JPH0530977B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a fuel control method for electronically controlling fuel supplied to an internal combustion engine for an automobile, and in particular to supplying suitable fuel during transient periods. This invention relates to a fuel control method that allows for

[Background of the invention]

Conventional electronic internal combustion engine control methods using microcomputers etc. store data necessary for controlling fuel supply amount and ignition timing in advance, and control the rotational speed of the internal combustion engine (hereinafter also referred to as engine). A system that controls fuel supply amount and ignition timing based on stored data read out in response to changes in
A feedback method is known in which the combustion pressure of the engine, the state of exhaust gas, etc. are detected, and the operating state of the engine is controlled accordingly.

However, the former method has a problem in that it cannot adapt to disturbances caused by changes in characteristics of the engine or control system, and the latter method has a problem in that a sufficient control response speed cannot be obtained.

For example, taking the fuel supply amount as an example, when the load changes suddenly, such as during acceleration, there is a problem that the fuel supply amount deviates from the target value due to fuel adhesion, evaporation delay of the adhering fuel, etc.

In order to avoid this, a correction means such as acceleration increase is provided to make the fuel amount match the target value, but it has not been able to achieve a sufficient effect.

[Purpose of the invention]

The purpose of the present invention is to eliminate the above-mentioned drawbacks of the prior art, and to solve the problem that when the load suddenly changes such as during acceleration, the fuel supply amount deviates from the target value due to fuel adhesion and evaporation delay of the adhering fuel. It is an object of the present invention to provide a control method for controlling fuel in an internal combustion engine that avoids problems and provides control with sufficiently few errors even in transient operating conditions.

[Summary of the invention]

In order to achieve this object, the present invention includes a plurality of sensors that detect the operating state of the internal combustion engine, and controls the amount of fuel supplied to the internal combustion engine according to the signals of the sensors, as shown in FIG. 17, for example. In the control device for an internal combustion engine, the amount of fuel supplied to the internal combustion engine is at least equal to the amount of adhering fuel adhering to the intake passage;
It is corrected based on a fuel adhesion rate representing the state in which the supplied fuel adheres to the intake passage and an evaporation time constant representing the evaporation state of the adhering fuel, and further based on air-fuel ratio information representing the air-fuel ratio concentration in the exhaust gas. The amount of adhering fuel is corrected.

In this way, responsiveness and fuel control accuracy can be improved by constructing a liquid film model of the fuel supplied to the engine and modifying the components of this model using information on the air-fuel ratio in exhaust gas. .

[Embodiments of the invention]

Embodiments of the control device for an internal combustion engine according to the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an embodiment of the present invention. In FIG. 1, an internal combustion engine 10 supplies a mixture of fuel and air controlled by a fuel supply device 12 such as a carburetor or a fuel injection device. Supplied. Exhaust gas is exhausted to the atmosphere through the exhaust pipe 14. A catalytic converter 16 is provided in the exhaust pipe 14 . Delivery system 12 is generally not capable of providing a desired response to fuel determining input parameters over the full range of engine operating conditions. Also, the supply device 12
Changes in engine operating parameters such as temperature therefore change the air/fuel ratio. Therefore, the air/fuel ratio provided by the supply system in response to the fuel determining input parameters generally deviates from a predetermined value during engine operation. In particular, the evaporation delay of fuel is large.
During acceleration at low temperatures, the deviation from the predetermined value increases.

The air/fuel ratio supplied by supply system 12 is selectively controlled in closed loop or open loop by electronic control unit 18. This control is performed by the exhaust pipe 14
Air-fuel ratio sensor 2 installed to detect the exhaust gas of
0 output, and also in response to outputs from various sensors, including the engine speed sensor, engine temperature sensor.

In FIG. 2, electronic control unit 18 takes the form of a digital computer and is adapted to provide a constant frequency pulse width modulated signal to supply system 12 to adjust the air/fuel ratio. External fixed storage ROM
The microprocessor 24 controls the operation of the dispensing device 12 by executing an operating program stored in the microprocessor 24 . The microprocessor 24 takes the form of a combinational module and includes, in addition to random access memory (RAM) and a clock, conventional counters, registers, accumulators, flag flip-flops, etc., such as the Motorola Microprocessor.
MC-6802, 68000, etc. can be adopted. or,
It may take the form of a microprocessor that utilizes external RAM and a clock oscillator.

Microprocessor 24 is combination module 2
The supply device 12 is controlled by executing the operating program stored in the ROM section 26. Module 26 includes an input/output interface and a programmable timer. Module 26 is
It may be a Motorola MC-6846 combination module. The input conditions on which open-loop, closed-loop control is based are provided to the input/output interface of module 26. An analog signal, such as an air/fuel ratio signal from sensor 20, is provided to a signal conditioner 32, the output of which is coupled to an A/D converter, multiplexer 34.

The particular analog state to be sampled and converted is controlled by microprocessor 24 according to the operating program via address lines from the input/output interface of module 26. When commanded, the addressed state is converted to digital form and sent to the input/output interface of module 26 and stored in the RAM at a ROM pointed location.

The duty cycle modulated output is provided by a conventional input/output interface circuit 36. This circuit includes an output counter that provides output pulses to the supply device 12, for example via a driver circuit 37. This includes the clock signal from clock driver 38 and the 100% clock signal from the timer section of module 26.
Receives Hz signal. The output counter portion of circuit 36 has a register into which a binary number representing the desired pulse width is periodically inserted.

At a frequency of 100Hz, the number in the register is sent to the gate of a down counter, and this down counter is
The time is measured by the output of the clock driver 38,
The output pulse of the output counter section has a required time equal to the time required for the down counter to read back to zero.
The output pulse is provided by a flip-flop which is set when the number in the register enters the gate of the down counter and is reset by the run signal from the down counter when the number counts to zero. Circuit 36 includes an input counter that gates clock pulses into the counter and receives speed pulses from the engine distributor.

In FIG. 2, a persistent storage device 40 is also provided having memory locations, in the form of a RAM powered from the vehicle's battery. The remainder of the system receives power through the ignition switch so that memory contents are retained even when engine 10 is stopped. Alternatively, persistent storage 40 may take the form of a storage device that is capable of retaining stored contents even when power is not applied.

Microprocessor 24, combination module 2
6. The input/output interface circuit 36 and the persistent storage device 40 are interconnected by an address bus, a data bus, and a control bus. The microprocessor 24 reads the data and sends it to the combination module 26.
The operation of the supply device 12 is controlled by executing the operation program given in the ROM section of the supply device 12.

The fuel supply device 12, as shown in FIG.
Air flow meter 301, fuel injection valve 302, throttle valve 3
03, consists of an intake pipe 304, inputs the signal of the air flow rate sensor 301 to the electronic control unit 18, controls the opening time of the fuel injection valve 302, and supplies fuel from the fuel pump 305 to the intake pipe 304. . It also has a bypass passage 306 that opens downstream of the throttle valve 303, and in the middle of the bypass passage,
A valve 307 is provided to control the amount of air. This valve 307 is driven by an electromagnetic solenoid 308, to which an output pulse having a frequency of 100 Hz is applied from the driver circuit 37.

In FIG. 4, the program begins at point 42 by first energizing and providing power to the various circuits by closing the ignition switch of engine 10 and then proceeds to step 44. This step
At 44, various elements in the system are initialized. Registers, flags, flip-flops (FF), counters and individual outputs are initialized. The process then proceeds to step 46, where the duty cycle memory DCM is initialized according to the duty cycles stored in the four memory locations KAM ₀ , . . . KAM ₃ of the persistent memory 40 .

As shown in Fig. 5, the duty cycle storage device stores information about engine speed and engine load.
Its location has been determined. Furthermore, the location of the duty cycle KAM of the persistent storage device 40 is determined with respect to the engine speed and engine load, as shown in FIG. Also,
Five memory locations of persistent storage 40
The fuel deposition rate storage device DXM is initialized according to the fuel deposition rates stored in XM ₀ , . . . XM ₄ . As shown in FIGS. 7 and 8, the locations of XM and DXM are determined based on the engine load. Also, the evaporation time constant storage device DVM is initialized according to the evaporation time constants stored in the four memory locations VM ₀ , . . . VM ₃ of the persistent storage device. VM, DVM are shown in Figures 9 and 10.
As shown in the figure, its location is determined relative to the temperature of the engine.

In step 46 these values are DCM ₀ ...DCM ₁₅
It is used to initialize each. KAM ₀
are placed in each of DCM ₀ − DCM ₂ , DCM ₄ − DCM ₆ , and KAM ₂ is placed in each of DCM ₈ − DCM ₁₀ , DCM ₁₂ − DCM ₁₄
placed in In Figure 11, at step 48
When initializing the DCM from KAM, the numerical value of the persistent storage device 40 is determined by the point 50. If the validity of the contents of the storage device 40 is lost due to the removal of the battery, etc., in step 54 KAM ₀ ~
KAM ₃ is initialized to the calibration value stored in the ROM section of module 26.

Even when initializing DXM from XM and DVM from VM, the validity of XM and VM is determined at point 50, and if the validity is lost, XM and VM are initialized from the ROM of module 26 at step 54. The calibration values stored in the unit will be initialized.

If the cooling water temperature is smaller than the constant Tw at point 52, it is corrected using the temperature bias value shown in FIG. From steps 56 and 58 the program leaves the routine and
Proceed to step 60 in Figure 4. The program is now set to enable interrupt routines. This is provided by setting an interrupt enable flag in microprocessor 24.

After step 60, the program changes to a background loop 62 that repeats continuously. This may include exhaust recirculation control functions and diagnostic alarm routines. By performing step 46, KAM
contains information for feed system adjustments over the engine operating range, which information can be used in an open-loop manner during open-loop modes of operation to provide more precise control of the air-fuel ratio of the mixture delivered to the engine 10 during warm-up. It forms part of the calibration value that is performed.

The timer portion of module 26 generates a 100 Hz interrupt signal that interrupts background loop routine 62. For each interrupt, a 0.01 second interrupt routine is recorded at step 64 in FIG. 13 and the process proceeds to step 66. The pulse width of a register in the output counter portion of circuit 36 is shifted into the output counter to issue a control pulse. This pulse provides the desired duty cycle signal to adjust the electromagnetic solenoid 308 of the supply system 12 to provide the desired mixture to the engine 10.

The program proceeds to step 68 and a read routine is executed. During this routine, the individual inputs are recorded in RAM at ROM specified locations, the engine speed determined via the input counter section of circuit 36, the various inputs of the A/D converter, etc.
recorded in each ROM specified location in RAM. Next, proceed to step 70, KAM,
Memory locations corresponding to the current engine operating points of DCM, VM, DVM, XM, and DXM are determined.

Points for completing index number routine 70
Proceeding to step 118, the engine speed RPM stored from RAM in step 68 is read from RAM;
It is compared with a reference speed value SRPM stored in ROM. This SRPM is smaller than the idle speed, but
greater than cranking speed. RPM＜SRPM
If , it means that the engine has not been started, and the process proceeds to step 120, the operation inhibit mode.
At the RAM location specified by the ROM, the determination width of the pulse width modulation signal for controlling the supply device 12, which is stored to store the control pulse width, is set to approximately zero and the duty cycle is Set the signal to zero percent.

If at point 118 it is determined that RPM > SRPM and the engine is running, proceed to point 122 and a wide open condition (WOT) exists;
A determination is made as to whether an increase in power is required.
This is accomplished by sampling information stored in ROM specified memory locations within RAM. Here, the state of the fully open switch is stored during step 68. If the engine is fully open, the program cycle proceeds to step 124, rich operating mode. The pulse width that provides the necessary duty cycle for power increase is determined,
Assigned to remember control pulse width
Stored in RAM memory location.

If the engine is not running at full throttle, proceed to point 126. An elapsed time counter monitoring time since engine start is compared to a predetermined time representing a time reference before implementing closed loop operation. This timer may take the form of a counter that is set to zero at initialization step 44 and is incremented at point 126 of the program every 0.01 ms interrupt period so that the number of interrupt times represents elapsed time. If the elapsed time is less than a predetermined value, an open loop mode routine is executed at step 128 to determine the open loop pulse width. If the time standard is met at point 126, proceed to point 130,
Operating conditions for the air-fuel ratio sensor 20 are determined. The system determines the operation of sensor 20 by parameters such as sensor temperature, sensor impedance, etc. If it does not work, proceed to step 128. If it is, go to point 134 and step 68 to set the engine temperature stored in RAM to ROM.
is compared with a predetermined calibration value stored in . If the engine temperature is lower than the calibrated value, the step
Proceed to 128. If it is greater than the calibrated value, step 136 is entered and a closed loop routine is executed to determine the control signal pulse width and store the pulse width in the predetermined assigned RAM location. At step 138, the pulse width is read from the RAM and placed in binary form into a register in the output counter section of the input/output circuit 36. This value is then inserted into the down counter at step 66 during the next 0.01 ms interrupt period to send a pulse output to the solenoid with the desired increment. When a control pulse is issued, it activates the air/fuel ratio control solenoid.
Operates every 0.01ms interrupt period to adjust the supply device.

The open loop mode routine at step 128 is as shown in FIG. The routine enters step 140, and in step 142 a pulse width correction value is obtained from the lookup table in the ROM portion of combination module 26. This correction factor can be a function of a single parameter, such as engine temperature, but it can also be a function of load. This correction factor is provided in 72 memory locations, as shown in FIG. 12, and is addressed according to engine temperature and load values. In step 144,
The control pulse width stored in RAM is (DCM+
pulse width correction value). The program proceeds to step 146 where a new cell flag is set. At step 148, step 70 (Figure 13)
The index value determined by is placed in the RAM location.

Further, as shown in FIG. 15, the pulse width of the fuel injection valve is determined. At step 402, the air amount Ga and engine speed RPM stored in the ROM specified memory location in the RAM are sampled, and the basic injection pulse width Δt=K(Ga/RPM) is calculated.
Furthermore, in step 404, the pulse width Δt _p is calculated. The pulse width Δt _p is given by the following equation.

Δt _p =Δt−FM/DVM/(1−DXM)+DXM/DVM (1) Here, DXM is the fuel deposition rate, and DVM is the evaporation time constant, which are given in the initial setting at step 48. The injection valve 302 is opened for a period of Δt _p by an interrupt synchronized with engine rotation to supply fuel to the internal combustion engine. FM in equation (1) is the amount of liquid film, and is calculated by the following equation.

FM=FM+Δτ{DXM・Δt _p・N−FM/DVM} (2) Here, Δτ is a constant, and when the interrupt routine is executed every 0.01 s, Δτ=0.01 s.

FIG. 16 shows the closed loop mode 136 routine. Closed loop mode is entered at point 150,
Proceed to point 152, engine operating point is 0.01s ahead
Determine if it has changed since the interrupt. This was determined in step 70 (Figure 13).
Pull DCMINX from RAM and set it to the previous 0.01 second interrupt time as determined in step 70.
This is done by comparing DCMINX, ie DDCMINX=. DCMINX＝
If ODCMINX, the operating point of the engine, does not change, proceed to point 154. At point 154, a new self-flag flip-flop in microprocessor 24 (set during the open loop routine of step 146) is sampled. If this flag is set,
Unit 18 was operating in open loop mode at the time of the previous 0.01s interrupt. However, when this flag is reset, unit 18
Operates in closed loop mode during 0.01s interrupt time.

Assuming that we have changed the operating point from the previous 0.01 s interrupt period or that the unit 18 is operating from an open loop mode to a closed loop mode, we proceed to step 156. In step 156, specify ROM,
The integral control term portion INT of the closed-loop control signal stored in the RAM location is
The memory location determined in Figure 1 is set equal to the pulse width obtained from the duty cycle memory DCM. This pulse width has been learned during previous closed-loop operation as a value to adjust the feed system 12 to provide a predetermined air/fuel ratio. At step 158, a transit time delay counter is set to a value representing the transit time through engine 10.
This transport delay time can be determined from engine operating parameters including engine speed and air volume. It can be obtained from the lookup table of the ROM portion of the combination module 26 addressed by the operating parameters.

At step 160, the new self flag flip-flop is cleared, indicating that unit 18 is operating in closed loop mode. The program then proceeds to step 162 where the old index stored in RAM is set equal to the index determined in step 70.

From step 162, or point 154, the program advances to point 163. Sample the transit delay counter at point 163 to determine if the transit delay is complete. If the delay is not complete, step 164 decrements the counter and steps
At 166, the control pulse width is previously set equal to the integral control term INT of the closed loop pulse width set to the duty cycle memory value at step 156, leaving the closed loop mode routine. Thereafter, the process proceeds to step 138 in FIG. 13, where the duty cycle pulse width is set in the register of the output counter section of the input/output circuit 36.

Thus, in closed loop mode, whenever the operating point of the engine changes, a time delay counter is set so that this counter decrements to zero. In other words, even if the system reaches steady state, closed loop control is not performed for a certain period of time. step
Perform step 500 before step 163.

In step 500, the operations shown in FIG. 17 are performed. In step 502, based on equations (1) and (2), Δt _p ,
Perform FM calculations. Also, FM at this time,
DVM and DXM are stored in RAM in step 504.
As shown in FIG. 18, when the transport delay time is K, this is stored in RAM (K). In step 506, the second counter is set and the point is
508, if the counter is not zero, step 510
decrements the counter, and in step 512 all
Move the contents of RAM (K) to RAM (K-1). When the delay counter decreases to zero, the step
Proceed to 514 and set the RAM (O) air-fuel ratio setting value A/F.
is determined from the output of the air-fuel ratio sensor 20 (A/
F) Compare m, ΔA/F=A/F-(A/F)
Calculate m. At step 516, FM is modified based on ΔA/F. If ΔA/F>0, FM
is estimated to be small, so increase FM.
In other words, FM is corrected as follows: FM=FM+a ₁ ΔA/F (3). Also, in step 518, DXM,
DVM fixes will be made. When ΔA/F>0,
Since DXM is small and DVM is estimated to be small, DXM and DVM are corrected using DXM=DXM+a ₂ /ΔA/F (4) DVM=DVM+a ₃・ΔA/F (5). Thereafter, the process proceeds to step 512. In this way, if the operation is unsteady in closed loop mode, in step 500, (1),
FM, DVM, and DXM in equation (2) are corrected. RAM
(K) stores values related to engine load and temperature, so the DXM and DVM of the corresponding locations in FIGS. 8 and 10 are updated.

When the transport time delay counter has decreased to zero in step 163, the process proceeds to step 168. step
At 168, the output of sensor 20 is compared to the calibration value,
It is determined whether the detected air-fuel ratio A/F is larger or smaller than the calibration value, that is, whether it is rich or lean. If the mixture is rich, ie, the A/F is low, the program proceeds to step 170, where the integral term of the closed loop control signal stored in RAM is set equal to the previously stored integral term plus the integral step value. Thereafter, in step 172, the closed loop control pulse width is set equal to (integral term + proportional step value) determined in step 170.

If it is determined at step 168 that the mixture is lean, the program proceeds to step 174, where the integral term of the control signal stored in RAM is reduced by the integral step value. Then, in step 176, the closed loop pulse width is
Stored in RAM (integral term - proportional step value)
is set equal to Steps 168 to 176 are repeated every 0.01ms after the engine has operated at the same operating point for a period longer than the transport delay time.
At the rate determined by the integration step, depending on whether the A/F is large or small, it forms a closed loop pulse width that increases and decreases in a ramp fashion as shown in Figure 19, and finally The mixture changes between rich and lean relative to the calibrated value. At this time,
A proportional step of pulse width in the direction giving the calibration value is provided.

Compare the engine temperature with the calibration value K ₁ at point 178. If the temperature is higher than K ₁ , stop updating the DCM. If it is low, proceed to step 180 and read the memory index formed in step 70.
Update DCM. The DCM is updated according to the following equation.

DCMV _N = DCMV _N-1 + (DC-DCMV _N-1 )/T ₁ ...(6) Here, DCMV _N : New pulse width value to be inserted DCMV _N-1 : Prior to the corresponding memory location Value of pulse width DC: Finally determined control pulse width T ₁ : Filter time constant The temperature conditions are determined at points 182 and 184, and only when the engine temperature is normal, the control pulse width is determined at step 186.
Update KAM. This update is performed in the same way as in equation (6). DCM update time constant is 5-30 seconds,
The time constant for updating KAM is about 240 seconds. After step 186, the program exits the closed loop mode routine. As the engine continues to operate in closed-loop mode, the foregoing sequence beginning at step 150 is repeated continuously, and as the engine passes through various operating points, the DCM,
Each of the KAMs is updated according to the value on the control signal,
As a result, each memory location is updated with the values necessary to provide a predetermined air/fuel ratio for a particular engine operating point.

During closed loop operation, each time the engine operating point changes, the control pulse width is instantaneously preset to a value that provides a predetermined air/fuel ratio at the current value of the engine operating parameter. During open-loop operation, the supply device:
Adjusted according to the value held in KAM. This value indicates the average control pulse width required to provide a predetermined air-fuel ratio for a change in engine parameters.

Here, the update of DXM and DVM in equations (4) and (5) is
It may be performed in the same way as updating the DCMV in equation (6).
In other words, as shown in the following equations (7) and (8), DXM _N = DXM _N-1 + (a ₂ ∆A/F - DXM _N-1 )/T ₁
...(7) DVM _N = DVM _N-1 + ( _a3・ΔA/F-DVM _N-1 )/T ₁
...(8) Can be renewed. Also, XM in Figure 17,
VM correction is updated in the same way as equations (7) and (8), but the filter time constant T ₁ is set large in the case of DXM and DVM.

Air-fuel ratio A/F when engine temperature is normal
is set as shown in FIG. 19, for example. 19th
In the region indicated by the dashed line in the figure, the A/F is smaller than the stoichiometric air-fuel ratio curve D, and in this load region the rich operating mode of step 124 in FIG. 13 is entered, and the air-fuel ratio is under open-loop control. If the engine temperature is normal in the region of A/F>D, the closed loop mode is entered at step 136 in FIG. In Figure 19, even if the load changes from θ ₁ to θ ₂ , the index
DCMINX does not change.

Once the transport delay is completed in step 163 of FIG. 16, the air/fuel ratios are compared in step 168. At this time, the signal from the air-fuel ratio sensor 20 indicates the state of the supply device 12 before the transportation delay. Here, the air-fuel ratio sensor 20 is programmable and is configured to output deviation from a predetermined air-fuel ratio.

An embodiment of the air-fuel ratio sensor 20 is shown in FIG. A chamber 202 and an orifice 204 are provided on one side of an oxygen ion conductive solid electrolyte 200 . When the controller 206 causes a current I to flow through the electrolyte 200 in the direction of the arrow, oxygen within the chamber 202 is discharged to the outside of the chamber 202 based on the principle of an oxygen pump. On the other hand, orifice 20
4, oxygen in the exhaust gas flows into the chamber 202 by diffusion. The relationship between the current I and the voltage at this time changes depending on the oxygen concentration in the exhaust gas, that is, the A/F, as shown in FIG. Therefore, if the setting A/F in FIG.
compared with the calibration value stored in the section. It is determined whether the detected air-fuel ratio A/F is larger or smaller than the predetermined A/F shown in FIG. 19, that is, whether it is rich or thin relative to the predetermined concentration. At this time, in Fig. 19, when the load changes from θ ₁ to θ ₂ , the memory location of the DCM does not change, but the predetermined air-fuel ratio changes, so the current I
is controlled by taking into account transportation delay time. In this way, the air-fuel ratio sensor 20 determines the state of the supply device 12 before the transportation delay time.

When entering from the open loop state in the closed loop mode operation shown in Fig. 13, as in Fig. 22 a,
Closed loop control begins after the cell flag is reset and the time delay counter reaches zero. here,
If the operating point of the engine changes during closed-loop operation, the flag is reset as shown in FIG. 22b, but the counter is set, so that closed-loop control is executed after the transportation delay time. If the operating point does not change within this transport delay time, CLPW
=INT=DCM, the control pulse width CLPW is set equal to the closed loop pulse width integral control term INT, and INT is set equal to DCM.
At the next 0.01 second interruption, INT is corrected based on the signal from the air-fuel ratio sensor 20. When the temperature conditions are right,
The DCM is also updated as shown at point P in FIG. 22c.

In FIG. 13, closed loop control is performed every 0.01 second interrupt. This is generally less than the transportation delay time.

In Fig. 23, during closed loop control, as the load θ is a, the A/F setting value is correspondingly b.
Here is an example of a case where it changes as follows. In this case, the current of the sensor 20 is controlled so that the corresponding A/F becomes the c curve. If b matches c, the output of sensor 20 is constant. However, in reality,
Due to changes in the supply device 12 over time, b deviates from c. Therefore, after a delay of the transportation delay time T, the output of the sensor 20 suddenly changes as shown in d. Based on the signal d of this sensor 20, the integral term INT is
It is modified as follows. Even if the correction is completed at point C in FIG. 23e, the result appears as an output from the sensor 20 with a delay of T. Meanwhile, INT
goes too far.

In this way, the value of INT changes between a dark state and a light state. At this time, a proportional step of the pulse width may be applied in the direction of providing the calibration value. Here, even if the A/F setting value changes as shown in b at point D, the sensor 20 is also programmed as shown in c, so this b,
A change in c does not affect the output d of the sensor 20.

In FIG. 20, when I is controlled so that the voltage V becomes the set value, the value of I becomes a value proportional to the oxygen concentration in the exhaust gas, that is, the A/F. In this case, the output of the sensor 20 appears delayed by the transportation delay time T. Therefore, at this time, the output of the sensor 20 can be used to determine the flow rate of the supply device 12 before T time. The A/F setting value T hours ago is stored and compared with the output of the sensor 20 to calculate the proportional step. The result is T
It will appear after some time and be revised again. In this case, 0.01
A proportional step can be determined every second according to the deviation, and convergence to the set value is faster than in the case of only integral operation. The air-fuel ratio R of the supply device 12 at any operating point is twisted by ε from the set value Rc, as shown in equation (2).

R=Rc+ε...(9) where, ε: Wrinkling due to changes over time, etc. This wrinkling of ε appears on the air-fuel ratio sensor 20 after T seconds. In the air-fuel ratio sensor 20, a signal of R(t-T) appears at time t, so in order to find ε at the memory location, Rc
It is necessary to compare it with the value of (t-T). Therefore, it is necessary to store the value of Rc(t-T) in RAM. As shown in Figure 24, the RAM ROM
Store Rc (t-nΔt) at the specified address,
The transport delay time is determined from the lookup table in the ROM portion of the combination module 26, which is addressed by the engine operating parameters including engine speed and air volume, and the Rc is extracted from the RAM at the address in FIG. can be read and compared. In this way, ε is determined, DCM is updated, and KAM is further updated.

During unsteady operation, the signal from the air-fuel ratio sensor 20
Parameters related to the dynamic characteristics of the fuel supply device 12
Since EM, DVM, and DXM are modified from time to time,
Δt _p is corrected to the correct value. For example, when low-volatility gasoline is supplied to the supply device 12, as shown in FIG. 25, the change in ΔA/F is large during the initial acceleration/deceleration;
Since FM, DVM, and DXM are modified to match the properties of the fuel, ΔA/F is adjusted for the second acceleration/deceleration.
becomes smaller. Similarly, the response delay of the air flow sensor 301 can also be corrected. Also, since the model equation (1) for the amount of liquid film shown in equation (2) is determined,
Regardless of the operating pattern, since the hourly value of the liquid film amount is known, accurate correction can be made using equation (1).

Therefore, according to the embodiments described above, even if the engine load fluctuates frequently, it is possible to maintain a good air-fuel ratio at all times, including during transient fluctuations.

By the way, gasoline engines are the mainstream engine for automobiles.

Therefore, when the engine 10 in FIG. 1 is a gasoline engine, its ignition control is often also performed by the electronic control unit 18.

Therefore, an embodiment in which the present invention is applied to this ignition control will be described below.

When the internal combustion engine 10 is a spark ignition engine, the ignition timing differs from its target value during transient operation due to the response delay of the air flow sensor 301 and the like. The ignition timing is determined by the amount of air charged in the engine 10 and the engine speed, and during transient operation, it is determined by the following equation (9).

Ga=b ₁・P・N+b ₂・dp/dt (9) However, the air flow rate Ga and the filling air amount b ₁・P・N measured by the air flow sensor 301 are affected by the volume of the intake pipe. I'll be late. Therefore, the 26th
As shown in the figure, in step 601, P is calculated,
In step 602, the ignition timing is obtained from the ignition timing lookup table previously stored in the addressed memory location corresponding to P and N.

P=P+(Ga- _b1.P.N ).DELTA..tau. (10) The calculation of P in step 601 is performed using equation (10). Also, in step 604, b ₂ , or responsiveness, can be modified. Module 2 at step 603
DVM ₀ , DXM 0 , DXM ₀ , stored in the ROM section of 6
Calibration values such as DCM ₀ and learning-controlled DVM,
By comparing the values of DXM, DCM, etc.
Since it is known how much the air flow rate sensor 301 differs from the standard value, _b2 can be corrected based on this difference. Here, b ₂ = a ₅ (DVM ₀ − DVM) + b ₂ ... (11) is corrected. These steps are implemented with a 100Hz interrupt signal, as in FIG. If the difference between DVM ₀ and DVM is large in step 605, the response of the sensor 301 is significantly different from the standard value, but this is due to an abnormality such as the volatility of gasoline being extremely low.
lights up the lamp to notify the driver of an abnormality.

Therefore, according to this embodiment, correct ignition timing control can always be performed no matter what the engine operating pattern is.

Note that the present invention is not limited to the above-described embodiments, and similar effects and effects can be achieved even when applied to control of the exhaust gas recirculation amount of a gasoline engine, the fuel injection amount, and the injection timing of a diesel engine.

〔Effect of the invention〕

As explained above, according to the present invention, the amount of fuel supplied to the internal combustion engine can be determined by at least the amount of adhering fuel adhering to the intake passage, the fuel adhesion rate representing the state in which the supplied fuel is adhering to the intake passage, and the adhesion rate. A liquid film model of the fuel supplied to the engine is corrected based on the evaporation time constant representing the evaporation state of the fuel, and further corrects the amount of adhering fuel using air-fuel ratio information representing the air-fuel ratio concentration in the exhaust gas. By constructing this model and modifying the components of this model using air-fuel ratio information in exhaust gas, responsiveness and fuel control accuracy can be improved.

Therefore, when the load changes suddenly, such as during acceleration, the amount of fuel supplied can be adjusted in a transient operating state to avoid the problem of deviation from the target value due to fuel adhesion or delayed evaporation of adhering fuel. Even if there is a difference, control with sufficiently few errors can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of an internal combustion engine control system to which the present invention is applied, FIG. 2 is a block diagram showing an example of an electronic control unit, FIG. 3 is an explanatory diagram of a fuel supply system, and FIG. 4 is a block diagram showing an example of an electronic control unit. 1 is a flowchart showing the operation of an embodiment of a control device for an internal combustion engine according to the present invention, FIGS. 5 to 10 are explanatory diagrams of memory contents in an embodiment of the present invention, and FIG. Flowchart showing operation, 12th
13 to 17 are flowcharts showing the operation of the embodiment. FIG. 18 is an explanatory diagram of the memory contents in the embodiment. 20 is a characteristic diagram for explaining the operation of one embodiment, FIG. 20 is an explanatory diagram of the sensor, and FIG. 21 is a characteristic diagram of the sensor.
22a, b, c and 23a to 23e are characteristic diagrams showing the operation of an embodiment of the present invention, FIG. 24 is an explanatory diagram of memory contents of the embodiment, and FIG. FIG. 26, which is a characteristic diagram for explaining the effects of the embodiment, is a flowchart showing the operation of another embodiment of the present invention. 10... Internal combustion engine, 12... Fuel supply device, 1
4...exhaust pipe, 16...catalytic converter, 18...
...Electronic control unit, 20...Air-fuel ratio sensor, 2
4... Microprocessor, 26... Combination module, 36... Input/output interface circuit,
40... Persistent storage device.

Claims (1)

[Scope of Claims] 1. A control device for an internal combustion engine, comprising a plurality of sensors for detecting the operating state of the internal combustion engine, and controlling the amount of fuel supplied to the internal combustion engine according to signals from the sensors. The amount of fuel supplied to the internal combustion engine is determined from at least the amount of adhering fuel adhering to the intake passage, a fuel adhesion rate representing the state in which the supplied fuel adheres to the intake passage, and an evaporation time constant representing the evaporation state of the adhering fuel. A method for controlling fuel supplied to an internal combustion engine, characterized in that the amount of adhering fuel is corrected and further corrected using air-fuel ratio information representing an air-fuel ratio concentration in exhaust gas. 2. The method of controlling fuel supplied to an internal combustion engine according to claim 1, wherein the amount of adhering fuel corrected based on the air-fuel ratio information is updated and used for the next correction. 3. The method of controlling fuel supplied to an internal combustion engine according to claim 1, further comprising modifying the fuel adhesion rate and the evaporation time constant using the air-fuel ratio information. 4. In claim 3, the fuel adhesion rate and the evaporation time constant corrected based on the air-fuel ratio information are updated and used for the next correction of the fuel supplied to the internal combustion engine. Control method.