JP3777925B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
According to the present invention, the engine air-fuel ratio is feedback-controlled based on the air-fuel ratio correction coefficient so as to coincide with the stoichiometric air-fuel ratio, while the steady state between the engine air-fuel ratio and the stoichiometric air-fuel ratio is determined based on the behavior of the air-fuel ratio correction coefficient. The present invention relates to an air-fuel ratio control apparatus for an internal combustion engine that learns an air-fuel ratio learning value for compensating for misalignment and reflects the air-fuel ratio learning value in the feedback control.
[0002]
[Prior art]
In the internal combustion engine, so-called air-fuel ratio control, in which feedback control of the engine air-fuel ratio to the stoichiometric air-fuel ratio is performed as a well-known countermeasure against exhaust purification. In this air-fuel ratio control, generally, a deviation between the engine air-fuel ratio and the stoichiometric air-fuel ratio is detected from the signal of the oxygen sensor, and the transient deviation of the engine air-fuel ratio is compensated based on the detection result. Therefore, a feedback correction coefficient (hereinafter referred to as “air-fuel ratio correction coefficient”) is calculated.
[0003]
In such air-fuel ratio control, the difference between the engine air-fuel ratio and the stoichiometric air-fuel ratio is caused by individual engine differences, individual characteristics in the fuel injection valve injection characteristics or sensor output characteristics for detecting the intake air amount, and the like. A steady deviation may occur. Therefore, a correction value for compensating for this deviation is learned based on the behavior of the air-fuel ratio correction coefficient, and the learned correction value (hereinafter referred to as “air-fuel ratio learning value”) is combined with the air-fuel ratio correction coefficient. Therefore, it is applied to feedback control of engine air-fuel ratio. As described above, the steady deviation of the engine air-fuel ratio is compensated by the learned value of the air-fuel ratio, so that the transient deviation of the engine air-fuel ratio caused by the acceleration / deceleration of the internal combustion engine or the like is preferably corrected by the air-fuel ratio correction coefficient. As a result, it is possible to compensate, and as a result, the accuracy of air-fuel ratio control can be improved.
[0004]
By the way, the engine air / fuel ratio shift tendency as described above varies depending on the engine operating state (particularly the engine load state). For this reason, the learning related to the engine air-fuel ratio as described above is performed for each of a plurality of learning regions divided according to the engine operating state, and the air-fuel ratio learning value is obtained separately as a value for each of these learning regions. Yes.
[0005]
On the other hand, in an internal combustion engine, the evaporated fuel generated in the fuel tank is temporarily stored in the canister, and the stored evaporated fuel is introduced into the engine intake system at a predetermined time and then burned in the engine combustion chamber. Control for so-called purge control is performed.
[0006]
If such purge control and learning on the engine air-fuel ratio are performed at the same time, the air-fuel ratio learning value will fluctuate depending on the concentration and amount of evaporated fuel introduced from the canister into the engine intake system. The correction value for compensating for the steady deviation of the air-fuel ratio is no longer appropriate.
[0007]
For this reason, in an internal combustion engine that executes such purge control and learning of the engine air-fuel ratio, learning of the engine air-fuel ratio is completed, and the learned value of the air-fuel ratio can compensate for a steady deviation of the engine air-fuel ratio. Each of these processes is executed at different times, for example, purge control is started after being set.
[0008]
Here, considering that there is a limit in the fuel storage capacity in the canister, it is desirable to complete the learning of the engine air-fuel ratio at an early stage and start the purge control at an earlier stage. However, simply completing the learning of the engine air-fuel ratio can ensure the fuel storage capacity of the canister, but the accuracy of the air-fuel ratio learning cannot be avoided, and the accuracy of the air-fuel ratio control is naturally low. .
[0009]
Therefore, conventionally, as seen in Japanese Patent Application Laid-Open No. 7-253040, it is determined that learning has been completed during a previous engine operation in a certain learning region, and in the learning region even during the current engine operation. Similarly, when it is determined that learning has been completed, all learning in other learning areas where learning has been completed during the previous engine operation is omitted, and purge control is immediately started in each of these learning areas. I am doing so.
[0010]
[Problems to be solved by the invention]
As described above, when learning is continuously completed in a certain learning region, the learning start time of the purge control can be certainly advanced by omitting learning in other learning regions, and the execution frequency is increased. Will be able to.
[0011]
However, on the other hand, in the conventional apparatus in which learning is appropriately omitted as described above, for example, the injection characteristic of the fuel injection valve changes over time only in a specific learning region, and accordingly, Even if the engine air-fuel ratio may shift, if learning is continuously completed in another learning region, learning in that specific learning region is not performed, and thus the air-fuel ratio learning value is updated. Nor. For this reason, if learning is omitted for a long time in the specific learning region, the learning result in the learning region, that is, the reliability of the air-fuel ratio learning value is lowered, and the accuracy of the air-fuel ratio control is lowered. It will be.
[0012]
Such inconveniences are not limited to changes over time in the injection characteristics of the fuel injection valve, but due to changes in the output characteristics of various sensors such as individual engine differences and intake air amount sensors, etc. The same can occur when the learning area changes differently over time.
[0013]
The present invention has been made in view of these conventional situations, and the purpose thereof is to suitably compensate for the deviation even when the deviation of the engine air-fuel ratio changes with time in each learning region. It is another object of the present invention to provide an air-fuel ratio control apparatus for an internal combustion engine that can ensure a higher execution frequency of the purge control even when the purge control is performed after the learning of the engine air-fuel ratio is completed. .
[0014]
[Means for Solving the Problems]
The means for achieving the above object and the effects thereof will be described below.
In the first aspect of the present invention, control means for feedback-controlling the engine air-fuel ratio so as to coincide with the stoichiometric air-fuel ratio based on the air-fuel ratio correction coefficient, and the engine air-fuel ratio based on the behavior of the air-fuel ratio correction coefficient. An air-fuel ratio learning value for compensating for a steady deviation from the air-fuel ratio is learned for each of a plurality of learning regions divided according to the engine operating state, and the same air-fuel ratio learning value is reflected in the feedback control. In the air-fuel ratio control apparatus for an internal combustion engine, comprising: learning means; and determination means for determining each learning region when learning by the learning means is completed when a predetermined learning completion condition is satisfied after engine startup. Setting means for setting a learning completion history for each learning region as having a history when the learning completion determination is made at least once by the determination means; and the setting Learning completion history in a certain learning area is so and a relaxing means to relieve the learning completion conditions when determining the learning completion in the learning region when it is set that there history by stages.
[0015]
In the above configuration, the air-fuel ratio learning value is learned for each learning region regardless of whether there is a learning completion history, and the learned value is reflected in the feedback control of the engine air-fuel ratio. Is compensated for each learning region. In addition, in this learning, the learning completion history is set as having a history, and the learning area that does not require a large update of the air-fuel ratio learning value is completed compared to the learning area that is set as having no history. Since the conditions are relaxed, learning is completed at an earlier stage. For this reason, when the purge control is performed after the learning is completed, the control is started early.
[0016]
Therefore, according to the above-described configuration of the invention described in claim 1, even if the engine air-fuel ratio shift varies with time in each learning region, the shift can be suitably compensated. In addition, even when the purge control is performed after the learning of the engine air-fuel ratio is completed, it is possible to secure a higher execution frequency of the purge control.
[0017]
Further, as a specific aspect of the configuration in which the learning completion condition is relaxed as described above, according to the invention described in claim 2, in the air-fuel ratio control apparatus for an internal combustion engine according to claim 1, The learning means updates the air-fuel ratio learning value so that a deviation between an average value of the air-fuel ratio correction coefficient and a reference value of the air-fuel ratio correction coefficient corresponding to the theoretical air-fuel ratio decreases, and the determination means Is that the learning completion condition is that the deviation exists within a predetermined determination range, and when the learning completion history is set as history by the setting means, the relaxation means It is possible to adopt a configuration in which the determination range is set larger than when no history is set.
[0018]
Further, in addition to such a configuration, according to the invention described in claim 3, in the air-fuel ratio control apparatus for an internal combustion engine according to claim 1 or 2, the learning means includes an average value of the air-fuel ratio correction coefficient and a theoretical sky. The air-fuel ratio learning value is updated so that a deviation from a reference value of the air-fuel ratio correction coefficient corresponding to the fuel ratio decreases, and the determination means continues the predetermined determination period within the predetermined determination range. The learning completion condition is set to “no history” when the learning completion history is set as “with history” by the setting means. It is also possible to adopt a configuration in which the determination period is set to be shorter than that.
[0019]
According to a fourth aspect of the present invention, in the air-fuel ratio control apparatus for an internal combustion engine according to the second or third aspect, the setting unit is configured to perform the learning completion history when the deviation is outside a predetermined range during the learning of the learning unit. And the learning completion history is set as no history.
[0020]
For example, when components related to air-fuel ratio control, such as a fuel injection valve or a sensor for detecting the intake air amount, are replaced, the reliability of the air-fuel ratio learned value learned so far is greatly reduced. The learned value is not able to properly compensate for the steady deviation of the engine air-fuel ratio. In such a case, even if the learning value of the air-fuel ratio is reflected in the feedback control, the steady-state deviation of the engine air-fuel ratio cannot be compensated, so that the air-fuel ratio correction coefficient greatly deviates from the vicinity of the reference value. become.
[0021]
In the configuration of the invention described in claim 4, when the learning regarding the engine air-fuel ratio is being executed and the average value of the deviation between the air-fuel ratio correction coefficient and its reference value is outside the predetermined range, the learning completion history is erased. Thus, the learning completion history is set as no history. Therefore, the air-fuel ratio learning value is surely learned without relaxing the learning completion condition, and this deviation is suitable even when the steady deviation of the engine air-fuel ratio changes greatly as described above. Will be able to compensate.
[0022]
According to a fifth aspect of the present invention, in the air-fuel ratio control apparatus for an internal combustion engine according to any one of the first to fourth aspects, the learning means completes the learning by the setting means when learning the air-fuel ratio learning value. When the history is set to have a history, the update amount of the air-fuel ratio learning value is set smaller than when the learning completion history is set to have no history.
[0023]
In each learning region where the learning completion history is set to have history, the learning is completed at least once, and the air-fuel ratio learning value is larger than the original value that can appropriately compensate for the engine air-fuel ratio deviation. There is no deviation. Accordingly, when the air-fuel ratio learning value is updated, it is desirable to update the learning value gradually so that even a slight change in the engine air-fuel ratio with time can be compensated appropriately.
[0024]
In this regard, according to the above-described configuration of the invention described in claim 5, when the learning completion history is set to have a history, the update amount of the air-fuel ratio learning value is set smaller than when the history is not set. Therefore, the learning value of the air-fuel ratio can be updated more precisely, and the deviation of the engine air-fuel ratio slightly changed with time as described above can be compensated finely.
[0025]
According to a sixth aspect of the present invention, in the air-fuel ratio control apparatus for an internal combustion engine according to any one of the first to fifth aspects, the control means sets the learning completion history as having a history by the setting means. At least one of the skip amount when the air-fuel ratio correction coefficient is skip-controlled and the integral amount when the integral control is performed is set larger than when the learning completion history is set to no history.
[0026]
In order to appropriately learn the air-fuel ratio learning value, it is desirable to set both the skip amount and the integral amount in the air-fuel ratio correction coefficient to be small so as to suppress fluctuations in the air-fuel ratio correction coefficient. However, if these skip amount and integral amount are set to be uniformly small, the behavior of the air-fuel ratio correction coefficient becomes slow, and the period during which the engine air-fuel ratio deviates from the stoichiometric air-fuel ratio is prolonged during learning of the air-fuel ratio learning value. As a result, there is a concern about the temporary deterioration of exhaust properties.
[0027]
In this regard, in the above-described configuration of the invention described in claim 6, when the learning completion history is set to have a history, and it is sufficient to slightly update the air-fuel ratio learning value, the air-fuel ratio correction coefficient is skip-controlled. Since at least one of the skip amount at the time of integration and the integration amount at the time of integral control is set larger than when no history is set, the exhaust property temporarily accompanying the learning of the engine air-fuel ratio as described above Deterioration can be suppressed.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of an air-fuel ratio control apparatus for an internal combustion engine according to the present invention will be described with reference to FIGS.
[0029]
FIG. 1 schematically shows the configuration of an air-fuel ratio control apparatus for an internal combustion engine according to the present embodiment.
As shown in the figure, the intake passage 11 of the internal combustion engine 10 is provided with a fuel injection valve 12 that injects fuel into the intake passage 11. Further, a throttle valve 13 for metering intake air is provided upstream of the fuel injection valve 12 in the intake passage 11, and an air cleaner 14 is further provided upstream thereof. The fuel injected from the fuel injection valve 12 is mixed with the intake air that has passed through the air cleaner 14 and then introduced into the engine combustion chamber (not shown) of the internal combustion engine 10. The exhaust generated by the combustion of the air-fuel mixture is discharged to the outside through the exhaust passage 15 from the engine combustion chamber. The exhaust passage 15 is provided with a catalytic converter 16 for purifying the exhaust.
[0030]
An air flow meter 17 is attached between the throttle valve 13 and the air cleaner 14 in the intake passage 11, and the amount of intake air passing through the intake passage 11 is detected by the air flow meter 17. On the other hand, an oxygen sensor 18 is attached upstream of the catalytic converter 16 in the exhaust passage 15, and the oxygen concentration in the exhaust gas is detected by the oxygen sensor 18. Further, a rotation speed sensor 19 is provided in the vicinity of the output shaft (not shown) of the internal combustion engine 10, and the rotation speed sensor 19 detects the rotation speed of the output shaft, that is, the engine rotation speed.
[0031]
The detection signals of these sensors 17 to 19 are input to an electronic control unit 20 that executes control related to engine operation such as air-fuel ratio control. The electronic control unit 20 includes a CPU 22 that performs arithmetic processing and an input / output circuit (both not shown), and a memory 22 that stores data necessary for various controls.
[0032]
Further, the internal combustion engine 10 includes an evaporated fuel processing mechanism 30 for performing purge control. The evaporative fuel processing mechanism 30 is provided in the middle of the canister 32 connected to the fuel tank 31 and the air cleaner 14, the purge passage 33 connecting the canister 32 and the intake passage 11, and in the middle of the purge passage 33. The flow control valve 34 is controlled to be opened and closed through the electronic control unit 20.
[0033]
The evaporated fuel generated in the fuel tank 31 is introduced into the canister 32 and once adsorbed by the fuel adsorbent inside. On the other hand, when the air is introduced into the canister 32 through the air cleaner 14 as the flow control valve 34 is opened, the fuel adsorbed on the fuel adsorbent of the canister 32 is detached from the adsorbent again. The separated evaporated fuel is introduced into the intake passage 11 through the purge passage 33 and then burned in the engine combustion chamber. The electronic control unit 20 appropriately controls the opening degree of the flow rate control valve 34 so as to minimize the influence of such purge control on the air-fuel ratio control of the internal combustion engine 10.
[0034]
Next, details of the air-fuel ratio control executed by the electronic control unit 20 will be described.
First, the procedure for calculating the fuel injection time TAU of the fuel injection valve 12 will be described with reference to the flowchart shown in FIG. A series of processing shown in this flowchart is repeatedly executed by the electronic control device 20 every predetermined control period.
[0035]
In this series of processing, first, the basic fuel injection time TP is calculated (step 110). This basic fuel injection time TP is the ratio between the fuel injection amount of the fuel injection valve 12 and the intake air amount, that is, the intake air amount detected by the air cleaner 14 and the rotational speed so that the engine air-fuel ratio becomes the stoichiometric air-fuel ratio. It is calculated based on the engine speed detected by the sensor 19.
[0036]
Next, a correction coefficient KT1 for performing post-startup increase correction, warm-up increase correction, and the like, and a correction coefficient KT2 for performing purge concentration correction, etc. are calculated (step 112). Then, the fuel injection time TAU is calculated according to the following arithmetic expression (1) (step 114).
[0037]
TAU ← TP ・ KT1 ・ (FAF + KGj + KT2) (1)
In the above equation (1), “FAF” is a feedback correction coefficient (hereinafter referred to as “air-fuel ratio correction coefficient”) for compensating for the deviation through feedback control when the engine air-fuel ratio transits transiently from the stoichiometric air-fuel ratio. It is said). As is well known, the air-fuel ratio correction coefficient FAF has a reference value of “1.0”, and is smaller than “1.0” when the engine air-fuel ratio is richer than the stoichiometric air-fuel ratio. When it is leaner than the air-fuel ratio, it is set to be larger than “1.0”. Accordingly, the fuel injection amount of the fuel injection valve 12 is corrected to decrease when the engine air-fuel ratio is richer than the stoichiometric air-fuel ratio, and is corrected to increase when the engine air-fuel ratio is leaner than the stoichiometric air-fuel ratio. Become.
[0038]
In the above equation (1), “KGj” is a correction coefficient (hereinafter referred to as “air-fuel ratio learning value”) for compensating for the deviation when the engine air-fuel ratio steadily deviates from the theoretical air-fuel ratio. It is.
[0039]
As described above, the basic fuel injection time TP is set so that the engine air-fuel ratio matches the stoichiometric air-fuel ratio, but the individual difference of the internal combustion engine 10, the injection characteristics of the fuel injection valve 12, or the air flow meter 17 There are cases where the engine air-fuel ratio deviates from the original time at which the engine air-fuel ratio can be made to coincide with the stoichiometric air-fuel ratio due to individual differences in output characteristics, and further changes with time of these individual differences. The air-fuel ratio learning value KGj is a correction value for compensating for such a deviation in the basic fuel injection time TP. The air-fuel ratio learning value KGj is set separately as a value for each of a plurality of areas j (hereinafter referred to as “learning areas”) divided according to the intake air amount.
[0040]
After calculating the fuel injection time TAU based on the above equation (1), this series of processes is temporarily terminated.
Next, the procedure for calculating the air-fuel ratio correction coefficient FAF will be described with reference to the flowcharts shown in FIGS. A series of processing shown in this flowchart is repeatedly executed by the electronic control device 20 every predetermined control period.
[0041]
In this series of processing, it is first determined whether or not an engine air-fuel ratio feedback (F / B) control condition is satisfied (step 210). Here, it is determined that the feedback control condition is satisfied, for example, when the engine cooling water temperature has risen to a predetermined temperature or more, for example, not when the engine is started, and so on.
[0042]
When it is determined that the feedback control condition is satisfied (step 210: YES), the learning region j corresponding to the current intake air amount is determined (step 212). Then, it is determined whether or not the learning completion history flag XKGRj corresponding to the determined learning region j is set to “ON” (step 214).
[0043]
This learning completion history flag XKGRj is a flag indicating a learning completion history related to the air-fuel ratio learning value KGj, and it is a condition that learning of the air-fuel ratio learning value KGj has been completed at least once by the time of the current engine operation. Is set to “On”. The setting contents of the learning completion history flag XKGRj are stored and held in the memory 22 even after the engine operation is stopped and the power supply to the electronic control device 20 is stopped.
[0044]
If it is determined that the learning completion history flag XKGRj is “OFF” (step 214: NO), the skip amount S when the air-fuel ratio correction coefficient FAF is skip-controlled and the coefficient FAF are integrated. The integration amount K at that time is set equal to the second skip amount S2 and the second integration amount K2, respectively (steps 217 and 218). The second skip amount S2 and the second integral amount K2 are calculated based on the engine operating state such as the intake air amount in the above steps 217 and 218.
[0045]
On the other hand, when it is determined that the learning completion history flag XKGRj is “ON” (step 214: YES), the skip amount S and the integral amount K are the first skip amount S1 and the first integral amount K1. They are set equal (steps 215 and 216). The first skip amount S1 and the first integral amount K1 are calculated in steps 215 and 216 based on the engine operating state in the same manner as the second skip amount S2 and the second integral amount K2.
[0046]
Here, the first skip amount S1 is calculated as a value that is always larger than the second skip amount S2 if the engine operating state is the same. Similarly, the first integral amount K1 is always calculated as a value that is always larger than the second integral amount K2 if the engine operating state is the same.
[0047]
After the skip amount S and the integral amount K are thus set, it is next determined whether or not the current engine air-fuel ratio is rich (step 219). As shown in FIG. 5A, the detection signal (voltage signal) V of the oxygen sensor 18 is higher than the predetermined voltage V1 when the engine air-fuel ratio is rich and the oxygen concentration in the exhaust gas is less than the predetermined concentration. When the engine air-fuel ratio is lean and the oxygen concentration in the exhaust gas is equal to or higher than the predetermined concentration, the engine air-fuel ratio becomes lower than the predetermined voltage V1. The electronic control unit 20 refers to the detection signal V of the oxygen sensor 18 at the time of the determination, and determines whether or not the engine air-fuel ratio is rich.
[0048]
When it is determined that the engine air-fuel ratio is rich (step 219: YES), it is further determined whether or not the engine air-fuel ratio was lean in the previous control cycle, that is, the engine air-fuel ratio is changed from lean in the current control cycle. It is determined whether or not the rich has been switched (step 220 in FIG. 4).
[0049]
If it is determined that the engine air-fuel ratio has changed from lean to rich in the current control cycle (step 220: YES), the current air-fuel ratio correction coefficient FAF is the value FAFL when the engine air-fuel ratio has changed from lean to rich. Is stored in the memory 22 (step 222). Thereafter, as shown in FIG. 5B, the skip amount S is subtracted from the current air-fuel ratio correction coefficient FAF, and the subtraction value (FAF-S) is set as a new air-fuel ratio correction coefficient FAF ( Step 224).
[0050]
On the other hand, when it is determined that the engine air-fuel ratio is still rich in the previous control cycle, that is, when it is determined that the engine air-fuel ratio is continuously rich (step 220: NO), FIG. As shown, the integration amount K is subtracted from the current air-fuel ratio correction coefficient FAF, and the subtraction value (FAF-K) is set as a new air-fuel ratio correction coefficient FAF (step 226).
[0051]
On the other hand, when it is determined that the current engine air-fuel ratio is lean (step 219: NO), it is then determined whether the engine air-fuel ratio was rich in the previous control cycle, that is, in the current control cycle. It is determined whether the fuel ratio has been switched from rich to lean (step 230 in FIG. 4).
[0052]
If it is determined that the engine air-fuel ratio has changed from rich to lean in the current control cycle (step 230: YES), the current air-fuel ratio correction coefficient FAF is the value FAFR when the engine air-fuel ratio has changed from rich to lean. Is stored in the memory 22 (step 232). Thereafter, the skip amount S is added to the current air-fuel ratio correction coefficient FAF, and the added value (FAF + S) is set as a new air-fuel ratio correction coefficient FAF (step 234).
[0053]
On the other hand, when it is determined that the engine air-fuel ratio was lean even in the previous control cycle, that is, when it is determined that the engine air-fuel ratio is continuously lean (step 230: NO), the current air-fuel ratio correction coefficient The integral amount K is added to FAF, and the added value (FAF + K) is set as a new air-fuel ratio correction coefficient FAF (step 236).
[0054]
FIG. 6 shows an example of how the air-fuel ratio correction coefficient FAF is set through the above steps 224, 234, 226, 236. FIG. 10A shows an example of how the air-fuel ratio correction coefficient FAF changes when there is a learning completion history of the air-fuel ratio learning value KGj (the learning completion history flag XKGRj is “ON”). In addition, there are shown examples of changes in the air-fuel ratio correction coefficient FAF when there is no learning completion history (the learning completion history flag XKGRj is “off”).
[0055]
When there is no learning completion history of the air-fuel ratio learning value KGj, the skip amount S and the integral amount K are set relatively small, so that the fluctuation of the air-fuel ratio correction coefficient FAF becomes slow. For this reason, although the period during which the actual engine air-fuel ratio deviates from the stoichiometric air-fuel ratio tends to be prolonged, the air-fuel ratio correction coefficient FAF does not deviate significantly from the vicinity of the reference value “1.0”.
[0056]
On the other hand, when there is a learning completion history of the air-fuel ratio learning value KGj, the skip amount S and the integral amount K are set to be relatively large, so that the change speed of the air-fuel ratio correction coefficient FAF increases. For this reason, compared with the case where there is no learning completion history, the fluctuation range when the air-fuel ratio correction coefficient FAF fluctuates around the reference value “1.0” is large, but the actual engine air-fuel ratio is the stoichiometric air-fuel ratio. The period of deviation is shortened.
[0057]
After the air-fuel ratio correction coefficient FAF is skip-controlled through the previous steps 224 and 234, the average value FAFAV of the air-fuel ratio correction coefficient FAF is calculated according to the following arithmetic expression (2).
[0058]
FAFAV ← (FAFL + FAFR) / 2 (2)
Next, the skip flag XSKIP indicating that the current control cycle is the timing at which the engine air-fuel ratio changes from lean to rich or from rich to lean (skip timing) is set to “on” (step 242). The process is temporarily terminated.
[0059]
On the other hand, when it is determined in step 210 of FIG. 3 that the feedback control condition is not satisfied (step 210: NO), the air-fuel ratio correction coefficient FAF is set to “1.0” to perform open-loop control of the engine air-fuel ratio. After that (step 250), the average value FAFAV of the air-fuel ratio correction coefficient FAF is also set to “1.0” (step 252).
[0060]
After the process of step 252 or the processes of the previous steps 226 and 236 are executed, the series of processes is temporarily ended.
Next, the procedure for calculating the air-fuel ratio learning value KGj will be described with reference to the flowcharts shown in FIGS. A series of processing shown in this flowchart is repeatedly executed by the electronic control device 20 every predetermined control period.
[0061]
In this series of processing, it is first determined whether or not a learning condition is satisfied (step 310). Here, the engine air-fuel ratio feedback control is being executed (the feedback control condition is satisfied), the internal combustion engine 10 is not in the acceleration / deceleration state, the intake air amount is stable, and so on. If it is, it is determined that the learning condition is satisfied.
[0062]
Next, a learning region j corresponding to the current intake air amount is determined (step 312), and whether or not there is a learning completion history in the learning region j, that is, the learning completion history flag XKGRj is set to “on”. It is determined whether or not there is (step 314).
[0063]
If it is determined that there is a learning completion history (step 314: YES), the determination value α and the determination period β used when determining the completion of learning and the air-fuel ratio learning value KGj are updated. The update amount DKG is set equal to the predetermined values α1, β1, and DKG1 (steps 320, 322, and 324). These predetermined values α1, β1, and DKG1 are all values that are set in response to the learning completion history.
[0064]
On the other hand, when it is determined that there is no learning completion history (step 314: NO), the determination value α, the determination period β, and the update amount DKG are set equal to the predetermined values α2, β2, and DKG2, respectively (step 314). 330, 332, 334). Each of these predetermined values α2, β2, DKG2 is a value set corresponding to the absence of the learning completion history, and the following equations (3) between the predetermined values α1, β1, DKG1 It is set so that the magnitude relationship shown in (5) is satisfied.
[0065]
α1> α2 (3)
β1 <β2 (4)
DKG1 <DKG2 (5)
Accordingly, the determination value α is set to a relatively large value as compared with the case where there is no learning completion history through the processing of each of the above steps 320, 322, 330 and 332, and the determination period β is set to the learning completion history. When there is, the period is set to be relatively shorter than when there is no.
[0066]
The predetermined values DKG1 and DKG2 relating to the update amount DKG are both set based on the absolute value | FAFAV−1.0 | between the average value FAFAV of the air-fuel ratio correction coefficient FAF and “1.0”. . FIG. 9 shows the relationship between the predetermined values DKG1 and DKG2 and the absolute deviation | FAFAV-1.0 |. The relationship between the two is stored in the memory 22 of the electronic control unit 20 as function data.
[0067]
As shown in the figure, each of the predetermined values DKG1 and DKG2 is set to a smaller value as the absolute deviation | FAFAV-1.0 | becomes smaller in order to ensure the convergence and accuracy of the air-fuel ratio learned value KGj. Is done. Accordingly, the update amount DKG is set to a smaller value as the absolute deviation | FAFAV-1.0 | becomes smaller through the above steps 324 and 334, and is compared with the case where there is a learning completion history as compared with the case where there is no learning completion history. Will be set to a small value.
[0068]
When the determination value α, the determination period β, and the update amount DKG are set in this manner, whether or not the learning region j has changed between the previous control cycle and the current control cycle, in other words, Then, it is determined whether or not the learning region j determined in the current control cycle is different from the learning region j determined in the previous control cycle (step 340). If it is determined that the learning area j has not changed (step 340: NO), it is further determined whether or not the skip flag XSKIP is set to “ON” (step 342).
[0069]
If it is determined that the skip flag XSKIP is set to “ON”, that is, if it is determined that the current control cycle is the skip timing (step 342: YES), the skip flag XSKIP is set to “ After being set to “OFF” (step 344), the skip counter value CSKIP is incremented (step 346). This skip counter value CSKIP is a counter value for counting the number of times that the skip control of the air-fuel ratio correction coefficient FAF is performed while the learning region j is maintained in the same region.
[0070]
Then, it is determined whether or not the skip counter value CSKIP is equal to or greater than the predetermined value KCSKIP (step 348 in FIG. 8). If it is determined that the skip counter value CSKIP is equal to or greater than the predetermined value KCSKIP (step 348: YES), In step 349, it is determined whether or not purge control is being executed, in other words, whether or not the flow control valve 34 is open (step 349).
[0071]
In the present embodiment, the purge control is permitted to be executed on the condition that learning is completed in the current learning region j. That is, when the internal combustion engine 10 is started and feedback control of the engine air-fuel ratio is started, learning of the air-fuel ratio learning value KGj is first performed, and after the learning is completed, purge control is performed in the current learning region j. Execution is allowed. Further, if the learning region j changes, learning of the air-fuel ratio learning value KGj is similarly performed in the learning region j after the change, and execution of purge control in the learning region j is permitted after completion of learning. Become.
[0072]
If it is determined that purge control is not being executed (step 349: NO), it is then determined whether the absolute deviation | FAFAV-1.0 | is equal to or less than a predetermined value γ (step 349). 350). This predetermined value γ is for evaluating the reliability of the air-fuel ratio learned value KGj. That is, when the absolute deviation | FAFAV-1.0 | is larger than the predetermined value γ, the steady deviation of the engine air-fuel ratio is not compensated for by the air-fuel ratio learning value KGj, so the average of the air-fuel ratio correction coefficient FAF is It can be considered that the value FAFAV is greatly deviated from the vicinity of “1.0”, and it can be determined that the reliability of the air-fuel ratio learning value KGj is lowered.
[0073]
If the absolute deviation | FAFAV−1.0 | is larger than the predetermined value γ, in other words, the deviation between the average value FAFAV of the air-fuel ratio correction coefficient FAF and the reference value “1.0” of the air-fuel ratio correction coefficient FAF. When (FAFAV-1.0) is outside the predetermined range (−γ ≦ FAFAV−1.0 ≦ γ) (step 350: NO), the learning completion history flag XKGRj in the current learning region j is “off”. (Step 390).
[0074]
After this process is executed, or when it is determined that the absolute deviation | FAFAV−1.0 | is equal to or smaller than the predetermined value γ (step 350: YES), the average value FAFAV of the air-fuel ratio correction coefficient FAF is the determination value. It is determined whether or not it is within a determination range (1.0−α ≦ FAFAV ≦ 1.0 + α) determined by α (steps 352 and 354).
[0075]
When the average value FAFAV of the air-fuel ratio correction coefficient FAF is within this determination range (steps 352 and 354: NO), the duration counter value CTIME is incremented (step 380). This continuation time counter value CTIME is a counter value that measures the time during which the average value FAFAV of the air-fuel ratio correction coefficient FAF is continuously within the above-mentioned determination range. Then, it is determined whether or not the duration counter value CTIME exceeds the determination period β (step 382). If the determination period β does not exceed the determination period β (step 382: NO), a series of processing is once performed. Is terminated.
[0076]
On the other hand, when the duration counter value CTIME exceeds the determination period β (step 382: YES), the learning completion flag XKGj indicating that the learning of the air-fuel ratio learning value KGj is completed during the current engine operation is “ On "is set (step 384). Further, after the learning completion history flag XKGRj is set to “ON” (step 386), a series of processing is once ended.
[0077]
Here, since the determination value α is set to a relatively large value when there is a learning completion history, the determination range (1.0−α ≦ FAFAV ≦ 1.0 + α) also has a learning completion history. In some cases, it will be relatively enlarged. On the other hand, for the determination period β, the synchronization period β is set to a relatively short period when there is a learning completion history, and both the skip amount S and the integral amount K are set to be relatively large. Therefore, the period during which learning completion is determined is set to be relatively short. Therefore, compared with the case where there is no learning completion history, the condition for determining completion of learning is relaxed, and it is determined that learning has been completed earlier.
[0078]
On the other hand, if it is determined that the average value FAFAV of the air-fuel ratio correction coefficient FAF is out of the determination range and is larger than the value within the same range (step 352: YES), or the average value FAFAV of the air-fuel ratio correction coefficient FAF Is out of the determination range and is determined to be a value smaller than the value within the same range (step 354: YES), the air-fuel ratio learning value KGj is updated based on the update amount DKG. .
[0079]
That is, when it is determined that the average value FAFAV of the air-fuel ratio correction coefficient FAF is larger than the value within the determination range (step 352: YES), after the duration counter value CTIME is cleared ( In step 370), the updated amount DKG is added to the current air-fuel ratio learned value KGj, and the added value (KGj + DKG) is set as a new air-fuel ratio learned value KGj (step 372). On the other hand, when it is determined that the average value FAFAV of the air-fuel ratio correction coefficient FAF is smaller than the value within the determination range (step 354: YES), after the duration counter value CTIME is cleared (step 360) The update amount DKG is subtracted from the current air-fuel ratio learning value KGj, and the subtraction value (KGj−DKG) is set as a new air-fuel ratio learning value KGj (step 362).
[0080]
Here, since the update amount DKG is set to a relatively small value as compared with the case where there is a learning completion history, the air-fuel ratio learned value KGj updated through the steps 362 and 372 is Even when the tendency of deviation of the engine air-fuel ratio slightly changes with time, this can be compensated appropriately. Even if the update amount DKG is set to be small as described above, the learning is completed at least once, and the air-fuel ratio learned value KGj deviates greatly from the original value that can appropriately compensate for the engine air-fuel ratio deviation. Therefore, there is no possibility that the convergence of the air-fuel ratio learned value KGj will be greatly reduced.
[0081]
After updating the air-fuel ratio learned value KGj in the above steps 362 and 372, this series of processing is terminated. Further, when it is determined that the current control cycle is not the skip timing (step 342: NO), when it is determined that the skip counter value CSKIP is less than the predetermined value KCSKIP (step 348: NO), purge control is executed. Even when it is determined that the process is in progress (step 349: YES), this series of processes is temporarily terminated. Further, when it is determined that the learning condition is not satisfied (step 310: NO), or when it is determined that the learning region j has changed (step 340: YES), the skip counter value CSKIP is cleared. After this (step 316), this series of processing is once ended.
[0082]
According to the air-fuel ratio control apparatus for an internal combustion engine according to the present embodiment described above, the following operational effects can be achieved.
(1) When the engine operation is started, the learning mode is different, but the air-fuel ratio learned value KGj is learned for each learning region j regardless of the history of learning completion, and the learned value KGj is used as the engine air-fuel ratio. This is reflected in the feedback control. For this reason, even if the engine air-fuel ratio shift varies with time in each learning region, this shift is compensated in each learning region j.
[0083]
Further, among the learning regions j, a region that has a learning completion history and that does not require a large update of the air-fuel ratio learning value KGj is compared with a region that has no learning completion history and is determined by the determination value α. While expanding the range, the learning completion condition is relaxed by setting the determination period β for determining completion of learning short. For this reason, for the learning region j having the learning completion history, the learning is completed at an earlier stage, and the purge control is also started early.
[0084]
Therefore, even if the engine air-fuel ratio shift varies with time in each learning region j, the shift can be compensated appropriately, and the frequency of execution of the purge control performed after the completion of learning can be reduced. More can be secured.
[0085]
(2) Further, when there is a learning completion history, the update amount DKG when updating the air-fuel ratio learning value KGj is set smaller than when there is no history, so the engine air-fuel ratio shift tendency Even when the air pressure slightly changes over time, the air-fuel ratio learning value KGj can be set appropriately in accordance with this. Therefore, the air-fuel ratio learning value KGj can be updated more precisely, and the engine air-fuel ratio shift slightly changed with time as described above can be compensated finely.
[0086]
(3) Furthermore, since the skip amount S and the integral amount K are set larger when there is a learning completion history than when there is no history, the change rate of the air-fuel ratio correction coefficient FAF increases. The period during which the engine air-fuel ratio deviates from the stoichiometric air-fuel ratio is shortened. Accordingly, it is possible to suppress the temporary deterioration of the exhaust properties accompanying the learning of the engine air-fuel ratio.
[0087]
(4) Further, when learning the engine air-fuel ratio, the steady-state deviation of the engine air-fuel ratio changes greatly, so that the average value FAFAV of the air-fuel ratio correction coefficient FAF and the reference value “1.0 of the air-fuel ratio correction coefficient FAF are calculated. ”(FAFAV−1.0) is out of the predetermined range (−γ ≦ FAFAV−1.0 ≦ γ), the learning completion history flag XKGRj in the current learning region j is set to“ off ”. Like to do. Accordingly, in this case, the learning completion condition is not relaxed, the update amount DKG is changed to a larger value, and the skip amount S and the integral amount K are changed to smaller values. The learning value KGj is reliably relearned.
[0088]
Accordingly, when the fuel injection valve 12 or the air flow meter 17 is replaced, or the setting content of the air-fuel ratio learning value KGj stored in the memory 22 is deleted by battery replacement or the like, the engine air-fuel ratio is kept constant. Even when the deviation greatly changes, the deviation can be compensated appropriately and promptly.
[0089]
The embodiment of the present invention described above can be implemented by changing the configuration as follows.
In the above embodiment, when the deviation between the average value FAFAV of the air-fuel ratio correction coefficient FAF and the reference value “1.0” is out of the predetermined range, only the learning completion history flag XKGRj in the learning region j is set to “off”. However, the learning completion history flag XKGRj in all the learning regions j may be set to “off”.
[0090]
In the above embodiment, the purge control has been described as being permitted after the learning of the engine air / fuel ratio is completed. Here, for example, immediately after the engine operation is started, the amount of evaporated fuel in the canister 32 may rise to the vicinity of the storage limit amount. For this reason, there is a case where it is better to reduce the evaporated fuel once it is reduced to a level far below its storage limit.
[0091]
Therefore, for example, when a predetermined condition is satisfied, the purge control is executed before the learning of the engine air-fuel ratio is completed, the evaporated fuel in the canister 32 is reduced, the learning is executed, and the learning is performed. You may make it permit execution of purge control again after completion. Even when the execution of the purge control is permitted as described above, the operational effects shown in the embodiment can be obtained.
[0092]
In the above embodiment, the determination value α, the determination period β, and the predetermined values DKG1 and DKG2 related to the update amount DKG are set depending on whether or not there is a learning completion history, but for example, learning in a certain learning region j The number of learning completions in the same learning region j after the completion history flag XKGRj is set to “on” is counted as a value for evaluating the reliability of the learning result (air-fuel ratio learning value KGj), and in addition to the presence or absence of the learning completion The predetermined values α, β, DKG1, and DKG2 may be variably set according to the number of learning completions.
[0093]
In the above embodiment, when there is a learning completion history, the determination range determined by the determination value α is expanded and the determination period β for determining completion of learning is set to be short. Only one of the expansion of the determination range and the shortening of the determination period β may be performed.
[0094]
In the above embodiment, when there is a learning completion history, both the skip amount S and the integral amount K are set larger than when there is no history, but only one of these amounts is set. You may make it set relatively large, when there exists a learning completion log | history.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram showing a configuration of an internal combustion engine and an air-fuel ratio control device thereof according to an embodiment.
FIG. 2 is a flowchart showing a processing procedure for calculating a fuel injection time.
FIG. 3 is a flowchart showing a processing procedure for calculating an air-fuel ratio correction coefficient.
FIG. 4 is a flowchart showing a processing procedure for calculating an air-fuel ratio correction coefficient.
FIG. 5 is a timing chart showing an example of how the oxygen sensor detection signal and the air-fuel ratio correction coefficient are changed.
FIG. 6 is a graph showing an example of how the air-fuel ratio correction coefficient changes.
FIG. 7 is a flowchart showing a processing procedure for calculating an air-fuel ratio learning value.
FIG. 8 is a flowchart showing a processing procedure for calculating an air-fuel ratio learning value.
FIG. 9 is a graph showing a function map for setting the update amount of the air-fuel ratio learning value.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Internal combustion engine, 11 ... Intake passage, 12 ... Fuel injection valve, 13 ... Throttle valve, 14 ... Air cleaner, 15 ... Exhaust passage, 16 ... Catalytic converter, 17 ... Air flow meter, 18 ... Oxygen sensor, 19 ... Rotation speed sensor 20 ... an electronic control unit, 22 ... a memory, 30 ... a fuel vapor processing mechanism, 31 ... a fuel tank, 32 ... a canister, 33 ... a purge passage, 34 ... a flow control valve.

Claims (6)

Control means for feedback-controlling the engine air-fuel ratio so as to match the stoichiometric air-fuel ratio based on the air-fuel ratio correction coefficient, and a steady state between the engine air-fuel ratio and the stoichiometric air-fuel ratio based on the behavior of the air-fuel ratio correction coefficient A learning means for learning an air-fuel ratio learning value for compensating for the deviation for each of a plurality of learning regions divided according to an engine operating state and reflecting the air-fuel ratio learning value in the feedback control, and a predetermined value after engine startup An air-fuel ratio control apparatus for an internal combustion engine comprising: learning completed by the learning means when the learning completion condition is satisfied; and determination means for determining for each learning region,
A setting means for setting a learning completion history for each learning region as having a history when the learning completion determination is made at least once by the determination means;
And a relaxation means for relaxing the learning completion condition when determining the completion of learning in the learning area when the learning completion history in the certain learning area is set as having a history by the setting means. An air-fuel ratio control apparatus for an internal combustion engine. The air-fuel ratio control apparatus for an internal combustion engine according to claim 1,
The learning means updates the air-fuel ratio learning value so that a deviation between an average value of the air-fuel ratio correction coefficient and a reference value of the air-fuel ratio correction coefficient corresponding to the theoretical air-fuel ratio decreases,
The determination means sets the deviation to be within a predetermined determination range as the learning completion condition,
The relaxation means sets the determination range to be larger when the learning completion history is set as history with the setting means than when the learning completion history is set as no history. An air-fuel ratio control apparatus for an internal combustion engine characterized by the above. The air-fuel ratio control apparatus for an internal combustion engine according to claim 1 or 2,
The learning means updates the air-fuel ratio learning value so that a deviation between an average value of the air-fuel ratio correction coefficient and a reference value of the air-fuel ratio correction coefficient corresponding to the theoretical air-fuel ratio decreases,
The determination means uses the deviation as a learning completion condition that the deviation continuously exists within a predetermined determination range for a predetermined determination period.
The relaxation means sets the determination period shorter when the learning completion history is set as having a history by the setting means than when the learning completion history is set as having no history. An air-fuel ratio control device for an internal combustion engine. The air-fuel ratio control apparatus for an internal combustion engine according to claim 2 or 3,
The setting means is configured to erase the learning completion history and set the learning completion history as no history when the deviation is outside a predetermined range during the learning of the learning means. Fuel ratio control device. The air-fuel ratio control apparatus for an internal combustion engine according to any one of claims 1 to 4,
When the learning means learns the air-fuel ratio learning value, the air-fuel ratio is more different when the learning completion history is set to have no history than when the learning completion history is set to no history. An air-fuel ratio control apparatus for an internal combustion engine, wherein the learning value update amount is set small. The air-fuel ratio control apparatus for an internal combustion engine according to any one of claims 1 to 5,
When the learning completion history is set as history by the setting means, the control means uses at least one of a skip amount when performing the skip control of the air-fuel ratio correction coefficient and an integration amount when performing the integral control as the learning completion history. Is set to be larger than when no history is set.

Images