JPH11210527A - Air-fuel ratio controller for engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent deterioration in controllability and exhaust emission by determining a predetermined deteriorating condition of exhaust flow rate on the basis of an engine operating condition and suppressing air-fuel feedback control if deterioration of estimation accuracy of an observer is predicted from rapid increase of unnecessary time accompanying reduction of the exhaust flow rate. SOLUTION: From a detection value of an air-fuel ratio detecting means 22, an air-fuel ratio (x) for each cylinder in an engine 1 is estimated by means of an estimating means 30b, and an air-fuel ratio is feedback controlled in each cylinder by means of a regulating mean 16 so that each air-fuel ratio estimation value (x) becomes a target air-fuel ratio. In this process, a basic regulation quantity, which corresponds to the target air-fuel ratio, of the regulating means 16 is set on the basis of an engine operating condition in a basic regulation quantity setting unit 30a, and by means of a feedback correction unit 30c, the basic regulation quantity is corrected so that the air-fuel ratio estimation value (x) for each cylinder becomes the target air-fuel ratio. If a determining means 30d determines that an exhaust flow rate of the engine 1 is dropped below a predetermined value, correction of the basic regulation quantity is suppressed by means of suppressing means 30e.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention belongs to the technical field of an air-fuel ratio control device which performs feedback control of the air-fuel ratio of intake air supplied to a multi-cylinder engine for each cylinder.

[0002]

2. Description of the Related Art Conventionally, as an air-fuel ratio control device for an engine of this type, as disclosed in Japanese Patent Application Laid-Open Nos. 8-232719 and 8-291736, an air-fuel ratio is set in an exhaust pipe assembly. There is known a sensor provided with a sensor for detecting the air-fuel ratio of each cylinder based on a detection value of the sensor by modern control theory.

In such an air-fuel ratio control device, generally,
A theoretical model that prescribes the behavior of the air-fuel ratio in the exhaust system of the engine is built in advance, and an observer that observes the air-fuel ratio of each cylinder of the engine is set based on this theoretical model. The observer estimates the air-fuel ratio of each cylinder based on the collective air-fuel ratio obtained from the value detected by the O2 sensor. Then, feedback control is performed according to, for example, a PID control rule so that the estimated air-fuel ratio of each cylinder matches the target air-fuel ratio, thereby converging the air-fuel ratio of the intake air to the target air-fuel ratio with high accuracy for each cylinder. Therefore, it is possible to eliminate a difference in air-fuel ratio between the cylinders due to a variation in the injection amount of the injector, etc., and further, to significantly reduce the control gain as compared with a case of controlling the air-fuel ratio of all the cylinders collectively. Since the value can be set large and accurately, it is possible to perform an extremely high response air-fuel ratio control.

[0004]

Generally, when a linear O2 sensor is provided in an exhaust pipe assembly of a multi-cylinder engine, the time required for the combustion gas discharged from each cylinder to reach the linear O2 sensor is reduced. Is a dead time in the feedback control system, and this dead time changes depending on the exhaust flow rate.Therefore, the change in the exhaust flow rate due to the change in the operating state of the engine has a considerable effect on the estimation accuracy of the air-fuel ratio by the observer. Is known to have an effect.

As shown in FIG. 8, for example, when the amount of air taken into the engine becomes less than a predetermined value when the vehicle is decelerated,
It has been found by the inventor of the present application that the dead time sharply increases as the exhaust flow velocity decreases. In such a case, in the above-described conventional air-fuel ratio control apparatus, the estimation error of the air-fuel ratio by the observer becomes considerably large due to a sudden increase in the dead time, and the controllability is deteriorated due to hunting and the like due to the large control gain. This causes a problem that exhaust emission is deteriorated.

In particular, when so-called fuel cut control for stopping the supply of fuel during deceleration of a vehicle or the like is performed, the air-fuel ratio of each cylinder suddenly changes to the lean side in addition to the decrease in the exhaust flow velocity as described above. However, even if the process shifts to the air-fuel ratio control thereafter, there is a problem that the estimation accuracy of the observer becomes extremely low.

[0007] The present invention has been made in view of the above points, and an object of the present invention is to devise an air-fuel ratio control procedure even when the air-fuel ratio estimation accuracy is reduced as described above. The purpose of the present invention is to prevent deterioration in controllability and the accompanying deterioration in exhaust emissions.

[0008]

In order to achieve the above object, the present invention provides a means for determining a predetermined decrease in the exhaust flow velocity based on the operating state of the engine. When the estimation accuracy of the observer is expected to decrease due to a sudden increase in the dead time accompanying the decrease, the feedback control of the air-fuel ratio is suppressed.

More specifically, according to the first aspect of the present invention, as shown in FIG. 2, an air-fuel ratio detecting means 22 provided in a collection portion of an exhaust passage of an engine 1 having a plurality of cylinders, Estimating means 3 for estimating the air-fuel ratio x (k + 1) for each cylinder of the engine 1 based on the value y (k) detected by the detecting means 22
0b, adjusting means 16 for adjusting the air-fuel ratio for each cylinder by adjusting the intake air amount or the fuel supply amount to the engine 1,
The air-fuel ratio estimated value x (k
The present invention is directed to an air-fuel ratio control device including control means for feedback-controlling the air-fuel ratio for each cylinder by the adjusting means 16 so that +1) becomes the target air-fuel ratio. The control means includes: a basic adjustment amount setting section 30a for setting a basic adjustment amount Tbase of the adjustment means 16 corresponding to the target air-fuel ratio based on the operating state of the engine 1;
A feedback correction unit 30c for correcting the basic adjustment amount Tbase so that the air-fuel ratio estimated value x (k + 1) for each cylinder based on 0b becomes the target air-fuel ratio. A determination unit for determining that the exhaust flow velocity of the engine is reduced to a predetermined value or less; and a feedback correction unit when the determination unit determines that the exhaust flow velocity is reduced to a predetermined value or less. Suppression means 30e for suppressing the correction of the basic adjustment amount Tbase by 30c is provided.

According to this configuration, the air-fuel ratio can be feedback-controlled for each cylinder based on the air-fuel ratio estimated value x (k + 1) for each cylinder estimated by the estimating means 30b. The difference in fuel ratio between cylinders can be eliminated, and the control gain can be set to be much larger than in the case where the air-fuel ratios of all cylinders are controlled collectively, so that extremely responsive control can be performed.

When the exhaust flow velocity from the engine 1 falls below a predetermined value, for example, when the vehicle decelerates, the predetermined decrease state of the exhaust flow velocity is determined by the determining means 30d based on the operating state of the engine 1. According to the determination result, the correction by the feedback correction unit 30c is suppressed by the suppression unit 30e, so that the estimated air-fuel ratio value x (k + 1)
The air-fuel ratio feedback control based on is controlled. That is, when the exhaust flow velocity from the engine 1 is reduced to a predetermined value or less, and the estimation error of the cylinder-by-cylinder air-fuel ratio x (k + 1) by the estimation unit 30b increases due to a sudden change in the dead time, By suppressing the air-fuel ratio feedback control, it is possible to prevent deterioration in controllability and exhaust emission even when the control gain is set large as described above.

According to a second aspect of the present invention, the feedback correction unit according to the first aspect of the present invention controls at least a proportional control operation and an integral control operation based on a deviation between the estimated air-fuel ratio value and the target air-fuel ratio. And the suppression means suppresses the correction by the feedback correction unit by reducing the control gain in the control law.

With this, the content of the feedback control is embodied, and the basic adjustment amount is corrected by the feedback correction unit according to a control law having at least a proportional control operation and an integral control operation, so that the air-fuel ratio estimated value is set to the target air-fuel ratio. The fuel ratio can be converged. In addition, the suppression unit can suppress the feedback control by reducing the control gain in the control law, thereby suppressing the correction by the feedback correction unit.

According to the third aspect of the present invention, the first or second aspect is provided.
The control means in the invention described above performs fuel cut control for stopping the supply of fuel to the engine in accordance with a predetermined operating condition of the engine, and the suppression means performs a predetermined period after the end of the fuel cut control, It is assumed that the correction by the feedback correction unit is suppressed.

That is, in general, during the fuel cut control, the dead time becomes extremely large due to the decrease in the exhaust flow velocity. Therefore, immediately after returning from the fuel cut control to the normal air / fuel ratio control, the air / fuel ratio of the exhaust pipe collecting portion is reduced. Due to the influence of the fuel cut control, a lean state different from the actual air-fuel ratio of each cylinder is obtained, and therefore, the estimated value of the air-fuel ratio of each cylinder based on the above-mentioned air-fuel ratio of the collecting portion becomes uncertain. Therefore, in the present invention, the air-fuel ratio feedback control based on the uncertain air-fuel ratio estimated value can be suppressed by suppressing the correction by the feedback correction unit for a predetermined period after the fuel cut control.

According to a fourth aspect of the present invention, the feedback correction unit according to the third aspect of the invention performs a proportional control operation, an integral control operation, and a differential control operation based on a deviation between the estimated air-fuel ratio value and the target air-fuel ratio. The control means corrects the basic adjustment amount in accordance with the PID control rule, and the control means adjusts the control gain in the PID control rule for a predetermined period after the end of the fuel cut control, by changing the control gain of the integral control operation to that of another control operation. A configuration is provided that includes a gain correction unit that corrects to a value larger than the control gain.

That is, in general, during the fuel cut control, the dead time becomes extremely large due to the decrease in the exhaust flow velocity. Therefore, immediately after returning from the fuel cut control to the normal air / fuel ratio control, the air / fuel ratio of the exhaust pipe collecting portion is maintained. Due to the influence of the fuel cut control, a lean state different from the actual air-fuel ratio of each cylinder is obtained, and therefore, the estimated value of the air-fuel ratio of each cylinder based on the above-mentioned air-fuel ratio of the collecting portion becomes uncertain. Therefore, a proportional control operation that directly reflects the current deviation between the air-fuel ratio estimated value and the target air-fuel ratio in feedback correction and a differential control operation that directly reflects the change degree of the current deviation in feedback correction are also performed. It becomes uncertain.

On the other hand, the integral value of the past deviation is reflected in the integral control operation, and the influence of the fuel cut control is relatively small. In addition, the integrated value of the past deviation reflects a variation in the air-fuel ratio caused by, for example, a variation in the injectors of each cylinder. Thus, in the present invention, the control gain of the integral control operation is made higher than the control gains of the other control operations, and the uncertain air-fuel ratio feedback control is suppressed by the PID control having a high rate of the integral control operation, while the cylinder of the air-fuel ratio is controlled. The gap can be effectively eliminated.

According to a fifth aspect of the present invention, an air-fuel ratio detecting means provided in a collection portion of an exhaust passage of an engine having a plurality of cylinders, and each of the cylinders of the engine based on a value detected by the air-fuel ratio detecting means. Estimating means for estimating the air-fuel ratio, adjusting means for adjusting the air-fuel ratio for each cylinder by adjusting the amount of intake air or fuel supply to the engine, and the estimated value of the air-fuel ratio for each cylinder by the estimating means is the target air-fuel ratio. The present invention is directed to an air-fuel ratio control device including control means for feedback-controlling the air-fuel ratio for each cylinder by the adjusting means so as to obtain a fuel ratio. The control means includes a basic adjustment amount setting unit for setting a basic adjustment amount of the adjustment means corresponding to the target air-fuel ratio based on the operating state of the engine, and an air-fuel ratio estimated value for each cylinder by the estimation means. And a feedback correction unit that corrects the basic adjustment amount so that the target air-fuel ratio becomes the target air-fuel ratio. A determination unit that determines that the exhaust flow velocity of the engine has decreased to a predetermined value or less, and an exhaust flow velocity The detection sensitivity correction means for increasing the detection sensitivity of the air-fuel ratio detection means when it is determined that the air-fuel ratio falls below a predetermined value is determined.

According to this configuration, an air-fuel ratio control with extremely high response can be performed as in the first aspect of the invention.
Further, for example, when the exhaust flow velocity from the engine drops below a predetermined value during deceleration of the vehicle or the like, the detection sensitivity of the air-fuel ratio detection unit is increased by the detection sensitivity correction unit according to the result of the determination by the determination unit. In other words, for example, the sampling time for detection by the air-fuel ratio detecting means is shortened to increase the detection sensitivity, so that even when the exhaust flow rate from the engine drops below a predetermined value and the dead time changes rapidly, Since the air-fuel ratio of the exhaust pipe collecting portion can be accurately detected without being delayed by the change, deterioration of controllability and deterioration of exhaust emission can be prevented.

Incidentally, in order to increase the detection sensitivity of the air-fuel ratio detection means, the filter gain may be increased in addition to shortening the detection sampling time.

[0022]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows an air-fuel ratio control apparatus according to the present invention
1 shows an embodiment in which the present invention is applied to a four-cylinder four-stroke gasoline engine 1.
(Only one is shown), a cylinder head 4 mounted on the upper surface of the cylinder block 3, and a piston 5 fitted reciprocally in each cylinder 2. A combustion chamber 6 surrounded by a piston 5 and a cylinder head 3 is formed in each of the cylinders 2. An ignition plug 7 is provided at an upper portion of the combustion chamber 6, and the ignition plug 7 is connected to an ignition circuit 8 including an igniter and the like.

Reference numeral 10 denotes an intake passage for supplying intake air (air) to the combustion chamber 6 of each cylinder 2.
The upstream end is connected to the air cleaner 11, while the downstream end is connected to the combustion chamber 6 via the intake valve 12. In the intake passage 10, a thermal airflow sensor 13 for detecting an amount of intake air to be taken into the engine 1, a throttle valve 14 for restricting the intake passage 10, a surge tank 15,
Four injectors (adjustment means) 16, 16,... (Only one is shown in the figure) for injecting fuel for each cylinder are arranged in order from the upstream side. The throttle valve 14
The throttle opening is controlled based on the accelerator opening and the like by an actuator 14a which is mechanically disconnected from the accelerator pedal and operates by receiving a control signal from the ECU 30 described later. Has become.

On the other hand, reference numeral 20 denotes an exhaust passage for discharging exhaust gas from the combustion chamber 6, the upstream end of which is connected to the combustion chamber 6 via an exhaust valve 21. The upstream side of the exhaust passage 20 is constituted by an exhaust manifold 20a composed of four exhaust pipes which are individually communicated with the combustion chamber 6 of each cylinder 2, and the exhaust manifold of the exhaust manifold 20a has an exhaust gas collecting section. A linear O2 sensor 22 is provided as air-fuel ratio detecting means for detecting the air-fuel ratio based on the oxygen concentration in the inside. A catalytic converter 23 composed of a three-way catalyst for purifying exhaust gas is provided downstream of the linear O2 sensor 22.

Further, the engine 1 is provided with a crank angle sensor 26 such as an electromagnetic pickup for detecting the rotation angle of the crankshaft 25. The engine 1 is rotated based on a pulse signal output from the crank angle sensor. The number is to be detected. Although not shown, a water temperature sensor is provided adjacent to the water jacket of the engine 1 and detects a cooling water temperature.

In FIG. 1, reference numeral 30 denotes an ECU (Electronic Control Uniform) constituted by a microcomputer or the like.
t). The ECU 30 includes an airflow sensor 1
3. Each output signal from the linear O2 sensor 22, the crank angle sensor 26, the accelerator sensor 28 and the like is input. On the other hand, from the ECU 30, a signal for controlling the throttle opening is output to the actuator 14a of the throttle valve 14, an ignition control signal is output to the ignition circuit 8, and the injector 16a for each cylinder is output. , A pulse signal for controlling fuel injection is output to the four cylinders 2, 2,..., The first cylinder, the third cylinder, the fourth cylinder,.
The ignition is performed in the order of the second cylinder.

In the ECU 30, as shown in the block diagram of FIG. 2, the basic injection amount (basic adjustment amount) Tbase of the injector 16 corresponding to the target air-fuel ratio is set by the basic adjustment amount setting section 30a. On the other hand, the air-fuel ratio y (k) of the exhaust pipe assembly detected by the linear O2 sensor 22
The air-fuel ratio x (k + 1) of each cylinder of the engine 1 is estimated by an observer (estimating means) 30b described later based on the estimated air-fuel ratio x (k + 1). PID controller (feedback correction unit) 3 so as to converge
The basic injection amount Tbase is corrected by 0c. And
By inputting the corrected injection amount Tout to the injector 16 and adjusting the fuel injection amount, the air-fuel ratio for each cylinder can be feedback-controlled with high accuracy. As is well known, the PID controller 30b calculates an air-fuel ratio estimated value x (k + 1)
The basic injection amount Tbase is corrected in accordance with a PID control law having a proportional control operation, an integral control operation, and a differential control operation based on a deviation between the PID and the target air-fuel ratio.
Control means is constituted by the controller 30c and the basic adjustment amount setting section 30a.

Next, a specific procedure of the air-fuel ratio control by the ECU 30 will be described with reference to flowcharts shown in FIGS.

First, in the main flow of the air-fuel ratio control shown in FIG.
An output signal from each sensor such as 3 is input and stored in the RAM, and a target air-fuel ratio is set in a subsequent step SA2. The target air-fuel ratio is set to change at a predetermined cycle around the stoichiometric air-fuel ratio A / F = 14.7 except for a specific case such as when the engine is started. In step SA3, the basic injection amount Tbase of the fuel corresponding to the target air-fuel ratio is read from a map set in advance based on the engine speed and the intake charging efficiency and set. The intake charging efficiency of the engine 1 is calculated based on the intake air amount and the engine speed.

At Step SA4, the linear O2 sensor 2
2 based on the output signal from the air-fuel ratio y
Calculate (k). The details of the calculation of the air / fuel ratio y (k) will be described with reference to FIG.
Then, the dead time D is calculated from a map set in advance based on the engine speed and the intake air amount. That is, the dead time D is the time required for exhaust gas discharged from a certain cylinder to flow through the exhaust pipe of the exhaust manifold 20a and arrive at the linear O2 sensor 22, and is directly dependent on the exhaust flow rate. It is. Therefore, a test using an experimental vehicle in which various conditions such as the volume of the exhaust pipe were set in the same way as the actual vehicle was used to previously store a map in which the dead time D was set in association with the engine speed and the intake charging efficiency as described above in the ROM. In step B1, the dead time D is calculated by reading from the map based on the engine speed and the intake charging efficiency.

Subsequently, in step SB2, the ignition cycle T of the engine 1 is calculated based on the engine speed. In step SB3, based on the dead time D and the ignition cycle T, the ignition timing T reaches the current linear O2 sensor 22. The number of the cylinder 2 from which the exhaust gas is mainly discharged is identified. That is, first, the cylinder identification coefficient csgt (k) indicating how many times the exhaust gas currently reaching the linear O2 sensor 22 has been discharged in the fuel injection cycle is defined. This csgt (k) is obtained by calculating the dead time D calculated in the current fuel injection cycle k by the ignition cycle T
, And the remainder is rounded up. For example, as shown in FIG. 5, the exhaust gas discharged in the 9th previous fuel injection cycle changes from the previous fuel injection cycle k-1 to the current fuel injection cycle k. Csgt (k) = 9 when the first area between the two has been reached.

Here, since the ignition order of each cylinder 2 of the engine 1 is determined to be the first cylinder, the third cylinder, the fourth cylinder, the second cylinder,..., The fuel injection cycle at the current time k is, for example, the first cylinder. Assuming that this is a cylinder fuel injection cycle, the fuel injection cycle nine times before corresponds to the fuel injection cycle of the second cylinder, that is, the opening timing of the exhaust valve 21 of the first cylinder. Thus, it is identified that the exhaust gas currently reaching the linear O2 sensor 22 is mainly discharged from the first cylinder.

Step SB4 following step SB3
Then, the current value and the previous value of the cylinder identification coefficient csgt (k) are compared, and if csgt (k) = csgt (k−1) is equal to YES, the process proceeds to step SB5, while if not equal, NO Proceed to step SB8. In step SB5, the detection address n of the linear O2 sensor 22 corresponding to the time when the exhaust gas arrives in the first area shown in FIG.
Is calculated. The detection address n is obtained by dividing the remainder obtained by dividing the dead time D by the ignition cycle T by the sampling time (for example, several milliseconds) of the linear O2 sensor 22.

Subsequently, at step SB6, the ECU 30
Of the linear O2 sensor 22 stored in the RAM is used, and the air-fuel ratio y (k) of the collecting portion is calculated based on this detected value. That is, the gas volume of the exhaust gas discharged from each cylinder 2 and reaching the exhaust pipe collecting portion periodically increases and decreases as shown in FIG. 6, but by employing the detection value of the detection address n, As shown by the point M in the figure, the detection value of the linear O2 sensor 22 when the exhaust gas volume discharged from a certain cylinder becomes the maximum can be adopted.
The air-fuel ratio y (k) of the collecting section corresponding to the cylinder can be detected with high accuracy.

In step SB7, the calculated collective air-fuel ratio y (k) is set as the collective air-fuel ratio y (k) # 1 corresponding to the first cylinder identified as described above, and is stored in the RAM of the ECU 30.
To memorize. Thus, the previous fuel injection cycle k-1
In the first region from to the current fuel injection cycle k, the cylinder that has just reached the exhaust pipe collection section and has the maximum gas volume is identified, and the collection section air-fuel ratio y ( Since k) can be detected with high accuracy, it is possible to increase the estimation accuracy when estimating the cylinder-by-cylinder air-fuel ratio by the observer 30b based on the collective air-fuel ratio as described later.

On the other hand, in step SB3, the cylinder identification coefficient
In step SB8, in which the current value of csgt (k) is not equal to the previous value and the result is NO, the current value and the previous value of csgt (k) are compared in magnitude and csgt (k)> csgt (k-1 If the current value is larger than YES in step), the process returns. If the current value is smaller than the previous value, the process returns to step SB9 and step S9.
Proceed to B10. That is, the case where the current value of the cylinder identification coefficient csgt (k) is larger than the previous value means that, for example, the engine 1 is in a deceleration operation state, the exhaust flow velocity is decreasing, and the dead time D is increasing. And in this case,
The peak of the exhaust gas volume does not appear in the first region of the current fuel injection cycle k, and the next fuel injection cycle k + 1
Therefore, in the current fuel injection cycle k, the calculation of the collective air-fuel ratio y (k) is not performed.

On the other hand, the case where the current value of the cylinder identification coefficient csgt (k) is equal to or less than the previous value means that, for example, the engine 1 is in an accelerating operation state and the exhaust flow velocity increases, and the dead time D decreases. In this case, in this case, the fuel injection cycle k-2 from the previous fuel injection cycle k-2 to the previous fuel injection cycle k-1
The peak of the exhaust gas volume from the first cylinder appears in a second region (see FIG. 5) until the first region, and in the first region, the peak of the exhaust gas from the third cylinder discharged from the third cylinder after the first cylinder. Exhaust gas volume peaks appear.

Therefore, in step SB8, the current value of the cylinder identification coefficient is determined to be less than or equal to the previous value, and the process proceeds to step S8.
In step B9 and step SB10, in the same manner as in step SB5 and step SB6, the collecting unit air-fuel ratio y (k) is calculated based on the detection value when the exhaust gas volume reaches a peak in the first region, and the subsequent step SB11 Then, the collective unit air-fuel ratio y (k) is stored in the RAM of the ECU 30 as the collective unit air-fuel ratio y (k) # 3 corresponding to the third cylinder.
Subsequently, in step SB12, the air-fuel ratio y (k) of the collecting portion is calculated based on the detected value when the exhaust gas volume reaches the peak in the second region in the same manner as described above.
At 13, this collective air-fuel ratio y (k) is stored in the RAM of the ECU 30 as a collective air-fuel ratio y (k) # 1 corresponding to the first cylinder.

Therefore, even if the dead time D changes due to a change in the operating state of the engine 1, the cylinder in which the gas volume at the exhaust pipe collecting section is peaked is identified, and the collecting section corresponding to the identified cylinder is identified. The air-fuel ratio y (k) can be detected with high accuracy. Then, for each fuel injection cycle, the collective unit air-fuel ratio corresponding to the discriminating cylinder replaced in the order of ignition is calculated, and y (k) # 1, y (k) # 2, y (k) #
3, and y (k) # 4 is stored and updated in the RAM for each corresponding cylinder. In this way, the cylinder at which the gas volume peaks at the exhaust pipe collecting section at that time is identified for each fuel injection cycle, and the collecting section air-fuel ratio y corresponding to the identified cylinder is identified.
Since (k) is stored and updated in the RAM, the signal input from the linear O2 sensor 22 can be stored for only two fuel injection cycles as shown in the first and second areas in FIG. I just need. In other words, the storage amount of the signal input from the linear O2 sensor 22 is smaller than that in the case where the collective air-fuel ratio (k) corresponding to the cylinder in which the fuel injection is performed at that time is calculated for each fuel injection cycle. This suffices to reduce the RAM capacity by half, so that the RAM capacity of the ECU 30 can be saved.

Next, returning to the main flow of FIG. 3, in step SA5 following step SA4 described above, the air-fuel ratio x (k + 1) of each cylinder is estimated by the observer 30b.
Here, the estimation of the air-fuel ratio for each cylinder by the observer 30b will be described below.

In general, when the observation of the state variables using the observer is applied to the estimation of the air-fuel ratio of each cylinder, first, how the air-fuel ratio of each cylinder is reflected in the air-fuel ratio of the collecting portion is described. Considering this, a theoretical model that represents the behavior of the air-fuel ratio at the junction is constructed. Specifically, the four cylinders 2, 2,... Of the engine 1 are ignited in a predetermined order such as the first cylinder, the third cylinder, the fourth cylinder, the second cylinder,. Therefore, the exhaust gas discharged from each of the cylinders 2 flows through each of the exhaust pipes of the exhaust manifold 20a and reaches the collecting portion in order. Therefore, it is considered that the ratio of the exhaust gas volume in the exhaust pipe collecting portion is that the exhaust gas from each cylinder 2 is mixed as shown in FIG. 7, and the collecting portion air-fuel ratio at a certain time k is different for each cylinder. Since it may be considered that the air-fuel ratio is determined by the air-fuel ratio and the ratio of the gas volume, the aggregate air-fuel ratio y (k) at the current time k is represented by the following equation.

[0043]

(Equation 1)

Where x (k), x (k-1), x (k-2), x (k-3)
Is the air-fuel ratio for each cylinder. Note that x (k), x (k-1),
As the x (k-2), x (k-3) and the air-fuel ratio y (k) of the collecting portion, the equivalence ratio, that is, the reciprocal of the excess air ratio λ is used.

The above c1, c2, c3, and c4 represent the gas volume of each cylinder as a weighted average, and represent the contribution ratio of each cylinder to the air-fuel ratio y (k) of the collecting section. The values of c1, c2, c3, and c4 change in response to changes in the exhaust flow speed due to changes in the engine speed and the intake charging efficiency, and can be specified experimentally.
For example, Table 1 shows the results obtained by attaching a linear O2 sensor to each cylinder, comparing the output values thereof with the output values from the linear O2 sensor provided in the exhaust pipe assembly, and performing multiple regression analysis.

[0046]

[Table 1]

By the way, when the above (Equation 1) is expressed as a determinant, the following (Equation 2) is obtained.
(k), x (k-1), x (k-2), x (k-3) are state variables, and y
It is an output equation showing the behavior of the air-fuel ratio of the collecting part when (k) is output.

[0048]

(Equation 2)

On the other hand, a state equation representing the behavior of the air-fuel ratio at the collecting portion is represented by the following (Equation 3).

[0050]

(Equation 3)

In the above (Equation 3), x (k + 1), which is the future value, cannot be given to the input. Therefore, it is assumed that the previous air-fuel ratio in the same cylinder is almost unchanged, and x (k +1) = x (k-3), so that
Is as shown in the following (Equation 4).

[0052]

(Equation 4)

As described above, the estimation of the air-fuel ratio for each cylinder by the observer 30b observes the state variable x (k) in the system expressed by the above-described equation (4) and the output equation (2). This results in the problem of the well-known Kalman filter. That is,

[0054]

(Equation 5)

From the above (Equation 5), X (k) is represented by the following recurrence formula.

[0056]

(Equation 6)

Here, P is a positive definite solution satisfying the Riccati equation. When the coefficient matrix is a constant, the following recurrence formula converges to obtain P. Note that C ^T is the transposed matrix of the matrix C.

[0058]

(Equation 7)

In the above (Equation 6) and (Equation 7), R,
Q is a weight matrix given as a design specification, and is an element that determines the convergence speed of the filter and stability against disturbance. In general, the larger the value of Q with respect to R, the faster the convergence speed, but the lower the stability against disturbances such as noise. ^AT is the transposed matrix of the matrix A.

From the above (Equation 6) and (Equation 7), the state variable x (k + 1) can be estimated based on the air-fuel ratio y (k) of the collecting section as in the following equation.

[0061]

(Equation 8)

Therefore, in step SA5 in the main flow of FIG. 3, c is obtained from a map set in advance as shown in Table 1 based on the engine speed and the intake charging efficiency.
The values of 1, c2, c3, and c4 are read, and the air-fuel ratio x (k + 1) is estimated based on the air-fuel ratio y (k) of the collecting section.

As described above, the cylinder-by-cylinder air-fuel ratio could be estimated by the observer 30b at step SA5.
It becomes possible to feedback control the air-fuel ratio every time.

Specifically, in step SA6 following step SA5, it is determined based on the engine speed or the like whether or not the fuel cut control for stopping the fuel supply at the time of deceleration of the vehicle or the like is being executed. If it is determined that the fuel cut is being performed, the process proceeds to step SA13. If it is determined that the fuel cut is not performed, the process proceeds to step SA7 to determine whether the intake air amount of the engine 1 is equal to or more than a predetermined value. I do. Here, the predetermined value is a predetermined intake air amount such that the degree of change of the dead time D sharply changes as shown in FIG.

In step SA7, the engine 1
If it is determined that the intake air amount is smaller than the predetermined value, the process proceeds to step SA12, while YES
Is determined, the process proceeds to step SA8, and it is determined whether a predetermined time has elapsed since the end of the fuel cut control. Then, if the predetermined time has not elapsed, NO
If so, the process proceeds to step SA11. If the predetermined time has elapsed, the process proceeds to step SA9 to perform normal PID gain setting. Subsequently, in step SA10, the feedback correction amount Tf / b is calculated by the PID controller 30c based on the normal PID gain setting, and the basic injection amount Tbase is corrected by the feedback correction amount Tf / b. I do.

Tf / b = PID {target air-fuel ratio-air-fuel ratio for each cylinder x (k + 1)} Tout = Tbase + Tf / b That is, an estimated air-fuel ratio x (k) for each cylinder estimated by the observer 30b. +1) and the target air-fuel ratio are multiplied by the control gain (P gain) of the proportional control operation and the control gain of the integral control operation (I
Gain) and the differential control term obtained by integrating the control gain (D gain) of the differential control operation with the differential value of the deviation to calculate a feedback correction amount Tf / b. The basic injection amount Tbase is corrected by the correction amount Tf / b, and the fuel injection amount is determined for each cylinder.

Here, since the cylinder-by-cylinder air-fuel ratio x (k + 1) can be estimated with high accuracy by the observer 30b as described above, it is necessary to set the P gain, I gain and D gain to extremely large values, respectively. And with this,
The feedback control of the air-fuel ratio for each cylinder can be made extremely responsive.

On the other hand, in step SA11 where it is determined that the predetermined time has not elapsed since the end of the fuel cut control in step SA8, the P gain and the D gain in the normal PID gain setting are reduced, and The control gain of the PID controller 30c is corrected so that the I gain becomes higher than the P gain and the D gain. Then, the process proceeds to step SA10 to execute feedback control, and thereafter returns.

That is, during the fuel cut control, the dead time D becomes extremely large due to the decrease in the exhaust flow velocity. Therefore, immediately after returning from the fuel cut control to the normal control, the air-fuel ratio y (k) of the collecting portion is reduced. Due to the influence of the control, a lean state different from the actual air-fuel ratio of each cylinder is established. For this reason, the air-fuel ratio estimated value x (k + 1) for each cylinder based on the gathering portion air-fuel ratio y (k) is also uncertain, and accordingly, the air-fuel ratio estimated value x (k + 1) The proportional control operation and the differential control operation which directly reflect the current deviation between the target and the target air-fuel ratio also become uncertain.

On the other hand, since the integral value of the past deviation is reflected in the integral control operation, the influence of the fuel cut control is relatively small, and the integral value of the past deviation is, for example, an injector for each cylinder. The variation in the air-fuel ratio resulting from the variation in the machine difference is reflected. Therefore, in step SA11, the I gain is made higher than the P gain and the D gain so as to increase the ratio of the integral control operation in the PID control.

Further, in step SA12 where it is determined that the intake air amount of the engine 1 is smaller than the predetermined value in step SA7 and the process proceeds to step SA12, all of the P gain, the I gain and the D gain are set with respect to the normal PID gain setting. The reduced setting is set, and the process proceeds to step SA10. That is, when the intake air amount of the engine 1 is smaller than a predetermined value,
As shown in FIG. 8, the dead time D sharply increases with a decrease in the exhaust flow velocity, and the estimation error of the cylinder-by-cylinder air-fuel ratio x (k) by the observer 30b increases.
The feedback correction based on (k) also becomes uncertain. Thus, in this case, the P, I, and D control gains in the PID controller 30c are all reduced to suppress feedback correction.

Further, step SA6 is followed by step SA6 in which it is determined that the fuel is being cut and YES.
In step 13, the feedback correction by the PID controller 30c is prohibited, the open loop control is performed by the basic operation unit 30a, and then the process returns. This is because in the fuel cut control, the injectors 16, 1
It is sufficient to stop the fuel supply from 6, 6,..., And it is not necessary to perform feedback correction of the air-fuel ratio.
Note that during the fuel cut control, the above step SA5
The estimation of the cylinder-by-cylinder air-fuel ratio may be omitted.

In the flowchart shown in FIG. 3, step SA3 is performed when the injector 1 corresponding to the target air-fuel ratio
The basic adjustment amount setting unit 30a for setting the basic injection amount Tbase of 6, 16,...
An observer 3 for estimating an air-fuel ratio x (k + 1) for each cylinder based on the collective air-fuel ratio y (k) detected by the sensor 22;
0b corresponds to the PID controller 30c that corrects the basic injection amount of each injector 16 based on the air-fuel ratio estimated value x (k + 1) by the observer 30b.

In step SA7, the determination means 30d for determining that the exhaust gas flow rate has decreased to a predetermined value or less based on the operating state of the engine 1, and in step SA1
1, SA12, and SA13 correspond to the suppression means 30e that suppresses the correction of the basic injection amount Tbase by the PID controller 30c when the determination means 30d determines that the exhaust flow velocity is in the predetermined reduced state. Step SA11 corresponds to the gain correction unit 30f that corrects the I gain to be larger than the P gain and the D gain in the PID controller 30c for a predetermined period after the fuel cut control.

Accordingly, in this embodiment, the air-fuel ratio estimated value x (k) for each cylinder estimated by the observer 30b
+1), it is possible to perform feedback control of the air-fuel ratio for each cylinder, so that it is possible to eliminate the difference in air-fuel ratio between cylinders and to control the air-fuel ratio of all cylinders collectively. In comparison, P, I,
Each control gain of D can be set extremely large,
Thereby, feedback control with extremely high response can be performed.

When the intake air amount of the engine 1 is smaller than a predetermined value and the exhaust flow rate is lower than a predetermined value, the dead time D accompanying the decrease in the exhaust flow rate causes a sudden change in the dead time D, so that the air-fuel ratio x ( Since the estimation error of (k + 1) becomes considerably large, each control gain of the PID controller 30c is reduced to suppress the feedback correction of the basic fuel injection amount Tbase. As a result, even if the control gains of P, I, and D are set large as described above, the controllability such as hunting is deteriorated due to the decrease in the estimation accuracy of the air-fuel ratio x (k + 1), and the exhaust gas associated therewith is deteriorated. Emission deterioration can be prevented.

During fuel cut control, PI
By prohibiting the feedback correction by the D controller 30c, even if the estimation accuracy of the observer 30b is extremely low due to the rapid change of the air-fuel ratio and the exhaust flow speed accompanying the fuel cut control, the controllability is deteriorated and the exhaust emission is deteriorated. Can be prevented.

Further, for a predetermined period after returning from the fuel cut control to the normal control, the P gain and the D gain are reduced, and the PID controller 30c is controlled so that the I gain becomes higher than the P gain and the D gain. By correcting the gain, the proportional control operation and the differential control operation that directly reflect the uncertainty of the air-fuel ratio estimated value due to the fuel cut control are suppressed, while the influence of the fuel cut control is relatively small and each cylinder is controlled. It is possible to increase the ratio of the integral control operation that well reflects the variation in the air-fuel ratio for each cylinder, thereby effectively eliminating the cylinder-to-cylinder difference in the air-fuel ratio while suppressing uncertain air-fuel ratio feedback control. .

(Other Embodiments) The present invention is not limited to the above embodiment, but includes various other embodiments. That is, in the above-described embodiment, the calculation of the air-fuel ratio y (k) in the exhaust pipe assembly shown in FIG. 4 is performed for each fuel injection cycle. However, the present invention is not limited to this. You may make it.

The air-fuel ratio control device of the present invention is also applicable to a so-called in-cylinder direct injection engine in which fuel is directly injected into a cylinder.

Further, in the above-described embodiment, it is determined that the exhaust flow velocity has decreased to a predetermined value or less based on the intake air amount of the engine 1. However, the present invention is not limited to this. The determination may be made based on the efficiency.

Furthermore, in the above embodiment, when it is expected that the dead time D increases due to the decrease in the exhaust flow velocity, the PI
Although the feedback correction by the D controller 30c is suppressed, the present invention is not limited to this. For example, it is also possible to provide detection sensitivity correction means for shortening the sampling time of detection by the linear O2 sensor 22 and increasing the detection sensitivity. In this way, even when the exhaust flow velocity from the engine 1 drops below a predetermined value and the dead time D suddenly changes, the air-fuel ratio y (k) of the exhaust pipe collecting portion is maintained without delay. Can be accurately detected,
Deterioration of controllability and deterioration of exhaust emission can be prevented.

In the above embodiment, since the integral control operation is for correcting the cylinder variation of the air-fuel ratio of each engine 1, the feedback correction amount of the integral control operation before the fuel cut control is performed before the fuel cut control is performed. After the fuel cut control is completed, the basic operation amount Tba is stored based on the stored feedback correction amount.
If se is corrected, controllability is improved.

[0084]

As described above, according to the air-fuel ratio control apparatus for an engine according to the first aspect of the present invention, the air-fuel ratio is feedback-controlled for each cylinder based on the estimated value of the air-fuel ratio for each cylinder by the estimating means. As a result, it is possible to eliminate the cylinder-to-cylinder difference in the air-fuel ratio and to increase the response of the control.
In the case where the dead time changes drastically as the exhaust flow velocity from the engine decreases and the estimation error of the air-fuel ratio of each cylinder by the estimation means becomes large, the above-described air-fuel ratio feedback control is suppressed by suppressing the air-fuel ratio feedback control. Even if the control gain is set large as described above, it is possible to prevent deterioration of controllability and deterioration of exhaust emission.

According to the third aspect of the present invention, uncertain air-fuel ratio feedback control can be suppressed even if the accuracy of estimating the air-fuel ratio decreases immediately after returning from the fuel cut control to the normal control.

According to the fourth aspect of the present invention, even if the estimation accuracy of the air-fuel ratio is lowered immediately after returning from the fuel cut control to the normal air-fuel ratio control, the air-fuel ratio feedback control is suppressed while the uncertain air-fuel ratio feedback control is suppressed. The difference in fuel ratio between cylinders can be effectively eliminated.

According to the fifth aspect of the present invention, similarly to the first aspect of the present invention, it is possible to eliminate a difference in air-fuel ratio between cylinders, and it is possible to perform extremely responsive control.
Even if the estimation error of the air-fuel ratio increases due to a sudden change in the dead time, the control sensitivity and the exhaust emission can be prevented from being deteriorated by increasing the detection sensitivity of the air-fuel ratio detection unit by the detection sensitivity correction unit.

[Brief description of the drawings]

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

FIG. 2 is a functional block diagram of the air-fuel ratio control device.

FIG. 3 is a flowchart illustrating a feedback control procedure of an air-fuel ratio.

FIG. 4 is a flowchart illustrating a procedure for identifying an exhaust cylinder from a dead time and calculating a collective air-fuel ratio corresponding to the identified cylinder.

FIG. 5 is an explanatory diagram of a method of identifying an exhaust cylinder based on a calculated dead time.

FIG. 6 is an explanatory diagram showing an increase / decrease characteristic of an exhaust gas volume discharged from a certain cylinder in an exhaust pipe collecting part.

FIG. 7 is an explanatory diagram showing a change in a ratio of an exhaust gas volume from each cylinder in an exhaust pipe assembly.

FIG. 8 is an explanatory diagram showing a relationship between an intake air amount of an engine and a dead time in air-fuel ratio feedback control.

[Explanation of symbols]

Tbase Basic injection amount (basic adjustment amount) x (k + 1) Air-fuel ratio estimated value for each cylinder 1 engine 2 cylinder 16 injector (adjustment means) 22 linear O2 sensor (air-fuel ratio detection means) 30a basic adjustment amount setting unit 30b Observer (estimating means) 30c PID controller (feedback correcting section) 30d determining means 30e suppressing means 30f gain correcting section

Claims (5)

[Claims] 1. An air-fuel ratio detecting means provided in a collection portion of an exhaust passage of an engine having a plurality of cylinders, and an estimation for estimating an air-fuel ratio for each cylinder of the engine based on a value detected by the air-fuel ratio detecting means. Means, adjusting means for adjusting the air-fuel ratio for each cylinder by adjusting the amount of intake air or fuel supply to the engine, and the estimated air-fuel ratio for each cylinder by the estimating means becomes the target air-fuel ratio. A control means for performing feedback control of an air-fuel ratio for each cylinder by the adjusting means, wherein the control means sets a basic adjustment amount of the adjusting means corresponding to the target air-fuel ratio to an operating state of the engine. And a feedback correction unit that corrects the basic adjustment amount such that the estimated value of the air-fuel ratio of each cylinder by the estimation unit becomes the target air-fuel ratio. Determining means for determining that the exhaust flow velocity of the engine has decreased to a predetermined value or less; and performing basic adjustment by the feedback correction unit when the determination means determines that the exhaust flow velocity has decreased to a predetermined value or less. An air-fuel ratio control device for an engine, further comprising suppression means for suppressing correction of an amount. 2. The apparatus according to claim 1, wherein the feedback correction unit corrects the basic adjustment amount based on a deviation between the estimated air-fuel ratio and the target air-fuel ratio according to a control law having at least a proportional control operation and an integral control operation. An air-fuel ratio control device for an engine, wherein the suppression means is configured to reduce a control gain in the control law, thereby suppressing correction by the feedback correction unit. 3. The control unit according to claim 1, wherein the control unit is configured to perform fuel cut control for stopping fuel supply to the engine in accordance with a predetermined operating condition of the engine. An air-fuel ratio control device for an engine, wherein the correction by the feedback correction unit is suppressed for a predetermined period after the end of the fuel cut control. 4. The feedback correction unit according to claim 3, wherein the feedback correction unit is configured to perform a basic adjustment amount based on a deviation between the estimated air-fuel ratio and the target air-fuel ratio according to a PID control law having a proportional control operation, an integral control operation, and a differential control operation. The control means adjusts the control gain in the PID control law for a predetermined period after the end of the fuel cut control so that the control gain of the integral control operation is larger than the control gains of the other control operations. An air-fuel ratio control device for an engine, comprising a gain correction unit for correcting the air-fuel ratio. 5. An air-fuel ratio detecting means provided in a collective portion of an exhaust passage of an engine having a plurality of cylinders, and an estimation for estimating an air-fuel ratio for each cylinder of the engine based on a value detected by the air-fuel ratio detecting means. Means, adjusting means for adjusting the air-fuel ratio for each cylinder by adjusting the amount of intake air or fuel supply to the engine, and the estimated air-fuel ratio for each cylinder by the estimating means becomes the target air-fuel ratio. A control means for performing feedback control of an air-fuel ratio for each cylinder by the adjusting means, wherein the control means sets a basic adjustment amount of the adjusting means corresponding to the target air-fuel ratio to an operating state of the engine. And a feedback correction unit that corrects the basic adjustment amount such that the estimated value of the air-fuel ratio of each cylinder by the estimation unit becomes the target air-fuel ratio. Determining means for determining that the exhaust flow velocity of the engine has dropped below a predetermined value; and detecting the air-fuel ratio detecting means when the determining means determines that the exhaust flow velocity has dropped to a predetermined value or less. An air-fuel ratio control device for an engine, comprising detection sensitivity correction means for increasing sensitivity.