JPS63189637A - Air-fuel ratio feedback control method for internal combustion engine - Google Patents

Abstract

PURPOSE:To compensate a change with aging, by continuously sensing a period of time during which a sensed value of an exhaust gas concentration shifts from a lean side to a rich side and a period of time during which it shifts from the rich side to the lean side and by changing a control constant based on a ratio of the periods of time in air-fuel ratio feedback control. CONSTITUTION:An electronic control unit 5 calculates a basic amount of fuel injection based on values sensed by a suction gas absolute pressure sensor 8 and an r.p.m. sensor 10 and compares a value sensed by an O2 sensor 13 with a reference value for effecting feedback control of an air-fuel ratio by proportional integral control. The electronic control unit 5 continuously measures a time t1 when a value sensed by the O2 sensor 13 changes from a lean side to a rich side with respect to the reference value and a time t2 when it changes from the rich side to the lean side. Thus, it senses a deviation of an air-fuel ratio based on a ratio between a period of time of t1-t2 and a period of time of t2-t1 and corrects the reference value changes so that the deviation is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to an air-fuel ratio feedback control method for an internal combustion engine, and in particular to an air-fuel ratio feedback control method for an internal combustion engine that compensates for changes over time in the output characteristics of an exhaust gas concentration detector disposed in the engine's exhaust system. The present invention relates to a fuel ratio feedback control method.

(Technical background and problems) Conventionally, the exhaust concentration (oxygen concentration) detected by an exhaust concentration detector (for example, 02 sensor) placed in the exhaust system of an internal combustion engine is compared with a predetermined reference value. Based on the results, the air-fuel ratio of the mixture supplied to the engine is controlled to the stoichiometric mixture ratio at which the three-way catalyst installed in the engine's exhaust system achieves the maximum conversion efficiency, thereby improving the exhaust gas characteristics. An air-fuel ratio feedback control method for an internal combustion engine is generally used (for example, Japanese Patent Application Laid-open No. 137633/1983).

The O3 sensor used for such air-fuel ratio control uses zirconium oxide or the like as a sensor element, and the amount of oxygen ions that permeate through the zirconium oxide is determined by the oxygen partial pressure in the atmosphere and the oxygen content in the exhaust gas. The oxygen concentration in the exhaust gas is detected by changing the output voltage of the 02 sensor in response to the change in pressure.

However, the output characteristics of the 0□ sensor configured as described above change over time, and especially after a vehicle equipped with the sensor has been running for a long time, its output characteristics deteriorate due to durability, and as a result, when the same condition It is known that the controlled air-fuel ratio shifts to the rich side compared to when shipped from the factory, even though air-fuel ratio feedback control is performed.

If no measures are taken against such changes in the characteristics of the O2 sensor over time, a problem will arise in that the operating performance, fuel efficiency, and exhaust gas characteristics of the engine will deteriorate.

(Objective of the Invention) The present invention has been made to solve the above-mentioned problems, and the present invention corrects the air-fuel ratio of the air-fuel mixture supplied to the engine according to the degree of change over time in the characteristics of the 02 sensor to achieve a target air-fuel ratio. An object of the present invention is to provide an air-fuel ratio feedback control method for an internal combustion engine that can achieve the following.

(Structure of the Invention) In order to achieve the above object, according to the present invention, an exhaust concentration detection value detected by an exhaust concentration detector disposed in the exhaust system of an internal combustion engine is compared with a reference value, and a The air-fuel ratio of the supplied air-fuel mixture is proportionally corrected by increasing or decreasing the air-fuel ratio by a first correction value when the detected exhaust gas concentration value changes from the rich side to the lean side or from the lean side to the rich side with respect to the reference value. control, and when the detected exhaust gas concentration value is on the lean side or rich side with respect to the reference value, the target air-fuel ratio is achieved by at least one of integral control, which increases or decreases the air-fuel ratio at predetermined time intervals using second correction values. In an air-fuel ratio feedback control method for an internal combustion engine that performs feedback control, a first time period from the time when the detected exhaust gas concentration value changes from the lean side to the rich side with respect to the reference value to the time when the detected exhaust gas concentration value changes from the rich side to the lean side. and a second time from the time when the detected exhaust gas concentration value changes from the rich side to the lean side with respect to the reference value to the time when it changes from the lean side to the rich side, and according to the ratio thus calculated. There is provided an air-fuel ratio feedback control method for an internal combustion engine, characterized in that the reference value is changed.

(Example) Hereinafter, an example of the present invention will be described in detail based on the accompanying drawings.

FIG. 1 is a block diagram showing the overall configuration of a fuel supply control device for an internal combustion engine to which the method of the present invention is applied. Reference numeral 1 indicates, for example, a four-cylinder internal combustion engine, and the engine 1
An intake pipe 2 communicating with the atmosphere is connected to the intake pipe 2.
A throttle valve 3 is provided in the middle. A throttle valve opening sensor 4 is connected to the throttle valve 3, which detects the valve opening θ〒H and outputs an electrical signal.

The detection signal is sent to an electronic control unit (hereinafter referred to as rEcUJ) 5 that controls the engine by executing arithmetic processing to calculate the air-fuel ratio and the like, as described below.

A fuel injection valve 6 is provided between the engine 1 and the throttle valve 3. The fuel injection valve 6 is connected to the engine 1.
One is provided for each cylinder, and is connected to a fuel pump (not shown), and the valve opening time for injecting fuel is controlled by a drive signal supplied from the ECU 3.

On the other hand, a pipe 7 is provided in the intake pipe 2 downstream of the throttle valve 3.
An intake pipe absolute pressure sensor 8 that detects the absolute pressure Pa^ in the intake pipe 2 is connected via E.
Sent to CU3.

An engine water temperature sensor 9, which is made of, for example, a thermistor, and detects the temperature Tw of the cooling water is provided on the peripheral wall of the cylinder of the engine 1 filled with cooling water, and its detection signal is sent to the ECU 3.

An engine rotation speed sensor (hereinafter referred to as "rNe sensor") 10 is attached to the camshaft or crank (not shown) of the engine 1, and this Ne sensor 10 generates a pulse signal ( TDC signal) is output, and this TDC signal is sent to the ECU 3.

A three-way catalyst 12 is connected to the exhaust pipe 11 of the engine 1, and the three-way catalyst 12 converts HClGo and NOx in the exhaust gas.
An 02 sensor 1, which is an exhaust gas concentration sensor, is installed in the exhaust pipe 11 on the upstream side of the three-way catalyst 12, which purifies the components.
3 is installed, the O2 sensor 13 detects the oxygen gas concentration in the exhaust gas, and the detection signal (VO2) is sent to the ECU.
Sent to 3.

Furthermore, a vehicle speed sensor 14 that detects the speed Sp of the vehicle is connected to the ECU 3, and a detection signal from the vehicle speed sensor 14 is sent to the ECU 3.

The ECU 3 inputs detection signals from the various sensors mentioned above,
The fuel injection time Tou〒 of the fuel injection valve 6 is calculated by the following equation.

Tou 〒=Ti @Ko read out from the memory device is the 02 feedback correction coefficient, which is calculated based on the 0□ feedback correction coefficient calculation subroutine to be described later. 2 and 8 are correction coefficients and correction variables respectively calculated according to the various engine parameter signals, and are based on detection signals from the various sensors described above to determine fuel efficiency characteristics, engine acceleration characteristics, etc. according to the engine operating state. A predetermined value is determined so as to optimize the various characteristics of.

ECU3 calculates the fuel injection time Tou as described above.
A drive signal for opening the fuel injection valve 6 is output based on 〒.

FIG. 2 is a diagram showing the circuit configuration inside the ECU 3 shown in FIG.
After waveform shaping in step 1, the central processing unit (rCPU
503 (hereinafter referred to as "rMe counter") 502
Also supplied. The Me counter 502 is the Ne sensor 1
It counts the time interval from the input of the previous TDC signal from 0 to the input of the current TDC signal, and its count value M
e is proportional to the reciprocal of the engine speed Ne.

The Me counter 502 transfers this count value Me to the data bus 5.
10 to the CPU 503.

On the other hand, throttle valve opening (θ〒H) sensor 4. Absolute pressure (
Pa^) sensor 8, engine water temperature (TW) sensor 9.
The output signals from the o2 sensor 13 and the vehicle speed (Sp) sensor 14 are respectively applied to a level correction circuit 504, and after being corrected to a predetermined voltage level in the circuit 504, the output signals are output from the CP.
Multiplexer 50 operates based on instructions from U503
5 is sequentially supplied to an analog-to-digital converter (A/D converter) 506. The converter 506 converts the output signals of the aforementioned sensors into digital signals, and supplies the digital signals to the CPU 503 via the data bus 510.

The CPU 503 further connects a read-only memory (hereinafter referred to as FROMJ) 507 and a random access memory (hereinafter referred to as "RAM") 50 via a data bus 510.
8 and a drive circuit 509. ROM507
stores various control programs executed by the CPU 503, the details of which will be described later, and various data such as correction coefficients and correction variables. Further, the RAM 508 temporarily stores calculation results etc. obtained by the execution of the various control programs by the CPU 503.

Then, the CPU 503 reads out or calculates the coefficient value or variable value corresponding to the output signal of each sensor described above from the ROM 507 according to the control program stored in the ROM 5074m, and adjusts the value of the fuel injection valve 6 based on the above calculation formula (1). The fuel injection time T out is calculated, and the value obtained by this calculation is supplied to the drive circuit 509 via the data bus 510 . The drive circuit 509 calculates the calculated fuel injection time To
The fuel injection valve 6 is opened over uT.

Next, a method of calculating 0 and the feedback correction coefficient Ko according to the temporal change in the characteristics of the O2 sensor according to the present invention will be explained.

As mentioned above, 0. The output characteristics of the sensor deteriorate due to endurance driving of the vehicle, and as a result, the feedback-controlled air-fuel ratio is biased towards the rich side or the lean side. This 0. The degree of deterioration of the sensor is 0.0% under stable engine running conditions. The time T*v required from the time when the sensor output voltage value vO□ (Fig. 3) changes from the lean side to the rich side with respect to the reference value v■F to the time when the sensor output voltage value vO□ (Fig. 3) changes from the rich side to the lean side
(between t2 and t3 and t5 and t6 in Figure 3) and the time required from the time when the output voltage value Vo2 changes from the rich side to the lean side with respect to the reference value V yutp to the time when the output voltage value Vo2 changes from the lean side to the rich side. TLV (Figure 3 t, ~t
KOX-(=Tt, v/T*v)
) (hereinafter TRV is referred to as "rich side transition time J" and TLV is referred to as "lean side transition time"), and this KOX value decreases according to the cumulative mileage of the vehicle, that is, the degree of durability deterioration. This has been experimentally confirmed.

Therefore, in the present invention, the air-fuel ratio is adjusted to the target air-fuel ratio by changing the reference value v■F according to the KOX value to correct the deviation of the air-fuel ratio toward the rich side or the lean side due to the deterioration of the 02 sensor. control accurately.

More specifically, in this embodiment, the value of KOX is set to a plurality of predetermined values KOX1 to KOX4 (KOX, >KOX, >KO
X3>KOX4) and set the reference value V according to this result.
IIIEF is changed as follows using correction values Δv寞□, ΔV* (FIG. 4).

(1) When KOX>KOX, it is estimated that the air-fuel ratio is largely biased toward the lean side, and a correction value Δv* is added to the reference value V large ip.

(2) When K OX 1 > K OX > K OXx, it is estimated that the air-fuel ratio is slightly biased towards the lean side,
A correction value ΔVex (<ΔV feces 2) is added to the reference value v■F.

(3) KOX,>KOX>KOX, (7) It is estimated that the air-fuel ratio is controlled to be approximately equal to the target (theoretical) air-fuel ratio, and the value of the reference value V trap i+p is held.

(4) KOX! > KOX > KOX, (7), it is estimated that the air-fuel ratio is slightly biased toward the rich side, and the reference value V is
Subtract the correction value Δvtribute□ from p.

(5) When KOX<KOX4, it is estimated that the air-fuel ratio is largely biased toward the rich side, and the correction value ΔvIIIi is subtracted from the reference value v■F.

FIGS. 5 and 6 show the reference value Vllll! for carrying out the method of changing the reference value Vlargeip described above. It is a program flowchart of the F setting subroutine.

First, in steps 3o to 36, it is determined whether the engine is in a predetermined operating state in which the output voltage Vo of the sensor 13 should be reversed at a normal cycle.

That is, in step 30, it is determined whether the temperature of the O2 sensor 13 is sufficiently high based on whether the engine water temperature Tw is greater than a predetermined value Tvox, and then in step 31, it is determined whether the engine is actually under feedback control. Discern. Furthermore, in steps 32 to 35, as a condition for the predetermined operating state, the engine rotation speed Ne is set to a predetermined value N.
eoxt, and NQOXH (step 32), and determine whether the intake pipe absolute pressure Pa^ is a predetermined value P
Determination of whether the value is between BOXL and P80XH (step 33), Determination of whether the vehicle speed SP indicating the cruising state of the vehicle is a value between the predetermined value SPOXL and 5poxo (step 34), and a determination as to whether the absolute value of the degree of change ΔPa^ of the absolute pressure indicating a stable cruising state is smaller than a predetermined width ΔP BOXA (step 35) are executed, respectively. Furthermore, in step 36, after these operating conditions are satisfied (steps 32 to 35)
After all the determination results are positive (Yes), it is determined whether this operating state has been continued for a certain period of time Tx.

Therefore, when all the determination results in steps 30 to 36 above are affirmative (Y6!I) (the operating state shown in FIG. 7), it is determined that the engine is in the predetermined operating state,
The program after step 38 is executed for the first time.

Incidentally, if the determination result in any one of steps 3o to 36 is negative (NO), the process proceeds to step 37, sets the averaging number count value nAV, which will be described later, to O, and ends this program.

In steps 38 to 52, the KOX value (=TLV/T large v) representing the degree of deterioration of the 02 sensor 13 described above
For determining Tt, the average value TLVAV of v time, and the average value TvAv of Tkv time are calculated.

Below, the average value TL from step 38 to step 52
VAV, 'rlVAV calculation method is shown in Figure 3.Ot
The explanation will be based on a timing chart of the sensor output voltage value vO□.

Now, let us consider a case where the determination result in step 36 becomes affirmative for the first time at time t1 in FIG. 3.

First, in step 38, it is determined whether the current output voltage value v02n of the 0□ sensor 13 is larger than the set reference value v large needle. At time t1 in Fig. 3, this discrimination result is negative (N
O), and the process proceeds to step 39, where it is determined whether the previous output voltage value vO□n-□ is larger than the set reference value VREF. At time t□, this discrimination result is also negative (No
), one program ends and moves to the next loop.

Even in subsequent loops, the output voltage value Vo is the set reference value V.
Until EF is exceeded (until time t2 in FIG. 3), the determination results in steps 38 and 39 are both negative (NO).

When the output voltage value v02 of the 02 sensor 13 rises and exceeds the set reference value V■F for the first time in this loop (Fig. 3 t
2), the determination result in step 38 is affirmative (Yes).
) and proceed to the second step 4°.

In step 40, the output voltage value V O,n in the previous loop
-, is larger than the set reference value V large ip is determined, and at time t2, this determination result is negative (No), and at this point in time the tox timer starts counting in step 43, which will be described later, in the first step 41. The count value tax at is set as the transition time Tox, and the process proceeds to the next step 42. In this step 42, the transition time T set in step 41 is
It is determined whether ox is within the allowable range TOXL to Toxo.

Since the tox timer has not yet started at time t2 in FIG. 3, the result of this determination is negative (No), and the process skips steps 50 to 52, which will be described later, and proceeds to step 43. By making the determination in step 42, it is possible to avoid extremely long transition times that occur during the transition period of the engine operating state, such as when the tox timer has already been activated before the time t, or between time t4 and time t5 in Figure 3. It is also possible to eliminate extremely short transition times due to noise generation as shown in FIG.

In step 43, the count value of the tox timer is reset and started, and in the next step 44, it is determined whether the averaging number count value nAV, which will be described later, is greater than or equal to a predetermined number of times NAV, and if the first determination result is affirmative (Yes). If so, the program from step 59 (FIG. 6) to be described later is executed, and if the answer is negative (NO), this program is ended.

In the next and subsequent loops, the output voltage value vO□ of Q2 sensor 13
as long as it exceeds the set standard value V*ip (Fig. 3 t2
to time t3), the determination results in steps 38 and 40 are both affirmative (Yes), and one program is ended.

Therefore, during this period, the count value of the tax timer is not reset to 1. It comes to represent the elapsed time from the point in time.

When the output voltage value Vo of the 0□ sensor 13 decreases and the output voltage value Vo crosses the set reference value v■F in this loop (at time t3 in FIG. 3), step 3 is performed again.
The determination result in step 8 is negative (No), and the process proceeds to step 39. In this loop (at time t3), the output voltage value Vo, n-1 in the previous loop is larger than the set reference value V trap tp, so the determination result is affirmative. (Yes) and next step 45
At this point, the count value of the tax timer tox(t,
-t, time) is set as the transition time Tox, and the process proceeds to the next step 46.

In step 46, similarly to step 42, step 45
The transition time Tox set in is within the allowable range WiToxL~T
It is determined whether or not it is within 0X)l, and the 1 determination result is negative (
No), steps 47 to 49 are skipped and the process proceeds to step 43.

When the determination result in step 46 is affirmative (Y18), the process proceeds to step 47, and the correction coefficient KN! :〒, multiplies by Kpsτ to obtain a new transition time Toxc. Correction coefficient KNI! 〒, KPBT are the Kst〒-Ne tables shown in Figs. 8 and 9, respectively, depending on the engine speed Ne and intake pipe absolute pressure Pa^
ps〒-Pa^ This is the value read from the table. In this way, in step 46, the correction coefficients K NET, Kpa
The reason why the transition time tox is corrected is that the transition time, that is, the reversal period of the 02 sensor itself changes significantly in accordance with changes in the engine rotational speed N e and the intake pipe absolute pressure Pa^.

The transition time TOSC corrected in step 47 is substituted into the following equation (2) in the next step 48, and as a result, 0
□The output voltage value Vo of the sensor 13 is the set reference value V*tp
The average value T dung VAVn of the rich side transition time Toxc (==T trap v) from the time when the lean side changes to the rich side to the time when the rich side changes to the lean side is calculated.

...(2◆ Here, Ttribute vAvn- is the previous value of the rich side transition time average value, COX is the averaging constant for calculating the average value, and Step 63, 65, 67, 69, which will be described later)゜71
The value COX corresponds to the degree of deterioration of the 02 sensor.

~COX, (however, 0 COX, ~COX, <256).

Returning to FIG. 5, in the next step 48, 1 is added to the averaging count value nAV representing the number of executions of transition time averaging based on the above-mentioned formula (2), and as described above, in step 48,
3, the count value of the tox timer is reset and started. In the next step 44, the count value of the tox timer is reset and started.
8 and Step 51, which will be described later, determines whether the averaging number count value nAV representing the number of times the transition time is averaged is greater than or equal to a predetermined number of times NAV. In addition, the predetermined number of times NA
V is set to a required value in steps 63, 65, 67, 69, and 71, which will be described later, depending on the degree of deterioration of the O2 sensor.

As described above, the transition time T between time points t2 and t3 in FIG.
ox (=T trap v) is set, and the rich side transition time average value T trap vAvn is calculated based on the corrected value Toxc of the transition time Tox.

The output voltage value Vo of the o2 sensor 13 is the set reference value V trap t
After going below p and after the time Di), step 38.39
Both determination results are negative (NO), and this program ends. Therefore, during this time, the count value of the tox timer is not reset and comes to represent the elapsed time from the 1 degree point.

0, when the output voltage value vo2 of the sensor 13 increases and crosses the set reference value V*ip in this loop (time t4 in FIG. 3), the determination result in step 38 is affirmative (Yes).
Then, the determination result in step 40 is negative (No).
Then, the process proceeds to step 41.

In step 41, the count value tax (t, -t, time) of the tox timer at this point (time point 14) is set as the transition time Tox (==TLv), and the process proceeds to the first step 42, where the transition time Tax is Tolerance range Toxt, ~Toxo
Next, in step 50, similarly to step 47, the transition time Tox is determined by a correction coefficient KNE corresponding to the engine rotational speed Ne and the intake pipe absolute pressure Pa^.
Corrected by T, K p a〒, new transition time Toxc
shall be.

Furthermore, in step 51, the corrected transition time Toxc is substituted into the following equation (3), and the output voltage value vO□ changes from the rich side to the lean side with respect to the set reference value v■F, and from the lean side to the rich side. The lean side transition time average value T t, , vAvn up to the time when the lean side transition time changes to is calculated.

COX 256-COXTLvAvn
= −X Toxc + 256X Tt, vAvn
-1...(3) Here, TLvAvn-x is the previous value of the lean side transition time average value, and COX is the same averaging constant as in the above-mentioned equation (2).

In step 52, 1 is added to the averaging count value nAV as in step 49, and in step 43
．． Proceed to step 44.

In this way, the calculation of the rich side and lean side shift time average values in steps 38 to 52 is repeated until the averaging count value nAV reaches the predetermined number of times NAV, which indicates the degree of deterioration of the 0□ sensor. Value KOX
This is to obtain (= TLVAV/TRVAV) more accurately.

If the calculation of the rich side and lean side transition time average values is performed a predetermined number of times NAV, and the determination result in step 44 is affirmative (Yes), steps 59 to 77 in FIG.
The reference value 'Vt according to the degree of deterioration of the 02 sensor 13
A change of 1p is made.

First, in steps 59 to 62, the value K OX (=Tt, vAv/T hole V) representing the degree of deterioration of the 02 sensor is determined.
AV) and the aforementioned predetermined values KOX1, KOX, , KOX,
, and KOX4 (KOX, >KOX,
>KOX, >KOX4). That is, in step 59, the rich side transition time average value TRVA after the completion of averaging a predetermined number of times is calculated.
In step 60, it is determined whether or not the value obtained by multiplying V by a predetermined value -KOXt is larger than the lean side transition time average value TLVAV after the completion of averaging for a predetermined number of times.
It is determined whether or not the value obtained by multiplying by a predetermined value KOX is larger than the average value TLVAV. On the other hand, step 61
In step 62, whether or not the value obtained by multiplying the average value T trap vhv by a predetermined value KOX4 is smaller than the average value TLVAV is determined.
X.

It is determined whether the multiplied value is smaller than the average value TLVAV.

Therefore, all the determination results in steps 59 to 62 are affirmative (
If the answer is Yes), the value KOX representing the degree of deterioration is between the predetermined values KOX and KOX, and in this case it is estimated that the air-fuel ratio is controlled to be approximately equal to the target air-fuel ratio, and step 63 is performed. Next, the averaging constant COX used in equations (2) and (3) and the predetermined number of averaging times NAv are set to standard values COX, NAV, respectively.

On the other hand, when the determination result in step 59 is negative (NO), since the KOX value is larger than the predetermined value KOX1, it is estimated that the air-fuel ratio is largely biased toward the lean side, and in step 64, the reference value V trap ip is The value in the previous loop of V*tvn-
Add the correction value Δv1 to 0 to get the reference value V* for this loop
tpn, and in the next step 65 the averaging constant COX
and the predetermined number of times NAV as COXl(> COXa
), NAVl(<NAVJ).

Moreover, when the determination result in step 60 is negative (NO),
The KOX values are predetermined values KOX1 and KOX.

Since the air-fuel ratio is between ΔV*
z) to get the reference value VI for this loop! =pn, and in the next step 67, the averaging constant COx and the predetermined number of times NAV are set as COX1 (COX, <COX2
< COX 1) , NAVI (NAVII
> NAVI > N AV, ).

Moreover, when the determination result in step 61 is negative (NO),
Since the KOX value is smaller than the predetermined value KOx4, it is estimated that the air-fuel ratio is largely biased toward the rich side, and step 6
8, the value of the reference value vuF in the previous loop V*yt=n-,
The correction value ΔV− is subtracted from the reference value V trap wvn for this loop, and in the next step 69, the averaging constant COx
and the predetermined number of times NAV, respectively COX, (=COX,)
, N AV4 (= N AVl).

Further, when the determination result in step 62 is negative (No),
Since the KOX value is between the predetermined value KOX and KOX4, it is estimated that the air-fuel ratio is slightly biased toward the rich side.
In step 70, the value V of the reference value V REF in the previous loop is
The correction value AV large is subtracted from the large yn-0 to set the reference value V+utpn in this loop, and in the first step 71, the averaging constant coX and the predetermined number of times NAV are respectively COX,
(=COX,), set to NAV3 (-=NAvt).

In this way, the averaging constant COX is set to a larger value (COXl
, COX, ), by setting the averaging predetermined number of times NAv to a smaller value (NAVl, NAV4), it is possible to speed up the averaging degree of the transition time average value T large vAvn, Tt, vAvn, and therefore, the time is quickly emptied. The fuel ratio can be controlled to the target air-fuel ratio.

In said step 64.66.68.70 the correction value Δ
The value V trap tpn in the current loop of the reference value V large tv changed by v large or ΔV- is greater than the predetermined upper limit value V*tpH in step 72, but is changed to a predetermined lower limit value in step 73. It is determined whether it is smaller than V*tpt,
When the determination results in steps 72 and 73 are both negative (No), the reference value V*ipn obtained in the current loop is set as the reference value VREF (step 74), and either one of the determination results is negative (No). If Yes), the reference value V*gpn-1 obtained up to the previous loop is set to the reference value V large ip (step 75).

In step 76, the thus changed reference value V is large! F is second
The data is stored in the RAM 508 in the figure, and in the next step 77, the averaging count value nAV is set to 0, and the program ends.

The reference value Vutp thus stored in the RAM 508 is used in steps 38, 39, and 40.

Furthermore, it is used in the 08 feedback correction coefficient calculation subroutine to be described later, thereby making it possible to bias the value of the correction coefficient Ko toward the rich side or the lean side.

FIG. 10 is a program flowchart of the 02 feedback correction coefficient calculation subroutine using the reference value VllKF changed according to the aging of the O sensor using the method described above.

First, it is determined whether the activation of the O2 sensor 13 has been completed (step 81). That is, the output voltage value of the O2 sensor 13 is determined to be at the activation starting point VX (for example, by the internal resistance detection method of the O2 sensor 13). 0.6v) and determines that it is activated when it reaches Vx.If the determination result is negative (No), the correction coefficient K
o, is set to 1.0 (step 82). On the other hand, if the determination result is affirmative (Yes), it is determined whether the engine is in the open control region (step 83).

This open control includes a high-load operating region, a low-speed region, an idle region, a high-speed region, a lean mixture region, etc. The high-load operating region includes, for example, a value in which the fuel injection time Tou〒 is larger than a predetermined value Twot. This is the area set to .

Here, T vat is a constant and is the lower limit value of the amount of fuel supplied necessary to enrich the air-fuel mixture during high-load operation such as when the throttle valve is fully open. The low rotation range is a range where the engine rotation speed Ne is below a predetermined value NLOP (for example, 700 rpm) and the intake pipe absolute pressure Pa is above a predetermined value PBIDL (for example, 360 mHg).

In the idle region, the engine rotation speed Na is a predetermined rotation speed NH.
OP (for example, 1000 rpm) and the absolute pressure Pa^ is lower than the predetermined pressure PBIDL, and the high rotation region is a region where the engine rotation speed Na is lower than the predetermined rotation speed N
The area is larger than Hop (for example, 3QQOrpm).

The air-fuel mixture lean region is a region in which the intake pipe absolute pressure Pa^ is smaller than the discrimination value P5Lts, which is set to a larger value as the engine speed Ne increases.

When the engine is in any of the above ranges, it is determined that the engine is being operated in the open control range, and in this case, the process proceeds to step 82 and the correction coefficient Ko is set to 1.0.

On the other hand, if the determination result in step 83 is negative (NO), it is determined that the engine is in the operating range where feedback control is required, and the process moves to closed loop control.
The output voltage value Vo, which is the output level of No. 3, is the reference value V large iF determined by the reference value v Ohno setting subroutine (Fig. 5 and Fig. 6) at the previous input of the TDC signal and the current input.
Step 84) If the determination result is affirmative (Yes), proportional (P term) control is performed (from step 85), and if negative (No), integral control is performed. (Step 90 onwards).

In step 85, the output voltage value vo2 of the O2 sensor 13 is the reference value Vll! It is determined whether or not the level is low with respect to P, and if the determination result is affirmative (Yes), a rich side proportional control correction value P large is read out from the Ne-PR child table according to the engine rotation speed Ne (step 86 )0 Then, in step 87, this correction value P hole is added to the previous same value of the correction coefficient Ko. The determination result in step 85 is negative (NO)
If so, set the lean side proportional control correction value Pt to No-PL.
Read the child table according to the engine speed Ne (step 88), and use the correction value PL thus read as the correction coefficient K.
o.

is subtracted from the previous equivalent value (step 89).

If the determination result in step 84 is negative (NO), integral control is performed as follows. First, in step 90, similarly to step 85, it is determined whether the output voltage value Vo of the 0□ sensor 13 is at a low level with respect to the reference value v■F. If this determination result is positive (Yes), T
1 is added to the value of the count number N of DC signal pulses (step 91), and it is determined whether the count number N+L has reached a predetermined value N+ (for example, 4) (step 92). As a result of this determination, If the count number NIL has not yet reached NI, the correction coefficient KO□ is kept at the value from the previous loop (step 93), and if the count number NIL has reached NI, the correction coefficient Ko is set to the engine rotational speed. A correction value Δk corresponding to Ne is added (step 94), and the number of pulses counted so far N is reset to 0 (step 95), and each time N reaches NI, a correction value Δk is added to the correction coefficient Ko. Add . On the other hand, if the determination result in step 90 is negative (No), 1 is added to the count number NIH of TDC signal pulses (step 96
), the count number N1)l is the predetermined value N! Step 97) 1 The determination result is negative (No).
In this case, the value of the correction coefficient Ko is kept at the value at the previous loop. If the result of the first determination is positive (Yes) in step 68), the correction value Δk is subtracted from the correction coefficient Ko. Reset the counted pulse number NIH to 0 (step 100), and set the NIH to
The correction value Δk is subtracted from the correction coefficient Ko every time the correction coefficient Ko is reached.

In this way, the reference value to be compared with the output voltage value vo of the 02 sensor is changed according to the degree of deterioration of the 02 sensor 13,
By applying the thus changed reference value VRIEF to the calculation of the o3 feedback correction coefficient Ko,
If the air-fuel ratio is biased towards the rich side due to deterioration of the 02 sensor 13, the time taken to determine the rich side of the air-fuel ratio by the 02 sensor can be made longer to reduce the value of the correction coefficient Ko. If it is biased toward the lean side, on the other hand, the lean side determination time can be made longer to increase the value of the correction coefficient Ko. As a result, the air-fuel ratio is corrected to the lean side or rich side depending on the degree of deterioration of the O2 sensor, and the target air-fuel ratio is achieved.

(Effects of the Invention) As detailed above, according to the present invention, the exhaust gas concentration detection value detected by the exhaust concentration detector disposed in the exhaust system of the internal combustion engine is compared with a reference value, and the Proportional control for increasing or decreasing the air-fuel ratio of the air-fuel mixture by a first correction value when the detected exhaust gas concentration value changes from a rich side to a lean side or from a lean side to a rich side with respect to the reference value; When the detected exhaust gas concentration value is on the lean side or the rich side with respect to the reference value, the air-fuel ratio is feedback-controlled to the target air-fuel ratio by at least one of integral control that increases or decreases the air-fuel ratio at predetermined time intervals using second correction values. In an air-fuel ratio feedback control method for an internal combustion engine, 1. A first time period from when the detected exhaust gas concentration value changes from the lean side to the rich side with respect to the reference value to when the detected exhaust gas concentration value changes from the rich side to the lean side, and when the detected exhaust gas concentration value changes from the rich side with respect to the reference value. The ratio with the second time from the time of change to the lean side to the time of change from the lean side to the rich side is determined, and the reference value is changed according to the ratio thus determined, so that the exhaust gas concentration is detected. Even if the characteristics of the air-fuel mixture change over time, the air-fuel ratio of the air-fuel mixture can be corrected to achieve the target air-fuel ratio.
As a result, the engine's driving performance, fuel efficiency, and exhaust gas characteristics can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the overall configuration of an air-fuel ratio control device that implements the method of the present invention, FIG. 2 is a block diagram showing the internal configuration of the electronic control unit shown in FIG. 1, and FIG. 0□A timing chart showing the change over time in the output voltage value Voa of the sensor 13. FIG. 4 is a graph showing the relationship between the value KOX representing the degree of deterioration of the 08 sensor 13 and the correction value Δv. FIGS. 5 and 6 is the reference value vIF according to the present invention
A program flowchart of the setting subroutine, FIG. 7 is a graph showing the engine operating range in which the program flowchart shown in FIGS. 5 and 6 is executed, and FIG. 8 is a graph showing the relationship between the correction coefficient Kst〒 and the engine speed No The graph, Figure 9, shows the correction coefficient Kpe〒 and the intake pipe absolute pressure PB.
Figure 10 is a graph showing the relationship between ^ and 02 feedback correction coefficient Ko. It is a program flowchart of a calculation subroutine. DESCRIPTION OF SYMBOLS 1... Internal combustion engine, 5... Electronic control unit (ECU), 8... Intake pipe absolute pressure (Pa^) sensor, 10... Engine speed (Ne) sensor, 13
...O2 sensor.

Claims (1)

[Claims] 1. Compare the detected exhaust concentration value detected by the exhaust concentration detector arranged in the exhaust system of the internal combustion engine with a reference value, and determine the air-fuel ratio of the air-fuel mixture supplied to the engine when the detected exhaust concentration value is the reference value. Proportional control that increases or decreases the air-fuel ratio by a first correction value when the ratio changes from a rich side to a lean side or from a lean side to a rich side, and the detected exhaust gas concentration value changes from a lean side to a rich side with respect to the reference value. At some point, in an air-fuel ratio feedback control method for an internal combustion engine, in which the air-fuel ratio is feedback-controlled to a target air-fuel ratio by at least one of integral control in which the air-fuel ratio is increased or decreased at predetermined time intervals by respective second correction values, the detected exhaust concentration value is a first time period from the time point when the value changes from the lean side to the rich side with respect to the reference value to the time point when the value changes from the rich side to the lean side, and the time when the detected exhaust gas concentration value changes from the rich side to the lean side with respect to the reference value. The air-fuel ratio feedback for an internal combustion engine, characterized in that the ratio between the time point and the second time point when the lean side changes to the rich side is determined, and the reference value is changed in accordance with the thus determined ratio. Control method.