JPH0465223B2 - - Google Patents

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to an air-fuel ratio learning control method for an internal combustion engine, and is particularly suitable for use in an automobile engine in which exhaust gas purification measures are taken using a three-way catalyst.
When the air-fuel ratio feedback condition is satisfied, the air-fuel ratio of the air-fuel mixture is feedback-controlled according to the deviation between the air-fuel ratio of the exhaust gas detected using the oxygen concentration sensor and the target air-fuel ratio, and the air-fuel mixture flows into the three-way catalyst. Controls the air-fuel ratio of exhaust gas and learns and stores an air-fuel ratio correction value, while prohibiting the air-fuel ratio feedback control and learning of the air-fuel ratio correction value when air-fuel ratio feedback conditions are not satisfied, including during high load;
The present invention relates to an improvement in an air-fuel ratio learning control method for an internal combustion engine in which the air-fuel ratio of an air-fuel mixture is feed-forward controlled using an air-fuel ratio set using a stored air-fuel ratio correction value.

[Conventional technology]

Internal combustion engines, especially automobile engines that use three-way catalysts to purify exhaust gas,
It is necessary to maintain the air-fuel ratio of the exhaust gas strictly near the stoichiometric air-fuel ratio, so an oxygen concentration sensor that detects the air-fuel ratio of the exhaust gas and an air-fuel ratio control means that controls the air-fuel ratio of the air-fuel mixture are used. When the air-fuel ratio feedback condition is met, the three-way catalyst BACKGROUND ART An air-fuel ratio control method for an internal combustion engine that controls the air-fuel ratio of inflowing exhaust gas has been put into practical use. In this type of air-fuel ratio control method, the air-fuel ratio is generally made richer than the stoichiometric air-fuel ratio when the engine is under high load in order to increase output and prevent a rise in exhaust gas temperature. When the air-fuel ratio is high, the air-fuel ratio feedback control according to the oxygen concentration sensor output as described above cannot be performed, so the air-fuel ratio feedback control is stopped and the amount determined in advance based on the engine speed, load, etc. The air-fuel ratio is controlled in a feed-forward manner by supplying fuel to the
No. 8427). At this time, when the air density is low, for example when climbing a high mountain, one way to reduce the amount of fuel and maintain the air-fuel ratio at a constant value is to use the air-fuel ratio feedback control using an oxygen concentration sensor. There is also an air-fuel ratio learning control method in which a fuel ratio correction value is learned and stored, and the amount of fuel set based on engine rotation and load is corrected using this air-fuel ratio correction value (Japanese Patent Laid-Open No. 57-44752). According to such an air-fuel ratio learning control method, the air-fuel ratio is learned and corrected according to environmental conditions or individual differences in various sensors for detecting engine operating conditions and changes over time, so good air-fuel ratio control can be achieved. It can be performed. However, the characteristics of the oxygen concentration sensor change when the temperature of the oxygen concentration sensor increases due to a rise in exhaust gas temperature, such as after the engine is operated at high speed and high load, and the output relative to the air-fuel ratio becomes unstable.
If air-fuel ratio feedback control is performed under these conditions, accurate air-fuel ratio feedback control will not be possible, and if an incorrect air-fuel ratio correction value is learned and stored under these conditions, the amount of fuel supplied during subsequent high loads will be reduced. was incorrectly corrected, causing the air-fuel ratio to go out of order. In order to solve these problems, it has been proposed to prohibit both air-fuel ratio feedback control and air-fuel ratio learning when the engine speed or engine load is high (Japanese Patent Laid-Open No. 44752/1989).

[Problem to be solved by the invention]

However, after a long period of high load,
Because the temperature of the oxygen concentration sensor is high,
Thereafter, even if the engine enters a low-load operating state, the temperature of the oxygen concentration sensor does not immediately drop, and remains higher than the appropriate temperature for a predetermined period. In other words, the output value of the oxygen concentration sensor continues to be in a state where it is not an accurate value for a predetermined period of time. Therefore, if the air-fuel ratio learning is restarted immediately after the high-load operation is stopped, as in the past, the problem arises that the output value of the oxygen concentration sensor is not normal, resulting in erroneous learning. I will. Therefore, it is conceivable to prohibit both the air-fuel ratio feedback control and the air-fuel ratio learning control for a predetermined period of time when the high-load state ceases to exist after the high-load state has continued for a predetermined period. In this case, the range of implementation of air-fuel ratio feedback control becomes narrower, resulting in a problem that the exhaust purification performance deteriorates. The present invention has been made to solve the above-mentioned conventional problems, and is an internal combustion engine capable of expanding the range of implementation of air-fuel ratio feedback control and improving exhaust purification performance without impairing the learning accuracy of the air-fuel ratio. The purpose of this invention is to provide an air-fuel ratio learning control method.

[Means to solve the problem]

The present invention performs three-way feedback control by controlling the air-fuel ratio of the air-fuel mixture according to the deviation between the air-fuel ratio of exhaust gas detected using an oxygen concentration sensor and the target air-fuel ratio when the air-fuel ratio feedback condition is satisfied. It controls the air-fuel ratio of the exhaust gas flowing into the catalyst, and learns and stores the air-fuel ratio correction value. On the other hand, when the air-fuel ratio feedback condition is not satisfied, including during high load, the air-fuel ratio feedback control and learning of the air-fuel ratio correction value are performed. In an air-fuel ratio learning control method for an internal combustion engine in which the air-fuel ratio of the air-fuel mixture is feed-forward controlled by the air-fuel ratio set using the air-fuel ratio correction value that has been prohibited and stored, the high-load state is After a period of continuation, when the high load condition is eliminated and the air-fuel ratio feedback condition is met, the air-fuel ratio feedback control is restarted first. Until the output becomes normal for a specified period of time,
The above object is achieved by continuing to prohibit learning of the air-fuel ratio correction value.

[Effect]

If the air-fuel ratio control is open-loop, for example, if the air-fuel ratio in the three-way catalyst continues to be rich, the O ₂ stored in the three-way catalyst will react with HC, etc., and then disappear. cannot be purified. On the other hand, when the lean state continues, _NOx cannot be purified at all. On the other hand, even if the air-fuel ratio deviates slightly from rich to lean overall, if the air-fuel ratio is increased or decreased by performing air-fuel ratio feedback control, the situation where O ₂ in the three-way catalyst is completely exhausted will disappear. Since the lean state will no longer continue, HC, _NOx , etc. can be purified. Therefore, the exhaust purification performance is significantly improved by increasing and decreasing the air-fuel ratio using air-fuel ratio feedback control, rather than using open-loop air-fuel ratio control. Therefore, in the present invention, even after the high load state has continued for a predetermined period, when the high load state is resolved and the air-fuel ratio feedback condition is satisfied, the air-fuel ratio feedback control is restarted first. The scope of implementation of fuel ratio feedback control has been expanded to improve exhaust purification performance. On the other hand, even after the start of air-fuel ratio feedback control, the above-mentioned non-high-load state continues for a predetermined period until the output of the oxygen concentration sensor becomes normal, and the air-fuel ratio correction value remains unchanged until the temperature of the oxygen concentration sensor falls sufficiently. By continuing to prohibit learning, learning accuracy is impaired and subsequent control is prevented from being adversely affected.

【Example】

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an electronically controlled fuel injection system that senses an intake air amount for an automobile engine will be described in detail below with reference to the drawings, in which an air-fuel ratio learning control method for an internal combustion engine according to the present invention is adopted. As shown in FIG. 1, this embodiment includes an air flow meter 12 for detecting the flow rate of intake air taken in by an air cleaner (not shown), an accelerator disposed in a throttle body 14, and an accelerator disposed in the driver's seat. A throttle valve 16 for controlling the flow rate of intake air that opens and closes in conjunction with a pedal (not shown), a surge tank 18 for preventing intake interference between cylinders, and an intake manifold 20. an injector 2 for injecting fuel toward the intake port of the engine 10, which is provided
2, an oxygen concentration sensor 26 disposed in the exhaust manifold 24 for detecting the air-fuel ratio from the residual oxygen concentration in the exhaust gas, and a distributor that rotates in conjunction with the rotation of the crankshaft of the engine 10. A taste distributor 28 having a shaft, and a cylinder discrimination sensor 3 built into the distributor 28 that outputs a cylinder discrimination signal and a rotation angle signal, respectively, in accordance with the rotation of the distributor shaft.
0 and rotation angle sensor 32, a cooling water temperature sensor 34 disposed on the engine block for detecting the engine cooling water temperature, and the air flow meter 1.
The basic injection amount per engine stroke is calculated according to the engine rotational speed determined from the intake air amount of the two outputs and the rotational angle signal of the rotational angle sensor 32 output, and this is calculated as the engine cooling amount of the cooling water temperature sensor 34 output. The air-fuel ratio is corrected according to the water temperature, etc., and when the air-fuel ratio feedback condition is satisfied, the air-fuel ratio of the mixture is adjusted according to the deviation between the air-fuel ratio of the exhaust gas detected using the oxygen concentration sensor 26 and the target air-fuel ratio. By performing feedback control, the air-fuel ratio of the exhaust gas flowing into the three-way catalyst is controlled, and the air-fuel ratio correction value is learned and stored. On the other hand, when the air-fuel ratio feedback condition is not satisfied, including when the load is higher than a predetermined load, the above-mentioned Prohibits air-fuel ratio feedback control and learning of the air-fuel ratio correction value, and determines the fuel injection amount to perform air-fuel ratio feed-off control of the air-fuel mixture according to the set air-fuel ratio using the stored air-fuel ratio correction value. An automobile engine 10 is provided with a digital control circuit 38 which outputs a valve opening time signal to the injector 22, determines the ignition timing according to the engine operating state, and outputs an ignition signal to the igniter-equipped coil 36. In the intake air amount sensing type electronically controlled fuel injection device, the digital control circuit 38
After the high load state continues for a predetermined period of time, when the high load state is resolved and the air-fuel ratio feedback condition is satisfied, the air-fuel ratio feedback control is restarted first, while the high load is no longer maintained. The prohibition of learning the air-fuel ratio correction value continues until the condition continues for a predetermined period of time. As shown in detail in FIG. 2, the digital control circuit 38 is a central processing unit (hereinafter referred to as MPU) consisting of a microprocessor that performs various arithmetic operations.
40, a multiplexer 46 for sequentially taking in the output of the air flow meter 12 inputted via a buffer 42, the output of the cooling water temperature sensor 34 inputted via a buffer 44, etc.; Analog-to-digital converter 48 for converting to a signal
and the output of the analog-to-digital converter 48
A first input/output port 50 for inputting into the MPU 40, the output of the oxygen concentration sensor 26 inputted via a buffer 52 and a comparator 54, and the cylinder discrimination sensor 30 and rotation angle sensor inputted via a shaping circuit 56. 32 output etc. as mentioned above.
A second input/output port 58 for inputting data into the MPU 40, a read-only memory (hereinafter referred to as ROM) 60 for storing programs or various constants, and a random memory for temporarily storing calculation data, etc. in the MPU 40. Access memory (hereinafter referred to as RAM) 62 and clock 64
and a first output port 68 for outputting the calculation results of the MPU 40 to the injector 22 via the drive circuit 66 at a predetermined timing.
Similarly, the calculation result in the MPU 40 is
and a second output port 72 for outputting to the igniter-equipped coil 36 at a predetermined timing via a drive circuit 70. The effects of the embodiment will be explained below. The flow of the main routine in this embodiment is shown in FIG. In the main routine of this embodiment, first, in step 101, initial settings of the input/output ports are performed.
Next, in step 102, RAM 62 is cleared and initial data is set. Next, the process proceeds to step 103, where the air-fuel ratio correction value is determined.
Determine whether or not to memorize M _Kt . Further, the process proceeds to step 104, where an intake air amount Q is calculated from the output of the air flow meter 12, an engine rotation speed N is calculated from the rotation angle signal output from the rotation angle sensor 32, and one revolution of the engine indicating the engine load is calculated. Calculate the intake air amount Q/N per unit. Next, the process proceeds to step 105 to calculate the ignition timing. Furthermore, the process proceeds to step 106, where it is determined whether the air-fuel ratio feedback condition is satisfied. If the determination result is positive, the process advances to step 107 and air-fuel ratio feedback processing is performed. On the other hand, if the determination result in step 106 is negative, the process proceeds to step 108 and replaces the air-fuel ratio feedback correction coefficient _Kt with the learned and stored air-fuel ratio correction value _MKt . After step 107 or 108, step 109
Then, the fuel injection time T is calculated using, for example, the following equation. T=(Q/ _{N )} × _{K B} The constant K _P is a fuel increase coefficient (0 during air-fuel ratio feedback control) determined for each engine condition. After completing step 109, return to step 103 and repeat steps 103 to 109. Specifically, the determination as to whether or not to store the air-fuel ratio correction value M _Kt in step 1-3 of the main routine is processed according to the routine shown in FIG. That is, first, in step 201, it is determined whether the flag _F500 , which is set every 500 milliseconds in a 4 millisecond interrupt routine as shown in FIG. 5 below, is set. If the determination result is positive, the process proceeds to step 202, where it is determined whether the intake air amount Q/N per revolution of the engine is a predetermined value, for example, 1.0/rev or more, which is a high load condition. When the determination result is negative, that is, when it is determined that the high load condition is not present, the process proceeds to step 203, and it is determined whether the engine speed N is 3000 rpm or more. When the judgment result in step 202 or 203 is positive, that is, when the high load and/or high rotation state is present, the process proceeds to step 204, and the duration of the high rotation and/or high load state is counted. Count up the counter _C1 by 1. Next, the process proceeds to step 205, where it is determined whether the count value of the counter _C1 is 2400 or more. If the judgment result is positive, that is, high rotation or /
If the high load state continues for 20 minutes or more, the process proceeds to step 206 and sets an air-fuel ratio learning stop flag F _MKT to prohibit learning of the air-fuel ratio correction value. After step 206 or the previous step
If the judgment result in step 205 is negative, step 207
, and a counter is set to obtain the delay time (for example, 10 minutes) until the air-fuel ratio learning resumes, which is determined according to the time it takes for the oxygen concentration sensor 26 to return to normal.
Put 1200 into _C2 . On the other hand, if the judgment result in step 203 is negative, that is, if there is neither a high load state nor a high rotation state, the process proceeds to step 208 and counts the time until the air-fuel ratio learning cancel flag F _MKT is set down. Count down the counter _C2 by 1. Next, the process proceeds to step 209, where it is determined whether the count value of the counter _C2 has become 0 or less. When the determination result is positive, that is, when the delay time until air-fuel ratio learning restarts has elapsed, step 210 is performed.
The process proceeds to Step 2, and sets 0 to counters C ₁ and C ₂ to clear counter C ₁ and prevent counter C ₂ from increasing to a negative value in preparation for the next high load or high rotation state. Next, the process proceeds to step 211, where the air-fuel ratio learning cancellation flag F _MKT is set. After the above-mentioned step 207 or 211 is completed, or when the judgment result in the above-mentioned step 209 is negative, the process proceeds to step 212, and the flag _F500 is lowered. After step 212 or step 201 above
If the result of the determination is negative, exit this routine and proceed to the next routine. The flag F ₅₀₀ is set every 500 milliseconds by an interrupt routine every 4 milliseconds as shown in FIG. That is, in step 301, the counter _C500 that counts the elapsed time is incremented by one. Next, the process proceeds to step 302, where it is determined whether the count value of counter _C500 is less than 125. When the judgment result is negative, that is, when 500 milliseconds have elapsed, the process proceeds to step 303, and the counter C ₅₀₀
Clear the flag F ₅₀₀ at step 304.
stand up. After step 304 or the previous step
When the determination result in step 302 is positive, this routine exits. Main routine steps shown in Figure 3 above
Specifically, the determination in step 106 as to whether or not the air-fuel ratio feedback condition is satisfied is processed according to a routine as shown in FIG. That is, first, in step 401, it is determined based on the output of the cooling water temperature sensor 34 whether or not the engine cooling water temperature is a predetermined value, for example, 70° C. or higher. If the determination result is positive, the process proceeds to step 402, where the engine rotation speed N is set to a predetermined value, e.g.
Determine whether the speed is 5500 rpm or less. If the judgment result is positive, proceed to step 403 and calculate the intake air amount Q/N per engine revolution.
is less than or equal to a predetermined value, for example 0.7/rev. When the determination result is positive, that is, when the engine has finished warming up, is in the low-medium rotation range, and is in the low-medium load range, the process proceeds to step 405, where air-fuel ratio feedback processing is executed. On the other hand, when the judgment result in any of the above steps 401 to 403 is negative, that is, the engine cooling water temperature is 70.
When the air-fuel ratio needs to be richer than the stoichiometric air-fuel ratio because fuel is difficult to mix with air at temperatures below ℃, or when the engine speed N is in a high rotation range exceeding 5500 rpm,
Intake air amount Q/N per engine revolution is 0.7
When the load is in a high load range exceeding /rev, the step
Proceeding to 406, the air-fuel ratio correction value M _Kt that has been learned and stored in advance is entered into the air-fuel ratio feedback correction coefficient K _t used in the calculation of equation (1) above. After completing step 405 or 406, this routine exits. Specifically, the air-fuel ratio feedback processing in step 405 of the routine shown in FIG. 6 is performed in the air-fuel ratio feedback processing routine in the main routine as shown in FIG. This is handled by an interrupt routine every 4 milliseconds. That is, in the air-fuel ratio feedback processing routine as shown in FIG.
In 501, it is determined whether the lean flag FL is 1 or not. When the determination result is positive, the process proceeds to step 502, where it is determined whether the rich flag FR is 0 or not. If the determination result is negative, the process proceeds to step 503, where it is determined whether the delay counter Cd is 2 or more. If the determination result is positive, proceed to step 504, reset the rich flag FR, and then
In step 505, the air-fuel ratio feedback correction coefficient is
Increase K _t by a relatively large value B. On the other hand, if the determination result in step 501 is negative, the process proceeds to step 506, where it is determined whether the rich flag FR is 1 or not. If the determination result is negative, the process proceeds to step 507, where it is determined whether the count value of the delay counter Cd is 20 or more. When the determination result of step 507 is negative, or when the determination result of step 502 is positive,
Proceeding to step 508, the air-fuel ratio feedback correction coefficient _Kt is increased by a relatively small value a. Also, whether the judgment result in step 506 above is positive or not.
Alternatively, if the determination result in step 503 is negative, the process proceeds to step 509, where the air-fuel ratio feedback correction coefficient _Kt is decreased by a. If the judgment result in step 507 is positive, the process proceeds to step 510, where the rich flag FR is set, and in step 511, the air-fuel ratio feedback correction coefficient _Kt is decreased by a relatively large value A. Next, the process proceeds to step 512, where it is determined whether the air-fuel ratio learning stop flag F _MKT is 0 or not. When the determination result is positive, that is, when the air-fuel ratio learning stop flag is off, the process advances to step 513.
Using the value that was significantly reduced by A in step 511 and restored by A/2 (to take the median value of the oscillations of the air-fuel ratio correction value), as shown in the following equation,
Correct the air-fuel ratio correction value M _Kt with a weight of 1/100,
Learn and remember. M _Kt ← [99×M _Kt + {K _t + (A/2)}]/100...(2) On the other hand, when the judgment result in step 512 is negative and the air-fuel ratio learning cancellation flag is set, The routine shown in FIG. 7 is exited without correcting the air-fuel ratio correction value M _Kt . In the interrupt routine every 4 milliseconds shown in FIG. 8, first, in step 601, it is determined whether the output voltage of the oxygen concentration sensor 26 is high or not, in accordance with the output of the oxygen concentration sensor 26. It is determined whether the fuel ratio is rich. If the determination result is positive, the process advances to step 602 and the lean flag FL is reset. Next, the process proceeds to step 603, where it is determined whether the rich flag FR is 0 or not. On the other hand, if the determination result in step 601 is negative, the process proceeds to step 604 and sets the lean flag FL. Next, the process proceeds to step 605, where it is determined whether the rich flag FR is 0 or not. If the determination result in step 605 is negative, or if the determination result in step 603 is positive, the process proceeds to step 606, where the delay counter Cd is incremented by one. On the other hand, if the judgment result in step 605 is positive or the judgment result in step 603 is negative, the process advances to step 607 and the delay counter Cd is cleared. After completing step 606 or 607, the routine exits and returns to the main routine via, for example, the routine shown in FIG. 5 above. Specifically, for example, if the output of the oxygen concentration sensor 26 is a high voltage and the air-fuel ratio is rich, the process proceeds to step 602 and the lean flag FL is reset. At this time, if the rich flag FR is 0, it means that the air-fuel ratio has just changed from lean to rich.
Delay counter Cd has become 0, and step 606
The delay counter Cd is counted up. When the air-fuel ratio feedback processing routine in the main routine shown in FIG. 7 is entered in this state, the lean flag FL is 0, so the flow goes to step 506, and the rich flag FR is also 0, so the delay counter Cd is counted. The numerical value is determined, and if the counted value is less than 20, the air-fuel ratio feedback correction coefficient _Kt is increased by a small value a in step 508 to further enrich the air-fuel ratio. On the other hand, when the count value of the delay counter Cd reaches 20, the process proceeds to step 510, where the rich flag FR is set, and at the same time, in step 511, the air-fuel ratio feedback correction coefficient _Kt is greatly decreased by a relatively large value A, and the After forcing the fuel ratio to lean,
If the air-fuel ratio learning cancellation flag is not set, in step 513, the air-fuel ratio correction value M _Kt is corrected with a weight of 1/100 by restoring the significantly reduced A by 1/2. On the other hand, when the output of the oxygen concentration sensor 26 changes from rich to lean, the above operation is performed in reverse. Here, the delay counter Cd is used because
This is to prevent malfunctions due to noise on the oxygen concentration sensor 26, and when changing the control from rich to lean or from lean to rich, the normal value a
The reason why the air-fuel ratio feedback correction coefficient _Kt is made to change greatly with larger values A and B is that the output of the oxygen concentration sensor 26 is controlled at the air-fuel ratio near the point where it switches between rich and lean by shortening the time. ,
Oxygen concentration sensor 2 detects small variations in fuel supply amount
This is to prevent unstable control due to the detection of 6. FIG. 9 shows an example of the state of change of the air-fuel ratio feedback correction coefficient _Kt in this embodiment. In the above embodiment, the present invention is applied to an automobile engine equipped with an intake air amount sensing type electronically controlled fuel injection device, but the scope of application of the present invention is not limited thereto. It is clear that the present invention is equally applicable to automobile engines with controlled fuel injection systems or internal combustion engines with general electronically controlled carburetors.

【Effect of the invention】

As described above, according to the present invention, the range of implementation of air-fuel ratio feedback control can be expanded and exhaust purification performance can be improved without impairing the learning accuracy of the air-fuel ratio correction value. In addition, since learning of the air-fuel ratio correction value is restarted after the temperature of the oxygen concentration sensor has sufficiently decreased, it is possible to reliably prevent learning accuracy from being impaired and adversely affecting subsequent control. Therefore, the air-fuel ratio feedback control can be performed reliably and accurately, and the exhaust gas purification performance can be improved, which is an excellent effect.

[Brief explanation of the drawing]

FIG. 1 is a partial block diagram showing the configuration of an embodiment of an intake air amount sensing type electronically controlled fuel injection device for an automobile engine in which the air-fuel ratio learning control method for an internal combustion engine according to the present invention is adopted. Cross-sectional view including, second
The figure is a block diagram showing the configuration of the digital control circuit used in the embodiment, FIG. 3 is a flowchart showing the overall structure of the main routine in the embodiment, and FIG. 4 is a block diagram showing the overall structure of the main routine in the embodiment. FIG. 5 is a flowchart showing in detail the part for determining whether or not to store the air-fuel ratio correction value; FIG. 5 is a flowchart showing a part of the interrupt routine every 4 milliseconds in the embodiment; FIG. FIG. 7 is a flowchart showing in detail the part of the main routine in the embodiment that determines whether the air-fuel ratio feedback condition is satisfied; FIG. 7 is a flowchart showing the air-fuel ratio feedback processing part of the main routine in detail; FIG. 8 is a flowchart showing a part of the interrupt routine every 4 milliseconds in the embodiment, and FIG. 9 is a diagram showing an example of the state of change of the air-fuel ratio feedback correction coefficient in the embodiment. DESCRIPTION OF SYMBOLS 10...Engine, 12...Air flow meter, 22...Injector, 26...Oxygen concentration sensor, 32...Rotation angle sensor, 34...Cooling water temperature sensor, 38...Digital control circuit.

Claims (1)

[Claims] 1. When the air-fuel ratio feedback condition is satisfied, the air-fuel ratio of the air-fuel mixture is feedback-controlled in accordance with the deviation between the air-fuel ratio of exhaust gas detected using an oxygen concentration sensor and the target air-fuel ratio. controls the air-fuel ratio of the exhaust gas flowing into the three-way catalyst, and learns and stores the air-fuel ratio correction value. On the other hand, when the air-fuel ratio feedback condition is not satisfied, including during high load, the air-fuel ratio feedback control and the air-fuel ratio correction value are performed. In an air-fuel ratio learning control method for an internal combustion engine, the air-fuel ratio of the mixture is feed-forward controlled using the air-fuel ratio set using the stored air-fuel ratio correction value. After the condition continues for a predetermined period, when the high load condition is eliminated and the air-fuel ratio feedback condition is satisfied, the air-fuel ratio feedback control is restarted first. An air-fuel ratio learning control method for an internal combustion engine, characterized in that the prohibition of learning the air-fuel ratio correction value continues until the output of the concentration sensor becomes normal for a predetermined period of time.