JP2009138676A - Air-fuel ratio control device for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately a control air-fuel ratio by correcting an output of an air-fuel ratio sensor 2 based on an output of an oxygen sensor 3 at a downstream side of a three-way catalyst device 1 in an air-fuel ratio control device for an internal combustion engine controlling a combustion air-fuel ratio to a desired air-fuel ratio based on the output of the air-fuel ratio sensor 2 at an upstream side of the three-way catalyst device 1. <P>SOLUTION: The output of the air-fuel ratio sensor 2 is corrected by a proportional term based on the deviation of the output of the oxygen sensor 3 while setting a reference on a theoretical air-fuel ratio and an integration term based on an integrated value of the deviation during operation setting the target air-fuel ratio on the theoretical air-fuel ratio. A fuel cut prohibition period during which operation for setting the target air-fuel ratio on the theoretical air-fuel ratio is continued until all of the integration term renewed by the set number of times every renewal timing continuously converge is provided. When the integration term smaller than the set number of times continuously converge and the engine is stopped, the fuel cut prohibition period is removed if all of the integration term renewed by the other set number of times smaller than the set number of times continuously converge from the first renewal timing in a next engine operation. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an air-fuel ratio control apparatus for an internal combustion engine.

It has been proposed that an air-fuel ratio sensor whose output varies linearly according to the air-fuel ratio of exhaust gas is disposed in the engine exhaust system, and the combustion air-fuel ratio is controlled to a desired air-fuel ratio based on the output of this air-fuel ratio sensor. Generally, a three-way catalyst device for purifying exhaust gas is arranged in the engine exhaust system. Since the three-way catalyst device purifies exhaust gas well when the air-fuel ratio of the exhaust gas is close to the stoichiometric air-fuel ratio, in general, the air-fuel ratio of the exhaust gas flowing into the three-way catalyst device is theoretically When the air-fuel ratio is leaner, the excess oxygen is stored, and when the air-fuel ratio of the exhaust gas flowing into the three-way catalyst device is richer than the stoichiometric air-fuel ratio, the insufficient oxygen is released, It has an O ₂ storage capability in which the air-fuel ratio of the exhaust gas is close to the theoretical air-fuel ratio.

Thereby, the air-fuel ratio sensor for detecting the combustion air-fuel ratio is arranged upstream of the three-way catalyst device so that the air-fuel ratio of the detected exhaust gas is not influenced by the O ₂ storage capacity of the three-way catalyst device. Yes. By the way, in order to make the air-fuel ratio control by the air-fuel ratio sensor accurate, an oxygen sensor whose output changes suddenly when the air-fuel ratio of the exhaust gas is close to the stoichiometric air-fuel ratio is disposed downstream of the three-way catalyst device. The deviation of the output of the air-fuel ratio sensor to the rich side or the lean side is corrected based on the output of the oxygen sensor whose change is moderated by the O ₂ storage capability.

  The correction of the output of the air-fuel ratio sensor based on the output of the oxygen sensor is generally performed by the reference output corresponding to the theoretical air-fuel ratio of the oxygen sensor and the actual output of the oxygen sensor in the operation where the target air-fuel ratio is the stoichiometric air-fuel ratio. The proportional term based on the deviation and the integral term based on the integrated value of the deviation are used. The integral term corrects the recent trending deviation of the output of the air-fuel ratio sensor, and the proportional term corrects the current deviation of the output of the air-fuel ratio sensor corrected by the integral term. For example, the integral term is updated so as to match the current state of the air-fuel ratio sensor with the update time as the time when the deviation of the oxygen sensor for the set number of times is calculated.

By the way, in an internal combustion engine, in order to reduce fuel consumption, fuel cut is frequently performed every time the engine is decelerated. When the fuel cut is performed, air flows into the three-way catalyst device as exhaust gas, and a large amount of oxygen in the air is occluded in the three-way catalyst device by the O ₂ storage capability. In order to enable good purification of exhaust gas even when the air-fuel ratio of the exhaust gas becomes leaner or richer than the stoichiometric air-fuel ratio, the three-way catalyst device has a maximum storable oxygen with O ₂ storage capacity It is preferable to store a desired amount of oxygen corresponding to about half of the amount. Thus, in general, immediately after the fuel cut, the combustion air-fuel ratio is made rich in order to release the oxygen stored in excess of the desired amount during the fuel cut so that only the desired amount of oxygen is stored. The enrichment control is performed.

  During fuel cut and enrichment control, the deviation of the output of the oxygen sensor with respect to the theoretical air-fuel ratio has a large absolute value as a matter of course and becomes a meaningless value. Air-fuel ratio control during fuel cut is unnecessary, and naturally, updating of the integral term that integrates the deviation of the output of the oxygen sensor is also prohibited. Further, during the enrichment control, it is necessary to perform an air-fuel ratio control in which the combustion air-fuel ratio is the desired rich air-fuel ratio. In this air-fuel ratio control, the air-fuel ratio sensor is controlled by a proportional term based on the deviation of the output of the oxygen sensor. The correction of the output is prohibited, and the update of the integral term that integrates the deviation of the output of the oxygen sensor is also prohibited.

  However, if the integral term is not updated during the fuel cut and the enrichment control immediately thereafter as described above, if the fuel cut is frequently performed from the start of the engine, the integral term is not updated at all in the current engine operation. In some cases, the output of the air-fuel ratio sensor cannot be corrected by the integral term suitable for the current state during the enrichment control and the subsequent operation of the theoretical air-fuel ratio, and accurate air-fuel ratio control becomes difficult.

  In order to solve this problem, it has been proposed to prohibit fuel cut until the integral term is updated several times and the integral term is almost converged (see Patent Document 1).

JP2007-198246 JP 2006-104978 A JP-A-9-68077 JP 61-190137 A

  In the air-fuel ratio control apparatus for an internal combustion engine described in Patent Document 1, if the deviation between the newly calculated integral term and the current integral term is less than the set value, the integral term is converged and learning of the integral term is performed. Once completed, the fuel cut prohibition is lifted. However, this deviation may be reduced by chance even though the integral term is not sufficiently converged, and once the integral term learning is completed just because this deviation is less than the set value, In some cases, an accurate integral term is not learned. At this time, accurate air-fuel ratio control cannot be performed.

  Accordingly, an object of the present invention is to provide an air-fuel ratio control apparatus for an internal combustion engine that controls the combustion air-fuel ratio to a desired air-fuel ratio based on the output of an air-fuel ratio sensor upstream of the three-way catalyst apparatus. Correcting the output of the air-fuel ratio sensor based on the output of the oxygen sensor enables accurate air-fuel ratio control.

  An air-fuel ratio control apparatus for an internal combustion engine according to claim 1 of the present invention is provided on the upstream side of a three-way catalyst device of an engine exhaust system, and an air-fuel ratio sensor whose output changes in accordance with the air-fuel ratio of exhaust gas, An oxygen sensor disposed downstream of the three-way catalyst device of the engine exhaust system and having an output that changes abruptly when the air-fuel ratio of the exhaust gas is close to the stoichiometric air-fuel ratio, and based on the output of the air-fuel ratio sensor In the air-fuel ratio control apparatus for an internal combustion engine that controls the combustion air-fuel ratio to a desired air-fuel ratio, the output of the air-fuel ratio sensor is the output of the oxygen sensor based on the stoichiometric air-fuel ratio during operation with the target air-fuel ratio as the stoichiometric air-fuel ratio. Corrected by a proportional term based on output deviation and an integral term based on the integrated value of the deviation, the integral term is updated at every update time, and all the integral terms updated a set number of times at each update time are continuous. Until the target converges When a fuel cut prohibition period for continuing operation with the fuel ratio being the stoichiometric air-fuel ratio is provided, and when the integral term less than the set number of times is continuously converged and the engine is stopped, the first renewal timing in the next engine operation The fuel cut prohibition period is canceled if all the integral terms that have been updated another set number less than the set number have converged continuously.

  An air-fuel ratio control apparatus for an internal combustion engine according to a second aspect of the present invention is the air-fuel ratio control apparatus for an internal combustion engine according to the first aspect, for calculating the integral term set at the beginning of the fuel cut prohibition period. The gain is changed at least once during the fuel cut prohibition period.

  According to the air-fuel ratio control apparatus for an internal combustion engine according to claim 1 of the present invention, the fuel cut prohibition period is provided, and during this period, the operation with the target air-fuel ratio as the stoichiometric air-fuel ratio is continuously performed, Since the integral term that corrects the tendency deviation of the output of the air-fuel ratio sensor is calculated and updated based on the integrated value of the deviation of the output of the oxygen sensor with respect to the theoretical air-fuel ratio, fuel cuts are frequently performed. Therefore, the output of the air-fuel ratio sensor can be corrected by an integral term suitable for the current situation, and the combustion air-fuel ratio is changed to the stoichiometric air-fuel ratio or in the enrichment control immediately after the fuel cut. It is possible to satisfactorily control the rich air-fuel ratio.

  In addition, since the fuel cut prohibition period continues until all the integral terms updated a set number of times at each update period converge continuously, the fuel cut prohibition period is canceled when the integral term converges only once. Thus, the learning of the integral term is not completed, and it is updated to an accurate integral term for correcting the gradual deviation of the air-fuel ratio sensor output, so that favorable air-fuel ratio control can be realized.

  Furthermore, when the integral term of less than the set number of times is continuously converged and the engine is stopped, in the next engine operation, if the fuel cut prohibition period is continued until the newly set number of integral terms continuously converges, it is short. When the engine stop is repeated over a period, fuel cut may not be performed at all and fuel consumption may deteriorate. However, according to this air-fuel ratio control apparatus, when the integral term less than the set number of times converges continuously and the engine is stopped, the integral term has converged to some extent. The fuel cut prohibition period is canceled if all the integral terms that have been updated for another set number of times less than the set number of times have converged continuously, so that the integral term is reduced to an accurate value. It can be renewed and fuel cut is easily performed.

  According to the air-fuel ratio control apparatus for an internal combustion engine according to claim 2 of the present invention, in the air-fuel ratio control apparatus for the internal combustion engine according to claim 1, the integral term set at the beginning of the fuel cut prohibition period is calculated. The gain is changed at least once during the fuel cut prohibition period, so that a large gain is used at the beginning of the fuel cut prohibition period and the integral term is heavily weighted. The air-fuel ratio is quickly converged to the stoichiometric air-fuel ratio by the air-fuel ratio control based on the output of the air-fuel ratio sensor corrected by the term, that is, the integral term is also converged early. Can be released.

  FIG. 1 is a schematic diagram showing an engine exhaust system, in which 1 is a three-way catalyst device, 2 is an air-fuel ratio sensor disposed upstream of the three-way catalyst device 1, and 3 is a three-way catalyst device 1. It is an oxygen sensor arranged on the downstream side. The air-fuel ratio sensor 2 and the oxygen sensor 3 both have an output voltage that changes according to the oxygen concentration in the exhaust gas, and the air-fuel ratio sensor 2 changes its output linearly corresponding to the air-fuel ratio of the exhaust gas. It is a linear output type, and the oxygen sensor 3 is a step output type in which the output changes abruptly when the air-fuel ratio of the exhaust gas is near the stoichiometric air-fuel ratio.

  The air-fuel ratio control apparatus according to the present invention feedback-controls the combustion air-fuel ratio to a desired air-fuel ratio based on the output of the air-fuel ratio sensor 2. Since the air-fuel ratio sensor 2 is arranged upstream of the three-way catalyst device 1 and is always exposed to unpurified exhaust gas, the output reliability is not so high, and the output goes to the rich side or the lean side. It may shift. Thereby, it is arrange | positioned in the downstream of the three-way catalyst apparatus 1, and generally detects whether the air-fuel ratio of exhaust gas is rich or lean, without being exposed to unpurified exhaust gas. The output of the air-fuel ratio sensor is corrected based on the output of the oxygen sensor 3 that is used for this purpose and has high output reliability.

The output V of the air-fuel ratio sensor 2 used for the feedback control of the combustion air-fuel ratio is corrected as described below to an output V ′ during operation in which the target combustion air-fuel ratio is the stoichiometric air-fuel ratio.
V ′ = V + P + I
Here, P is a proportional term obtained by multiplying a deviation d between a reference output (for example, 0.5 volts) with respect to the theoretical air-fuel ratio and an actual output of the oxygen sensor 3 by a predetermined gain Pg, and I is a deviation d of the set number of times. Is an integral term obtained by multiplying the integrated value by a predetermined gain Ig. In this way, the integral term I corrects the current tendency shift of the output of the air-fuel ratio sensor 2, and the proportional term P corrects the current shift of the output of the air-fuel ratio sensor 2 corrected by the integral term I. To be. Based on the output V ′ of the air / fuel ratio sensor 2 corrected in this way, the air / fuel ratio of the exhaust gas is accurately estimated, and the intake air amount detected by the air flow meter is adjusted so that the combustion air / fuel ratio becomes the stoichiometric air / fuel ratio. Thus, the fuel injection amount is feedback corrected.

The present air-fuel ratio control apparatus controls the combustion air-fuel ratio of the internal combustion engine to the stoichiometric air-fuel ratio (stoichiometric). However, as shown in the time chart of FIG. 2, in order to reduce fuel consumption during engine deceleration, etc. When the fuel cut F / C is performed, of course, the fuel injection is stopped, so that the air-fuel ratio control is not performed. During the fuel cut F / C, air containing a large amount of oxygen flows into the three-way catalyst device 1 as exhaust gas, and therefore the three-way catalyst device 1 has about the maximum storable oxygen amount of O ₂ storage capacity. More than half of the oxygen is occluded (if the fuel cut F / C time is long, the oxygen is occluded to the maximum storable oxygen amount). Thus, as it is, when the combustion air-fuel ratio becomes leaner than the stoichiometric air-fuel ratio after the fuel cut, the three-way catalyst device 1 cannot store excess oxygen well, and the NO _x purification capability is improved. It will decline.

Therefore, immediately after the fuel cut F / C, the combustion air-fuel ratio is controlled to the desired rich air-fuel ratio, and the oxygen stored in the three-way catalyst device 1 during the fuel cut is released to about half of the maximum storable oxygen amount, Thereafter, even if the combustion air-fuel ratio becomes lean or rich, the air-fuel ratio of the exhaust gas in the three-way catalyst device 1 can be made close to the theoretical air-fuel ratio, and the NO _x purification capacity, CO and It is preferable to maintain high HC purification capacity. When the enrichment control immediately after the fuel cut F / C is completed, the stoichiometric control with the target combustion air-fuel ratio as the stoichiometric air-fuel ratio is started.

  The deviation d between the reference output of the oxygen sensor 3 with respect to the stoichiometric air-fuel ratio and the actual output of the oxygen sensor 3 cannot be calculated during the fuel cut and during the enrichment control immediately thereafter. In the control, the output of the air-fuel ratio sensor 2 cannot be corrected by the proportional term P. Further, during the enrichment control, it is preferable to control the combustion air-fuel ratio to the desired rich air-fuel ratio by correcting the output of the air-fuel ratio sensor 2 with the integral term I. However, since the deviation d is not calculated, the integral term I Cannot be updated.

  Thus, if fuel cut is frequently performed from the start of the engine, the integral term I cannot be updated, and in the enrichment control and stoichiometric control, the integral term I that matches the current state is used for output correction of the air-fuel ratio sensor 2. Can not do it. The air-fuel ratio control apparatus calculates an integral term I that matches the current state of the air-fuel ratio sensor 2 according to the first flow charts shown in FIGS. 3 to 5, and improves the output of the air-fuel ratio sensor 2 in the enrichment control and the stoichiometric control. The combustion air-fuel ratio can be controlled to the desired air-fuel ratio.

  First, in step 101, it is determined whether or not the flag F is 1. The flag F is set to 1 to indicate the completion of learning of the integral term I, is set to 0 when the vehicle is new, and is stored in the backup RAM when the engine is stopped. Initially, the determination at step 101 is denied and the routine proceeds to step 102 where fuel cut is prohibited. Next, at step 104, it is determined whether a condition for calculating a deviation d between the reference output with respect to the stoichiometric air-fuel ratio and the current output of the oxygen sensor 3 is satisfied.

  When the fuel cut is prohibited, the determination in step 104 is affirmed except during engine warm-up including the oxygen sensor 3 immediately after the start, etc., the deviation d is calculated, and the routine proceeds to step 105. Of course, if the deviation d is calculated, the output of the air-fuel ratio sensor 2 is corrected using the proportional term P calculated based on the deviation d and the integral term I at the previous engine stop as the integral term. Then, perform stoichiometric control while fuel cut is prohibited.

  In step 105, the set number of deviations d for calculating the integral term is calculated, and it is determined whether or not it is time to update the integral term I based on the integrated value of these deviations d. When this determination is denied, the process is terminated as it is, but when it is time to update the integral term I, the routine proceeds to step 106. In step 106, it is determined again whether the flag F is 1. Initially, this determination is denied, and in step 107, the integral term is updated to the newly calculated integral term I. If the integral term I is updated in this way, thereafter, the output of the air-fuel ratio sensor 2 is corrected using the newly calculated proportional term P and the updated integral term I, and fuel cut is prohibited. Implement stoichiometric control.

  Next, in step 108, the first count value C1 is increased by 1, and in step 109, the second count value C2 is increased by 1. The first count value C1 and the second count value C2 are both 0 when a new vehicle is used.

Next, at step 110, the integrated value S of the difference between the current integral term I _i and the previous integral term I _i-1 is calculated. This integrated value S is also set to 0 when a new vehicle is used. Next, at step 111, it is determined whether or not the absolute value of the integrated value S is less than or equal to the set value b. Initially, since the change in the integral term I between the previous time and the current time is large, the integrated value S also increases, the determination in step 111 is denied, and the integrated value S is reset to 0 in step 114.

  When the change in the integral term I becomes small and the determination in step 111 is affirmed, the third count value C3 is increased by 1 in step 112, and the fourth count value C4 is increased by 1 in step 113. The third count value C3 and the fourth count value C4 are set to 0 when the vehicle is new, and are reset to 0 in steps 115 and 116 when the determination in step 111 is negative.

  In step 117, it is determined whether or not the first count value C1 is equal to or greater than the first set number N1, and in step 118, it is determined whether or not the second count value C2 is equal to or greater than the second set number N2. In step 119, it is determined whether or not the third count value C3 is greater than or equal to the third set number of times N3. In step 120, it is determined whether or not the fourth count value C4 is greater than or equal to the fourth set number of times N4. .

  If any of the determinations in steps 117 to 120 is denied, it is determined in step 122 whether or not the engine is to be stopped. If this determination is denied, the process ends as it is, but the determination in step 122 is affirmed. In step 123, the first count value C1 is stored in the backup RAM as the current C1, in step 124, the second count value C2 is reset to 0, and in step 125, the third count value C3 is set to the current C3. And stored in the backup RAM. In step 126, the fourth count value C4 is reset to zero.

  As described above, the first count value C1 indicates the number of integration updates of the integral term I regardless of whether the engine has stopped halfway. When the first set number N1 is 40 (for example, 40), the determination at step 117 is affirmative. Is done. On the other hand, the second count value C2 represents the number of updates of the integral term I in the current operation, and when the second set number N2 (for example, 20) is smaller than the first set number N1, the determination in step 118 is Affirmed.

Further, the third count value C3 is an integration in which the integrated value S of the difference between the current integral term I _i and the previous integral term I _i-1 is smaller than the set value b, regardless of whether the engine is stopped halfway. When the third set number of times N3 (for example, 10 times) is reached, the determination in step 119 is affirmed. On the other hand, the fourth count value C4 represents the number of consecutive times that the integrated value S of the difference between the current integral term I _i and the previous integral term I _i-1 in the current operation is smaller than the set value b. When the fourth set number N4 (for example, 5 times) is smaller than the third set number N3, the determination in step 120 is affirmed.

  Steps 117 to 120 are four conditions for determining that the learning of the integral value I has been completed. If all the determinations in steps 117 to 120 are affirmed, the flag F is set to 1 in step 121. As a result, the determination at step 101 is affirmed, and the prohibition of fuel cut is released at step 103. Thereby, fuel cut is performed at the time of engine deceleration or the like, and immediately after that, the above-described enrichment control is performed. In such a case, in step 104, during the fuel cut, during the enrichment control, and immediately after the completion of the enrichment control, the exhaust gas in which the target air-fuel ratio is made the stoichiometric air-fuel ratio by the stoichiometric control from the three-way catalyst device 1. As long as the flow does not flow out, the condition for calculating the deviation d is not satisfied.

When a new integral term I is calculated in step 105, the determination in step 106 is affirmed. Therefore, in step 127, the newly calculated integral term I _i and the current integral term I _i-1 are calculated. It is determined whether or not the absolute value of the difference is greater than the set value a. Only when this determination is affirmative, the integral term is updated to the newly calculated integral term I in step 128.

In other words, the current integral term I _i-1 is determined to be sufficiently converged according to the current state of the air-fuel ratio sensor 2 by affirming all the judgments in steps 117 to 120, and this integral term If the absolute value of the difference of the newly calculated integral term I _i with respect to I _i-1 is large, the current value of the air-fuel ratio sensor 2 is changed and the integral term is updated. However, the absolute value of the difference is small. At some time, assuming that the current state of the air-fuel ratio sensor 2 has not changed, the integral term is not updated and the air-fuel ratio control is stabilized.

  Thus, according to the first flowchart, even if the deviation between the current integral term and the previous integral term falls below the set value by chance, the learning is not completed as the integral term has converged. Until all the above four conditions are affirmed and updated to an accurate integral term, learning of the integral term is continued.

  The first condition of step 117 is that the integral term cumulative update count exceeds the first set count N1 regardless of whether the engine is stopped halfway. Thus, at least for accurate integral term learning, at least It is assumed that the integral term of the first set number N1 needs to be updated. However, regarding the number of updates, not only the first condition but also the second condition of step 118 requires that the number of updates of the integral term in the current operation exceed the second set number N2 which is smaller than the first set number N1. Has been.

  In order to be updated to an accurate integral term, it is necessary to update the integral value I at least by the first set number N1. However, if the integral term has to be updated by the first set number of times N1 in one operation, when the operation is repeated for a short time, the integral value learning is not completed and the fuel cut is prohibited. It cannot be released and fuel consumption will deteriorate. As a result, even if the engine is stopped, the first count value C1 of the integral term is stored, and in the next operation, the count is incremented every time the integral term I is updated from the stored first count value C1. Even if the number of updates in is less than the first set number of times N1, if the first count value C1 reaches the first set number of times N1, the determination in step 117 is affirmed so that the fuel cut prohibition can be easily cancelled. I have to.

  However, in this operation, even if the first count value C1 reaches the first set number N1 due to several updates of the integral value I, the current operation for determining completion of learning of the integral term I is now integrated. Since the value is not sufficiently reflected in the update of the value, unless the second count value C2 corresponding to the number of updates of the integral value I in the current operation reaches the second set number N2, the determination in step 118 is It is not affirmed. In other words, the second condition for determining the completion of learning of the integral value I is the number of times that the integral term must be updated at least in one operation so that the current operation is sufficiently reflected in the update of the integral term. It has established. Even when such a second condition is added, it is easier to cancel the fuel cut prohibition than in the case where the integral term has to be updated by the first set number N1 in one operation.

  In the first flowchart, the third and fourth conditions of steps 119 and 120 are set to determine the completion of learning of the integral term. However, the learning of the integral term is performed only by the aforementioned first and second conditions. The fuel cut prohibition may be canceled after judging completion. Further, the completion of learning of the integral term may be determined only by the first condition.

The third condition for determining completion of learning of the integral term in step 119 is that the current integral term I when the accumulated value S of the difference between the current integral term I _i and the previous integral term I _i-1 is smaller than the set value b. Regardless of whether _i has converged and the engine stopped halfway, the integral term I updated at each update time has continuously converged beyond the third set number N3. Thus, for accurate learning of the integral term, the updated integral term needs to converge continuously at least for the third set number of times N3. However, regarding the number of continuous convergence times, not only the third condition but also the fourth condition of step 120, the number of continuous convergence times of the integral term in the current operation exceeds the fourth set number N4 which is smaller than the third set number N3. Is required. In this way, the integrated value S of the difference between the current integral term and the previous integral term is used to determine whether or not the current integral term I has converged. Even when I increases or decreases, the absolute value of the integrated value S exceeds the set value b, and it is determined that the integral term has not converged.

  In order to be updated to an accurate integral term, it is necessary that the updated integral value I continuously converges at least by the third set number N3. However, if the integral term I must converge continuously for the third set number N3 in one operation, when the operation is repeated for a short period of time, the learning of the integral value is not completed forever. The prohibition of cut cannot be lifted, and the fuel consumption will deteriorate. Thereby, even if the engine is stopped, the third count value C3 of the integral term is stored, and in the next operation, it is counted every time the integral term I updated from the stored third count value C3 continuously converges. Even if the number of times of continuous convergence in this operation is less than the third set number N3, if the third count value C3 reaches the third set number N3, the determination in step 119 is affirmed. It makes it easier to cancel the fuel cut prohibition.

  However, in this operation, even if the integrated value I updated several times continuously converges and the third count value C3 reaches the third set number N3, this completes learning of the integral term I. Since the current operation to be determined is not sufficiently reflected in the update of the integral value, the fourth count value C4 corresponding to the continuous convergence number of the integral value I in the current operation reaches the fourth set number N4. Unless otherwise, the determination in step 120 is not affirmed. That is, the fourth condition for determining the completion of learning of the integral value I is that the integral term must converge at least continuously in one operation so that the current operation is sufficiently reflected in the update of the integral term. The number of updates is set.

  In the first flowchart, in order to determine the completion of the integral term learning, the first and second conditions described above are omitted, and the completion of the integral term learning is determined only by the third and fourth conditions in steps 119 and 120. The fuel cut prohibition may be canceled. Further, the completion of learning of the integral term may be determined only by the third condition.

  FIG. 6 is a second flowchart for accelerating the convergence of the integral term and accelerating the fuel cut prohibition release, and is performed simultaneously with the first flowchart. First, in step 201, it is determined whether or not the flag F in the first flowchart is 1. When the learning of the integral term I is not completed, this determination is denied and the routine proceeds to step 202.

  In step 202, it is determined whether or not the first count value C1 of the first flowchart, that is, whether or not the integral term integration update count has reached a set count N (for example, 20). At the beginning of learning of the integral term, this determination is denied, and in step 203, the gain Ig for calculating the integral term is set to a relatively large Ig1. Thereby, at the beginning of learning of the integral term, the large gain Ig1 is used and the integral term I is heavily weighted.

  On the other hand, when the first count value C1 reaches the set number N and the determination in step 202 is affirmed, in step 204, the gain Ig for calculating the integral term is set to a relatively small Ig2. Thereby, when the learning of the integral term approaches, the small gain Ig2 is used to reduce the weight of the integral term I. Further, when the flag F is set to 1 and learning of the integral term is completed, the gain is set to a gain Ig3 smaller than Ig2 in step 205.

  Thus, as shown in FIG. 9, the fuel cut prohibition period for learning the integral term starts at time t0. When the integral update count C1 of the integral term reaches the set count N at time t1, the gain is reduced. Thus, it is switched from Ig1 to Ig2. Thereby, in the air-fuel ratio control based on the output of the air-fuel ratio sensor corrected by the integral term, before time t1, the air-fuel ratio is largely corrected by the heavily weighted integral term, and after time t1, the air-fuel ratio is weighted small. Therefore, the integral term can be converged to the theoretical air-fuel ratio at time t2 without diverging the air-fuel ratio, that is, the integral term can be reliably converged early at time t2. it can.

  As a result, when a constant gain (for example, Ig2) is used in the integral value learning period (indicated by a dotted line), the integral term does not converge until time t3. However, the integral term I is accelerated by the control of the second flowchart. It is possible to reliably converge, and to cancel the fuel cut prohibition at an early stage and to suppress the deterioration of fuel consumption. If the flag F is set to 1 and learning of the integral term is completed, the gain is set to Ig3 smaller than Ig2, and the weight of the integral term is further reduced to suppress the fluctuation of the air-fuel ratio in the air-fuel ratio control. It has become.

  FIG. 7 is a third flowchart for accelerating the convergence of the integral term and accelerating the release of the fuel cut prohibition, and only the difference from the second flowchart will be described below.

  In this flowchart, in step 302, it is determined whether or not the gain Ig of the integral term I is relatively small Ig2. Initially, this determination is denied and the routine proceeds to step 303. In step 303, it is determined whether or not the absolute value E of the difference in the integral term I between the current time and the previous time is smaller than the set value e, that is, whether or not the integral term I has converged to some extent. At the beginning of learning of the integral term, this determination is denied, and in step 304, the gain Ig for calculating the integral term is set to a relatively large Ig1. If the integral term I converges to some extent, the determination in step 303 is affirmed, and in step 305, the gain Ig for calculating the integral term is set to a relatively small Ig2. When the gain is set to Ig2, the determination in step 302 is affirmed, and therefore the gain is not returned to Ig1.

Thus, as shown in FIG. 9, the initial fuel cut prohibition period of learning the integral term, the absolute value E ₁ of the difference between the integral term I ₁ of the current integral term I ₂ and the last time or set value e However, at time t1 ′, when the absolute value E ₂ of the difference between the current integral term I ₃ and the previous integral term I ₂ becomes smaller than the set value e, the gain is reduced from Ig1 to Ig2 Is switched to. Thereby, in the air-fuel ratio control based on the output of the air-fuel ratio sensor corrected by the integral term, before the time t1 ′, the air-fuel ratio is largely corrected by the heavily weighted integral term, and after the time t1 ′, the air-fuel ratio is Since it is corrected to be small by the small weighted integral term, it can be converged to the theoretical air-fuel ratio at time t2 without diverging the air-fuel ratio, that is, the integral term is surely converged early at time t2. be able to.

  FIG. 8 is a fourth flowchart for accelerating the convergence of the integral term and accelerating the release of the fuel cut prohibition. Only the difference from the third flowchart will be described below.

  In this flowchart, in step 403, an integrated value ΣP · Ga obtained by integrating the product of the proportional term P and the intake air amount Ga at the same time is set by the integration of the product of the current proportional term P and the current intake air amount Ga. It is determined whether or not the calculation has been reached after reaching the set number of times. This integrated value corresponds to the degree of convergence of the air-fuel ratio as a result of the air-fuel ratio control based on the output of the air-fuel ratio sensor corrected by the proportional term P and the integral term I. When the set number of times is not reached and the integrated value ΣP · Ga is not calculated, the determination at step 403 is denied and the process ends as it is. However, when the determination at step 403 is affirmed, the absolute value of the integrated value is determined at step 404. Is less than the set value g.

  When the determination in step 404 is negative, the air-fuel ratio has not converged so much to the stoichiometric air-fuel ratio. In step 405, the integrated value ΣP · Ga is reset to 0, and in step 406, the integral term is calculated. The gain Ig is a relatively large Ig1. When the air-fuel ratio converges to the stoichiometric air-fuel ratio to some extent, the determination in step 404 is affirmed, and in step 407, the gain Ig for calculating the integral term is set to a relatively small Ig2. Thus, similarly to the third flowchart, the air-fuel ratio can be converged to the theoretical air-fuel ratio early, and the integral term can be converged early.

It is the schematic which shows the exhaust system of the internal combustion engine which the air-fuel ratio control apparatus by this invention controls. It is a time chart which shows the target air fuel ratio of the air fuel ratio control apparatus by this invention. It is a part of 1st flowchart implemented by the air fuel ratio control apparatus by this invention. It is a part of 1st flowchart of FIG. It is a part of 1st flowchart of FIG. It is a 2nd flowchart for making an integral term converge early. It is a 3rd flowchart for making an integral term converge early. It is a 4th flowchart for making an integral term converge early. It is a time chart which shows the change of the integral term by control of a 2nd and 3rd flowchart.

Explanation of symbols

1 Three-way catalyst device 2 Air-fuel ratio sensor 3 Oxygen sensor

Claims (2)

  An air-fuel ratio sensor disposed upstream of the engine exhaust system three-way catalyst device and having an output varying in accordance with the air-fuel ratio of the exhaust gas, and an exhaust gas disposed downstream of the engine exhaust system three-way catalyst device An air-fuel ratio control apparatus for an internal combustion engine, comprising: an oxygen sensor whose output changes abruptly when the air-fuel ratio of gas is close to the theoretical air-fuel ratio, and controlling the combustion air-fuel ratio to a desired air-fuel ratio based on the output of the air-fuel ratio sensor The output of the air-fuel ratio sensor includes a proportional term based on a deviation of the output of the oxygen sensor with respect to the theoretical air-fuel ratio and an integral term based on the integrated value of the deviation during operation with the target air-fuel ratio as the theoretical air-fuel ratio. The integral term is updated at each update time, and the operation with the target air-fuel ratio as the stoichiometric air-fuel ratio is continued until all the integral terms updated a set number of times at each update time converge continuously. No fuel cut When a period is provided and the integral term less than the set number of times continuously converges and the engine is stopped, in the next engine operation, another set number of times smaller than the set number of times is updated from the first update time. The air-fuel ratio control apparatus for an internal combustion engine, wherein the fuel cut prohibition period is canceled if all the integral terms converge continuously.   2. The internal combustion engine according to claim 1, wherein a gain for calculating the integral term set at the beginning of the fuel cut prohibition period is changed to be small at least once during the fuel cut prohibition period. Air-fuel ratio control device.

Images