JPH06200809A - Air-fuel ratio control device of internal combustion engine - Google Patents

Abstract

(57) [Abstract] [Purpose] While performing feedback control, the correction value stored for each cylinder in each operating region eliminates the deviation of the air-fuel ratio between the cylinders, keeps the exhaust property constant, and purifies the exhaust gas with a three-way catalyst. Do efficiently. [Structure] When the feedback control condition of the air-fuel ratio by the exhaust gas oxygen sensor is satisfied (S11), the air-fuel ratio feedback correction value α is discharged from the lean side time TL α, the rich side time TR α, and the # 1 cylinder. The lean time TL # 1 and the rich time TR # 1 of the exhaust component of the exhaust gas are detected (S13 to 16), and KINJ0n = a × (AB−1-b) +1 (where A = TL α / TR α, B # 1 = TL # 1 / T
The correction coefficient for each cylinder is calculated by R # 1) (S19), and the fuel supply amount is calculated based on this.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an air-fuel ratio control system for an internal combustion engine, and more particularly to an air-fuel ratio control system for correcting an air-fuel ratio step difference for each cylinder.

[0002]

2. Description of the Related Art As an internal combustion engine equipped with this type of air-fuel ratio control device, there is a conventional one disclosed in, for example, Japanese Patent Application Laid-Open No. 3-149330. That is, an oxygen sensor is provided in the collecting portion of the exhaust pipe, air-fuel ratio feedback control is performed for each cylinder based on the signal of the oxygen sensor, and an air-fuel ratio feedback correction coefficient for each cylinder and air-fuel ratio feedback control for all cylinders are performed. There is a method in which the air-fuel ratio feedback correction coefficients are obtained, the variations are obtained, the basic fuel injection amount is corrected for each cylinder, and then the air-fuel ratio feedback control is performed in all cylinders.

[0003]

However, in such a conventional air-fuel ratio control device, the air-fuel ratio feedback control is performed only for a specific cylinder, and the oxygen sensor of the collecting portion is used to detect the air-fuel ratio of that cylinder. However, because the other cylinders are not feedback controlled,
The oxygen sensor of the collecting portion is affected by the air-fuel ratios of the other cylinders, and cannot accurately detect the air-fuel ratio of the cylinder to be detected.

Further, since feedback control is performed only in one cylinder, the responsiveness of the air-fuel ratio is poor, and the conversion (purification) of exhaust components in the catalyst is not sufficiently performed. Further, in the conventional example, the deviation of the air-fuel ratio is the characteristic variation of the injector, and the feedback correction value is one for each cylinder as the deterioration of the injector, but in reality, the variation of the intake air amount for each cylinder is different depending on the operating condition. In the example, it cannot be corrected enough.

By the way, the oxygen sensor is provided at the collecting portion of the exhaust passages (exhaust manifolds) of all cylinders of the engine or partial cylinders (for example, one bank in a V-type 8-cylinder engine). The concentration ratio is not separately detected for each cylinder, but the oxygen concentration ratio as an approximately average value of each cylinder is detected. However, the engine has a plurality of factors that cause variations in the air-fuel ratio among the cylinders. Therefore, when air-fuel ratio feedback control of all cylinders or partial cylinders is performed using the oxygen sensor as described above, Since the air-fuel ratio cannot be controlled individually to the target air-fuel ratio and fluctuations in the air-fuel ratio cannot be suppressed sufficiently, the exhaust properties are not constant and the capacity of the three-way catalyst to purify the exhaust is set to a large value. However, there is a problem in that the amount of the noble metal used in the three-way catalyst increases and the cost increases.

The factors that cause the air-fuel ratio variation among the cylinders include uneven intake distribution, fuel injection valve characteristic variations, cylinder charging efficiency differences, etc., which cause a rich air-fuel ratio in each cylinder.・ Lean occurs, but since the air-fuel ratio cannot be detected for each cylinder with conventional air-fuel ratio feedback control, even if the overall average can be controlled to a target air-fuel ratio, the air-fuel ratio variation among the cylinders The air-fuel ratio fluctuates around the target air-fuel ratio.

Further, in order to perform feedback control based on the detected oxygen concentration ratio as a substantially average value of each cylinder,
The responsiveness of the air-fuel ratio is poor, and exhaust gas components in the catalyst cannot be converted (purified) sufficiently. The present invention has been made in view of the above problems, and it is possible to correct the air-fuel ratio difference for each cylinder so that the fuel supply amount can be feedback-controlled so that the actual air-fuel ratio approaches the target air-fuel ratio. An object is to provide an air-fuel ratio control device for an internal combustion engine.

[0008]

Therefore, the present invention is based on FIG.
As shown in, the average air-fuel ratio detection means for detecting the averaged air-fuel ratio of the air-fuel mixture supplied to each cylinder from the predetermined components in the exhaust gas from which the exhaust gas from each cylinder is merged, and the average air-fuel ratio detection means An air-fuel ratio feedback control means for feedback-correcting the air-fuel ratio for each cylinder so as to approach the target air-fuel ratio based on the air-fuel ratio detected by The cylinder-by-cylinder air-fuel ratio detection means for detecting the air-fuel ratio for each cylinder based on the component is provided, and the air-fuel ratio detected by the average air-fuel ratio detection means is controlled to be rich and lean with respect to the target air-fuel ratio. A shift detecting means for detecting a shift in the air-fuel ratio of each cylinder by comparing the rich and lean times for each cylinder detected by the air-fuel ratio detecting means for each cylinder; And air-fuel ratio correction means for correcting the air-fuel ratio to each cylinder on the basis of the deviation to be, has a configuration provided with a.

Further, as shown in FIG. 2, a cylinder-by-cylinder correction value for correcting the air-fuel ratio for each cylinder based on the deviation detected by the deviation detecting means is provided for each of the operating regions divided into a plurality based on the engine operating state. The stored correction value for each cylinder and the correction value for each cylinder stored in the correction value storage means for each cylinder are set so that the air-fuel ratio detected by the average air-fuel ratio detection means is rich and lean with respect to the target air-fuel ratio. It may be configured to include learning correction means for correcting and rewriting by learning based on each controlled time and the cylinder rich / lean time detected by the cylinder air-fuel ratio detection means. .

[0010]

According to the above construction, the air-fuel ratio of each cylinder is set to the target air-fuel ratio based on the air-fuel ratio detected by the average air-fuel ratio detecting means based on the predetermined component in the exhaust where the exhaust from each cylinder is merged. Feedback correction is performed so as to approach.
Here, the deviation of the air-fuel ratio for each cylinder, each time the air-fuel ratio detected by the average air-fuel ratio detection means is controlled to rich and lean with respect to the target air-fuel ratio, and by the cylinder-by-cylinder air-fuel ratio detection means It is detected by comparing the detected rich and lean times for each cylinder.

Further, the air-fuel ratio correction means corrects the air-fuel ratio for each cylinder based on the deviation. Further, a cylinder-by-cylinder correction value for correcting the air-fuel ratio for each cylinder based on the deviation detected by the deviation detecting means is stored for each of a plurality of operating regions divided based on the engine operating state, and the cylinder-by-cylinder correction value is learned. May be corrected and rewritten.

Since the cylinder-by-cylinder correction value is learned based on time, even when the oxygen sensor deteriorates and the output (voltage) of the oxygen sensor fluctuates, the rich side and the lean side are different from each other. Since it is possible to secure the required accuracy only by measuring the time, for example, the configuration of determining the state of each cylinder by obtaining the area (time x voltage) between the rich side and the lean side using the output of the oxygen sensor, etc. It is possible to avoid the deterioration of accuracy as much as possible in comparison with.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. In FIG. 3, air is sucked into the 4-cylinder internal combustion engine 1 from an air cleaner 2 through an intake duct 3, a throttle valve 4 and an intake manifold 5. At the branch portion of the intake manifold 5, a fuel injection valve 6 is provided for each cylinder. The fuel injection valve 6 is an electromagnetic fuel injection valve that is energized by a solenoid to open and stop energized to be closed. The fuel injection valve 6 is energized and opened by a drive pulse signal from a control unit 12 described later, and a fuel not shown is shown. Fuel that is pumped and adjusted to a predetermined pressure by a pressure regulator is injected and supplied.

A spark plug 7 is provided in the combustion chamber of the engine 1 to ignite sparks to ignite and burn the air-fuel mixture. Exhaust gas is discharged from the engine 1 through the exhaust manifold 8, the exhaust duct 9, the three-way catalyst 10, and the muffler 11. The three-way catalyst 10 is an exhaust purification device that oxidizes CO and HC in the exhaust components and reduces NO _X to convert them into other harmless substances, and burns the air-fuel mixture at the stoichiometric air-fuel ratio. Sometimes both conversion efficiencies are the best.

The control unit 12 includes a CPU, RO
A microcomputer including an M, a RAM, an A / D converter, and an input / output interface is provided, and the input signals from various sensors are received and arithmetic processing is performed as described below to operate the fuel injection valve 6. Control. As the various sensors, a hot-wire type or flap type air flow meter 13 is provided in the intake duct 3 and outputs a voltage signal according to the intake air flow rate Q.

In the case of four cylinders provided with a crank angle sensor 14, a reference signal REF (reference signal) for each 180 ° crank angle and a position signal POS (unit signal) for each 1 ° or 2 ° crank angle are provided. Is output. Here, the engine rotation speed N can be calculated by measuring the cycle of the reference signal REF or the number of generated position signals POS within a predetermined time, and one of the reference signals REF is different from the other. The cylinder discrimination signal of the # 1 cylinder can be identified by the pulse width. A water temperature sensor 15 for detecting the cooling water temperature Tw of the water jacket of the engine 1 is also provided.

Further, the exhaust manifolds 8 of the # 1 to # 4 cylinders are assembled as shown in FIG. 4, and the exhaust is guided to the three-way catalyst 10 and the muffler 11 via the exhaust duct 9 communicating with each other. There is. And, in the exhaust manifold 8,
Oxygen sensors 21, 2 as cylinder-by-cylinder air-fuel ratio detecting means for detecting exhaust components of exhaust gas discharged from the # 1 to # 4 cylinders
2, 23 and 24 are provided.

Further, the exhaust duct 9 is provided with an oxygen sensor 31 as an average air-fuel ratio detecting means for detecting the air-fuel ratio of the air-fuel mixture sucked into the engine 1 via the oxygen concentration in the exhaust gas. Here, the CPU of the microcomputer incorporated in the control unit 12 performs the arithmetic processing according to the program on the ROM shown as the flowcharts in FIGS. 5, 6 and 9 to 12 to control the fuel injection.

The functions as the air-fuel ratio feedback control means, the deviation detection means, the air-fuel ratio correction means and the learning correction means are achieved by the above program. Further, as the cylinder-by-cylinder correction value storage means, a RAM with a backup of a microcomputer incorporated in the control unit 12 corresponds. FIG. 5 is a fuel injection amount calculation routine, that is, a routine for calculating the fuel injection amount to be injected and supplied to each cylinder, which is executed every predetermined time (for example, 10 ms).

Step 1 (indicated as S1 in the figure).
The same applies hereinafter), based on the intake air flow rate Q detected based on the signal from the air flow meter 13, the engine rotation speed N calculated based on the signal from the crank angle sensor 14, and the signal detected from the water temperature sensor 15. Input the water temperature Tw and the like. In step 2, the basic fuel supply amount Tp = K × Q / N (K is a constant) corresponding to the intake air flow rate per unit rotation is calculated from the intake air flow rate Q and the engine rotation speed N.

In step 3, various correction coefficients COEF
(= 1 + K _MR + K _Tw + K _AS + K _AI + ...) is the water temperature Tw
And so on. In step 4, the voltage correction amount Ts is set based on the battery voltage. This is to correct a change in the injection flow rate of the fuel injection valve 6 due to a change in the battery voltage.

In step 5, the air-fuel ratio feedback correction value LAMBDAα set in another routine is read. In step 6, the correction value KINJn for each cylinder, which is the learning value of the correction value for each cylinder corresponding to the cylinder in which the fuel injection is performed and is set in a learning routine described later, is read.

In step 7, the fuel supply amount Ti is calculated according to the following equation. Ti = Tp · COEF · LAMBDA · KINJn + Ts In step 8, the fuel supply amount Ti set in step 7
Is set in the output register for the calculation cylinder. As a result, when the fuel injection timing is synchronized with a predetermined engine rotation (for example, every one rotation), a drive pulse signal having a pulse width corresponding to the latest set fuel injection amount Ti is provided to the fuel cylinder provided in the calculation cylinder. Given to the injection valve 6,
Fuel injection is performed.

The fuel supply amount Ti calculated in step 7 is the fuel supply amount T when a plurality of cylinders simultaneously inject.
In the case of sequential injection, Ti = Tp · COEF · LAMBDA · KINJn · 2 + Ts, as a matter of course. Next, the program shown in the flowchart of FIG. 6 relates to the first embodiment of the present invention, and the air-fuel ratio feedback correction value α (initial value = 1.
0) for proportional-integral control based on the rich / lean detection of the air-fuel ratio by the oxygen sensor 31, and for correcting the basic fuel supply amount Tp for each cylinder for each operation region divided into a plurality based on the engine operating state. Cylinder correction value of KINJn
Is a program for setting.

The program shown in the flow chart of FIG. 6 is executed every one revolution of the engine 1. As will be described later, first, the program is executed for the # 1 cylinder. Step
At 11, it is determined whether or not the air-fuel ratio feedback control condition is currently satisfied. Here, the condition for stopping (clamping) the air-fuel ratio feedback control is, for example, the following case.

At start-up Low water temperature At engine high load operation At deceleration operation At idle operation When oxygen sensor 31 is abnormal And when the air-fuel ratio feedback control condition is established, the routine proceeds to step 12 to carry out learning. Determine the current operating range.

In step 12, the above-mentioned basic fuel supply amount T
Of the plurality of operating regions divided by p and engine speed N, the corresponding operating region is determined. The air-fuel ratio feedback correction value α and the cylinder-by-cylinder correction value KINJn are stored for each operating region. In step 13, as shown in FIG. 7, the time (lean side time) TL α during which the reduction correction of the air-fuel ratio feedback correction value α in the air-fuel ratio feedback control by the oxygen sensor 31 is corrected is read.

In step 14, immediately after reading the air-fuel ratio feedback correction value α in step 13, the oxygen sensor 21 provided in the exhaust manifold 8 of the # 1 cylinder detects the exhaust gas from the # 1 cylinder. The time (lean time) TL # 1 during which the exhaust component of the exhaust gas is in a lean state is read (see FIG. 7). In step 15, as shown in FIG. 7, the time (rich side time) TR α during which the increase correction of the air-fuel ratio feedback correction value α in the air-fuel ratio feedback control by the oxygen sensor 31 is corrected is read.

In step 16, the oxygen is discharged from the # 1 cylinder detected by the oxygen sensor 21 provided in the exhaust manifold 8 of the # 1 cylinder immediately after reading the air-fuel ratio feedback correction value α in step 13. The time during which the exhaust component of the exhaust gas is rich (rich time) TR # 1 is read (see FIG. 7). In step 17, the ratio A of TL α read in step 13 to TR α read in step 15
(= TL α / TR α) is calculated.

In step 18, the ratio B # between TL # 1 read in step 14 and TR # 1 read in step 16
1 (= TL # 1 / TR # 1) is calculated. In step 19, the correction coefficient KI for each cylinder for correcting the basic fuel supply amount Tp so as to eliminate the deviation of the air-fuel ratio related to the # 1 cylinder.
NJ0n is calculated according to the following equation. KINJ0n = a × (A / B # 1-b) +1 where the constant a in the above equation is the cylinder-specific correction coefficient KINJ0n
Is a slope with respect to A / B # 1, and the constant b is as described below. That is, due to the difference between the rich-to-lean response and the lean-to-rich response of the output of the oxygen sensor 21, the fuel injection amount correction is not always necessary when A / B # 1 = 1, B is set to correct the responsiveness of the sensor 21.

In step 19, the correction coefficient KINJ0n for each cylinder may be calculated according to the following equation. That is, KINJ0n = a * (A-B # 1-b) +1 where the constants a and b are set to correct the responsiveness of the oxygen sensor 21 as described above.

At step 20, the learning value KINJ of the cylinder-by-cylinder correction coefficient KINJ0n related to the operating region determined at step 12
The cylinder-by-cylinder correction value KINJn is updated and calculated according to the following equation so that n is learned in the direction of eliminating the air-fuel ratio deviation between the cylinders. KINJn = c * KINJn + (1-c) * KINJ0n where c is a coefficient for weighted averaging.

Then, the cylinder-by-cylinder correction value KINJn calculated in step 20 is updated in step 21 among the plurality of operating areas divided by the basic fuel supply amount Tp and the engine speed N. As data, the map data is updated. In step 22, in order to determine whether or not the learning has been completed for all cylinders, in this embodiment, since the engine has four cylinders, it is determined whether n representing the number of cylinders is four, and n When it is determined that ≠ 4, after n is incremented in step 23, the process returns to step 11, and the learning is performed for the # 2 to # 4 cylinders in the same manner as described above.

When it is determined in step 22 that n = 4, n is set to 1 in step 24, and this routine ends. When the engine has, for example, 6 cylinders, it is needless to say that this routine is terminated when it is determined that n = 6.
Therefore, according to the first embodiment, it is possible to detect the deviation of the air-fuel ratio of each cylinder at the same time while performing the feedback control by the collecting part oxygen sensor, so that the air-fuel ratio of each cylinder is accurately detected. It became possible to do. Further, since the air-fuel ratio of each cylinder is corrected while performing feedback control by the collecting part oxygen sensor, control can be performed without deteriorating the exhaust gas purification performance of the three-way catalyst 10.

That is, as shown by the dotted line in FIG.
In the feedback control by the collective oxygen sensor, the feedback correction is performed so that the mixing ratio becomes appropriate, but consider the case where the air-fuel ratio of the # 1 cylinder is improperly mixed and deviates to the rich side. In this case, the ratio A (= TL α / TR α) calculated in step 17 is approximately 1. On the other hand, the ratio B # 1 (= TL # 1 / TR # 1) calculated in step 18
Causes TR # 1 to be longer than TL # 1, so B # 1 is B
# 1 <1, so A / B # 1 becomes large, and # 1
The cylinder-by-cylinder correction coefficient KINJ0n is set small. Therefore, the fuel supply amount Ti for the # 1 cylinder is set small,
Leaning will be achieved.

Further, in the first embodiment, a predetermined cylinder,
Since the learning value is obtained by calculating the weighted average after calculating the correction value for the predetermined operation area, while securing the learning frequency,
There is also an effect that the learning value can be stabilized. In addition, the cylinder-by-cylinder correction value KINJn
Since correction or learning correction is made based on α, TL # 1 and TR # 1, etc., even if the oxygen sensor 21, 31 etc. deteriorates and the output (voltage) of the oxygen sensor fluctuates, the rich side Since the required accuracy can be ensured only by measuring the time between the lean side and the lean side, for example, using the output of the oxygen sensor, the area between the rich side and the lean side (time x voltage)
In comparison with the configuration for determining the state of each cylinder,
There is also a great effect that it is possible to avoid a decrease in precision as much as possible.

Next, the control according to the second embodiment of the present invention will be described. The fuel injection amount calculation routine is the same as that in the first embodiment, and therefore its explanation is omitted. Next, FIG.
Also, the program shown in the flowchart of FIG. 10 sets the cylinder-specific correction value KINJn for correcting the basic fuel supply amount Tp for each cylinder for each operation region divided into a plurality of sections based on the engine operating state, as described above. However, the steps having the same effects as those of the flowchart shown in FIG. 6 are denoted by the same step numbers and the description thereof will be omitted.

When it is judged at step 11 that the air-fuel ratio feedback control condition is satisfied, the routine proceeds to step 31, where m, which represents the number of arithmetic mean data to be described later, is set to 1. In step 32, the current operating area
It is determined whether or not it is in the same operating range as the operating range determined in 12, and only if they are the same, the process proceeds to step 33.

In step 33, the time (lean side time) TL α during which the air-fuel ratio feedback correction value α in the air-fuel ratio feedback control by the oxygen sensor 31 is reduced and corrected.
Read ₁ (m = 1 at TL α _m ). In step 34, the air-fuel ratio feedback correction value α
(Lean time) TL # 1 when the exhaust component of the exhaust gas discharged from the # 1 cylinder is detected by the oxygen sensor 21 provided in the exhaust manifold 8 of the # 1 cylinder immediately after reading Read ₁ (m = 1 in TL # 1 _m ).

In step 35, the time during which the air-fuel ratio feedback correction value α in the air-fuel ratio feedback control by the oxygen sensor 31 is increased and corrected (rich side time) TR α
Read ₁ (m = 1 in TR α _m ). In step 36, the air-fuel ratio feedback correction value α
Immediately after the reading of, the time (rich time) TR # 1 in which the exhaust component of the exhaust gas discharged from the # 1 cylinder is detected by the oxygen sensor 21 provided in the exhaust manifold 8 of the # 1 cylinder (rich time) Read ₁ (m = 1 in TR # 1 _m ).

At step 37, m representing the number of data of the arithmetic mean for a predetermined cylinder and a predetermined operating region is a predetermined value m _0.
It is determined whether or not the above. If it is determined that m ≧ m _0, it is determined that the data has been averaged enough to stabilize the learning described later, and the process proceeds to step 38.
On the other hand, if it is determined that m is less than m ₀ , m is incremented in step 39 and then the process returns to step 12.

In step 38, TL α _m , TR α _m , T
Arithmetic mean values TL α ₀ and TR α of L # 1 _m and TR # 1 _m
0 , TL # 1 ₀ and TR # 1 ₀ are calculated. That is, for example, TL α ₀ = (TL α ₁ TL α ₂ + ... TL α _m ) / m is calculated. In step 39, the ratio D (= TL α ₀ / T of the TL α ₀ and the TR α ₀ calculated in step 38 is calculated.
Calculate R α ₀ ).

In step 40, the ratio E # 1 (= TL # 1 ₀ between the TL # 1 ₀ and the TR # 1 ₀ calculated in step 38 is calculated.
/ TR # 1 ₀₎ is calculated. In step 40, a correction coefficient KINJ0n for each cylinder for correcting the basic fuel supply amount Tp is calculated according to the following equation so as to eliminate the deviation of the air-fuel ratio related to the # 1 cylinder. KINJ0n = a * (D / E # 1-b) +1 Here, the constants a and b in the above equation are the same as described above.

In step 40, the cylinder-by-cylinder correction coefficient KINJ0n may be calculated according to the following equation. That is, KINJ0n = a × (D-E # 1-b) +1 In step 42, the learning value KINJn of the cylinder-by-cylinder correction coefficient KINJ0n relating to the operating region determined in step 41 is in the direction of eliminating the deviation of the air-fuel ratio between the cylinders. As will be learned in step S1, the cylinder-specific correction value KINJn is updated and calculated according to the following equation.

KINJn = c × KINJn + (1-c) × KINJ0n where c is a coefficient for weighted averaging. Then, the cylinder-by-cylinder correction value KINJn calculated in step 42 is
Then, the map data is updated as the update data of the corresponding operating region among the plurality of operating regions divided by the basic fuel supply amount Tp and the engine speed N.

Therefore, also in the second embodiment, the same operation and effect as those of the first embodiment can be obtained. Further, in the second embodiment, the arithmetic mean of the detection times of the predetermined cylinder and the predetermined operation region is calculated, the correction value is calculated using the sufficiently stable detection time, and the weighted average is calculated to obtain the learning value. Therefore, there is an effect that it becomes possible to obtain a learning value having a very high stabilization.

Next, the control according to the third embodiment of the present invention will be described. The fuel injection amount calculation routine is the same as that in the first embodiment, and therefore its explanation is omitted. Next, FIG.
Also, the program shown in the flowchart of FIG. 12 sets the cylinder-specific correction value KINJn for correcting the basic fuel supply amount Tp for each cylinder for each operation region divided into a plurality based on the engine operating state, as described above. Although the program is for this purpose, steps having the same operations as those in the flowcharts shown in FIGS. 9 and 10 are denoted by the same step numbers and description thereof will be omitted.

In step 51, the learning value, which is either the learning value that is updated and newly stored in step 21, or the learning value that is already stored in step 21, is used for each cylinder in the updated predetermined operating region. Correction value KINJ ₁ , KINJ ₂ , KINJ ₃
And KINJ ₄ are each 1 or more, and whether or not the learning value (learning value by the oxygen sensor 31) Lα by the basic air-fuel ratio learning control performed as a separate routine is less than 1.

If YES is determined, the routine proceeds to step 52, where cylinder-by-cylinder correction values KINJ ₁ , KINJ ₂ , KINJ ₃ and
The minimum value of KINJ ₄ is calculated as KINJ _min .
In step 53, the cylinder-by-cylinder correction value KINJn as the update data of the corresponding operating region updated in step 21 and the learning value Lα by the basic air-fuel ratio learning control are corrected according to the following equation.

KINJn = KINJn- (KINJ _min -1) Lα = Lα + (KINJ _min -1) On the other hand, if NO in step 51, the step
Proceed to 54. In step 54, the cylinder correction values KINJ ₁ , KINJ
₂ , it is determined whether KINJ ₃ and KINJ ₄ are each less than 1 and the learning value Lα by the basic air-fuel ratio learning control is 1 or more.

If YES is determined, the routine proceeds to step 55, where cylinder-by-cylinder correction values KINJ ₁ , KINJ ₂ , KINJ ₃ and
The maximum value among KINJ ₄ is calculated as KINJ _max .
In step 56, the cylinder-by-cylinder correction value KINJn as the update data of the corresponding operating region updated in step 21 and the learning value Lα by the basic air-fuel ratio learning control are corrected according to the following equation.

KINJn = KINJn- (KINJ _max -1) Lα = Lα + (KINJ _max -1) On the other hand, if NO in step 54, the step
55 and 56 are jumped, and the correction value KINJn for each cylinder stored in step 21 and the learning value L by the basic air-fuel ratio learning control
Set as it is without correcting α.

Therefore, also in the third embodiment, the same operation and effect as those of the first embodiment can be obtained. Since the basic air-fuel ratio learning is added, the fuel supply amount Ti is calculated according to the following equation. Ti = Tp · COEF · LAMBDA · KINJn · Lα + Ts Furthermore, in the third embodiment, the arithmetic mean of the respective detection times is calculated for a predetermined cylinder and a predetermined operating region, and a correction value is obtained using a sufficiently stable detection time. Is calculated, and the learning value is obtained by further weighted averaging. Therefore, it is possible to obtain a learning value with extremely high stabilization.

Further, in steps 51 to 56, the correction by the cylinder-by-cylinder correction value KINJn is set to the minimum, and the cylinder-by-cylinder correction value KINJn and the learning value Lα by the basic air-fuel ratio learning control are also extremely greatly corrected. Is prevented. Here, both the cylinder-by-cylinder correction value KINJn and the learning value Lα by the basic air-fuel ratio learning control are both limited within a predetermined limit value, and when the limit value is exceeded, the cylinder is clamped to the limit value and a sufficient correction is made. However, in the third embodiment, this fear can be avoided, and the learning effect can be enhanced.

[0055]

As described above, according to the present invention, feedback correction is performed so that the air-fuel ratio approaches the target air-fuel ratio for each cylinder based on the air-fuel ratio related to the exhaust gas from which the exhaust gas from each cylinder is merged. At the same time, the air-fuel ratio correction means
The air-fuel ratio is corrected for each cylinder based on the air-fuel ratio related to the exhaust gas obtained by combining the exhaust gases from the respective cylinders and the rich time and lean time of the cylinder-specific air-fuel ratio detected by the cylinder-specific air-fuel ratio detecting means.

Further, in which operating range is corrected by the cylinder-by-cylinder correction value stored for each operating region that is corrected and updated by learning for each operating region so as to eliminate the deviation of the air-fuel ratio between the cylinders. Also, there is an effect that the correction for each cylinder is reduced, the feedback control of the air-fuel ratio is reliably performed, the exhaust property is constant, and the exhaust purification by the three-way catalyst is efficiently performed.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing the configuration of the present invention.

FIG. 2 is a functional block diagram showing the configuration of the present invention.

FIG. 3 is a system configuration diagram showing an overall configuration of a fuel supply device for an internal combustion engine according to an embodiment of the present invention.

FIG. 4 is a plan view for explaining a mounting position of the oxygen sensor in FIG.

FIG. 5 is a flow chart for explaining air-fuel ratio control in the embodiment of the present invention.

FIG. 6 is a flowchart for explaining control for setting a cylinder-specific correction value in the first embodiment of the present invention.

FIG. 7 is a time chart for explaining the sampling timing of the oxygen sensor output in the above embodiment.

FIG. 8 is a time chart for explaining the operation of the above embodiment.

FIG. 9 is a flowchart for explaining control for setting a cylinder-specific correction value in the second embodiment of the present invention.

FIG. 10 is a flowchart for explaining control for setting a cylinder-specific correction value according to the second embodiment of the present invention.

FIG. 11 is a flowchart for explaining control for setting a cylinder-specific correction value according to the third embodiment of the present invention.

FIG. 12 is a flowchart for explaining control for setting a cylinder-specific correction value according to the third embodiment of the present invention.

[Explanation of symbols]

 1 Internal Combustion Engine 6 Fuel Injection Valve 10 Three Way Catalyst 12 Control Unit 13 Air Flow Meter 14 Crank Angle Sensor 21 Oxygen Sensor 31 Oxygen Sensor

Claims (2)

[Claims] 1. An average air-fuel ratio detecting means for detecting an averaged air-fuel ratio of an air-fuel mixture supplied to each cylinder from a predetermined component in the exhaust gas obtained by combining the exhaust gases from the respective cylinders, and the average air-fuel ratio detecting means. An air-fuel ratio feedback control means for feedback-correcting the air-fuel ratio for each cylinder so as to approach the target air-fuel ratio based on the air-fuel ratio detected by The cylinder-by-cylinder air-fuel ratio detecting means for detecting the air-fuel ratio for each cylinder based on the component is provided, and each time the air-fuel ratio detected by the average air-fuel ratio detecting means is rich and lean with respect to the target air-fuel ratio. A shift detecting unit that compares the rich and lean times of each cylinder detected by the cylinder-specific air-fuel ratio detecting unit to detect a shift of the air-fuel ratio of each cylinder; Air-fuel ratio control system for an internal combustion engine, wherein the air-fuel ratio correction means for correcting the air-fuel ratio to each cylinder on the basis of the deviation is issued, that was provided. 2. A cylinder-by-cylinder correction value storage that stores a cylinder-by-cylinder correction value for correcting the air-fuel ratio for each cylinder based on the deviation detected by the deviation detection means for each of a plurality of operating regions divided based on the engine operating state. Means, and the cylinder-by-cylinder correction value stored in the cylinder-by-cylinder correction value storage means, at each time when the air-fuel ratio detected by the average air-fuel ratio detecting means is controlled to be rich and lean with respect to the target air-fuel ratio. And a learning correction unit that corrects and rewrites by learning based on the rich and lean times for each cylinder detected by the cylinder-by-cylinder air-fuel ratio detection unit. Air-fuel ratio control device for internal combustion engine.