JPH0942011A - Air-fuel ratio control device for internal combustion engine - Google Patents

Abstract

(57) 【Abstract】 PROBLEM TO BE SOLVED: To quickly and accurately estimate the concentration of evaporated fuel (evaporative gas), and to carry out stable air-fuel ratio feedback control. A canister 13 for adsorbing evaporated fuel is connected to a fuel tank 7, and the canister 13 and the engine 1 are connected to each other.
The purge valve 16 in the discharge passage 15 communicating with the intake pipe 2 of
Are arranged. The CPU 21 calculates the smoothed value by smoothing the past several times of the data of the feedback correction coefficient, and estimates the evaporation concentration based on the degree of change of the smoothed value to the fuel increase side or the fuel decrease side. Also, CPU
Reference numeral 21 adjusts the fuel injection amount by the injector 4 according to the feedback correction coefficient and the evaporation concentration (evaporation correction coefficient). In such a case, the CPU 21 predicts the change amount of the feedback correction coefficient according to the increase / decrease change amount of the evaporation concentration, and approximates the convergence time of the smoothed value to the convergence time of the feedback correction coefficient corresponding to the predicted change value. Correct the smoothed value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is an internal combustion engine in which fuel is injected by an injector and evaporated fuel (hereinafter referred to as evaporative gas) generated in a fuel tank is sucked into an intake system of an internal combustion engine for combustion. The present invention relates to an air-fuel ratio control device for an engine.

[0002]

2. Description of the Related Art As an air-fuel ratio control device of this type, an evaporated fuel release mechanism (evaporative purge mechanism) has been conventionally used. That is, a canister for adsorbing the evaporation gas generated in the fuel tank is connected to the fuel tank, and an electromagnetic purge valve is arranged in the discharge passage that connects the canister and the intake system of the internal combustion engine. Then, the evaporative gas adsorbed on the canister is discharged (purged) into the intake system of the internal combustion engine together with air as the purge valve opens.
It is mixed with the fuel injected by the injector and burned.

In such a case, the fuel injection amount by the injector is increased or decreased by the air-fuel ratio correction value (feedback correction coefficient FAF) according to the deviation between the actual air-fuel ratio and the target air-fuel ratio, and at the same time by the evaporative purge mechanism. The amount is corrected according to the amount of evaporative gas purge. At this time, it is an essential element to accurately detect the concentration of the evaporative gas (hereinafter referred to as the evaporative concentration) when correcting the reduction of the fuel injection amount according to the purge amount of the evaporative gas.

As a well-known method for detecting the evaporation concentration, a smoothed value obtained by smoothing the past several data of the air-fuel ratio correction value is used, and the degree of change of the smoothed value (from the reference value to the fuel increase side or the fuel decrease amount). There is known a technique of estimating the concentration by increasing or decreasing the evaporative concentration by a predetermined amount according to the degree of change to the side (also referred to as concentration learning). That is, when the air-fuel ratio is feedback-controlled, the air-fuel ratio correction value changes little by little between the increase amount (lean) and the decrease amount (rich) with respect to the reference value. Therefore, when detecting the evaporation concentration,
The smoothed smoothing of the air-fuel ratio correction value that changes in small steps is used to enable the evaporation concentration detection in a stable state.

In the above-described method for detecting the evaporation concentration, if the smoothed air-fuel ratio correction value is on the fuel reduction side (when the air-fuel ratio is near the rich side), the concentration is updated so as to increase and the air-fuel ratio correction is performed. If the rounded value is on the fuel increase side (when the air-fuel ratio is leaner), the concentration is updated to be smaller.

[0006]

However, the above-mentioned conventional techniques have the following problems. In other words, since the smoothed value changes with a predetermined delay with respect to the air-fuel ratio correction value, if the air-fuel ratio suddenly changes (abruptly changes to the rich side) due to the start of the evaporative purge, the evaporative purge will be delayed by a predetermined time. Follow the change in the air-fuel ratio correction value (in this case, change to the fuel reduction side). In this case, even if the air-fuel ratio correction value stabilizes at a predetermined value after the evaporation purge,
Due to the delay of the smoothed value, the fluctuation of the same value remains, and during that time, the evaporation concentration is excessively updated in the direction of increasing. Therefore, the evaporation concentration becomes excessively high, and immediately after that, it is necessary to update the concentration so as to decrease it. After that, the evaporation concentration repeats unnecessary changes (large ⇔ small) over several times, which causes overshoot.

The present invention has been made in view of the above problems, and its purpose is to quickly and accurately estimate the concentration of evaporated fuel (evaporative gas), and to obtain stable air-fuel ratio feedback. An object of the present invention is to provide an air-fuel ratio control device for an internal combustion engine that can perform control.

[0008]

In order to achieve the above object, the invention as set forth in claim 1 is such that, as shown in FIG. 15, an evaporative gas generated in a fuel tank is adsorbed by a canister and the canister is adsorbed. Evaporative Fuel Release Mechanism M2 for Ejecting Evaporative Fuel Adsorbed in the Air to the Intake System of Internal Combustion Engine M1
(Evaporative Purge Mechanism), Fuel Release Amount Adjusting Means M3 for Adjusting Evaporative Gas Release by Evaporative Fuel Release Mechanism M2, and Air-Fuel Ratio Sensor M4 for Detecting Air-Fuel Ratio of Mixture Supplying to Internal Combustion Engine M1. And an air-fuel ratio correction value calculation means M5 for calculating an air-fuel ratio correction value for eliminating the deviation between the air-fuel ratio detected by the air-fuel ratio sensor M4 and the target air-fuel ratio, and the air-fuel ratio correction value calculation means M5. Air-fuel ratio average value calculating means M6 for calculating the air-fuel ratio average value by smoothing the past several times of the corrected air-fuel ratio correction values.
When the evaporation amount is discharged by the fuel discharge amount adjusting means M3, the degree of change of the air-fuel ratio average value calculated by the air-fuel ratio average value calculating means M6 to the fuel increasing side or the fuel decreasing side (for example, from a reference value). Based on the deviation amount), the concentration of the evaporative gas released from the evaporated fuel release mechanism M2 is increased or decreased by a predetermined amount, thereby estimating the actual concentration value, and the air-fuel ratio correction value calculation unit M5.
Air-fuel ratio correction value calculated by
Air-fuel ratio control means M9 for adjusting the fuel injection amount by the injector M8 according to the concentration of the evaporative gas estimated by
In the air-fuel ratio control apparatus for an internal combustion engine, the change predicting the change amount of the air-fuel ratio correction value by the air-fuel ratio correction value calculating unit M5 in accordance with the increase / decrease change amount of the concentration of the evaporative gas by the concentration estimating unit M7. The amount prediction means M10 and the air-fuel ratio in order to bring the convergence time of the air-fuel ratio average value closer to the convergence time of the air-fuel ratio correction value corresponding to the change amount of the air-fuel ratio correction value predicted by the change amount prediction means M10. The gist is that the air-fuel ratio average value correction means M11 for correcting the average value is provided.

In short, according to the structure corresponding to the characteristic part of the present invention, the change amount of the air-fuel ratio correction value is predicted (change amount) according to the increase / decrease change amount of the evaporation concentration estimated by the concentration estimating means M7. Prediction means M10). That is, when the air-fuel ratio feedback control is performed, there is a predetermined correspondence between the change in the evaporation concentration and the change in the air-fuel ratio correction value at the time of the change, and the air-fuel ratio correction according to the concentration change amount is performed. The amount of change in value can be predicted.

On the other hand, in order to bring the convergence time of the air-fuel ratio correction value closer to the convergence time of the air-fuel ratio correction value, the air-fuel ratio average value is corrected (air-fuel ratio average value correction means M11). That is, since the average air-fuel ratio value is a smoothed value obtained by smoothing the past several times of the air-fuel ratio correction value (normalized value), the average air-fuel ratio value is usually a predetermined delay from the change in the air-fuel ratio correction value. Change with. Therefore, in the conventional technique, the evaporation concentration estimated according to the change state of the average value of the air-fuel ratio excessively changes even after the stabilization of the air-fuel ratio correction value, which causes overshoot. On the other hand, in this configuration, since the average value of the air-fuel ratio is predicted and corrected, the update of the excessive evaporation concentration (concentration learning) is avoided, the overshoot of the same concentration value is eliminated, and a stable air-fuel ratio is eliminated. Feedback control can be realized.

According to a second aspect of the invention, in the first aspect of the invention, the air-fuel ratio average value correcting means M11 is provided.
When the concentration value of the evaporative gas estimated by the concentration estimating means M7 increases, the change amount estimating means M10
When the air-fuel ratio average value is corrected to the fuel increase side based on the change amount of the air-fuel ratio correction value predicted by the above, and when the concentration value of the evaporative gas estimated by the concentration estimating means M7 decreases, the change is performed. The gist is to correct the air-fuel ratio average value to the fuel reduction side based on the amount of change in the air-fuel ratio correction value predicted by the amount predicting means M10.

That is, according to the structure of claim 2, when the evaporation concentration increases, the average air-fuel ratio is corrected to the fuel increase side based on the change amount of the air-fuel ratio correction value, and the evaporation concentration decreases. In the above, the air-fuel ratio average value is also corrected to the fuel reduction side based on the amount of change in the air-fuel ratio correction value, so that the convergence timing of the air-fuel ratio correction value and the air-fuel ratio average value is changed as described in claim 1 above. Get closer. as a result,
The evaporation concentration is detected without causing overshoot, and thus stable air-fuel ratio feedback control can be realized.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the air-fuel ratio control device of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing a schematic configuration of an air-fuel ratio control system for an internal combustion engine. In FIG. 1, a multi-cylinder internal combustion engine (hereinafter referred to as an engine) 1 is mounted on a vehicle, and an intake pipe 2 and an exhaust pipe 3 are connected to the engine 1. An electromagnetic injector 4 is provided at the inner end of the intake pipe 2, and a throttle valve 5 is provided upstream of the electromagnetic injector 4.
The exhaust pipe 3 is provided with an oxygen sensor 6 as an air-fuel ratio sensor that outputs a voltage signal according to the oxygen concentration in the exhaust gas.

A fuel supply system for supplying fuel to the injector 4 has a fuel tank 7, a fuel pump 8, a fuel filter 9 and a pressure regulating valve 10. Then, the fuel (gasoline) in the fuel tank 7 is sucked up by the fuel pump 8 and is pressure-fed to each injector 4 via the fuel filter 9. Further, the fuel supplied to each injector 4 is adjusted to a predetermined pressure by the pressure regulating valve 10.

Next, the structure of the evaporated fuel release mechanism (evaporative purge mechanism) will be described. A purge pipe 11 is provided above the fuel tank 7, and the purge pipe 11 is connected to the surge tank 12 of the intake pipe 2. Purge pipe 1
In the middle of 1, a canister 1 containing activated carbon as an adsorbent for adsorbing the evaporative gas generated in the fuel tank 7
3 are provided. The canister 13 is provided with an atmosphere opening hole 14 for introducing outside air. Purge pipe 1
Reference numeral 1 denotes a discharge passage 15 on the surge tank 12 side of the canister 13, and a purge valve 16 composed of a variable flow solenoid valve is provided in the discharge passage 15 in the middle thereof. In this embodiment, the purge valve 16 corresponds to the fuel discharge amount adjusting means.

In the purge valve 16, the valve body 17 is always biased by a spring (not shown) in the direction of closing the seat portion 18, but by exciting the coil 19, the valve body 17 is moved in the direction of opening the seat portion 18. It has become. That is, the purge valve 16 opens the discharge passage 15 by exciting the coil 19 and closes the discharge passage 15 by demagnetizing the coil 19. The opening / closing operation of the purge valve 16 is performed by the CPU described later.
The duty ratio is controlled based on the pulse width modulation by 21.

Therefore, if a control signal is supplied from the CPU 21 to the purge valve 16 to connect the canister 13 and the intake pipe 2 of the engine 1, fresh air is introduced into the canister 13 through the atmosphere opening hole 14, Fresh air canister 13
Ventilate the inside. At this time, the evaporation gas is sent from the intake pipe 2 into the cylinder of the engine 1 to perform canister purge, and the adsorption function of the canister 13 can be recovered. As shown in the characteristic diagram of FIG. 2, the amount of purge air accompanying the introduction of fresh air is adjusted according to the duty ratio of the pulse signal supplied from the CPU 21 to the purge valve 16. FIG. 2 shows the characteristics when the negative pressure in the intake pipe 2 is constant. From this characteristic diagram, it can be seen that the purge air amount increases almost linearly as the duty ratio of the purge valve 16 increases from 0%.

Further, the throttle valve 5 has a throttle sensor 22 for detecting the opening of the valve 5, and the surge tank 12 has a pressure (absolute pressure) of intake air passing through the throttle valve 5.
An intake pressure sensor 23 for detecting the temperature is provided, and a water temperature sensor 24 for detecting the temperature of the cooling water is provided in the cylinder block of the engine 1. In addition to the throttle opening signal, the intake pressure signal, and the cooling water temperature signal from each of the above-mentioned sensors, the CPU 21 has an engine speed signal from a speed sensor (not shown) and an intake temperature from an intake temperature sensor (not shown). A signal and an atmospheric pressure signal from an atmospheric pressure sensor (not shown) are input.

The CPU 21 calculates the intake pressure PM, the cooling water temperature THW, the engine speed NE, the intake temperature THA, the atmospheric pressure PA, etc. on the basis of the respective detection signals, and uses these data as RA.
It is temporarily stored in M26. A backup RAM that stores and retains data even when the power is cut off in a part of the RAM 26
(Not shown) is configured. Note that, for example, instead of the intake pressure signal from the intake pressure sensor 23, the intake air amount signal from the intake air amount sensor is input to the CPU 21, or the intake pressure signal before engine start is used as the atmospheric pressure signal in the CPU 2.
You can also type in 1.

Further, the ROM 25 stores calculation programs and various maps for controlling the operation of the entire engine. Then, the CPU 21 executes the air-fuel ratio control based on the calculation program and the map in the ROM 25.
In other words, the CPU 21 is rich in the oxygen sensor 6
Feedback correction factor FA based on the lean judgment result
F (air-fuel ratio correction value) is calculated and the canister 13
To calculate the purge correction coefficient FPRG according to the purge amount of the evaporative gas into the intake pipe 2. Then, the basic injection time Tp calculated according to the engine operating state (rotation speed, intake pressure, etc.) is corrected by these correction factors, etc., to obtain the final injection time τ, and the injector 4 at a predetermined injection timing.
Fuel injection by. In the present embodiment, C
The PU 21 constitutes an air-fuel ratio correction value calculation means, an air-fuel ratio average value calculation means, a concentration estimation means, an air-fuel ratio control means, a change amount prediction means, and an air-fuel ratio average value correction means.

The operation of the air-fuel ratio control device configured as described above will be described below with reference to FIGS. 4 to 14.
In the flow chart used in this embodiment,
4 and 5 show an air-fuel ratio learning control routine as a base routine by the CPU 21, and FIGS. 6 and 7 show a purge rate calculation routine as a subroutine of FIGS.
The evaporation concentration calculation routine is shown. 8 shows an air-fuel ratio feedback control routine, FIG. 9 shows a fuel injection control routine, and FIG. 10 shows a purge valve control routine. These routines are executed by the CPU 21 at a predetermined interrupt timing.

First, the overall control operation by the routines of FIGS. 4 to 10 will be briefly described. In the routine of FIGS. 4 and 5, steps 102 and 103 and steps 116 and 117 are performed.
Learning is executed at. At this time, the air-fuel ratio deviation amount for each operating state of the engine 1 is obtained, and the learning correction value FLRN for correcting the deviation amount is stored in the backup RAM of the RAM 26. When the learning completion condition is not satisfied for a predetermined time during the learning execution, the learning correction value FLR
It is considered that N sticking or the like has occurred, learning is temporarily stopped, and purge control is forcibly executed to ensure a minimum purge amount. In addition, when the learning completion condition is satisfied within the predetermined time, the purge control is performed based on the purge rate RPRG and the evaporation concentration FLPRG calculated according to the air-fuel ratio.

Further, during the purge control shown in FIGS.
The routines of FIGS. 6 and 7 are executed, and the purge rate RPRG and the evaporation concentration FLPRG (estimated concentration) are calculated according to the air-fuel ratio. Here, the purge rate RPRG (%) indicates the ratio of the purge flow rate GPRG of the evaporative gas to the intake air amount GA in the intake pipe 2 (RPRG = GPRG / G
A) and evaporation concentration FLPRG (%) indicate the ratio of the fuel contained in the evaporation gas per 1% of the purge rate. Furthermore, during the purge control as well, the purge valve 16 is driven at a predetermined duty ratio by the routine of FIG.

Further, in the routine of FIG. 8, the feedback correction coefficient FAF is calculated. In the routine of FIG. 9, the basic injection time Tp is calculated and the basic injection time Tp is calculated.
On the other hand, feedback correction, air-fuel ratio learning correction, and purge correction according to the evaporation purge are performed to calculate the final injection time τ by the injector 4.

The following conditions (a) to (f) are mainly set as the execution conditions (feedback conditions) for the air-fuel ratio feedback in the following routines. Suppose the conditions are met. (B) Do not start. (B) The fuel is not being cut. (C) Cooling water temperature THW ≧ 40 ° C.
To be. (D) τ> τ min (however, τ
min is the minimum injection time of the injector 4). (E) The oxygen sensor 6 is in an active state. (F) Not under high load and high rotation.

Further, in the following routine, the feedback correction coefficient FAF is smoothed for each skip or every predetermined time, and the value is used as the smoothed value FAFAV. Specifically, the smoothed value FAFAV is calculated by the following equation (1).

FAFAV = {FAFAVi-1 .n + FAF. (256-n)} / 256 (1) Here, the subscript "i-1" indicates the previous value of the smoothed value FAFAV, The symbol n is a constant for detecting the degree of smoothing. That is, the smoothed value FAFAV is calculated by smoothing the data of the past several times of the feedback correction coefficient FAF (air-fuel ratio correction value), and in the present embodiment, the smoothed value FAFAV is the “air-fuel ratio average value”. Value.

The absolute value of the difference between the smoothed value FAFAV and the reference value (= 1) of the feedback correction coefficient FAF is used as the deviation ΔFAF of the feedback correction coefficient FAF (ΔFAF = | FAFAV-1 |).

Specific processing contents of each routine will be described below in order from the air-fuel ratio learning control routine of FIGS. When the CPU 21 is powered on and the routine is started, the CPU 21 first executes step 10
At 1, the air-fuel ratio learning condition is determined. This air-fuel ratio learning condition includes the above-mentioned feedback condition and water temperature condition (THW
> 80 ° C.) and the like. If the learning condition is satisfied, the CPU 21 executes air-fuel ratio learning (update of the learning correction value FLRN) in step 102, and then determines in step 103 whether or not the initial learning completion condition is satisfied. . The initial learning completion condition in the present embodiment means the deviation ΔFAF (= of the feedback correction coefficient FAF.
| FAFAV-1 |) is stable within 2% (the smoothed value FAFAV is stable with respect to the reference value), indicating that the feedback correction coefficient FAF has been skipped 12 times.

At the time of initial learning, the CPU 21 determines in step 104 whether or not the elapsed time from the start of initial learning is within a predetermined time (60 seconds in this embodiment). That is, the learning correction value FLRN is generally provided with upper and lower limits, and when the learning correction value FLRN sticks to the upper and lower limits, the air-fuel ratio does not converge to the target value, so that the affirmative determination is made in step 104. Then, if the determination in step 104 is affirmative, the CPU 21 proceeds to step 105, and in steps 105 to 108, the purge rate R
Executes fixed PRG control.

Specifically, the CPU 21 causes the step 105
In step 106, the purge execution flag XPRG is set to "1", and the purge rate RPRG is fixed to a predetermined value (for example, RPRG = 1%) in step 106. The purge execution flag XPRG is a flag for determining whether or not the evaporative purge by the purge valve 16 is executed, and XPRG =
“1” indicates that purging is executed, and XPRG = “0”
Indicates that purging is not executed. In the above process, the purge rate RPRG is set as the purge execution flag XPRG is set.
Evaporative purging is performed to obtain the minimum purge by the fixed control of.

Then, when step 107 is satisfied,
That is, when step 106 is continued for 40 seconds, the CPU 2
1 resets the purge execution flag XPRG to “0” in step 108, and then returns to step 102.
After that, the CPU 21 executes steps 102 to 104 again, and when the sticking of the learning correction value FLRN is eliminated and step 103 is satisfied, the process proceeds to step 109.

On the other hand, the CPU 21 calculates the purge rate RPRG in step 109, and calculates the evaporation concentration FLPRG in step 110. Here, step 10
9 corresponds to the purge rate calculation routine of FIG. 6, and step 110 corresponds to the evaporation concentration calculation routine of FIG. 7, the details of which will be described later. At this time, since the air-fuel ratio has converged in the stable region, the evaporation concentration FLPRG depends on the purge rate RPRG, and the evaporation concentration FLPRG can be detected accurately.

Next, the CPU 21 executes steps 111 to 1
The purge continuation condition shown in 13 is determined. Specifically, CP
U21 determines in step 111 whether the intake air temperature THA is higher than 50 ° C. At this time, if THA> 50 ° C., the CPU 21 determines that the heat received in the fuel tank 7 increases and the amount of evaporative gas generation increases, and the process returns to step 109 to continue the purge control. Further, the CPU 21 determines in step 112 whether the evaporation concentration FLPRG is a value larger than 1%. At this time, FLPRG> 1%
If so, the CPU 21 determines that the amount of evaporation gas adsorbed by the canister 13 is large, and returns to step 109 to continue the purge control. Further, the CPU 21 executes step 1
In step 13, it is determined whether the elapsed time from the start of purging is within 120 seconds. If it is within 120 seconds, step 109
Return to and continue the purge control. That is, step 111
If any of the determinations to 113 is affirmatively determined, it is considered that the purge is necessary, and the purge control is performed with priority over the air-fuel ratio learning.

When all of the conditions of steps 111 to 113 are negatively determined, the CPU 21 determines that the purge is not necessary and proceeds to step 114 of FIG. 5 to reset the purge execution flag XPRG to "0" and continue. In step 115, the purge rate RPRG is reset to 0%.

Thereafter, the CPU 21 executes air-fuel ratio learning (updates the learning correction value FLRN) in step 116, and determines in step 117 whether or not the regular learning completion condition is satisfied. Here, the regular learning completion condition indicates that the feedback correction coefficient FAF has been skipped six times when the deviation ΔFAF is stable within 2%.

Further, at the time of regular learning, the CPU 21 determines in step 118 whether or not the elapsed time from the start of regular learning is within a predetermined time (40 seconds in this embodiment). When step 118 is affirmatively determined, the CPU 21 executes the control of fixing the purge rate RPRG in steps 119 to 122. That is, the CPU 21
Sets "1" to the purge execution flag XPRG in step 119, and fixes the purge rate RPRG to a predetermined value (eg, RPRG = 1%) in step 120. When it is determined in step 121 that 40 seconds have elapsed, the CPU 21 resets the purge execution flag XPRG to "0" in step 122, and then step 116.
Return to After that, the CPU 21 again executes steps 116 to 1.
18 is executed, and when step 117 is satisfied, the process returns to step 109 in FIG.

After that, the CPU 21 executes the above-mentioned step 10
9 to 122 are repeatedly executed. In short, the purge control including the data detection of the evaporation concentration FLPRG and the like greatly affects the accuracy of the injection amount correction. Therefore, in order to realize highly accurate air-fuel ratio control, the air-fuel ratio learning and the purge control are executed in a non-overlapping period, and the evaporation concentration FLPRG is set only in the stable air-fuel ratio region (stable region of the feedback correction coefficient FAF). Etc. data detection is performed. And
Evaporation concentration FLPRG detection period (Step 10
(When the learning completion condition of 3, 117 is satisfied), precise purge control according to the same concentration value is performed, and the evaporation concentration FLP is reduced.
During a period in which RG cannot be detected (when the learning completion condition of steps 103 and 117 is not satisfied), relatively rough purge control is performed as a temporary process.

Next, the purge rate calculation routine of FIG. 6 will be described. In FIG. 6, the CPU 21 executes step 201.
In step 202, it is determined whether or not the above-mentioned feedback condition is satisfied, and in step 202, it is determined whether or not the cooling water temperature THW> 80 ° C. Then, steps 201 and 20
When either of the two is negatively determined, the CPU 21 resets the purge execution flag XPRG to “0” in step 203, and also, in step 204, the purge rate RPRG.
Is reset to 0% and this routine ends.

If both steps 201 and 202 are affirmatively determined, the CPU 21 sets the purge execution flag XPRG to "1" in step 205, and then calculates the purge rate RPRG in steps 206 to 210. Specifically, the CPU 21 determines in step 206 that the deviation ΔFAF> 5.
It is determined whether or not it is%, and in step 207 the deviation ΔFA
It is determined whether F> 10%. In such a case, ΔF
If AF ≦ 5%, the CPU 21 proceeds to step 208 and increases the value of the purge rate RPRG by 0.05%. If 5% <ΔFAF ≦ 10%, the CPU 21 proceeds to step 209 and holds the purge rate RPRG at the value at that time. If ΔFAF> 10%, CPU21
Proceeds to step 210 and sets the value of the purge rate RPRG to 0.
05% reduction.

Finally, in step 211, the CPU 21 checks whether or not the purge rate RPRG is within the upper limit set in FIG. 3, and if the value exceeds the upper limit, holds it at the upper limit. 3 is a full-open purge rate map determined by the engine speed NE and the engine load (in this embodiment, the intake pressure PM, but may be the intake air amount or the throttle opening). The maximum purge rate when the duty ratio of the valve 16 = 100% is shown.

On the other hand, in the evaporation concentration calculation routine of FIG. 7, the CPU 21 first determines in step 301 whether the purge execution flag XPRG is "1". Then, if XPRG = "0", the CPU 21 ends the routine as it is. If XPRG = “1”, the CPU 21 determines in step 302 the deviation amount (= F) from the reference value (= 1) for the FAF smoothed value FAFAV.
AFAV-1). Where the averaged value FAFAV
Is a numerical value calculated by the above equation (1).

Also. The CPU 21 continues to step 303
The update width α of the evaporation concentration FLPRG is set to “2%”, and the update width α is multiplied by the purge rate RPRG to calculate the predicted change amount β of the feedback correction coefficient FAF (β = α · RPRG).

After that, the CPU 21 starts from step 304.
In step 308, the evaporation concentration FLPRG is calculated according to the deviation amount of the smoothed value FAFAV from the reference value, and in step 309 and 310, the smoothed value FAFAV is corrected according to the predicted change amount β of the feedback correction coefficient FAF. .

More specifically, the CPU 21 causes the step 304.
And the deviation amount of the above-mentioned smoothed value FAFAV (FAFAV-1)
Is greater than 2% (determination level for determining that the air-fuel ratio is leaner) or not. Also C
The PU 21 determines in step 305 that the amount of deviation (FAFAV
It is determined whether -1) is smaller than -2% (a determination level for determining that the air-fuel ratio is near the rich side).

In such a case, (FAFAV-1)> 2% (YES in step 304) means that the actual evaporation concentration is lower than the currently estimated evaporation concentration FLPRG, and the CPU 21 Step 306
The value of the evaporation concentration FLPRG is decreased by α (0.2%). Further, in the subsequent step 309, the CPU 21 decreases the smoothed value FAFAV by the predicted change amount β of the FAF corresponding to the decrease in the evaporation concentration FLPRG.

On the other hand, (FAFAV-1) <-2% (YES in step 305) means that the actual evaporation concentration is higher than the currently estimated evaporation concentration FLPRG, and the CPU 21 In step 307, the value of the evaporation concentration FLPRG is increased by α (0.2%).
Further, in the subsequent step 310, the CPU 21 increases the moderation value FAFAV by the predicted change amount β of the FAF corresponding to the increase in the evaporation concentration FLPRG.

Further, -2% ≤ (FAFAV-1) ≤2
% (Steps 304 and 305 are both N
O), actual evaporation concentration and current evaporation concentration FLP
It means that the RG and the
Holds the evaporation concentration FLPRG at the value at that time in step 308. After the calculation of the evaporation concentration, the CPU2
In step 1, 31, it is determined whether the evaporation concentration FLPRG is within the upper and lower limit value of 0 to 25%, and this routine is ended.

That is, according to the routine of FIG. 7, when the evaporation concentration FLPRG is increased or decreased, the feedback correction coefficient FA is increased according to the increase or decrease.
The motion of F is predicted, and the predicted amount (β) is the smoothed value FA
It will be reflected in FAV.

Next, the air-fuel ratio feedback control routine of FIG. 8 will be described. This routine is executed by the CPU 21 by interrupting every 4 msec. In FIG. 8, the CPU 21 first determines in step 401 whether or not the above-described feedback condition is satisfied. Then, if the feedback condition is not satisfied, the CPU 21 proceeds to step 402, where the feedback correction coefficient FAF =
1.0. When the feedback condition is satisfied, the CPU 21 proceeds to step 403, compares the oxygen sensor output with the predetermined determination level, and outputs them for the predetermined time H, respectively.
The air-fuel ratio flag XOXR is operated with a delay of I (msec). For example, when the output of the oxygen sensor 6 is on the rich side, XOXR = “1”, and on the lean side, XOXR =.
It is set to “0”.

Next, the CPU 21 proceeds to step 404 and operates the value of the feedback correction coefficient FAF based on the air-fuel ratio flag XOXR. That is, the air-fuel ratio flag X
When the OXR changes from "0" to "1" or from "1" to "0", the value of the feedback correction coefficient FAF is skipped by a predetermined amount, and the air-fuel ratio flag XOXR continues to "1" or "0". At this time, integral control of the feedback correction coefficient FAF is performed. Then, the CPU 21 proceeds to the next step 405 to check the upper and lower limits of the value of the feedback correction coefficient FAF, and thereafter ends this routine.

Next, the fuel injection control routine of FIG. 9 will be described. This routine is executed by the CPU 21 by interrupting every 4 msec. In FIG. 9, the CPU
21 calculates the basic injection time Tp according to the engine speed NE and the intake pressure PM based on the data stored as a map in the ROM 25 in step 501. Next, in step 502, the CPU 21 calculates a basic correction coefficient Fc corresponding to the correction coefficient (cooling water temperature, post-starting amount increase, intake air temperature, etc.) regarding the operating state of the engine 1, the feedback correction coefficient FAF, and the learning correction value FLRN. To do. Further, in the subsequent step 503, the CPU 21 calculates the purge correction coefficient FPRG by multiplying the evaporation concentration FLPRG calculated in the routine of FIG. 6 and the purge rate RPRG calculated in the routine of FIG. 5 (FPRG = FLP).
RG / RPRG).

Thereafter, the CPU 21 calculates the final injection time τ based on the basic injection time Tp, the basic correction coefficient Fc, the purge correction coefficient FPRG, and the invalid injection time Tv at step 504 (τ = Tp · (Fc− FPRG) + T
v). Then, the CPU 21 carries out fuel injection by the injector 4 at a predetermined fuel injection timing based on the final injection time τ.

Next, the purge valve control routine of FIG. 10 will be described. This routine is 100ms by CPU21
It is executed by a time interrupt every ec. In FIG. 10, the CPU 21 determines in step 601 whether or not the purge execution flag XPRG is “1” and in step 602 whether or not the feedback execution condition is satisfied (high load increase is not in progress). Distinguishing things,
It is also possible to determine that the oxygen sensor 6 is operating normally).
If XPRG = “0”, the CPU 21 proceeds to step 611 and sets the duty ratio for driving the purge valve 16 to 0%. On the other hand, when XPRG = “1”, the CPU 21 executes the duty ratio calculation process in steps 603 to 607 if the feedback condition is satisfied, and the duty ratio calculation process in steps 608 to 610 if the feedback condition is not satisfied. Perform arithmetic processing.

Specifically, in steps 603 to 607, the CPU 21 first reads the intake pressure PM in step 603 and also reads the engine speed NE in step 604. Then, in the following step 605, the CPU 21 multiplies the predetermined coefficient Ka, the engine speed NE and the intake pressure PM to calculate the intake air amount GA (GA = Ka · N).
E ・ PM).

Further, the CPU 21 determines in step 606 the intake air amount GA and the purge rate R obtained in the routine of FIG.
Multiply with PRG to calculate the purge flow rate GPRG (G
PRG = GA · RPRG). Subsequently, in step 607, the CPU 21 makes the duty ratio map of FIG. 11 based on the two parameters of the purge flow rate GPRG and the pressure difference between the atmospheric pressure PA and the intake pressure PM (hereinafter, this pressure difference is referred to as a gauge pressure). The drive duty ratio of the purge valve 16 is obtained by using this. When the value of each parameter takes an intermediate value of map values, the duty ratio is obtained by interpolation.

On the other hand, in steps 608-610,
The CPU 21 determines the intake pressure PM (absolute pressure) in step 608.
And the engine speed N in step 609.
Read E. Then, the CPU 21 continues to step 61.
At 0, the drive duty ratio of the purge valve 16 is obtained using the duty ratio map of FIG. 12 based on the two parameters of the engine speed NE and the intake pressure PM. FIG.
According to the map of No. 2, the engine is fully closed (duty ratio = 0%) or fully opened (duty ratio = 99.
6%) is selected.

Thereafter, the CPU 21 proceeds to step 612 and
The purge valve 16 is driven at the above duty ratio. That is, according to the routine of FIG. 10, when the feedback control is executed, the opening / closing control of the purge valve 16 is performed according to the purge rate RPRG according to FIG. 6 and the engine operating state (FIG. 11).
Purge control using the map of), and during open control,
Since the purge rate RPRG cannot be manipulated, the purge valve 16 is controlled at a fixed duty ratio (fully open or fully closed) according to the engine operating state (purge control using the map of FIG. 12).

Next, the operation peculiar to the present embodiment obtained by each of the above routines will be described more specifically with reference to the time charts of FIGS. 13 and 14. Note that FIG. 13 shows the FAF when the evaporation concentration FLPRG is increased by α%.
It is explanatory drawing which shows the prediction principle of a change, and the purge correction coefficient FPR accompanying the change of the said evaporation concentration FLPRG is shown in the same figure.
G, feedback correction coefficient FAF and final injection time τ
The movement of is shown. Further, FIG. 14 shows the evaporation concentration FLPRG.
14A and 14B show the movements of the feedback correction coefficient FAF and its smoothed value FAFAV, respectively, and FIG. 14A shows the movement in the present embodiment and FIG. 14B shows the movement in the conventional form shown for comparison.

That is, the entire period shown in FIG. 13 is the evaporation purge period, and at the time t1 in the figure, the evaporation concentration FLPR is
When G is forcibly increased by α%, the purge correction coefficient F
PRG is increased by β (= α · RPRG). At this time, since the fuel injection amount (final injection time τ) is temporarily reduced, the air-fuel ratio becomes lean and the feedback correction coefficient FAF increases to compensate for it. Note that, in FIG. 13, it is assumed that the engine operating state such as the water temperature and the accelerator opening is constant, and for the sake of convenience, after the time t1, the evaporation concentration FLPR is increased.
G holds a constant value.

After the increase of the evaporation concentration FLPRG, time t1
From t2 to t2, the feedback correction coefficient FAF gradually increases. Then, at time t2, the final injection time τ returns to the value before the change in the evaporation concentration while converging to a value corresponding to the increase in the concentration. In this case, the feedback correction coefficient F
The amount of change in AF matches the change width (predicted change value β) of the purge correction coefficient FPRG. That is, when the evaporation concentration FLPRG changes as described above, the amount of change in the feedback correction coefficient FAF thereafter can be predicted.

On the other hand, in FIGS. 14A and 14B, the evaporation purge is started at times t11 and t21, respectively, and the evaporation concentration FLPRG is converged to the actual concentration value at times t12 and t22. Specifically, in FIG. 14B showing the conventional form as a comparative example, since the above-described change prediction of the feedback correction coefficient FAF is not performed, the F
The movement of the smoothed value FAFAV is delayed with respect to the change of AF. As a result, the evaporation concentration FLPRG calculated based on the deviation amount of the smoothed value FAFAV is overshot (similarly, the purge correction coefficient FPRG also overshoots). Further, the feedback correction coefficient FAF is also disturbed by this overshoot, and the time required for density detection and stabilization of FAF becomes long.

On the other hand, FIG. 14 showing the present embodiment.
In (a), since the change in the above-described feedback correction coefficient FAF is predicted, the convergence time of the FAF and the convergence time of the smoothed value FAFAV substantially match. As a result, it is possible to quickly detect the concentration without causing the overshoot of the evaporation concentration FLPRG. Also, the feedback correction coefficient FAF is quickly stabilized. As described above in detail, in the air-fuel ratio control device of the present embodiment, the change amount of the feedback correction coefficient FAF (air-fuel ratio correction value) is predicted according to the increase / decrease change amount of the evaporation concentration FLPRG (FIG. 7, Change prediction value β in step 303). Further, the smoothing value FAFAV is corrected so that the convergence time of the feedback correction coefficient FAF becomes closer to the convergence time of the smoothed value FAFAV (average value of the air-fuel ratio). More specifically, when the evaporation concentration FLPRG increases, the moderation value FAFAV is corrected to the fuel increase side based on the predicted change value β of the feedback correction coefficient FAF, and when the evaporation concentration FLPRG decreases, The smoothed value FAFAV is corrected to the fuel reduction side based on the predicted change value β (steps 309 and 310 in FIG. 7).

According to this structure, the excess evaporation concentration F
Update of LPRG (concentration learning) can be avoided, and overshoot of the same concentration value is eliminated. As a result, the evaporation concentration FLPRG can be estimated quickly, and stable air-fuel ratio feedback control can be realized. Further, even when a large amount of evaporation purge is performed and the change in the air-fuel ratio becomes large, the overshoot of the evaporation gas concentration can be avoided.

As another configuration of this embodiment, a learning completion condition for determining completion of air-fuel ratio learning is set (step 103 in FIG. 4, step 11 in FIG. 5).
7) If the learning completion condition is not satisfied for a predetermined time or more, the air-fuel ratio learning is temporarily interrupted and the evaporative purge is forcibly performed (step 105 in FIG. 4).
˜108, steps 119 to 122 in FIG. 5). With this configuration, the problem that the evaporation purge is stopped for a long time and the evaporation gas cannot be adsorbed by the canister 13 is solved.

The present invention can be embodied in the following modes other than the above embodiment. In the above-described embodiment, the air-fuel ratio control device is embodied by combining the air-fuel ratio learning process and the duty control of the purge valve 16, but it can be arbitrarily changed as long as it satisfies the requirements of the present invention. For example, it can be embodied in combination with a device that simply performs the air-fuel ratio learning and the evaporative purge alternately at predetermined intervals, or a device that simply turns on and off the purge valve 16.

[0068]

According to the first and second aspects of the present invention,
It has an excellent effect that the concentration of the evaporated fuel (evaporative gas) can be estimated quickly and accurately, and thus stable air-fuel ratio feedback control can be performed. Further, even when a large amount of evaporation purge is performed and the change in the air-fuel ratio becomes large, the overshoot of the evaporation gas concentration can be avoided.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an air-fuel ratio control device in the present embodiment.

FIG. 2 is a diagram showing a characteristic of a purge air amount with respect to a duty ratio.

FIG. 3 is a map showing the upper limit of the purge rate when the purge valve is fully opened.

FIG. 4 is a flowchart showing an air-fuel ratio learning control routine.

FIG. 5 is a flowchart showing an air-fuel ratio learning control routine similar to FIG.

FIG. 6 is a flowchart showing a purge rate calculation routine.

FIG. 7 is a flowchart showing an evaporation concentration calculation routine.

FIG. 8 is a flowchart showing an air-fuel ratio feedback control routine.

FIG. 9 is a flowchart showing a fuel injection control routine.

FIG. 10 is a flowchart showing a purge valve control routine.

FIG. 11 is a map for obtaining a duty ratio.

FIG. 12 is a map for obtaining a duty ratio.

FIG. 13 is a time chart for explaining the operation of the embodiment.

FIG. 14 is a time chart for explaining the operation of the embodiment.

FIG. 15 is a block diagram corresponding to a claim.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Engine (multi-cylinder internal combustion engine), 2 ... Intake pipe, 4 ... Injector, 6 ... Oxygen sensor as an air-fuel ratio sensor, 7
... fuel tank, 13 ... canister, 16 ... purge valve as fuel release amount adjusting means, 21 ... air-fuel ratio correction value calculating means, air-fuel ratio average value calculating means, concentration estimating means, air-fuel ratio controlling means, change amount predicting means, A CPU as an air-fuel ratio average value correction means.

Claims (2)

[Claims] 1. A vaporized fuel releasing mechanism for adsorbing vaporized fuel generated in a fuel tank by a canister and releasing the vaporized fuel adsorbed by the canister to an intake system of an internal combustion engine; and the vaporized fuel releasing mechanism. A fuel discharge amount adjusting means for adjusting the discharge amount of the evaporated fuel, an air-fuel ratio sensor for detecting an air-fuel ratio of the air-fuel mixture supplied to the internal combustion engine, an air-fuel ratio detected by the air-fuel ratio sensor and a target air-fuel ratio. An air-fuel ratio correction value calculating means for calculating an air-fuel ratio correction value for eliminating a deviation from the fuel ratio, and an air-fuel ratio by smoothing data of the past several times of the air-fuel ratio correction value calculated by the air-fuel ratio correction value calculating means. An air-fuel ratio average value calculating means for calculating an average value, and when the evaporated fuel is released by the fuel release amount adjusting means, an actual air-fuel ratio average value calculated by the air-fuel ratio average value calculating means is used. The fuel injection amount by the injector according to the concentration estimating means for estimating the degree value, the air-fuel ratio correction value calculated by the air-fuel ratio correction value calculating means, and the concentration of the evaporated fuel estimated by the concentration estimating means. In an air-fuel ratio control device for an internal combustion engine having an air-fuel ratio control means, in accordance with the amount of increase or decrease in the concentration of evaporated fuel by the concentration estimating means, the change amount of the air-fuel ratio correction value by the air-fuel ratio correction value calculating means A change amount predicting means for predicting, in order to bring the convergence time of the air-fuel ratio average value close to the convergence time of the air-fuel ratio correction value corresponding to the change amount of the air-fuel ratio correction value predicted by the change amount predicting means, An air-fuel ratio control device for an internal combustion engine, comprising: an air-fuel ratio average value correction means for correcting an average fuel ratio value. 2. The air-fuel ratio average value correcting means determines the change amount of the air-fuel ratio correction value predicted by the change amount predicting means when the concentration value of the evaporated fuel estimated by the concentration estimating means increases. Correct the average value of the air-fuel ratio based on the fuel increase side based on the air-fuel ratio, and when the concentration value of the evaporated fuel estimated by the concentration estimating means decreases, the air-fuel ratio correction value estimated by the change amount estimating means The air-fuel ratio control device for an internal combustion engine according to claim 1, wherein the average value of the air-fuel ratio is corrected to the fuel reduction side based on the amount of change.