JP2501566B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

Description

TECHNICAL FIELD The present invention relates to a device for controlling an air-fuel ratio of an automobile engine.

(Prior Art) In recent years, demands for engines for automobiles have become more sophisticated, and there is a tendency to achieve high levels of contradictory issues such as emission reduction, high output, and low fuel consumption.

In order to address these issues, combustion control under ultra-lean air-fuel ratio has been attempted. For example, as such, "Internal Combustion Engine, Vol. 23, No. 12," October 1984, pages 33-40, Sankaido There is a lean burner described in. In this device,
Based on the output of the lean sensor capable of detecting the air-fuel ratio in a wide range from rich to lean, feedback control of the air-fuel ratio is performed up to the ultra-lean air-fuel ratio region to achieve the above requirement.

In this case, unlike the characteristic that the stoichiometric air-fuel ratio is constant in steady running, a lean air-fuel ratio is used as the target value even in some acceleration regions. For example, in the normal acceleration range, the air-fuel ratio is 2
2.5, the air-fuel ratio is 21.5 in the steady running range and 15.5 when idling. Also, the output air-fuel ratio is
Try to secure vehicle power performance using 12-13. NOx tends to decrease extremely as the air-fuel ratio shifts to such a lean range, which is in line with the recent reduction in NOx emissions. On the other hand, on the other hand, the air-fuel ratio applicable range that can satisfy both the NOx emission level for satisfying the exhaust gas regulation and the allowable torque fluctuation level is narrow, and precise air-fuel ratio control is required.

(Problems to be Solved by the Invention) However, in such a conventional air-fuel ratio control device, although the above-mentioned corresponding effects can be obtained by the lean combustion, the response delay of the air-fuel ratio sensor must be taken into consideration. Since the configuration is not provided, an error may occur in the detection of the air-fuel ratio and the accuracy of the air-fuel ratio control may deteriorate. Especially, when the air-fuel ratio varies depending on the operating conditions, such a detection error becomes remarkable. As a result, the control accuracy is lowered, and the engine is operated on the richer or leaner side than the target air-fuel ratio, resulting in deterioration of exhaust emission characteristics and deterioration of drivability such as torque fluctuation.

(Object of the invention) Therefore, the present invention corrects the current output of the air-fuel ratio sensor by a predetermined delay correction calculation to calculate the air-fuel ratio of the air-fuel mixture in the present combustion stroke and shifts the value of the target air-fuel ratio. By adjusting the combustion phases of the two so that the combustion timing will be the same for the air-fuel ratio of the air-fuel mixture in this combustion stroke, the accuracy of feedback control of the air-fuel ratio will be improved to further improve exhaust emission and operability. In addition to improving the time constant of the output of the air-fuel ratio sensor by calculating the time constant of the output of the air-fuel ratio sensor based on the engine speed and the basic injection pulse width, It is intended to further improve the accuracy of the feedback correction coefficient of the air-fuel ratio by setting the value to an appropriate value depending on each case such as high load.

(Means for Solving Problems) In order to achieve the above object, an air-fuel ratio control apparatus for an internal combustion engine according to the present invention is an air-fuel ratio detecting apparatus for detecting an air-fuel ratio of an intake air-fuel mixture, as shown in the basic conceptual diagram of FIG. The current outputs of the fuel ratio detecting means a, the operating state detecting means b for detecting the operating state of the engine, and the air-fuel ratio detecting means a are corrected by a predetermined delay correction calculation to calculate the air-fuel ratio of the air-fuel mixture in the combustion stroke this time. In the air-fuel ratio control device for an internal combustion engine, which comprises a correcting means c for controlling the engine and a target setting means d for setting a target air-fuel ratio based on the operating state of the engine, the value of the target air-fuel ratio set by the target setting means d. Is shifted to match the combustion phases of the air-fuel mixture in the current combustion stroke calculated by the correction means c with the same timing, and the engine speed and A time constant calculation means f for calculating the time constant of the output of the air-fuel ratio detection means based on the main injection pulse width, a time constant calculated by the time constant calculation means f, and a combustion phase when the combustion phase is matched by the phase matching means e. A control unit g that calculates a control value for feedback-correcting the air-fuel ratio based on the target air-fuel ratio and the calculated air-fuel ratio of the correction unit c, and intake air or combustion supply amount is changed based on the output of the control unit g. And an operating means h for operating the fuel ratio.

(Operation) In the present invention, the current output of the air-fuel ratio sensor is corrected by a predetermined delay correction calculation, the air-fuel ratio of the air-fuel mixture in this combustion stroke is calculated, and the value of the target air-fuel ratio is shifted. The combustion phases of the two can be adjusted so as to have the same timing with respect to the air-fuel ratio of the air-fuel mixture in the combustion process. Then, the control value for feedback-correcting the air-fuel ratio is appropriately calculated based on the target air-fuel ratio when the combustion phases are matched and the air-fuel ratio corrected by the correction calculation. Therefore, the feedback control of the air-fuel ratio is executed with high accuracy, and the deterioration of exhaust emission characteristics and the deterioration of drivability such as torque fluctuation are prevented. Furthermore, since the time constant of the output of the air-fuel ratio sensor is calculated based on the engine speed and the basic injection pulse width, the time constant of the output of the air-fuel ratio sensor is low load, low load high rotation, high load, etc. The value becomes appropriate depending on the case, and the accuracy of the feedback correction coefficient of the air-fuel ratio is further improved.

(Example) Hereinafter, the present invention will be described with reference to the drawings.

2 to 6 are views showing the first embodiment of the present invention. First, the configuration will be described. In FIG. 2, reference numeral 1 is an engine, and intake air is supplied to each cylinder through an intake pipe 2.
Fuel is injected from the fuel tank 3 by the fuel pump 4, the fuel filter 5, the fuel damper 6, and the fuel supply pipe 7 by the injector (operating means) 8 based on the injection signal Si. Further, the fuel temperature Tf is detected by a fuel temperature sensor 9 provided in the fuel supply pipe 7, and the surplus fuel is returned to the fuel tank 3 via a fuel pressure regulator 10 and a fuel return pipe 11. Then, the air-fuel mixture in the cylinder is ignited and exploded by the discharge action of the ignition plug 12, and becomes exhaust gas and is exhausted through the exhaust pipe 13.

The flow rate Qa of the intake air amount is detected by the air flow meter 14 and controlled by the throttle valve 15 of the intake pipe 2. The opening Cv of the throttle valve 15 is detected by the throttle valve opening sensor 16, and the engine 1
The rotation speed N of is detected by the crank angle sensor 17. Further, the temperature Tw of the cooling water flowing through the water jacket is detected by the water temperature sensor 18. Further, the oxygen concentration in the exhaust gas is detected by an oxygen sensor (air-fuel ratio detecting means) 19, and the oxygen sensor 19 is of a type whose output Vi changes uniquely for a wide range of air-fuel ratios from rich to lean regions, etc. Is used.

The air flow meter 14, the throttle valve opening sensor 16 and the crank angle sensor 17 constitute an operating state detecting means 20, and outputs from the operating state detecting means 20, the fuel temperature sensor 9, the water temperature sensor 18 and the oxygen sensor 19 are It is input to the control unit 21. The control unit 21 has a function as a correction means, a target setting means, a phase adjustment means, a time constant calculation means and a control means, and as shown in detail in FIG. 3, a CPU 22, a ROM 23, a RAM 24, an A / D converter. And an I / O port, which are connected to each other by a common bus 27. The A / D converter 25 converts each signal Qa, Tf, Tw and vi input as an analog signal into a digital signal, and in accordance with an instruction from the CPU 22, the CPU 22 or RA is supplied at a predetermined time.
Output to M24. CPU22 takes in the external data which is necessary according to the program written in ROM23,
Further, while exchanging data with the RAM 24, the processing value necessary for air-fuel ratio control is arithmetically processed, and the processed data is output to the I / O port 26 as necessary. The signals from the sensor groups 16 and 17 are input to the I / O port 26, and the I / O
The injection signal Si is output from the port 26. ROM23 is CPU22
The RAM 24 stores the calculation program in, and the RAM 24 stores the data used for the calculation in the form of a map or the like. A part of the RAM 24 is composed of, for example, a non-volatile memory, and retains the stored contents (learned value and the like) even after the calculation is stopped.

Next, the operation will be described. First, the reason why a response delay occurs in the output of the air-fuel ratio sensor as pointed out in the conventional problems will be described.

Generally, since the air-fuel ratio sensor detects the concentration of the air-fuel ratio by diffusion of exhaust gas and gas exchange, it is conceivable that the air-fuel ratio sensor has a certain delay element in the detection. This lag element changes depending on the diffusion coefficient and the gas exchange rate in the vicinity of the air-fuel ratio sensor, and the factors of the change include (1) engine speed, (2) load, and (3) change width. Although the order of the delay element is considered to be multidimensional, it can be approximated as a substantially first-order delay system. Therefore, when the correction by the difference method is applied as the method for correcting the primary delay line, the corrected air-fuel ratio A / F is expressed by the following equation.

A / F = (A / F ₀ ) + α (A / F ₀ −A / F ₁ ), where A / F ₀ : current air-fuel ratio A / F ₁ : previous air-fuel ratio α: correction coefficient (this α corresponds to the approximate first-order lag time constant of the air-fuel ratio sensor.) To explain this in more detail, this first-order lag system is expressed by the following equation when expressed using a differential equation. Becomes

However, y _n : current time y _n-1 : previous time (that is, y _n before Δt time) T: time constant of air-fuel ratio Δt: sample interval (time difference between A / F ₀ and A / F ₁ ) Therefore, actually The output behavior of the air-fuel ratio sensor is shown as a first-order lag output of the time constant T as shown in FIG. That is, the response delay of the air-fuel ratio sensor output can be corrected by using the difference method.

FIG. 5 is a flow chart showing a program for air-fuel ratio control based on the above-mentioned basic principle, and P _{1 to} P _{7 in the} figure show respective steps of the flow. This program is executed once every predetermined time. First, the engine speed N is read at P ₁ , and the P ₂ intake air amount Q _a is read. The rotation speed N is calculated by measuring the interval time of the reference signal (signal for every 360 °) from the crank angle sensor 10 or measuring the number of pulses of the position signal (signal for every 1 °) at a predetermined time. Then
The basic pulse width (basic injection quantity) Tp is calculated according to the following equation by P _3.

However, K: constant then calculates various correction coefficients COEF at P _4, and calculates an invalid pulse width Ts at P _5. Here, COEF is a fuel delay correction coefficient, which corrects the fuel amount during a transition. The value is determined by the vaporization of the fuel and the wall flow rate, and is specifically calculated by the magnitude of acceleration / deceleration, the engine warm-up state and the operating state, whether or not after the start. Further, the various correction coefficients COEF and the invalid pulse width Ts are various correction coefficients for correcting the basic pulse width Tp, but since they have little relation to the present invention, detailed description thereof will be omitted.

Next, at P ₆ , the injection pulse width Ti is calculated according to the following equation.

Ti = Tp × COEF × ALPHA + Ts ...... However, ALPHA: Feedback correction coefficient of air-fuel ratio The calculation of the feedback correction coefficient ALPHA of the air-fuel ratio will be described in detail in a program described later in FIG.

Further, at P ₇ , the injection pulse width Ti is stored in the output register of the I / O port 26, and the injection signal Si having the fuel injection pulse width corresponding to this Ti is output to the injector 8 at the predetermined crank angle. The process ends.

FIG. 6 is a flow chart showing a program for calculating the feedback correction coefficient ALPHA of the air-fuel ratio, and this program is executed for each stroke of combustion of the engine. First, at P ₁₁ , the target air-fuel ratio TA under the operating conditions at that time is determined from the predetermined table map using N, Tp, and Tw as parameters.
Look up F ₀ {TAF ₀ = func (N, Tp, Tw)}. However, this cooling water temperature Tw is read by another routine not shown. Next, at P ₁₂ , the output of the target air-fuel ratio for two revolutions is shifted to combine the phase with the output of the air-fuel ratio sensor, and at P ₁₃ , the previous read value A / F ₀ of the air-fuel ratio sensor is changed to A / F ₁
Rewrite the memory as. Next, at P ₁₄ , the output of the air-fuel ratio sensor is A / D converted, and the A / D converted value is stored in the memory as A / F ₀ .

Moreover, to calculate the feedback correction coefficient ALPHA in the air-fuel ratio by applying a P ₁₅ to P foregoing difference method in ₁₈ below. That is, in P ₁₅ , the next constant T is calculated according to the following equation, and in P _{16 the} difference correction coefficient α (α = K / N where K: constant, N: engine speed) is calculated, and at P ₁₆ The output A / F of the air-fuel ratio sensor after delay correction is calculated according to the above-mentioned formula.

T = func (TP, N) …… Then, it calculates the feedback correction coefficient ALPHA in the air-fuel ratio by PI (proportional-integral) control to the following equation at P _17, the process ends.

ALPHA = Kp (TAF ₄ −A / F) ＋ ΣK ₁ (TAF ₄ −A / F) …… However, TAF ₄ : Output after phase adjustment calculated in P ₁₂ A / F: Value calculated in P ₁₇ Kp: Constant As described above, in this embodiment, the response delay of the output of the air-fuel ratio sensor is appropriately corrected by using the difference method, and the accurate air-fuel ratio feedback correction coefficient ALPHA is calculated. Therefore, it is possible to improve the drivability by preventing deterioration of exhaust emission characteristics and torque fluctuation.

Further, in the present embodiment, the time constant T of the air-fuel ratio is appropriately calculated as a function of the basic pulse width Tp and the engine speed N, so that as shown in FIG. The time constant T of the air-fuel ratio sensor is set to a low load (solid line in the figure),
It can be set to a more appropriate value depending on the case of low load high rotation (broken line in the figure) and high load (dashed line in the figure), and the accuracy of the feedback correction coefficient ALPHA of the air-fuel ratio can be further improved. .

Although the delay element of the output of the air-fuel ratio sensor is corrected by using the difference method in the present embodiment, the present invention is not limited to this. The point is that it is only necessary to appropriately correct the delay element, and, of course, a mode in which differentiation is used instead of the difference method is also possible. In this case, it is possible to perform more precise correction.

Further, in this embodiment, as a feedback control method,
Although PI (proportional / integral) control is adopted, the present invention is not limited to this, and other aspects such as P (proportional) control and PID
It may be performed by (proportional / integral / derivative) control. That is,
For P control to the following equation eighth expression of the aforementioned step P _17, in the case of PID control may be changed first equation to the following equation.

_{ALPHA = Kp (TAF 4 -A /} F) ...... ALPHA = Kp (TAF 4 -A / F) + ΣK 1 (TAF 4 -A / F) + K D Δ ...... However, delta: the current (TAF ₄ -A / F) -Previous (TAF _4- A / F) FIGS. 7 to 10 are diagrams showing a second embodiment of the present invention, and the hardware configuration in this embodiment is the same as that in the first embodiment. Therefore, the hardware configuration is omitted.

First, the basic idea of this embodiment will be described. Generally, the original characteristics of the air-fuel ratio sensor are output only on the lean side of the theoretical air-fuel ratio (λ = 1), but the structure of the air-fuel ratio sensor according to the present invention is such that the pump current (diffusion current) Ip is flown into zirconia and the surface O ₂ produces an ion is small state, when the sudden change output using the output sudden change characteristics in the vicinity zirconia original O ₂ Ionzero is pouring Ip so as to maintain the balance, to retrieve the output in the rich side than the stoichiometric air-fuel ratio (See FIG. 7 (a)). This output characteristic changes depending on the O ₂ concentration in the lean range and the CO concentration in the rich range rather than the theoretical air-fuel ratio, but the O ₂ concentration in the lean range changes the excess air ratio λ (λ = air-fuel ratio / theoretical air-fuel ratio). As shown in FIG. 9 (b), the output Ip is proportional to the fuel ratio) and is 2 at λ = 1 with respect to the excess air ratio λ at the fuel air ratio F / A (the reciprocal of the air fuel ratio A / F). It can be represented by a broken line in a book.

On the other hand, the constant K of the above-mentioned equation can be expressed by the following equation if the target fuel-air ratio is TFA and the coefficient of the flow rate characteristic of the injector is K _INJ .

K = K _INJ × TFA …… From the above, the output characteristics of the air-fuel ratio sensor are uniformly converted to the fuel-air ratio F / A from the rich range to the lean range, and the sensor output characteristics are linearized. If feedback control (hereinafter referred to as F / B control) is executed so that the air ratio is TFA, whether the current air-fuel ratio is rich or lean with respect to the target value, or the position of the target air-fuel ratio TFA is It is possible to make the feedback control equal to the rich side and the lean side regardless of the side of the logical air-fuel ratio.

However, in the conventional air-fuel ratio F / B control method, the rich side and the lean side of the target air-fuel ratio are controlled with the same feedback gain so that the output of the air-fuel ratio sensor itself becomes the target air-fuel ratio. Therefore, the result of the F / B control may shift from the target value to cause an offset, which may deteriorate the control accuracy and deteriorate the drivability and the exhaust emission characteristic. Further, in order to avoid such a decrease in control accuracy, it is conceivable to change the feedback gain between the rich side and the lean side of the target air-fuel ratio, but the feedback gain may be changed depending on the location (target air-fuel ratio). , The calculation processing for determining the gain becomes complicated. Therefore, there is a possibility that the cost and the performance may deteriorate due to the increase in the program capacity and the calculation time.

Therefore, in this embodiment, the output of the air-fuel ratio sensor is set to the fuel-air ratio F /
Paying attention to the fact that there are two broken line characteristics on the scale of A, a signal that is completely linear with respect to the fuel air ratio F / A is obtained as shown by the broken line in FIG. F / B based
By performing the control, the control gains are set to the same gain on both the rich side and the lean side of the target value to avoid the control offset and improve the control accuracy. When performing the delay correction before performing the linearization correction, the slope sensitivity differs between the rich side and the lean side relative to the theoretical air-fuel ratio, so the gain of the delay correction is increased between the rich side and the lean side. If it is not corrected by changing the value, an error will occur. Therefore, it is necessary to perform delay correction after linearization.

8 and 9 are flowcharts showing an air-fuel ratio control program based on the above-mentioned basic principle. In the description of the present embodiment, steps for performing the same processing as those in the first embodiment are designated by the same reference numerals and the description thereof will be omitted. . 8 and 9 of this embodiment are shown in FIG. 5 of the first embodiment,
Each corresponds to FIG.

After passing through P ₂ in the program of FIG. 10, the basic pulse width Tp ′ is calculated according to the following equation at P ₃₁ , and the predetermined table map with Tp ′, N, Tw as parameters at P ₃₂ Target air-fuel ratio TFA {TFA ＝ func (Tp ′, N, T
w)} is looked up.

However, K _INJ is a constant determined by the flow rate characteristics of the injector, and then the basic fuel pulse width T according to the target fuel-air ratio at P ₃₃
Calculate p according to the following formula.

FIG. 9 is a flow chart showing a program for calculating the feedback correction coefficient ALPHA of the air-fuel ratio, and this program is executed for each stroke of combustion of the engine. First, in P ₄₁ , the signals of TFA for four strokes corresponding to the intake stroke to the exhaust stroke are shifted to perform the phase combination with the air-fuel ratio sensor. That is, in the routine in which the calculation is performed for each stroke signal like this program, the output of the air-fuel ratio with respect to the target fuel-air ratio TFA obtained at the time of injection can be detected is that the fuel is sucked and compressed in the intake stroke, It is when the exhaust gas is discharged in the explosion and exhaust stroke and reaches the place where the carbon sensor (air-fuel ratio sensor) 19 is attached. Therefore, here, the signal of TFA is shifted by four strokes corresponding to the intake stroke to the exhaust stroke to perform the phase combination with the air-fuel ratio sensor signal. Then, the output of the air-fuel ratio sensor is A / D converted by the P _42,
The A / D converted value is stored in memory as F / A ₀ . In P ₄₃ , the value of KFBA ₀ is changed to KFBA ₁ to obtain the previous linearizer result.
To transfer, to compute a linearization correction factor KHOS by F / A ₀ to A / D conversion according to the following equation with P ₄₄ (see FIG. 10).

_{KHOS = func (F / A 0} ) ...... then linearize the KFBA ₀ according to the following equation at P _45.

KFBA ₀ = KHOS × F / A ₀ ...... Therefore, as shown in Fig. 7 (b) above, the output characteristics of the air-fuel ratio sensor (for the air-fuel ratio F / A (Polygonal line characteristic) can be displayed as one linear straight line from the rich region to the lean region, and the output of KFBA ₀ can be made linear with respect to F / A as shown by the chain line in the figure.

In the following P _{46 to} P _49, the delay correction of the air-fuel ratio sensor is corrected using the difference method as in the first embodiment, and the air-fuel ratio feedback correction coefficient ALPHF is calculated. That is, P ₄₆
The differential correction coefficient (corresponding to the sampling interval of the air-fuel ratio) α is calculated according to the formula below, and the output of the air-fuel ratio sensor after delay correction is linearized according to the formula below on P ₄₇
Calculate KFBA.

However, T: primary delay time constant of the air-fuel ratio sensor m: constant KBFA = KBFA ₀ + α (KBFA ₀ −KBFA ₁ ) …… Then, at P ₄₈ , the phase correction signal TFA ₄ of the target fuel-air ratio TFA and the air-fuel ratio sensor Fuel-air ratio KFBA obtained by linearizing the output of
The difference Δ with is calculated according to the following formula, and the air-fuel ratio feedback correction coefficient ALPHA is calculated according to the following formula on P ₄₉ .
Finishing the process Δ = TFA ₄ −KFBA …… ALPHA = Kp · Δ + Σ (K ₁ · Δ) In this way, in the present embodiment, the relationship between the output of the air-fuel ratio sensor and its fuel-air ratio is two. Paying attention to the fact that it becomes a broken line characteristic,
A signal that is completely linear with respect to the fuel-air ratio F / A is obtained as shown by the broken line in FIG. 7 (b), and based on this signal, a precise F / A is calculated.
Along with the B control, the response delay of the air-fuel ratio sensor is also appropriately corrected to provide an accurate air-fuel ratio feedback correction coefficient AL.
Since PHA is calculated, the effects of both can be combined to realize extremely accurate air-fuel ratio F / B control, and exhaust emission characteristics and drivability can be further improved.

In this embodiment, the linearization of the air-fuel ratio and the correction of the response delay of the air-fuel ratio sensor according to the present invention are performed at the same time. However, the present invention is not limited to this, and only the linearization of the air-fuel ratio is performed alone. Needless to say, it may be applied to the air-fuel ratio control. In this way, if the linearization of this embodiment is applied to the conventional air-fuel ratio control device, etc., no hardware changes or special sensors are required, and there is no change in the control gain, and only software support is required to achieve control accuracy. Can be improved.

(Effect) According to the present invention, the current output of the air-fuel ratio sensor is corrected by a predetermined delay correction calculation to calculate the air-fuel ratio of the air-fuel mixture in the current combustion stroke and shift the value of the target air-fuel ratio. Since the combustion phases of the two are matched so that the combustion timing is the same as the air-fuel ratio of the air-fuel mixture in this combustion stroke, the accuracy of feedback control of the air-fuel ratio can be improved and torque fluctuations can be prevented. Exhaust emissions and drivability can be further improved.

Also, since the time constant of the output of the air-fuel ratio sensor is calculated based on the engine speed and the basic injection pulse width, the time constant of the output of the air-fuel ratio sensor can be set to low load, low load high rotation, high load, etc. The value can be set to an appropriate value depending on each case, and the accuracy of the air-fuel ratio feedback correction coefficient can be further improved.

[Brief description of drawings]

FIG. 1 is a basic conceptual diagram of the present invention, FIGS. 2 to 6 are diagrams showing a first embodiment of the present invention, FIG. 2 is an overall configuration diagram thereof, and FIG. 3 is a circuit configuration diagram of its control unit. FIG. 4 is a diagram showing characteristics of the air-fuel ratio sensor, FIG. 5 is a flow chart showing a program for the air-fuel ratio control, FIG. 6 is a flow chart showing a program for calculating the air-fuel ratio feedback correction coefficient, and 7 to FIG. 10 is a diagram showing a second embodiment of the present invention, FIG. 7 is a diagram showing characteristics of the air-fuel ratio sensor, FIG. 8 is a flow chart showing a program for the air-fuel ratio control, and FIG. A flowchart showing a program for calculating an air-fuel ratio feedback correction coefficient,
FIG. 10 is a diagram showing the characteristics of the linearization correction coefficient. 1 ... Engine, 8 ... Injector (operating means), 19
...... Oxygen sensor (air-fuel ratio detecting means), 20 ...... Operating state detecting means, 21 ...... Control unit (air-fuel ratio correcting means, target setting means, phase adjusting means, time constant calculating means, control means).

Claims (1)

(57) [Claims] Claims: 1. A) air-fuel ratio detecting means for detecting the air-fuel ratio of the intake air-fuel mixture; b) operating state detecting means for detecting the operating state of the engine; and c) delaying the current output of the air-fuel ratio detecting means by a predetermined delay. An internal combustion engine provided with a correction unit that corrects by a correction calculation to calculate the air-fuel ratio of the air-fuel mixture in the current combustion stroke, and d) a target setting unit that sets a target air-fuel ratio based on the operating state of the engine. In the air-fuel ratio control device, e) shift the value of the target air-fuel ratio set by the target setting means so that the same timing is obtained with respect to the air-fuel ratio of the air-fuel mixture in the present combustion stroke calculated by the correction means. Phase matching means for matching the combustion phases of both, f) time constant calculating means for calculating the time constant of the output of the air-fuel ratio detecting means based on the engine speed and the basic injection pulse width, and g) time constant calculating means. Therefore, control means for calculating a control value for feedback-correcting the air-fuel ratio based on the calculated time constant, the target air-fuel ratio when the combustion phases are matched by the phase matching means, and the calculated air-fuel ratio of the correction means; and h) control means An air-fuel ratio control device for an internal combustion engine, comprising: an operating unit that operates an air-fuel ratio by changing a supply amount of intake air or fuel based on an output of the air-fuel ratio.