JP3024463B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

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 device for an internal combustion engine, and more particularly to a technique for diagnosing a failure in a fuel supply system based on an air-fuel ratio learning correction value.

[0002]

2. Description of the Related Art An electronically controlled fuel injection device for an internal combustion engine having an air-fuel ratio learning correction function has been known. The air-fuel ratio learning correction compares the exhaust air-fuel ratio detected by the oxygen sensor with the target air-fuel ratio, and adjusts the air-fuel ratio feedback correction coefficient in a direction to bring the actual air-fuel ratio closer to the target air-fuel ratio by proportional / integral control or the like. On the other hand, while performing variable control, a correction request level based on the air-fuel ratio feedback correction coefficient is learned and stored as an air-fuel ratio learning correction value, and the fuel injection amount is corrected and set by the air-fuel ratio learning correction value together with the air-fuel ratio feedback correction coefficient. It is to let.

Since the air-fuel ratio learning correction value indicates a correction request level required to compensate for a deviation of the air-fuel ratio from the target air-fuel ratio as described above, the air-fuel ratio learning correction value is abnormal. To indicate the correct correction request level,
It can be estimated that a large air-fuel ratio deviation has occurred due to some failure in the fuel supply system such as the fuel injection valve.

Therefore, various diagnostic devices for diagnosing a failure in the fuel supply system based on the level of the air-fuel ratio learning correction value have been proposed (see Japanese Patent Application Laid-Open No. Hei 5-163982).

[0005]

However, in the conventional failure diagnosis of the fuel supply system using the air-fuel ratio learning correction value, when a large difference in the air-fuel ratio suddenly occurs due to the failure of the fuel supply system, such diagnosis is performed. There has been a problem that failures cannot be reliably diagnosed. That is, when an air-fuel ratio deviation occurs, the air-fuel ratio feedback correction coefficient is gradually changed to eliminate the air-fuel ratio deviation.
If the air-fuel ratio deviation is abrupt and large, the output of the oxygen sensor remains stuck in a rich or lean state due to the response delay of the air-fuel ratio feedback control, which is completely different from the normal control cycle. It will show control characteristics.

For this reason, the reset operation of the feedback correction value (periodical clamp control), which is to be executed when the period of the feedback control (rich / lean inversion period) becomes abnormally long, is as described above. The air-fuel ratio feedback correction value is returned to the initial value once when a sudden and large air-fuel ratio deviation occurs, whereby the correction request level can be reflected in the air-fuel ratio learning correction value. Since the air-fuel ratio learning correction value does not change to a level indicating the air-fuel ratio deviation actually generated,
In spite of the occurrence of a failure that causes a deviation of the air-fuel ratio beyond a predetermined value in the fuel supply system, there is a problem that failure determination is not performed.

That is, when the air-fuel ratio deviation gradually increases, the air-fuel ratio learning can be performed with sufficient responsiveness to the air-fuel ratio deviation. It is possible to make a failure determination at the stage where the air-fuel ratio learning value exceeds the reference value and gradually changes with the expansion of the value. On the other hand, when a large air-fuel ratio deviation occurs suddenly, the air-fuel ratio learning does not easily proceed by the above-described reset operation, and there is a concern that the intended failure diagnosis cannot be performed.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and the failure diagnosis of the fuel supply system based on the air-fuel ratio learning correction value can be reliably performed even when a sudden and large air-fuel ratio deviation occurs. The purpose is to be.

[0009]

An air-fuel ratio control apparatus for an internal combustion engine according to the present invention is configured as shown in FIG. In FIG. 1, the air-fuel ratio detection means detects the air-fuel ratio of the engine intake air-fuel mixture, and the air-fuel ratio feedback correction means
The air-fuel ratio feedback correction value is variably controlled so that the air-fuel ratio detected by the air-fuel ratio detecting means approaches the target air-fuel ratio.

The air-fuel ratio learning means learns a correction request level based on the air-fuel ratio feedback correction value variably controlled by the air-fuel ratio feedback correction means as an air-fuel ratio learning correction value. Further, the feedback correction initial setting unit forcibly performs the initial setting of the air-fuel ratio feedback correction value based on a cycle abnormality of the air-fuel ratio feedback control by the air-fuel ratio feedback correcting unit.

The learning value updating means at the time of the initial setting is adapted to execute the learning correction value of the air-fuel ratio based on the detected air-fuel ratio on which the transition to the initial setting control is based upon the initial setting control by the feedback correction initial setting means. To update. on the other hand,
The fuel supply amount correction means corrects and controls the fuel supply amount by the fuel supply means based on the air-fuel ratio feedback correction value and the air-fuel ratio learning correction value.

Further, the diagnosis means performs a failure diagnosis of the fuel supply system based on a comparison between the air-fuel ratio learning correction value and a reference value, and outputs a failure diagnosis signal. Here, the diagnostic means includes:
When the failure of the fuel supply system is diagnosed based on the comparison between the air-fuel ratio learning correction value and the reference value, and the number of times of the initial setting control by the feedback correction initial setting means is a predetermined number, the failure is finally determined. It is preferable to make a determination.

Further, the air-fuel ratio learning correction value learned by the air-fuel ratio learning means is calculated based on an air-fuel ratio learning correction value for each of a plurality of operating regions and an entire region reflection learning value applied to all operating regions. In such a case, it is preferable that the air-fuel ratio learning correction value updated by the initialization-time learning value updating means and the air-fuel ratio learning correction value used by the diagnostic means are set as the full range reflection learning value.

[0014]

According to the air-fuel ratio control device having such a configuration, the air-fuel ratio feedback correction value is variably controlled so that the detected actual air-fuel ratio approaches the target air-fuel ratio, while the correction request level based on the air-fuel ratio feedback correction value is adjusted. Is learned as the air-fuel ratio learning correction value.

Further, when a period abnormality of the air-fuel ratio feedback control occurs, the initial setting of the air-fuel ratio feedback correction value is forcibly performed. Then, at the time of the initial setting control, the air-fuel ratio learning correction value is updated based on the detected air-fuel ratio on which the transition to the initial setting control is based, regardless of the correction level based on the air-fuel ratio feedback correction value.

Therefore, a sharp and large air-fuel ratio deviation occurs,
If the feedback control cycle becomes abnormal due to the response delay of the air-fuel ratio feedback control and the air-fuel ratio learning with respect to the air-fuel ratio deviation, the air-fuel ratio feedback correction value is initialized and the air-fuel ratio learning correction value is set based on the cycle abnormality. Be updated. For this reason, even if the true correction request level based on the air-fuel ratio feedback correction value cannot be fully reflected on the air-fuel ratio learning correction value by the initial setting control associated with the period abnormality, the air-fuel ratio learning By updating the correction value, it becomes possible to bring the air-fuel ratio learning correction value closer to the correction request level.

As described above, the failure diagnosis of the fuel supply system is performed based on the comparison between the air-fuel ratio learning correction value, which is also updated during the initial setting control due to the cycle abnormality, and the reference value. It is. Here, together with the diagnosis based on the comparison between the air-fuel ratio learning correction value and the reference value, the number of times of the initial setting control is discriminated, and the air-fuel ratio deviation indicated by the air-fuel ratio learning correction value causes a period abnormality to occur. It is preferable to confirm whether or not the learning has been performed in accordance with the deviation.

If the air-fuel ratio learning correction value is divided into an air-fuel ratio learning correction value for each of the plurality of operating regions and an entire region reflection learning value applied to the entire operating region, the air-fuel ratio feedback correction is performed. The update accompanying the initial setting control of the correction value is performed with respect to the whole-area reflection learning value, and the failure diagnosis of the fuel supply system is performed based on the whole-area reflection learning value. The whole-range reflection learning value is applied in the whole operation region and has a high learning frequency, so that it is possible to reflect the correction request level with good response, thereby enabling early failure diagnosis.

[0019]

Embodiments of the present invention will be described below. In FIG. 2 showing the system configuration of the embodiment, an air flow meter 7 is shown.
Detects an air flow rate Qa drawn into the engine 1 via an air cleaner (not shown), and the idle switch 9 detects a fully closed position of the throttle valve 8. The crank angle sensor 10 built in the distributor outputs a signal for each unit crank angle and a signal Ref for each reference crank angle position, and the water temperature sensor 11 detects a cooling water temperature Tw of the engine.

An oxygen sensor as air-fuel ratio detecting means
Numeral 12 is a sensor that changes its output in response to the oxygen concentration in the exhaust gas, which is closely related to the air-fuel ratio of the engine intake air-fuel mixture, and has a characteristic that its output changes suddenly at the stoichiometric air-fuel ratio. It is. Further, a knock sensor 13 for detecting knocking vibration of the engine, a vehicle speed sensor 14 for detecting a traveling speed of a vehicle on which the engine is mounted, and the like are provided. To be entered.

In this embodiment, the control unit 21 serves as air-fuel ratio feedback correction means, air-fuel ratio learning means, feedback correction initial setting means, initial value learning value updating means, fuel supply amount correction means, and diagnosis means. Is provided by software as shown in flowcharts of FIGS. 3 to 7 described later. Fuel is supplied to the engine 1 by an electromagnetic fuel injection valve 4 provided at an intake port 3.
(Fuel supply means). Fuel injection valve 4
Is driven to open in response to an injection pulse signal from the control unit 21, and injects and supplies fuel adjusted to a predetermined pressure. The injection amount is adjusted by the valve opening time (injection pulse width).

The control unit 21 calculates the basic fuel injection amount Tp based on the intake air flow rate Qa and the engine speed Ne such that the ratio (air-fuel ratio) to the intake air amount becomes a target value (target air-fuel ratio). (Tp = K · Qa / Ne:
K is a constant), and usually the air-fuel ratio (base air-fuel ratio) determined by the basic fuel injection amount Tp is near the stoichiometric air-fuel ratio which is the target air-fuel ratio.

The exhaust pipe 5 is provided with a three-way catalyst 6 for treating harmful exhaust components (CO, HC, NOx) discharged from the engine 1. The three-way catalyst 6 keeps the conversion efficiency of exhaust harmful components at the best when the atmosphere of the catalyst is in a narrow range centered on the stoichiometric air-fuel ratio. Therefore, the control unit 21 controls the fuel injection amount based on the output signal from the oxygen sensor 12 so that the actual air-fuel ratio is maintained near the stoichiometric air-fuel ratio at which the three-way catalyst 6 can sufficiently exhibit its purification ability. Is feedback corrected.

Here, the state of the air-fuel ratio feedback control based on the output of the oxygen sensor 12 will be described with reference to the flowchart of FIG. The routine shown in the flowchart of FIG. 3 is executed for each reference angle signal Ref. First, in S1, it is determined whether or not the clamp condition of the air-fuel ratio feedback control is satisfied.

The clamp conditions are, for example, at the time of cooling, at the time of starting, at the time of deceleration, at the time of idling, and the like. When these operating conditions are satisfied, the air-fuel ratio feedback control is not performed. Furthermore, if the rich signal or the lean signal of the oxygen sensor 12 continues for a predetermined time or more (when a cycle is abnormal) during the execution of the air-fuel ratio feedback control without satisfying the clamp condition, the air-fuel ratio is forcibly removed. Period monitor clamp control (initial setting control) for temporarily returning the fuel ratio feedback correction coefficient α to an initial value is executed.

In the period monitor clamp control, the air-fuel ratio feedback correction coefficient α (air-fuel ratio feedback correction value) is gradually increased from the value at that time to the initial value (100%).
%), And feedback control is resumed from a predetermined initial state after a predetermined time has elapsed. When the clamp condition is satisfied, the process proceeds to S23, and the air-fuel ratio feedback correction coefficient α is reset to an initial value. When the clamp condition is not satisfied and the air-fuel ratio feedback control is performed, the process proceeds to S2. Oxygen sensor
From the 12 outputs, it is determined whether or not the actual air-fuel ratio is rich with respect to the stoichiometric air-fuel ratio.

If the air-fuel ratio is richer than the stoichiometric air-fuel ratio, the process proceeds to S3, in which it is determined whether or not the rich determination was made last time. Is determined. If the previous time is not the rich state but the reversal from lean to rich, the process proceeds to S5, where the basic fuel injection amount Tp and the engine speed Ne are set.
Is read out from a map having the following parameters as parameters corresponding to the current operation condition, and in the next S6, the current correction coefficient α is set to α _MAX as the maximum value. That is, in the lean determination state, as will be described later, the control for gradually increasing the correction coefficient α is performed, and the control is shifted to the decrease control with the inversion to the rich state. The correction coefficient α immediately before the shift of the maximum value indicates the maximum value.

At S7, the step amount P obtained at S5 is calculated.
R is subtracted from the previous correction coefficient α, and the result of the subtraction is updated and set as a new correction coefficient α. On the other hand, if it is determined in S3 that the rich determination was also made last time, the process proceeds to S8.
To the integral I in proportion to the fuel injection amount Ti described later.
Calculate R. In the next step S9, the integral IR is subtracted from the correction coefficient α up to the previous time, and the result of the subtraction is updated and set as a new correction coefficient α. Therefore, until the rich state with respect to the stoichiometric air-fuel ratio is canceled and the air condition becomes lean, the correction coefficient α gradually decreases in accordance with the integral IR every time this routine is executed.

Similarly, when it is determined that the actual air-fuel ratio is lean with respect to the stoichiometric air-fuel ratio, if it is an inversion (S4), the correction coefficient α at that time is set to the minimum value α _MIN (S11). ), The correction coefficient α is obtained by a step PL (S10) set from the basic fuel injection amount Tp and the engine speed Ne.
Is increased (S12). On the other hand, when the lean state is continued, the integral IL (S
13), the correction coefficient α is controlled to increase (S14).

The above steps PR and PL and integral I
The air-fuel ratio feedback correction coefficient α is set so that the actual air-fuel ratio approaches the stoichiometric air-fuel ratio, which is the target air-fuel ratio, using R and IL.
, The air-fuel ratio of the air-fuel mixture formed according to the amount of fuel injected from the fuel injection valve 4 is stabilized near the stoichiometric air-fuel ratio. Further, in the present embodiment, as shown in the flowchart of FIG. 4, there is provided a region-by-region air-fuel ratio learning function of learning a correction request level based on the air-fuel ratio feedback correction coefficient α for each of a plurality of operation regions. .

In the flowchart of FIG. 4, first, S
In steps S15 to S18, an air-fuel ratio learning condition is determined. In S15, it is determined whether or not the vehicle is stably located in the same area on the air-fuel ratio learning map divided into a plurality of areas based on the basic fuel injection amount Tp and the engine speed Ne.
At 16, it is confirmed that the air-fuel ratio feedback control is being performed.

In S17, the normal operation state of the oxygen sensor 12 is determined by confirming that the deviation between the maximum value and the minimum value of the output of the oxygen sensor 12 is equal to or greater than a predetermined value. In S18, it is determined whether or not the output of the oxygen sensor 12 has been sampled a predetermined number of times or more. And the above S15 ~
When S18 in learning conditions are satisfied all, the process proceeds to S19, the maximum value alpha _{MA X} of correction factor obtained in S6, S11 of the flowchart of FIG. 3 alpha, using the minimum value alpha _MIN, the correction coefficient alpha The shift amount ε from the control center (100%) is obtained according to the following equation.

The displacement amount _{ε = (α MAX + α MIN} ) / 2-1 Next, in S20, determines on the basis of the learning area and the basic fuel injection amount Tp and the engine speed Ne at that time, the air-fuel ratio learning map The region-specific air-fuel ratio learning correction value KBLRCA (initial value = 100%) stored in the area is read. Then, in S21, the deviation ε is added to the air-fuel ratio learning correction value KBLRCA of the area read from the map.
Is added as the new air-fuel ratio learning correction value KBLRCA of the area, and the data on the map is rewritten.

However, the region-specific air-fuel ratio learning correction value KBL
RCA is within the predetermined upper and lower limit values (for example, 90%
~ 110%). In the present embodiment, the above-described region-specific air-fuel ratio learning correction value KBLRCA is used.
In addition, the whole region reflection learning value KBL applied in the whole operation region
The RCB is separately learned, and the fuel injection amount Ti (fuel supply amount) corresponding to the final injection pulse width is determined according to the following equation.

Ti = Tp × CO × (α + KBLRCA + KBLRCB−2) + Ts In the above equation, CO is various correction coefficients set based on the cooling water temperature Tw and the like, and Ts is an invalid pulse width set according to the battery voltage. is there. The whole range reflection learning value KBLRCB
Is shown in the flowchart of FIG.

In the flowchart of FIG. 5, first, S
At 31, it is determined whether or not a learning permission condition for the whole range reflection learning value KBLRCB is satisfied. It is preferable that the learning permission condition basically matches the execution permission condition of the air-fuel ratio feedback control, and the additional condition is that the canister is not being purged. Then, when the learning permission condition is satisfied, the process proceeds to S32, and it is determined whether or not the condition of the period monitor clamp control based on the rich period abnormality is satisfied. That is, in the present embodiment, as described above, when the rich signal or the lean signal of the oxygen sensor 12 continues for a predetermined time (for example, 30 seconds) during the air-fuel ratio feedback control (for example, when the cycle is abnormal), the forcible operation is performed. The correction coefficient α is temporarily returned to the initial value. In S32, it is determined whether or not the period monitor clamp control is in the clamp control due to the abnormality of the rich continuation time.

Then, at the time of the clamp control based on the abnormality of the rich time (at the time of the initial setting control), the process proceeds to S33 to perform a subtraction process of the entire region reflection learning value KBLRCB. That is,
When the air-fuel ratio is stuck in a rich state, it is estimated that the air-fuel ratio has not been fully corrected due to the response delay of the air-fuel ratio feedback control and the air-fuel ratio learning and the limitation by the upper and lower limits of the correction value. Based on the rich detection that is the basis of the initial setting control, the entire range reflection learning value KBL
The RCB is reduced, and the correction request in the leaning direction is reflected on the entire region reflection learning value KBLRRCB.

Similarly, if it is determined in S34 that the period monitor clamp control is being executed due to an abnormality in the lean continuation time, the process proceeds to S35, where the all-region reflection learning value KBLRCB is set.
Increase. The whole range reflection learning value KBLRCB reflects a correction request that cannot be absorbed by the maximum and minimum values of the region-specific air-fuel ratio learning correction value KBLRCA during the air-fuel ratio feedback control, as described later. Once the period monitor clamp control for returning α to the initial value is performed, the air-fuel ratio feedback correction coefficient α
Therefore, the desired air-fuel ratio learning cannot be performed because the correction request level cannot be directly known.

However, for example, when the air-fuel ratio sticks to the lean state and the process shifts to the period monitor clamp control (initial setting), there is a request to update the correction level by the learning correction value to at least the rich side. Therefore, in the present embodiment, when the process shifts to the clamp control based on the period abnormality, the entire region reflection learning value KB
The LRCB is updated, and the learning correction request estimated by shifting to the clamp control is reflected in the entire-range reflecting learning value KBLRCB.

Therefore, a sharp and large air-fuel ratio deviation occurs,
Even in the case where the air-fuel ratio sticks to rich or lean, and the control shifts to the period monitor clamp control without learning the correction request level, the learning correction request due to the air-fuel ratio deviation is added to the entire region reflection learning value KBLRCB. It can be reflected, and of course, the whole range reflection learning value KBLRCB
Is applied to the entire range, and when the period monitor clamp control is executed, the update is performed sequentially regardless of the operation range. Therefore, the correction request level for the air-fuel ratio deviation should be reflected early. Is possible.

On the other hand, the normal whole range reflection learning value KBLRCB
Learning control is performed after S36. In S36, a value obtained by subtracting 100%, which is the initial value of each correction value, from the sum of the air-fuel ratio feedback correction coefficient α and the region-specific air-fuel ratio learning correction value KBLRCA corresponding to the current operating condition,
The region-based air-fuel ratio learning correction value KBLRCA is compared with an upper limit value KBLGH # (for example, 110%) that is regulated.

Here, α + KBLRCA-1 is the upper limit K
If BLGH # (eg, 110%) or more, the region-specific air-fuel ratio learning correction value KBLRCA is set to the upper limit KBL.
Even if it is changed to GH #, it indicates that there is an air-fuel ratio error toward the lean side that cannot be absorbed only by the region-specific air-fuel ratio learning correction value KBLRCA. Therefore, α + K
BLRCA-1 is the upper limit KBLGH # (for example, 110%)
If it is equal to or greater than the above, the process proceeds to S37, and the entire region reflection learning value KB
LRCB is added and updated, and the air-fuel ratio error remaining is reflected in the entire range learning value KBLRCB to reduce the air-fuel ratio error.

Incidentally, the whole range reflection learning value KB in S37 is described.
The addition of LRCB is preferably configured to add a predetermined ratio of the difference between (α + KBLRCA-1) and the upper limit KBLGH # (for example, 110%), but the addition may be performed using a fixed value. Similarly, in S38, α + KBLRC
It is determined whether or not A-1 is lower than a lower limit value KBLGL # (for example, 90%). If the lower limit value is lower than the lower limit value KBLGL #, the process proceeds to S39, and the entire region reflection learning value KBLRRCB is performed.
, The portion (lean air-fuel ratio error) remaining as a lean-side correction request is reflected in the entire region reflection learning value KBLRCB.

Α + KBLRCA-1 is the upper limit K
When the air-fuel ratio is within the range between BLGH # and the lower limit KBLGL #, it is considered that the condition is such that the air-fuel ratio learning correction required by the region-based air-fuel ratio learning correction value KBLRCA can be sufficient and sufficient. Then, at this time, in order to avoid that an excessive correction is made by the entire-area reflection learning value KBLRCB, the process proceeds to S40, in which direction the entire-area reflection learning value KBLRRC is learned and updated with respect to the initial value (100%). Is determined.

If the global reflection learning value KBLRRC is learned to be updated to a value equal to or greater than the initial value, the process proceeds to S41, where the global reflection learning value KBLRRC is subtracted so as to approach the initial value, and the update learning is performed to a value lower than the initial value. If yes, the process proceeds to S42, and the addition processing of the entire range reflection learning value KBLRCB is performed so as to approach the initial value. If the deviation of the base air-fuel ratio gradually increases, the above-mentioned steps S36 to S36 are performed.
With the learning control of 42, the air-fuel ratio learning correction value KBL for each area
The average level of the correction request that cannot be completely corrected by the RCA is reflected on the entire range reflection learning value KBLRCB.

Here, the cycle monitor clamp control is
This will be described with reference to the flowchart of FIG. In the flowchart of FIG. 6, first, in S51, the rich air-fuel ratio of the actual air-fuel ratio to the stoichiometric air-fuel ratio is determined based on the output of the oxygen sensor 12.
Determine Lean. Here, when the rich state of the air-fuel ratio is determined, it is determined in S52 whether or not the previous state was the rich state. Then, if the state is continuously rich, it is determined in S53 whether or not the time measured by the timer for measuring the rich or lean continuation time is equal to or longer than a predetermined time (for example, 30 seconds) as described later.

When it is determined that the timer measurement time is equal to or longer than the predetermined time and the lean state is maintained for the predetermined time or longer, the process proceeds to S54, and the cycle monitor clamp control (the reset control of the air-fuel ratio feedback correction coefficient α) is performed. ) And count the number of period monitor clamp controls performed based on the abnormality of the rich continuation time.

In step S55, the timer is cleared so that the duration of the rich or lean state is newly measured. On the other hand, when it is determined in S52 that the previous air-fuel ratio state was lean, it is at the time of the reversal from the lean state to the rich state, so that by clearing the timer in S56, only the duration of the rich or lean state is counted. To be done.

Similarly, when it is determined in S51 that the air-fuel ratio is lean, it is determined in S57 whether or not the vehicle is in a continuous lean state.
It is determined whether or not the continuation time exceeds a predetermined time (S58). If the lean state continues for more than the predetermined time, the cycle monitor clamp control based on the lean cycle abnormality is executed and the control is performed. The number of times of the clamp control due to the lean cycle abnormality is counted (S59). Then, when the monitor cycle clamp control is executed, the timer is cleared and the lean / rich continuation time is counted again (S60).

Next, the failure diagnosis of the fuel supply system using the whole range reflection learning value KBLRCB learned as described above will be described with reference to the flowchart of FIG. In the flowchart of FIG. 7, first, it is determined in S71 whether the diagnosis permission condition is satisfied. The diagnosis permission condition is, for example, a condition that the cooling water temperature Tw is equal to or higher than a predetermined temperature and that the output of the oxygen sensor 12 indicates a normal function state of the sensor.

When the condition for permitting diagnosis is satisfied, S
Proceeding to 72, it is determined whether the entire range reflection learning value KBLRCB is stuck to a lower limit value HLRMIN # (for example, 85%) which is a reference value. Here, the whole range reflection learning value KBLR
When it is determined that CB is stuck to the lower limit value HLRMIN #, it is predicted that there is a deviation in the base air-fuel ratio that cannot be corrected even by the region-specific air-fuel ratio learning correction value KBLRCA and the entire region reflection learning value KBLRCB. In step S73, the number of executions of the periodic monitor clamp control based on the abnormality of the rich
It is determined whether or not M1 # (three times) or more.

Then, the entire range reflection learning value KBLRCB is stuck to the lower limit value HLRMIN #, and the number of executions of the period monitor clamp control based on the abnormality of the rich continuation time is equal to or more than a predetermined number of times CMOIM1 # (three times). In this case, it is determined that a large air-fuel ratio shift to the rich side exceeding the normally predicted variation in the base air-fuel ratio has occurred, and a failure determination signal is output in S74.

The failure determination signal is used as a trigger signal for turning on a failure warning lamp in a vehicle on which the engine is mounted, and provides the driver with any failure (such as a supply system component such as a fuel injection valve or an air flow meter) in the fuel supply system. A warning is issued that combustion at the target air-fuel ratio cannot be performed due to a failure of a control system component or an abnormality such as leakage of intake air.

Similarly, in step S75, the whole range reflection learning value KBLR
Upper limit value HLRMAX # (for example, 131 where CB is a reference value)
%), And when it is determined in S76 that the number of period monitor clamp controls based on the abnormality of the lean period is equal to or greater than the predetermined number CMOIM1 #, a large empty space to the lean side is obtained. It is considered that a failure has occurred in the fuel supply system that causes the fuel ratio deviation, and the process proceeds to S74 to perform a failure determination.

Here, the whole range reflection learning value KBLRCB
Is the correction coefficient α and the region-specific air-fuel ratio learning correction value KBLRC
Even when the period monitor clamp control in which the original correction request cannot be read from A is executed, it is updated irrespective of the operation area, so that a sharp and large air-fuel ratio deviation occurs, and the air-fuel ratio deviation cannot be fully learned. Even when the correction coefficient α is returned to the initial value as it is, the entire region reflection learning value KBLRCB can be updated with a good response in a direction in which the air-fuel ratio deviation is eliminated. Can be diagnosed based on the entire range reflection learning value KBLRCB.

In the failure diagnosis, it is checked whether the frequency of the period monitor clamp control indicates the occurrence of the period abnormality corresponding to the direction of the sticking of the learning value, together with the sticking of the whole region reflection learning value KBLRCB. Therefore, it is possible to avoid an erroneous diagnosis in a case where the sticking is temporarily generated due to a learning failure of the entire region reflection learning value KBLRCB.

[0057]

As described above, according to the present invention,
The air-fuel ratio learning value is also updated when control is performed to return the air-fuel ratio feedback correction value to the initial value once an abnormality occurs in the air-fuel ratio feedback control cycle. Even when the correction request based on the feedback correction value cannot be completely learned by the initial setting control, the air-fuel ratio deviation can be reflected in the air-fuel ratio learning value. This has the effect of improving the diagnostic performance of the failed fuel supply system failure diagnosis.

Further, the reliability of the failure diagnosis based on the air-fuel ratio learning correction value can be improved by determining the number of times of the initial setting control based on the period abnormality together with the air-fuel ratio learning value. Further, when the air-fuel ratio learning value is divided into a region-specific air-fuel ratio learning value that is learned for each of a plurality of operation regions and an entire region reflection learning value that is applied to all operation regions, the entire region air-fuel ratio learning value is set to the above-described value. By updating along with the initial setting control and diagnosing a failure of the fuel supply system based on the whole area reflection learning value, it is possible to frequently learn a correction request predicted according to the initial setting control. As a result, earlier failure diagnosis is possible.

[Brief description of the drawings]

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

FIG. 2 is a schematic diagram showing a system configuration of an embodiment.

FIG. 3 is a flowchart illustrating air-fuel ratio feedback control according to the embodiment.

FIG. 4 is a flowchart showing region-specific air-fuel ratio learning according to the embodiment.

FIG. 5 is a flowchart showing the whole range air-fuel ratio learning of the embodiment.

FIG. 6 is a flowchart illustrating cycle monitor clamp control according to the embodiment.

FIG. 7 is a flowchart illustrating diagnostic control according to the embodiment.

[Explanation of symbols]

 Reference Signs List 1 internal combustion engine 4 fuel injection valve 6 three-way catalyst 7 air flow meter 10 crank angle sensor 12 oxygen sensor 21 control unit

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁷ , DB name) F02D 41/14 310

Claims (3)

(57) [Claims] An air-fuel ratio detecting means for detecting an air-fuel ratio of an engine intake air-fuel mixture, and air for variably controlling an air-fuel ratio feedback correction value so that an air-fuel ratio detected by the air-fuel ratio detecting means approaches a target air-fuel ratio. Fuel-fuel-feedback correction means, air-fuel-ratio learning means for learning a correction request level based on the air-fuel-ratio feedback correction value variably controlled by the air-fuel-ratio feedback correction means as an air-fuel-ratio learning correction value, and air by the air-fuel-ratio feedback correction means. Feedback correction initial setting means for forcibly initializing the air-fuel ratio feedback correction value based on a cycle abnormality of the fuel ratio feedback control; and Initially updating the air-fuel ratio learning correction value based on the detected air-fuel ratio based on Setting-time learning value updating means; fuel supply amount correction means for correcting and controlling the fuel supply amount by fuel supply means based on the air-fuel ratio feedback correction value and the air-fuel ratio learning correction value; and the air-fuel ratio learning correction value and reference Diagnostic means for diagnosing a failure in the fuel supply system based on a comparison with a value and outputting a failure diagnostic signal; and an air-fuel ratio control device for an internal combustion engine. And a diagnostic means for diagnosing a failure in a fuel supply system based on a comparison between the air-fuel ratio learning correction value and a reference value, and determining a predetermined number of initial setting controls by the feedback correction initial setting means. 2. The air-fuel ratio control apparatus for an internal combustion engine according to claim 1, wherein a failure determination is finally performed when the number of times is equal to the number of times. 3. An air-fuel ratio learning correction value learned by said air-fuel ratio learning means is calculated from an air-fuel ratio learning correction value for each of a plurality of operating regions and an entire region reflection learning value applied to all operating regions. The air-fuel ratio learning correction value updated by the initialization-time learning value updating means and the air-fuel ratio learning correction value used by the diagnostic means are the whole-range reflection learning values. An air-fuel ratio control device for an internal combustion engine according to any one of the above.