JP3467881B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an air-fuel ratio control device for an internal combustion engine, and more particularly to a device for controlling the learning of the air-fuel ratio while switching the air-fuel ratio between near the stoichiometric air-fuel ratio and lean air-fuel ratio. Regarding improvement.

[0002]

2. Description of the Related Art As a conventional air-fuel ratio control system for an internal combustion engine, for example, there is a technique disclosed in Japanese Patent Application Laid-Open No. 2-67441. This is a vaporized fuel evaporation that prevents vaporized fuel from evaporating to the outside air by temporarily adsorbing the vaporized fuel generated in the fuel tank etc. to the canister and sucking the adsorbed vaporized fuel to the engine during engine operation. While equipped with an air-fuel ratio learning control device that corrects the air-fuel ratio control error due to product-to-product variation and changes over time, in order to prevent the air-fuel ratio learning from being affected by purging, It is forbidden to update the learning value.

When the learned value monotonously increases or decreases and reaches the upper and lower limit values, it is determined that the air-fuel ratio control device has some abnormality. When the fuel is removed (hereinafter referred to as “purge”) and is being sucked into the engine, the air-fuel ratio may become too rich to be corrected by the feedback control of the fuel supply amount due to the increase in the fuel. The learning becomes small by the control, and the learning value reaches the lower limit depending on the degree of enrichment. Stability is higher during lean operation (diluted air-fuel ratio control) than during stoichiometric operation (theoretical air-fuel ratio feedback control).
The allowable setting range of the air-fuel ratio that allows operation with a small amount of Ox emission is narrow, and the change in the air-fuel ratio with respect to the change in the fuel supply amount is large. It is common to prohibit lean driving if you end up in. Considering that the allowable range of the air-fuel ratio during lean operation is narrower than that during stoichiometric operation as described above, set the upper and lower limits so that lean operation is performed only when the learned value is within a narrower range. It will be.

[0004]

However, in such a conventional air-fuel ratio control system for an internal combustion engine, the lean operation is performed by comparing the upper and lower limit values, which are preset fixed values, with the learning value. If the air-fuel ratio becomes richer than the target value to the extent that it cannot be considered abnormal due to product variations or changes over time, the learning value is constantly in a small state, so the concentration is too high. There was a problem that the lower limit value could be easily reached by a slight disturbance such as no purge gas, and lean operation was prohibited even though it was not necessary.

The present invention has been made in view of the above-mentioned conventional problems, and by improving the learning control system,
An object of the present invention is to provide an air-fuel ratio control device for an internal combustion engine, which solves the problem.

[0006]

Therefore, as shown in FIG. 1, the invention according to claim 1 temporarily adsorbs the evaporated fuel generated in the fuel tank by the adsorbing means, and the adsorbing means adsorbs the engine intake air. And an operating state detecting means for detecting the operating state of the engine, and an evaporative fuel evaporation preventing device for communicating with the system to separate the evaporated fuel adsorbed by the adsorbing means and guide it to the engine intake system for treatment. Air-fuel ratio detection means for detecting the air-fuel ratio of the intake air-fuel mixture, and the basic control value of the air-fuel ratio so that the actual air-fuel ratio of the engine intake air-fuel mixture detected by the air-fuel ratio detection means approaches the theoretical air-fuel ratio. An air-fuel ratio feedback control means for performing feedback control of the air-fuel ratio by increasing or decreasing with a correction value, and a learning value updated and modified to reduce a deviation of the air-fuel ratio feedback correction value from a reference value. And a lean air-fuel ratio control means for controlling the air-fuel ratio of the actual engine intake air-fuel mixture to a target lean air-fuel ratio under predetermined operating conditions. In the air-fuel ratio control device for an internal combustion engine, the air-fuel ratio learning means has two learning values depending on whether or not the evaporated fuel adsorbed by the adsorption means is being desorbed when performing the air-fuel ratio learning. However, when the difference between the learning value when the vehicle is not leaving and the learning value when the vehicle is leaving is larger than a predetermined value, a lean air-fuel ratio control inhibiting means is provided for inhibiting the lean air-fuel ratio control.

Further, the invention according to claim 2 is to set the predetermined value, which is used when the two lean values are compared in the lean air-fuel ratio control prohibiting means, to be variable based on the load of the engine. Characterize. According to a third aspect of the present invention, the accumulated desorption amount detecting means for detecting the accumulated desorption amount of the evaporated fuel from the adsorbing means, and the desorption of the evaporated fuel based on the accumulated desorption amount are temporarily prohibited, It is characterized in that learning control is performed by using the learning value when not leaving.

[0008]

According to the first aspect of the present invention, independent learning control is performed by using different learning values for non-purging of vaporized fuel and purging. Then, when the air-fuel ratio is greatly enriched by the purge and the learning value for purging decreases and the difference from the learning value for non-purging is a predetermined value or more, the influence on the air-fuel ratio of the purge gas is large, so Even if the lean air-fuel ratio control conditions are satisfied, the lean air-fuel ratio control is prohibited.

According to the second aspect of the present invention, the purge rate changes according to the intake negative pressure that changes with the load of the engine, and the effect on the air-fuel ratio also changes depending on the purge rate. Therefore, a condition for prohibiting the lean air-fuel ratio control is set. By setting it variably according to the load, an appropriate prohibition condition can be obtained. In the invention according to claim 3, the amount of evaporated fuel adsorbed to the adsorbing means can be estimated by the accumulated desorption amount from the adsorbing means,
Since the purge rate also changes depending on the amount of evaporated fuel, an appropriate prohibition condition can always be obtained by variably setting the condition for prohibiting the lean air-fuel ratio control depending on the cumulative desorption amount.

[0010]

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 2, an intake passage 2 of an engine 1 is provided with an air flow meter 5 for detecting an intake air flow rate Q sucked through an air cleaner 4 and a throttle valve 6 for controlling the intake air flow rate Q in conjunction with an accelerator pedal. ing. An electromagnetic fuel injection valve 7 for injecting and supplying fuel for each cylinder is provided in a manifold portion downstream of the throttle valve 6.

Further, in the exhaust passage 3 of the engine 1, an oxygen sensor 8 is provided as an air-fuel ratio detecting means for detecting the air-fuel ratio of the intake air-fuel mixture by detecting the oxygen concentration in the exhaust gas at the manifold collecting portion, and the downstream thereof. On the side, a three-way catalyst 9 is provided as an exhaust gas purification catalyst that purifies the exhaust gas by maximizing the oxidation of CO and HC and the reduction of NO _x in the exhaust gas near the stoichiometric air-fuel ratio.

The distributor 10 has a built-in crank angle sensor 11, and the control unit 50 counts a crank unit angle signal output from the crank angle sensor 11 in synchronization with engine rotation for a certain period of time. Alternatively, the cycle of the crank reference angle signal is measured to detect the engine rotation speed Ne. The control unit 50 calculates a fuel amount corresponding to the target air-fuel ratio from the fuel injection valve 7 based on the values detected by the various sensors by a method described later, and has a pulse width corresponding to the fuel amount. The injection pulse signal is output to the fuel injection valve 7. Fuel injection valve 7
The valve is driven by the injection pulse signal to inject and supply the fuel, which is pressure-fed from a fuel pump (not shown) and is controlled to a predetermined pressure by the pressure regulator. The air-fuel ratio is controlled by controlling the injection amount.

By the way, a purge passage 13 is provided which connects the space above the liquid surface of the fuel tank 12 and the downstream portion of the throttle valve 6 of the intake passage 2 of the engine 1, and the purge passage 13 is provided in the purge passage 13. A canister 14 as an adsorbing means capable of temporarily adsorbing the evaporated fuel generated in 12 or the like is interposed. A purge control valve 15 is provided downstream of the canister 14 in the purge passage 13, and the purge control valve 15 is opened based on a signal from the control unit 50 when the engine 1 is in a predetermined operating state. As a result, the canister 14 is connected to the engine 1
Intake negative pressure is introduced, the adsorbed fuel vapor is released from the canister 14, and the fuel vapor (hereinafter,
It is called purge gas. ) Is sucked.
This structure is the evaporated fuel evaporation prevention device.

The throttle valve 6 is provided with a throttle sensor 60 for detecting the opening of the throttle valve 6 and an idle switch 61 which is turned on when the throttle valve 6 is fully closed, and the engine 1 detects the cooling water temperature. A water temperature sensor 62 is provided, and a vehicle speed sensor 63 that detects the vehicle speed is provided. The control unit 50 having functions as air-fuel ratio feedback control means, air-fuel ratio learning means, lean air-fuel ratio control means, lean air-fuel ratio control prohibiting means includes CPU, ROM, RAM, A /
The microcomputer is provided with a D converter, an input / output interface, and the like, receives input signals from various sensors, and performs arithmetic processing according to a program on the ROM shown in each flowchart.

A lean operation approval / disapproval determination routine in this embodiment is shown in FIGS. In this embodiment, the learning performed during non-purging is referred to as base learning, and the learning performed during purging is referred to as map learning. In addition, since purging is prohibited during base learning, the same learning value should be updated / referenced in all operating regions as the base learning value in order to prevent the purge prohibition time from increasing due to the learning not slowly converging. Therefore, as long as the learning condition is satisfied, the base learning proceeds regardless of the driving state. On the other hand, as the map learning value, a plurality of learning values are assigned in a map according to the driving state. In the flow shown in the figure, when a predetermined operating condition is satisfied by making a determination for each operating condition, and a determination is made as to the degree of progress of each learning, when both base learning and map learning have converged. , The determination is made by comparing the base learning value and the map learning value. here,
Of the map learning values assigned in a map according to the driving state, the learning value to be compared with the base learning value is specified in advance, and the difference between the base learning value and the specified map learning value is The lean operation is permitted when the value is within the predetermined value, and the lean operation is not permitted in other cases.
It is desirable to specify a learning value corresponding to an operation region in which lean operation is frequently performed in normal operation, as a map learning value for performing comparison determination.

The flow contents will be specifically described below with reference to FIG. 3. When the idle switch 61 is in a non-idle state (S1, S2) when the state is not detected, the cooling water temperature TW detected by the water temperature sensor 62 is set to the lower limit. When it is in the range between the value TWL and the upper limit set value TWH (S3, S4), the basic fuel injection amount TP detected as the representative value of the load is the lower limit set value TPL.
When the engine speed NE detected by the crank angle sensor 11 is in the range of the upper limit set value TPH (S5, S6), the engine speed NE is in the range of the lower limit set value NEL to the upper limit set value NEH (S
7, S8), when the throttle valve opening TVO detected by the throttle sensor 60 is equal to or less than the set value TVH (S9, S10),
The vehicle speed VSP detected by the vehicle speed sensor 63 is the set value VSPL.
When it is above (S11, S12), when the vehicle speed change amount ΔVSP is less than or equal to the set value DVH (S13, S14), when the base learning convergence flag FBSLTD has a value indicating a convergence state (S15, S16). , Map learning convergence flag FLRNT
When D is a value indicating the converged state (S17, S18),
The region i (i = j) designated as described above among the operating regions set by dividing the base learning value LBSBU into a plurality of parts
1, j2, ...) difference from the map learning value TALLPi
When (LBSBU-TLALPi) is smaller than the set value LENALP, when all the above conditions are met,
Lean operation permission flag FLEAN that permits lean operation
Is set to 1 (S19), otherwise the lean operation permission flag FLEAN is set to 0 (S20).

Next, a learning routine as an air-fuel ratio learning means for setting the learning value shown in FIGS. 4 and 5 will be described. For both the base learning value and the map learning value, a learning counter that is incremented each time the learning value is updated is provided, and it is determined that the learning has converged when the learning counter exceeds a predetermined value. Since there is only one base learning value, the learning counter also has only one corresponding CBSLTD, and when this value exceeds the predetermined value NBSLTD, the convergence determination flag FBSLTD is set. A learning counter is assigned to each learning value for the map learning value, but only the counter corresponding to the learning region designated in advance performs the convergence determination. As a result, it is possible to shift to the lean operation without waiting for the learning value in the unused area to converge during the lean operation. The learning counters of all the designated areas have a predetermined value NLRNTD.
If so, the map learning convergence determination flag FLRNTD is set. Further, switching between base learning and map learning is performed by checking the base learning convergence flag FBSLTD.

The flow contents will be specifically described below with reference to FIGS. 4 and 5. First, the learning region for updating the map learning value by inputting the basic fuel injection amount TP and the engine speed NE in the current operating state. APLC is obtained, and the map learning value TALLPi is searched from the learning area APLC (S
31-S33). Next, when the cooling water temperature TWN is equal to or higher than the learning permission lower limit water temperature TWLCL and lower than the learning permission upper limit water temperature TWLCH, and it is determined that the air-fuel ratio feedback control is being performed, the feedback correction coefficient α and the output value VO of the oxygen sensor 8 are obtained. ₂ and input the output value VO ₂ reference value S
It is determined whether or not the magnitude relationship with LO ₂ is inverted with respect to the previous value (S34 to S38).

When it is judged that the air-fuel ratio has been inverted, it means that the increasing / decreasing direction of the air-fuel ratio feedback correction coefficient α has been inverted. Therefore, after the α inversion counter CJRC is incremented, Value AL
P2, basic fuel injection amount TP2, engine rotation speed NE2 are replaced with previous values ALP1, TP1, NE1, respectively, and α at the latest reversal, basic fuel injection amount TP, engine rotation speed NE are updated to the latest values ALP2, TP2, respectively. Replace with NE2 (S39 to S41).

Further, the α inversion counter CJRC becomes 2 or more, and the absolute value of the difference between the previous value and the latest value of each value at the time of inversion | ALP2-ALP1 |, | TP2-
TP1 |, | NE2-NE1 | are set values AL21, respectively.
When TP21 and NE21 or less and the value of the base learning convergence determination flag FBSLTD is the value (1) indicating the convergence state of the base learning value, the map learning value TALLPi of the searched learning area is calculated as follows. Correct and update according to the formula (S42 to S47).

TALLPi (updated value) = TLALPi
(Previous value) + G2 · {ALP1 + ALP2) /-1} Here, the value {ALP1 + ALP2) /-1} is the average value of the value of the air-fuel ratio feedback correction coefficient α at the previous reversal and the value at the current reversal This is a deviation from the value 1, and the value obtained by multiplying the value by the weight G2 is added to the previous learned value to correct and update.

In this way, the map learning value TALLPi
After the learning is completed, the map learning counter CLRNTDi of the corresponding area i is incremented (S48). When the value of the flag FBSLTD is the value (0) indicating the non-convergence state of the base learning value in S46, the searched base learning value LBSBU is modified and updated by the following equation (S49).

LBSBU = LBSBU + G3. {(AL
P1 + ALP2) / 2−1} The learning method is also the same as the learning method of the map learning value. Then, the base learning counter CBSLTD is incremented (S50). After such learning, the cumulative rotation number counter CO21NV of the engine during the reversal of α is reset to 0, and the output value VO2 of the oxygen sensor 8 at the present reversal is set to the previous value VO2 for the next calculation. Replace with OLD (S51,
S52). When it is determined in step 35 that the cooling water temperature TWN is not within the predetermined range, when it is determined in S36 that the air-fuel ratio feedback control is not in progress, and when it is determined in S38 that the oxygen sensor 8 is not reversing. Resets the α inversion counter CJRC, and at the same time, the counter CO2
After integrating 1 NV (S59, S60), proceed to S52.

Then, the base learning counter CBSLTD
It is determined whether or not the value of has reached the base learning convergence determination number NBSLTD. When it is determined that the value of has reached the base learning convergence determination number, that is, the base learning has converged, the base learning convergence determination flag FBSL
When TD is set to 1 and it is determined that NBSLTD is not reached and the base learning is not converged, the base learning convergence determination flag FBSLTD is set to 0 (S53 to S5
Five) . As described above, this value is used in S46 when selecting the base learning and the map learning, and also used as the condition for the lean operation permission / prohibition determination in FIG.

Further, it is determined whether or not the value of the map learning counter CLRNTDi has reached the map learning convergence determination number NLRNTD in all of the designated areas for map learning, and the value has been reached, that is, the map learning in all areas has converged. If it is determined that the map learning convergence determination flag FL
If RNTD is set to 1, and if it is determined that the convergence of the map learning in all areas is not completed otherwise,
The map learning convergence determination flag FLRNTD is set to 0 (S56 to S58). As described above, this value is used as the condition for the lean operation approval / disapproval determination in FIG.

FIG. 6 shows a routine for calculating the fuel-air ratio correction coefficient for lean operation. Based on the figure,
The lean operation permission flag FLEAN set in FIG.
(S61), the target fuel-air ratio TDML is searched from the lean fuel-air ratio map when the lean operation is 1, which is 1, (S62), and then the fuel-air ratio correction coefficient DML is set to the target TDML.
Gradually decreases by ΔDML until it reaches (S6
3) On the other hand, the value of the lean operation permission flag FLEAN is 0.
If so, the target fuel-air ratio TDML is searched from the stoichiometric fuel-air ratio map (S64), and then the fuel-air ratio correction coefficient DML is set by gradually increasing by ΔDML until reaching the target TDML (S65).

When the value of the fuel-air ratio correction coefficient DML is 1.0, the feedback control is performed to the stoichiometric air-fuel ratio, so the air-fuel ratio feedback correction coefficient α is changed to perform the feedback control. When the value of the fuel-air ratio correction coefficient DML is set to a value other than 1.0, the feedback control based on the air-fuel ratio detection value of the oxygen sensor 8 is not performed, and the air-fuel ratio feedback correction coefficient α is clamped to 1.0. Then, feedforward control is performed so as to approach the target fuel-air ratio TDML (S66, S67).

FIG. 7 shows how the fuel-air ratio gradually changes. FIG. 8 is a routine showing the opening / closing control of the purge control valve. Explaining based on the figure, the idle switch 61 is on (S71, S72), and the cooling water temperature TW is TWCP.
Within the range of L ≦ TW ≦ TWCPH (S73, S74),
The basic fuel injection amount TP, which is the engine load, is TPCPL ≦ TP
Within the range of ≤TPCPH, the purge operation condition (S75, S76) is satisfied, and the base learning convergence flag FB
When SLTD is set to 1 and it is determined that the base learning has converged, the purge control valve 15 is opened to perform purging, and in other cases, the purge control valve 15 is closed to prohibit purging (S77 ~ S79).

FIG. 9 is a diagram showing changes over time of the purge gas concentration, the learned value and the air-fuel ratio when this embodiment is applied. After the start (t = 0), when the learning value update condition is satisfied, first, base learning is performed. During this period, purging is prohibited. In the figure, the air-fuel ratio deviates to the lean side due to product variations, but the air-fuel ratio converges to the theoretical air-fuel ratio as the base learning value increases due to learning control. When the base learning value is updated a predetermined number of times, the mode is switched to map learning, and the purge is permitted at the same time. If other purge permission conditions are satisfied at this time, the purge is started. The map learning value becomes smaller due to the effect of enrichment by the purge gas, and as the time elapses, the map learning value becomes smaller to a certain value and then becomes larger on the contrary. When the learning progresses and the difference between the base learning value and the map learning value becomes smaller than the predetermined value LEANALP, the lean operation is permitted, and if all other permission conditions are satisfied at this time, the lean operation is started.

FIG. 10 shows a portion in place of step 17 of the first embodiment shown in FIG. 3 in the lean operation permission / prohibition judging routine according to the second embodiment (embodiment of the invention of claim 2). Is. In this embodiment, a predetermined value LENLPi used at the time of comparison is allocated in a map shape corresponding to each of the map learning values of each designated area where the comparison with the base learning value is performed. The ratio of the purge gas to the intake air flow rate (hereinafter referred to as the purge ratio) varies depending on the operating state, and when the load is high and the suction negative pressure is small, the purge ratio also becomes small. In this case, since the purge rate is small, even if the concentration of the purge gas is high, the influence on the air-fuel ratio is smaller than that at the time of low load. For this reason, if the vehicle is left in an idle state after a long period of cruising operation and an extreme operation such as when the vehicle suddenly starts (high-load operation) is performed, the fuel vapor will be charged during idle, but in the low-load range. Even if the learning is not performed and the learning is performed under a high load, the purge rate is small, and therefore the richness is not so great. Therefore, the map learning value is not too small and the lean is permitted.

When the predetermined value LENALP is reduced in order to prohibit lean operation during the above operation,
The conditions may be too strict for the low load region, and it may be difficult to shift to lean operation. Therefore, the map setting is made so that the predetermined value LENALPi corresponding to the driving region can be variably set so that such an extreme driving can be dealt with. Specifically, the predetermined value L
ENALPi is set to be small in the high load region and large in the low load region in order to cope with the demand.

FIG. 11 shows a purge cumulative amount calculation routine according to the third embodiment (embodiment of the invention of claim 3). That is, since the base learning is performed while purging is prohibited, in order to make the limit on the purge amount as small as possible, in the first embodiment, once the convergence is achieved, the base learning is not performed until the engine is stopped. did. However, it is possible that the environment will change significantly due to long-term driving (for example, when moving from lowland to highland),
The base learning can be performed even after the first convergence judgment.

Explaining in detail with reference to the flow chart, first, the throttle valve opening TVO and the engine speed NE are input, and the throttle opening area ATVO is ATVO = ATH.
Calculate as {1-cos (TVO)} (S81, S82)
. Next, the calculated throttle opening area ATVO
The intake negative pressure (boost pressure) is searched from the map based on the engine rotational speed NE and the purge amount ΔPRGQ per unit time for the searched intake negative pressure (mmHg) is searched from the map (S83, S84). ).

The purge amount ΔPRGQ per unit time calculated in this way is integrated to calculate the cumulative purge amount PRGQ during purging, and it is judged whether or not this value has reached a predetermined value PGRBSL. Sometimes, the learning counter CBSLTD for determining the convergence of the base learning is cleared (S85 to S8
7). As a result, the base learning convergence determination flag FBSL
TD is also cleared (S53, S54 in FIG. 5), as a result of which purging is prohibited (S15 → S19 in FIG. 3), the base learning is restarted.

[0035]

As described above, according to the first aspect of the present invention, the learning value for non-purging and the learning value for purging are separately provided, and when the difference between them is equal to or more than a predetermined value. Since the lean air-fuel ratio control is prohibited, the lean air-fuel ratio control is prohibited only when the effect of the purge gas on the air-fuel ratio is large, and the lean air-fuel ratio control prohibition condition can be kept to the necessary minimum. Exhaust purification performance can be improved.

According to the second aspect of the present invention, the condition for prohibiting the lean air-fuel ratio control is set variably depending on the load, so that an appropriate prohibition condition corresponding to the purge rate that changes according to the load can be obtained. In the invention according to claim 3, since the condition for prohibiting the lean air-fuel ratio control is set by the accumulated desorption amount from the adsorbing means, an appropriate prohibition condition corresponding to the purge rate changing to the accumulated desorption amount can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram according to the present invention.

FIG. 2 is an overall configuration diagram of an embodiment according to the present invention.

FIG. 3 is a flowchart showing a lean operation approval / disapproval determination routine in the embodiment.

FIG. 4 is a flowchart showing a first stage of an air-fuel ratio learning routine in the embodiment.

FIG. 5 is a flowchart showing the latter stage of the air-fuel ratio learning routine.

FIG. 6 is a flowchart showing a fuel-air ratio correction coefficient calculation routine in the embodiment.

FIG. 7 is a time chart showing how the above fuel-air ratio correction coefficient gradually changes.

FIG. 8 is a flowchart showing a control routine of the purge control valve in the above embodiment.

FIG. 9 is a time chart showing changes in purge gas concentration, learned value, and air-fuel ratio in the same embodiment.

FIG. 10 is a flowchart showing only a changed portion of the lean operation permission / prohibition determination routine according to the second embodiment from the first embodiment.

FIG. 11 is a flowchart showing a cumulative departure amount calculation routine according to the third embodiment.

[Explanation of symbols]

1 organization 2 Intake passage 3 exhaust passage 5 Air flow meter 7 Fuel injection valve 8 oxygen sensor 11 Crank angle sensor 50 control unit 60 Throttle sensor 61 Idle switch

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁷ Identification code FI F02D 45/00 340 F02D 45/00 340F F02M 25/08 301 F02M 25/08 301J (56) References JP-A-61-112755 ( JP, A) JP 3-121232 (JP, A) JP 2-67441 (JP, A) JP 6-221231 (JP, A) JP 6-280644 (JP, A) JP HEI 7-217470 (JP, A) JP-A-63-183450 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) F02D 41/02 F02D 41/14 F02D 45/00 F02M 25 / 08

Claims (3)

(57) [Claims] Claim: What is claimed is: 1. Evaporative fuel generated in a fuel tank is temporarily adsorbed by an adsorbing means, the adsorbing means is communicated with an engine intake system, and the evaporated fuel adsorbed by the adsorbing means is separated to remove the engine fuel intake system. On the other hand, an evaporative fuel transpiration prevention device is provided to guide and process the engine, while an operating state detecting means for detecting an engine operating state, an air-fuel ratio detecting means for detecting an air-fuel ratio of an engine intake air-fuel mixture, and an air-fuel ratio detecting means are provided. Air-fuel ratio feedback control means for feedback-controlling the air-fuel ratio by increasing or decreasing the basic control value of the air-fuel ratio by the air-fuel ratio feedback correction value so that the actual air-fuel ratio of the engine intake air-fuel mixture to be detected approaches the stoichiometric air-fuel ratio. An air-fuel ratio learning means for correcting the basic control value by using a learning value updated and corrected so as to reduce the deviation of the air-fuel ratio feedback correction value from the reference value; and a predetermined operating condition. And a lean air-fuel ratio control means for controlling the actual air-fuel ratio of the engine intake air-fuel mixture to a target lean air-fuel ratio, and an air-fuel ratio control device for an internal combustion engine, comprising: When performing the fuel ratio learning, there are two learning values depending on whether or not the evaporated fuel adsorbed by the adsorbing means is being desorbed, and the difference between the learning value at the time of non-departure and the learning value at the time of desorption. An air-fuel ratio control device for an internal combustion engine, comprising lean air-fuel ratio control inhibiting means for inhibiting the lean air-fuel ratio control when the value is larger than a predetermined value. 2. The lean air-fuel ratio control prohibiting means, wherein the predetermined value used when comparing two learning values is variably set based on the load of the engine. Air-fuel ratio controller for internal combustion engine. 3. A cumulative desorption amount detecting means for detecting a cumulative desorption amount of the evaporated fuel from the adsorbing means, and a temporary prohibition of the desorption of the evaporated fuel based on the cumulative desorption amount,
The air-fuel ratio control device for an internal combustion engine according to claim 1 or 2, wherein learning control is performed using the learning value during non-separation during this period.