to lean, in step 195 the integration constant K _i is added to the air-fuel ratio feedback correction coefficient FAF. To do.

FIG. 1 shows a routine executed every 360 ° CA. At step 100, the current intake pipe pressure PM _NEW to 36
Deviation of intake pipe pressure by subtracting intake pipe pressure PM _OLD before 0 ° CA
Calculate DLPM. In the next step 102, it is determined whether or not acceleration is being performed by determining whether or not the deviation DLPM of the intake pipe pressure exceeds a positive predetermined value (for example, 2 mmHg). When it is determined that the intake pipe pressure deviation DLPM exceeds a predetermined positive value and the vehicle is accelerating, the O ₂ sensor output OX is compared with a reference level (for example, 0.45 V) in step 104 to obtain O
₂ Determine whether or not the sensor output OX shows the latched state based on the theoretical air-fuel ratio. When it is determined that the O ₂ sensor output OX indicates the air-fuel ratio rich state, the count value CAC is incremented in step 106, and the O ₂ sensor output OX becomes less than the reference level to indicate the air-fuel ratio lean state. When it is determined that the count value CAC is decremented in step 108.

At the next steps 110 and 114, whether the air-fuel ratio shows a latch tendency from the stoichiometric air-fuel ratio by judging whether or not the count value CAC has become a value outside the first predetermined range (50 to -50). , Determine if it shows a lean tendency.
That is, when it is determined in step 110 that the count value CAC exceeds the upper limit value (50) of the first predetermined range, that is, when it is determined that the air-fuel ratio shows the latch tendency, the BU- After reducing the acceleration increase coefficient KAC stored in the RAM by a predetermined value (for example, 0.1), the count value CAC is set to 0 in step 118. The initial value of the acceleration increase coefficient KAC is set to 1.0 and BU-RA
Remembered by M. Further, in step 114, it is determined whether the count value CAC is less than the lower limit value (-50) of the first predetermined range, and when it is determined that the count value CAC is less than the lower limit value of the first predetermined range, the air-fuel ratio Is biased to the lean side of the theoretical air-fuel ratio and the air-fuel ratio shows a lean tendency, and at step 116, the acceleration increase coefficient KAC stored in the BU-RAM is increased by a predetermined value (for example, 0.1). At the subsequent step 118, the count value CAC is set to 0. When the count value CAC is within the first predetermined range, the routine proceeds to the routine of FIG. 8 without correcting the acceleration increase coefficient KAC.

If it is determined in step 102 that the intake pipe pressure deviation DLPM is less than or equal to the positive predetermined value, the intake pipe pressure deviation DLPM is negative in the step 120 (for example, −2 mmHg).
It is determined whether or not the vehicle is decelerating by determining whether or not it is less than. When it is determined that the deviation DLPM of the intake pipe pressure is equal to or greater than the negative predetermined value, it is determined that the engine is in a steady operation state and the routine proceeds to the routine shown in FIG. When it is judged that the condition is present, the process proceeds to step 122.
In step 122, the O ₂ sensor output OX is compared with the above-described determination level to determine whether the O ₂ sensor output indicates a latch state or a lean state based on the stoichiometric air-fuel ratio. When it is determined that the O ₂ sensor output indicates the air-fuel ratio rich state, the count value CDC is incremented in step 124, and when it is determined that the O ₂ sensor output OX indicates the air-fuel ratio lean state, step 12
At 6, the count value CDC is decremented.

At the next steps 128 and 132, the count value CDC
Is outside the second predetermined range (for example, 50 to -50), the air-fuel ratio is biased toward the latch side from the stoichiometric air-fuel ratio, indicating that the air-fuel ratio has a tendency to reach or is empty. It is determined whether the fuel ratio is leaner than the theoretical air-fuel ratio and the air-fuel ratio shows a lean tendency. That is, in step 128, it is determined whether the count value CAC exceeds the upper limit value (50) of the second predetermined range to determine whether or not the air-fuel ratio has a latch tendency, and the air-fuel ratio reaches the latch value. If it is determined that the trend is shown, step 13
At 0, the deceleration reduction coefficient KDC stored in the BU-RAM is increased by a predetermined value (for example, 0.1), and then at step 136 the count value CDC is set to 0. The initial value of this deceleration reduction coefficient KDC is set to 1.0 and stored in the BU-RAM. Further, in step 132, it is judged whether or not the air-fuel ratio shows a lean tendency by judging whether or not the count value CDC is less than the lower limit value (−50) of the second predetermined range, and the air-fuel ratio shows a lean tendency. If it is determined that
In step 4, the deceleration reduction coefficient KDC stored in the BU-RAM is reduced by a predetermined value (for example, 0.1), and then in step 136 the count value CDC is set to zero. On the other hand, if it is determined in step 128 and step 132 that the count value CDC is within the second predetermined range, the deceleration reduction coefficient KDC
The process proceeds to the routine shown in FIG.

Fig. 6 shows changes in the O ₂ sensor output, changes in the count values CAC and CDC, changes in the acceleration increase coefficient KAC, and changes in the deceleration decrease coefficient KDC, along with changes in vehicle speed and intake pipe pressure PM, when controlled as described above. Show.

FIG. 8 shows a routine for calculating the fuel injection time TAU. At Step 200, the engine speed NE, the intake pipe pressure PM and the engine cooling water temperature THW are taken in, and Step 2
At 02, the basic fuel injection time TP is calculated based on the engine speed NE and the intake pipe pressure PM. Next step 204
Then, the map shown in FIGS. 9 and 10 is used to calculate the increase / decrease time f ₁ according to the engine speed NE and the increase / decrease time f ₂ according to the engine cooling water temperature THW, and in step 206, the increase / decrease time f By adding ₁ and f ₂ , the engine speed NE and the engine coolant temperature TH can be calculated as shown in the following equation (1).
The increase / decrease amount time f (NE, THW) according to W is calculated.

f (NE, THW) = f ₁ + f ₂ (1) In the next step 208, the deviation DLPM of the intake pipe pressure calculated in step 100 of FIG. 1 and the increase / decrease time f (NE, THW) are used. Basic fuel injection time during transition according to the following equation (2)
Calculate TPAEW.

TPAEW = DLPM · f (NE, THW) (2) Here, the basic fuel injection time TPAEW during transition is positive because DLPM> 0 during acceleration, and DLPM <0 during deceleration, so basic fuel injection during transition Time TPAEW goes negative.

In step 210, it is determined whether or not acceleration is in progress by determining whether the deviation DLPM of the intake pipe pressure exceeds a positive predetermined value (for example, 2 mmHg). If it is determined that acceleration is in progress, in step 214. After reading the acceleration increasing coefficient KAC stored in the BU-RAM and setting it as K, the routine proceeds to step 220.
On the other hand, when it is determined that the intake pipe pressure deviation DLPM is less than the positive predetermined value, the intake pipe pressure deviation D
Decide whether or not deceleration is being performed by determining whether LPM is less than a predetermined negative value (for example, −2 mmHg). If it is determined that deceleration is being made, the deceleration / decrease amount stored in BU-RAM at step 216. After reading the coefficient KDC and setting it as K, step 220
Go to. On the other hand, in step 212, the intake pipe pressure deviation D
When LPM is determined to be equal to or greater than the negative predetermined value, that is, when the deviation DLPM of the intake pipe pressure takes a value between the positive predetermined value and the negative predetermined value, it is determined that the steady operation state is in progress, and in step 218. After setting the value of K to 0, the process proceeds to step 220.

In step 220, the basic fuel injection time TP, the value K set as described above, the transient basic fuel injection time TPAEW, the air-fuel ratio feedback correction coefficient FAF calculated in the routine of FIG. The fuel injection time TAU is calculated according to the following equation using the correction coefficient F determined by the water temperature and the like.

TAU = (TP + K * TPAEW) * FAF * F (3) Then, in a routine not shown, it is judged whether or not the fuel injection timing is reached, and when it is judged that the fuel injection timing is reached, the time corresponding to the fuel injection time TAU is driven. Synchronous fuel injection synchronized with the crank angle is executed by setting the down counter of the circuit 94 and opening the fuel injection valve until the value of the down counter becomes zero. Here, since the value of K is set to 0 during steady operation, the fuel injection amount is controlled according to the basic fuel injection time TP, the air-fuel ratio feedback back correction coefficient FAF and the correction coefficient F, and during acceleration the basic fuel during transition Since the injection time TPAEW has a positive value, the amount of fuel K / TPAEW is increased relative to the basic fuel injection time TP, and during deceleration the basic fuel injection time TPAEW during transition has a negative value KDC / TPAE
The amount of fuel W is reduced with respect to the basic fuel injection amount.

Transient basic fuel injection time TP when controlled as above
AEW, acceleration increase value KAC / TPAEW, deceleration decrease value KDC / TPAEW,
The basic fuel injection time TP, the fuel injection time TAU, and the fuel injection time after learning are shown in FIGS. 11 (2) and (3). It should be noted that FIG. 11 (1) shows a change in intake pipe pressure.

Next, a second embodiment of the present invention will be described with reference to FIG. In this example, the basic fuel injection time TPAEW _i during transient
The acceleration state and the deceleration state are determined based on the. Therefore, in FIG. 12, parts corresponding to those in FIG.

In step 230, it is determined whether or not the basic fuel injection time TPAEW _i during transition exceeds a predetermined positive value (for example, 100 μsec). When it is judged that the transient basic fuel injection time TPAEW _i exceeds the positive predetermined value, it is judged that the vehicle is accelerating and the routine proceeds to step 104, where it is judged that the transient basic fuel injection time TPAEW _i is less than the positive predetermined value. In step 236, the basic fuel injection time TPAEW _i during transition is set to a negative predetermined value (for example,
It is determined whether or not deceleration is being performed by determining whether it is less than 100 μsec). Steady operation when the transient basic fuel injection time TPAEW _i proceeds to step 122 is determined to negative than the predetermined value it is determined that during deceleration when the transient basic fuel injection time TPAEW _i is determined to be less than a predetermined negative value Judging as the state
14 Proceed to the routine shown in FIG. Step 232 and Step 234
Is for processing the same as steps 110 and 114 in FIG. 1, and in step 232, it is determined whether or not the absolute value of the count value CAC is a predetermined value (eg, 50) or more. It is determined whether or not the air-fuel ratio deviates from the stoichiometric air-fuel ratio, and when it is judged that the air-fuel ratio deviates from the stoichiometric air-fuel ratio, the count value CAC is determined in step 234 to determine whether the It is determined whether the fuel ratio is biased to the rich side or the lean side with respect to the theoretical air-fuel ratio. When it is determined that the air-fuel ratio is biased to the lean side, the acceleration increase coefficient KAC is corrected in step 116, and when it is determined that the air-fuel ratio is biased to the latch side, step 112 is used. Correct the KAC. Further, since the steps 238 and 204 are the same as the steps 128 and 130 in FIG. 1, their description will be omitted.

The count value CAC, the acceleration increase coefficient KAC, the transitional basic fuel injection time TPAEW, and the change in the O ₂ sensor output OX when controlled as described above are changed with the change in the intake pipe pressure PM and the vehicle speed.
Shown in Figure 13.

FIG. 14 shows the fuel injection time TAU calculation routine of the present embodiment, in which the engine speed NE and the intake pipe pressure PM are taken in at step 250, and the basic fuel injection time TP _NEW at the present time is calculated at step 252. . At the next step 254, 360 ° CA from the current basic fuel injection time TP _NEW.
The deviation ΔTP of the basic fuel injection time is calculated by subtracting the previous basic fuel injection time TP _OLD . And step
Deviation ΔTP at 256 and previous transient basic fuel injection time T
Using the value obtained by damping PAEW _i-1 with the damping coefficient C (0 <C <1), the basic fuel injection time TPAE during transition according to the following formula
Calculate W _i .

TPAEW _i ← ΔTP + TPAEW _i-1 · C (4) In step 258, whether or not acceleration is in progress by determining whether or not the transient basic fuel injection time TPAEW _i exceeds a positive predetermined value (for example, 100 μsec). If it is during acceleration, the value of the acceleration increase coefficient KAC is set to K in step 260, and then the process proceeds to step 268. In step 262, it is determined whether or not deceleration is being performed by determining whether or not the transient basic fuel injection time TPAEW _i is less than a negative predetermined value (for example, −100 μsec),
When it is determined that the vehicle is decelerating, the value of the deceleration reduction coefficient KDC is set to K in step 264. On the other hand, step 262
When it is determined that the normal operation is being performed in step 266,
At 0, the value of K is set to zero. At step 268, the basic fuel injection time is calculated according to the following equation.

TAU ← (TP _NEW + K ・ TPAEW _i ) ・ FAF ・ F (5) Then, in the fuel injection amount control routine (not shown), it is judged whether or not it is the injection timing, and when it is judged that the injection timing is reached, the fuel injection time TAU is set. Fuel injection is executed by opening the fuel injection valve for a corresponding period of time.

Changes in the basic fuel injection time TPAEW during transition when the fuel injection amount is controlled as described above, changes in the basic fuel injection time TP,
The change in fuel injection time TAU is shown in FIG. 15 together with the change in intake pipe pressure.

Next, a third embodiment of this embodiment will be described. The present embodiment is intended to prevent erroneous learning when the present invention is applied to a fuel injection amount control device that controls the fuel injection amount according to the following formula.

TAU = (TP + K · TPAEW) · KG · FAF · F (6) where KG is the air-fuel ratio feedback correction coefficient FAF a predetermined number of times so that the average value FAFAV of the air-fuel ratio feedback correction coefficient FAF is within a predetermined range. It is a learning value that is corrected each time it skips, and is increased when the average value FAFAV is equal to or higher than the upper limit value within the predetermined range, and is decreased when the average value FAFAV is equal to or lower than the lower limit value within the predetermined range. This learning value KG is applied to the above equation (6) to obtain the fuel injection time TAU, and the fuel injection amount is learned and controlled by this.
Such learning control is described in, for example, JP-A-60-233328 and JP-A-61-16243.

By the way, in the learning control according to the above equation (6), the learning value learned according to the magnitude of the average value of the learning value (acceleration increasing coefficient KAC and deceleration decreasing coefficient KDC) and the air-fuel ratio feedback back correction coefficient updated at the transition time. Since two learning values such as KG are used, for example, if the acceleration increase coefficient is large, the fuel increase value during acceleration increases and the air-fuel ratio becomes rich. At this time, the learning value KG is learned. As shown in FIG. 16, erroneous learning is performed so that the average value of the air-fuel ratio feedback correction coefficient becomes small and the learning value KG becomes small. For this reason,
The air-fuel ratio becomes lean because the learning value KG is small when the fuel increase value KAC / TPAEW becomes 0 after shifting to the steady operation state. Also, if the learning value KG is small and the vehicle shifts to the acceleration operation state, it is determined that the fuel increase value is small and the acceleration increase coefficient KA
It is erroneously learned so that C becomes larger. In addition, although the acceleration increase coefficient has been described above, the same applies to the deceleration decrease coefficient.

Therefore, in the present embodiment, the transition time and the steady time are discriminated, learning of the learning value KG is prohibited during the transition to learn the transition learning value, and learning of the transient learning value is prohibited during the steady state to learn the learning value KG.
The learning of the transient learning value and the learning value KG are not executed at the same time by performing the learning of the air-fuel ratio to prevent erroneous learning and to shift from the transient operating state to the steady operating state and I try to prevent disturbance.

The third embodiment will be described in detail below with reference to the drawings. 17th
In the figure, the steps 103 and 121 and the step 123 for setting and resetting the learning permission flag F are added to the 360 ° CA routine of FIG. Therefore, in FIG. 17, steps in the vicinity of the added step are described with the same reference numerals as in FIG. 1, and other steps that are the same as in FIG. 1 are omitted. When it is determined in step 102 that the vehicle is in the accelerated state, the learning permission flag F is reset in step 103. If it is determined in step 120 that the vehicle is in the decelerated state, the learning permission flag F is reset.
Then, when the steady state is judged by the steps 102 and 120, the learning permission flag F is set at step 123.

FIG. 18 shows a learning routine for learning the learning value KG, and in step 270, the arithmetic mean value FAFAV of the maximum and minimum values of the air-fuel ratio feedback back correction coefficient FAF every time the air-fuel ratio feedback back correction coefficient FAF skips a predetermined number of times. That is, the average value FAFAV of the air-fuel ratio feedback back correction coefficient is calculated from A, B, C ... Shown in FIG. 19 (2) according to the following equation (7).

In the next step 272, it is determined whether or not the learning condition is satisfied, for example, by determining whether the engine cooling water temperature exceeds a predetermined value (for example, 80 ° C.) during air-fuel ratio feedback control. . When it is determined that the learning condition is satisfied, it is determined in step 274 whether the learning permission flag F is set. When the learning permission flag F is set, the average value FAFAV of the air-fuel ratio feedback back correction coefficient is within a predetermined range in steps 276 and 278 (for example, 0.98≤FAFAV).
≦ 1.02) is determined, and when the value is within the predetermined range, the learning value KG is not learned and this routine is ended. On the other hand, when the average value FAFAV exceeds the upper limit of the predetermined range, the learning value KG is increased in step 282 by a predetermined value (for example, 0.02), and the average value FAFAV
If is less than the lower limit value of the predetermined range, the learning value KG is reduced by a predetermined value (for example, 0.02) in step 280.

As a result, when the acceleration / deceleration state is determined, the acceleration increase coefficient KAC and the deceleration decrease coefficient K are determined by the routine of FIG.
When DC is learned and is determined to be in a steady state, the learning value KG is learned by the learning routine shown in FIG. As a result, the transient learning value and the learning value KG are not learned at the same time, which prevents erroneous learning and prevents deterioration of the exhaust emission. Note that the intake pipe pressure PM and the acceleration increase coefficient KAC / TPAE when controlled as described above
W, O ₂ sensor output, count value CAC, acceleration increase coefficient KAC,
FIG. 20 shows changes in the air-fuel ratio feedback correction coefficient FAF and the learning value KG.

In the above-described embodiment, an example in which the learning permission flag F determines the area for learning each learning value has been described. However, the learning permission flag in step 274 in FIG. 18 is used instead of the determination as to whether or not the learning permission flag is set. It is also possible to provide a step for determining whether it is a transient time or a steady time so that the learning value KG is learned only in the steady time. Further, a plurality of learning values KG may be provided corresponding to a plurality of areas divided according to the magnitude of the load. The basic fuel injection time during transition TPAEW
May be determined only by the deviation DLPM of the intake pipe pressure.

The third embodiment can be applied to the second embodiment which determines the acceleration / deceleration state by TPAEW _i instead of DLPM, and the fuel injection routine of FIG. 8 is used in the second embodiment. The fuel injection routine of FIG. 14 can be used in the first embodiment. Further, although the example in which the present invention is applied to the engine that calculates the basic fuel injection time by the intake pipe pressure and the engine rotation speed has been described above, the intake air per rotation of the engine is provided with the air flow meter that detects the intake air amount. The present invention can be applied to an engine that calculates the basic fuel injection time from the amount. Further, the count value at the time was described O ₂ sensor output Ritsuchi For an example of calculating the time of the deviation by Deikurimento the count value counted value when incremented vital O ₂ sensor output lean when the O ₂ sensor output Ritsuchi in the above The deviation may be obtained by incrementing the count value when decrementing and leaning the O ₂ sensor output. The count value and the O ₂ sensor output is lean when the O ₂ sensor output is incremented another count value and when showing the lean and when showing the Ritsuchi O ₂ sensor output indicates a Ritsuchi The deviation from the count value at the time of calculation may be calculated to calculate the deviation of the above time. Also, in the above, an example in which the deviation is obtained by counting for each predetermined crank angle (360 ° CA) has been described, but the deviation may be obtained by counting for each predetermined time.

[Brief description of drawings]

FIG. 1 is a flow chart showing a routine for updating the acceleration increase coefficient and the deceleration decrease coefficient in the first embodiment of the present invention,
FIG. 2 is a diagram showing a change in a feed back correction coefficient and an air-fuel ratio in the conventional art, FIG. 3 is a diagram showing a change frequency in the conventional air-fuel ratio, and FIG. 4 is a fuel injection amount learning to which the present invention is applicable. FIG. 5 is a schematic diagram of an engine equipped with a control device, FIG. 5 is a block diagram showing details of the microcomputer shown in FIG. 4, and FIG. 6 is a diagram showing changes in the acceleration increase coefficient and deceleration decrease coefficient in the first embodiment. FIG. 7 is a flowchart showing a routine for calculating an air-fuel ratio feedback back correction coefficient, FIG.
FIG. 11 is a flow chart showing a routine for calculating a fuel injection time, and FIGS. 9 and 10 are diagrams showing an increase / decrease amount time corresponding to the engine speed and an increase / decrease amount time corresponding to the engine cooling water temperature, respectively. Is a diagram showing changes in the basic fuel injection time at transition and the fuel injection time in the first embodiment,
FIG. 12 is a flow chart showing a routine for correcting the acceleration increase coefficient and the deceleration decrease coefficient in the second embodiment of the present invention.
FIG. 13 is a diagram showing changes in the acceleration amount increase coefficient and basic fuel injection time in the second embodiment, and FIG. 14 is a flow chart showing a routine for calculating the fuel injection time in the second embodiment, and FIG. Is a diagram showing changes in the basic fuel injection time at transition and fuel injection time in the second embodiment, and FIG. 16 is a diagram showing changes in acceleration increase coefficient and learning value.
FIG. 18 is a flow chart of a routine including steps for setting and resetting the learning permission flag in the third embodiment, and FIG. 18 is a flow chart showing a learning routine in the third embodiment.
Figure (1) is a diagram showing changes in the air-fuel ratio signal, Figure 19 (2)
Is a diagram showing the change in the air-fuel ratio feedback back correction coefficient,
FIG. 20 is a diagram showing changes in the acceleration amount increase coefficient, learning value and the like in the third embodiment. 6 ... Pressure sensor, 8 ... Slot valve 24 ... Fuel injection valve, 30 ... O ₂ sensor, 44 ... Micro computer.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-54-132021 (JP, A) JP-A-59-128944 (JP, A) JP-A-59-194056 (JP, A)

Claims (1)

[Claims] 1. A transient state determining means for determining a transient state,
In the transient state, fuel injection means for injecting by increasing or decreasing the amount of fuel determined by the correction coefficient during the transient, an oxygen concentration sensor for detecting the residual oxygen concentration in the exhaust gas, and the oxygen concentration sensor output from the target air-fuel ratio Deviation calculating means for calculating the deviation between the time when the latch state is shown and the time when the oxygen concentration sensor output shows the lean state from the target air-fuel ratio, and the deviation is outside the predetermined range in the transient state. A fuel injection amount learning control device for an internal combustion engine, comprising: a correction unit that corrects the correction coefficient during the transition so that the deviation when the value becomes a value falls within the predetermined range.