JPS6123376B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for feedback controlling the air-fuel ratio state of an internal combustion engine using signals from an air-fuel ratio sensor.

In an electronically controlled fuel injection type (BFI type) internal combustion engine, the basic control is to control the intake air amount Q of the engine.
is measured with an air flow sensor and the basic injection time of the fuel injector is determined by detecting the engine rotational speed N as _{τ B =} Q/N・K (K is a constant)
The fuel injection amount is determined by calculation and fuel injection control is performed. Such an EFI internal combustion engine further includes an air-fuel ratio sensor (A/F sensor) for detecting the air-fuel ratio state of the engine, such as an oxygen concentration detector for detecting the argon concentration in the exhaust gas, in the exhaust system of the internal combustion engine. The basic fuel injection time is multiplied by the air-fuel ratio feedback coefficient f (A/F) determined according to the output signal of the A/F sensor, thereby controlling the fuel supply amount or air supply amount and adjusting the air-fuel ratio state of the engine. Air-fuel ratio feedback control is generally performed.

Such an air-fuel ratio feedback control system is provided with a function of canceling air-fuel ratio feedback control when the A/F sensor fails or is inactive. In other words, when the rich signal and lean signal, which indicate that the air-fuel ratio is on the rich side or on the lean side, respectively, continue for a predetermined time or longer as output signals of the A/F sensor, the feedback coefficient f(A/F
F) is set to "1" to open the air-fuel ratio feedback control.

The reason why the air-fuel ratio feedback control of an EFI internal combustion engine is in the open state is because of the above-mentioned A/
In addition to the failure or inactive state of the F sensor, the open state also occurs, for example, when the feedback control cannot follow up during deceleration and the rich state continues for a predetermined period of time or more (hereinafter referred to as "rich open"). In the case of a rich open occurring during deceleration or the like, the A/F sensor must output a lean signal next time to restart feedback control immediately.

By the way, when an EFI internal combustion engine with air-fuel ratio feedback control in the open state is operated at high altitudes, i.e. in areas with low air density, the airflow sensor detects as if a larger amount of air is being sucked in than the amount of air actually used for combustion. As a result, the basic control point for air-fuel ratio control shifts to the rich side. In other words, when operating an EFI internal combustion engine with air-fuel ratio feedback control at high altitudes, the feedback coefficient f
(A/F) is a much smaller value than the value in lowlands, for example, about 0.85.

Therefore, when feedback control becomes open loop due to deceleration at high altitudes, the feedback coefficient is "1" during that time and the basic air-fuel ratio is in an overrich state, so the A/F sensor is not functioning normally. Even if the A/F sensor functions properly, the A/F sensor may not output a lean signal and the air-fuel ratio feedback control may not be restarted, or even if a lean signal is output and the control is restarted, the correction amount is too large and the above-mentioned predetermined In many cases, the correction cannot be completed within the time limit, and therefore, the next lean signal is not output, resulting in an open loop again. As described above, there is a problem in that the air-fuel ratio feedback control of the internal combustion engine cannot be performed smoothly at high altitudes.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide an air-fuel ratio feedback control method for an internal combustion engine that can reliably perform smooth air-fuel ratio feedback control even in areas with low air density such as highlands.

In order to solve the above-mentioned problems, the present invention changes the air-fuel ratio feedback coefficient involved in calculating the fuel injection amount according to the output signal of the air-fuel ratio sensor that detects the air-fuel ratio state of the internal combustion engine. In an air-fuel ratio feedback control method for an internal combustion engine in which the air-fuel ratio state of an engine is feedback-controlled by adjusting the amount of fuel supplied to the engine according to the amount of fuel supplied to the engine, a rich signal indicating that the air-fuel ratio state is on the rich side is detected when the air-fuel ratio is on the rich side. When the sensor continuously outputs the air-fuel ratio for a predetermined time or more, the air-fuel ratio feedback coefficient is set to a constant value and the feedback control is stopped, and when the feedback control stopping action continues for the predetermined time, the air-fuel ratio of the engine is temporarily changed. When the air-fuel ratio sensor outputs a lean signal within a predetermined time after the forced control to the lean side, the feedback control is performed so that the air-fuel ratio feedback coefficient is less than the constant value. Provided is an air-fuel ratio feedback control method for an internal combustion engine, characterized in that the air-fuel ratio feedback control method restarts from a smaller value.

The present invention will be explained in detail below using the drawings.

FIG. 1 shows a block diagram of a fuel injection control system for an internal combustion engine as an embodiment of the present invention. In the figure, 10 is an air flow sensor that generates an intake air amount signal having an analog voltage level corresponding to the intake air amount of the engine, and 11 is a water temperature sensor that generates a water temperature signal having an analog voltage level corresponding to the cooling water temperature. 12 is a crank angle sensor that generates a pulse at each predetermined crank angle position of the engine, for example, every 30° or 180° from a crank angle position of 0°, and 13 is installed in the exhaust system of the engine to detect exhaust gas. This is an A/F sensor that outputs a signal depending on the concentration of a specific component, for example, an oxygen component. This A/F sensor detects when the engine's air-fuel ratio is on the rich side (rich side) of the stoichiometric air-fuel ratio.
It outputs a rich signal of about 1V, and a lean signal of about 0.1 to 0.2V when it is on the lean side. In FIG. 1, 14 is a fuel injection solenoid valve provided for each cylinder or a plurality of cylinders in the intake system of the engine. Signals from each of the above-mentioned sensors 10 to 13, as well as sensors not shown, are sent to an electronic control unit 15 including a microcomputer, which calculates the optimum fuel supply amount according to the operating state of the engine, and finally controls the fuel injection valve. The air-fuel ratio control is performed by controlling the injection time of 14.

Next, the configuration of this electronic control unit 15 will be explained. An analog multiplexer 17 selects a signal from one of the sensors 10 and 11, as well as a sensor (not shown), and an analog signal selected by the multiplexer 16. A/D converter that converts the signal into a digital signal, 1
8 is a gate circuit which is controlled to open and close by a pulse sent from the crank angle sensor 12, for example every 30 degrees of crank angle, and which controls the passage of a clock applied via a line 19; A count and latch circuit 21 that counts and stores data corresponding to the reciprocal of the engine rotational speed, compares the output signal level of the A/F sensor 13 with the reference signal level and determines "1", "0'' rich and lean signals are formed, a reversal point pulse is formed at the reversal point of the rich signal and lean signal, and the timer circuit 23 and interrupt latch circuit 2
The waveform processing circuits sent to the waveforms 4 and 4 are respectively shown. The timer circuit 23 is a counter circuit that is reset by a pulse applied from the input/output port 22 at the reversal point and starts counting the clock applied via the line 19, and corresponds to the duration T of the rich signal lean signal. It is configured to form data to be sent to the input port 25. Furthermore, in FIG. 1, 26 is one of the fuel injection valves 14.
The output port 27, in which the output data corresponding to the injection time τ of 2 times is stored, forms an injection signal having a pulse width according to the output data, and generates the injection signal in synchronization with the crank angle pulse fed through the line 28. An output control circuit outputs an injection signal, and 29 represents a power amplification circuit for the injection signal. A multiplexer 16, an A/D converter 17, a count and latch circuit 20, an input/output port 22, an input port 25, an interrupt latch circuit 24, and an output port 26 are connected to RAM, ROM, etc. that constitute the microcomputer via a common bus 30. memory 31,
It is connected to a central processing unit (CPU) 32 and an I/O and memory control device 33 including a clock generation circuit.

A calculation program related to fuel injection time and the like is stored in the ROM of the memory 31, and when an initialization signal is applied to the CPU 32, calculation processing based on this calculation program is started. In this case, the signals from sensors 10 and 11 are
Multiplexer 16 and A/D converter 1 in response to a time-synchronized interrupt signal generated at predetermined intervals.
7 is controlled, they are converted into intake air amount data Q and water temperature data, respectively, and are stored in the RAM of the memory 31.
stored within. Also, count and latch circuit 2
The data 1/N regarding the rotational speed stored in 0 is stored in the RAM of the memory 30 at every predetermined crank angle.

When a crankshaft synchronization interrupt signal is sent from the interrupt latch circuit 24 at every predetermined crank angle position, the CPU 31 performs arithmetic processing regarding the fuel injection time τ. That is, τ=Q/N・K・f(A/F)・α. Performs arithmetic processing of R+τ _o . However, in the above formula, f(A/
F), α are the air-fuel ratio feedback coefficients,
B is the correction coefficient thereof, B is another correction coefficient such as warm-up correction, and τ _o is the invalid injection time correction coefficient of the fuel injection valve. The output data regarding the calculated fuel injection time τ is sent to the output control circuit 27 via the output port 26 as described above, and becomes a fuel injection signal to control the opening time of the fuel injection valve 14. As a result, the engine air-fuel ratio feedback control is performed.

The air-fuel ratio feedback coefficient f (A/F) and its correction coefficient α are calculated based on the signal from the A/F sensor 13. This will be explained using the flowchart shown in the figure and the waveform diagram shown in FIG.

When an inversion point pulse is applied from the waveform processing circuit 21 to the interrupt latch circuit 24 at the inversion point from the lean signal to the rich signal, the CPU 32 performs the operation shown in FIG. 2A as an interrupt process (hereinafter referred to as L-R interrupt process). )do. That is, first, in step 40, the counter of the timer circuit 23 is reset, and in step 41, the counter starts counting. Next, in step 42, the lean oven flag (L-OPN flag) is turned off. In addition, in the following description, "setting" or "turning on" various flags means setting a pre-specified bit for each flag in RAM to a specific value, for example, "1".
Setting the flag to "off" means setting the bit to "0", for example. Therefore, by checking each flag, the CPU can know what kind of control is currently being executed, and can use this information as a condition for branching.

Now, in step 43, the feedback coefficient f(A/F) is decreased by a constant C, thereby performing skip control of the air-fuel ratio at the time of lean/rich reversal (section a in FIG. 4).
Next, in step 44, the air-fuel ratio correction coefficient α is set to α=1, the LR interrupt processing is completed, and the process returns to the main routine.

When the waveform processing circuit 21 applies an inversion point pulse to the interrupt latch circuit 24 at the inversion point from the rich signal to the lean signal, the CPU 32 performs the operation shown in FIG. )do. That is, first, in steps 45 and 46, the timer circuit 23 is reset and started, and in step 47, it is determined whether a forced lean flag (CMPL-L flag), which will be described later, is on. If the forced lean flag is off, it is determined in step 48 whether a weighting flag (W flag), which will be described later, is on. If the weighting franc is off, proceed to step 49, set f(A/F) = f(A/F) + D (where D is a constant), and perform skip control of the air-fuel ratio during lean reversal (b in Figure 4). part). Normally, the process passes through step 49 and proceeds to step 50 where α=
1, and further steps 51, 52, 53, 54.
, the Rich temporary open flag (R
-FOPN flag) After turning off the forced lean flag, waiting flag, and rich open flag (R-OPN flag) described later, R-
Ends the L interrupt processing and returns to the main routine.
If it is determined in step 47 that the forced lean flag is on, or if it is determined in step 48 that the weighting flag is on, the process proceeds to step 55, where f(A/F) is made equal to the constant β. The process then proceeds to step 50.

After the L-R interrupt processing or the R-L interrupt processing is performed, a time interrupt signal is applied to the interrupt latch circuit 24 at predetermined intervals, thereby causing the CPU
32 executes the interrupt calculation process shown in FIG.
That is, first, in step 60, the timer circuit 2
3, the counting data, that is, the duration T of the rich signal or lean signal is read. Next, in step 61, it is determined whether a rich signal or a lean signal is currently being output. If the R-L interrupt processing has been performed and the lean signal has been output, the process proceeds to step 62, where it is determined whether the duration T of the lean signal is greater than a predetermined value _T10 . If T<T ₁₀ , proceed to step 63 and calculate the feedback coefficient f(A/
After increasing F) by a predetermined value B and setting α=1 in step 64, this interrupt processing is terminated and the process returns to the main routine. When the program is executed in a loop of repeating steps 60, 61, 62, 63, and 64 after the next time interrupt process, f(A/F) gradually increases as shown in C in FIG. The injection amount gradually increases and the air-fuel ratio is controlled in the rich direction. When the duration of the lean signal becomes very long and T≧T ₁₀ in step 62, the process proceeds to step 65, where the lean open flag is turned on, and in the next step 66, f(A/F) is turned on.
= 1, the air-fuel ratio feedback control is stopped, an open state is entered, step 64 is executed, and the interrupt process ends. The time interrupt processing beyond step 62 is repeated until the next LR interrupt processing is performed.

If it is determined in step 61 that the rich signal is being output, that is, if this time interrupt is performed after the LR interrupt processing, the process proceeds to step 67, where it is determined whether the rich temporary open flag is on or not. will be judged. Since the rich temporary open flag is normally off, the process proceeds to step 68, where it is determined whether the rich signal duration T is longer than the predetermined time _T1 . If T< _T1 , the feedback coefficient f(A/F) is set to the predetermined value A in step 69.
α=
After setting it to 1, the interrupt processing ends and the program is executed repeatedly in a loop of steps 60, 61, 67, 68, 69, and 64, and f
(A/F) gradually decreases as shown in d of FIG. 4, the fuel injection amount gradually decreases, and the air-fuel ratio is controlled in the lean direction. Normally, this time interrupt processing is repeated until the next RL interrupt processing is performed.

When the output e of the A/F sensor 13 takes on a pattern that alternately repeats a rich signal and a lean signal as shown in part _e1 of FIG. 4, the air-fuel ratio feedback coefficient f(A/F) is as described above, , as shown in b, c, a, and d of FIG. 4, the air-fuel ratio is alternately increased and decreased, thereby performing feedback control of the air-fuel ratio.
If the A/F sensor output continues to be a rich signal only, the system performs a special operation. Below A/
A case in which the F sensor 13 continuously outputs a rich signal from, for example, point g in FIG. 4 will be described. First, when the duration time T becomes T≧T ₁ and interrupt processing is performed, the program goes to step 6.
8, the process branches to step 70, where a rich temporary open operation is performed. That is, in step 70, the rich temporary open flag is turned on, and then in step 66, f(A/F) is set to 1, feedback control is stopped, and open control is started. Since the rich temporary open flag is turned on, in the subsequent time interrupt process, a branch is made from step 67 to step 71, and in step 71 it is determined whether the forced lean flag is turned on. Since the forced lean flag is off at this point, the process proceeds to step 72, where the rich signal duration T is compared with a predetermined time _T2 . If T< _T2 , the process proceeds to steps 66 and 64, where f(A/F)=1 and α=1, and the feedback stop by the rich temporary open operation is continued. The duration of this rich temporary open operation, i.e. T _2
-T1 is normally set to 1 to 5 seconds in order to reconfirm whether the air-fuel ratio is really rich in the open loop. In the case of T≧T ₂ , that is, the A/F sensor 13
If T is continuously outputting a rich signal for more than ₂ hours, perform a forced lean operation. That is, in step 73, the forced lean flag is turned on, in step 74, f(A/F) is set to 1, and in step 75, α is set to 0.8. As a result, the fuel injection time τ becomes shorter, and the air-fuel ratio is forcibly controlled in the lean direction. Since the forced lean flag is on, the subsequent time interrupt processing branches from step 71 to step 76, and in step 76, the rich signal, the duration T, and the predetermined time _T3 are connected. A comparison is made. If T< _T3 , proceed to step 74 and 75, set f(A/F)=1, α=0.8,
Continue forced lean operation. The period of this forced lean operation is set to a relatively short period of time, approximately T ₃ -T ₂ =0.1 to 1.0 seconds, in order to prevent deterioration of the engine operating feeling. When T≧T ₃ , the process proceeds to step 78, where it is determined whether the weighting flag is on. At this point, the weighting flag is off, so the process branches to step 79 to turn on the weighting flag, and at the same time sets f(A/F)=1 at step 66, and then proceeds to step 66.
In step 4, the correction coefficient α is returned to α=1, the forced lean operation is ended, and a waiting state is established.
Because the waiting flag is on,
In the subsequent time interrupt processing, step 78
A branch is then taken to step 80, in which a comparison is made between the duration T of the rich signal and a predetermined time _T4 . T<
If T ₄ , proceed to step 66 and 64 and f
Set (A/F)=1, α=1, and continue the waiting state. This waiting period is T ₄ − _{T 3} =
It is set to about 1 to 5 seconds. If T≧T ₄ , that is, the rich signal is in spite of the forced lean operation.
T If it continues for ₄ hours or more, proceed from step 80 to step 81, and set the waiting flag 81.
is off. Next, the process proceeds to step 77, where the forced lean flag is turned off, and the process further proceeds to step 82. After turning on the rich open flag in step 82, step 66 and further step 6
Proceeding to step 4, f(A/F)=1 and α=1, the rich open flag is operated, and the feedback control continues to be stopped.

The operation of this embodiment has been described above for the case where the rich signal continues to appear, but once the lean signal is output from the A/F sensor 13, the R-L interrupt processing routine is executed as described above. is executed and feedback control is restarted. However, in that case, the initial value of the feedback coefficient f(A/F) is f(A/F)=1 when the feedback control is restarted from the rich temporary open operation and the rich open operation state, and the initial value of the feedback coefficient f(A/F) is 1. When feedback control is restarted from the waiting state, step 55 in FIG. 2B is performed.
, so f(A/F)=β (however, β is usually
0.9).

Next, the effects of this embodiment will be explained. As shown in FIG. 4, according to the system of this embodiment, even if the rich signal continues for T ₁ hour or more and enters the provisional rich open state h, and the air-fuel ratio is controlled to be open, it will continue for T ₂ hours. After the time has elapsed, the atmosphere of the A/F sensor is forcibly made lean by controlling α to approximately 0.8 and entering a forced lean state i. As a result, if the A/F sensor function is normal, a lean signal will be obtained, making it possible to determine whether the A/F sensor is malfunctioning, inactive, or normal. Become. Furthermore, in the system of this embodiment, when the lean signal is output and the feedback control is restarted, the feedback coefficient f (A/F) is set equal to the constant β.
Since it is set to a value of about 0.9, which is smaller than 1,
A desired feedback coefficient value can be reached in a short time. Therefore, if the air-fuel ratio state when the feedback control is stopped is on the rich side due to the influence of altitude, etc., according to the conventional system,
The amount of correction when restarting feedback control becomes large and it is not possible to make a sufficient correction within a predetermined period, and as a result, the system returns to open-loop control again. However, in the system of this embodiment, the amount of correction is Since it is set to a small value, the above-mentioned inconvenience does not occur, and feedback control can be reliably restarted. Furthermore, in the system of this embodiment,
T Even if the forced lean state i ends after ₃ hours have passed.
T It remains in waiting state j until ₄ hours have elapsed, waiting for a lean signal to occur. This is because when the air-fuel ratio is controlled by adjusting the amount of fuel supplied from the fuel injection valve installed in the intake system, the A/F sensor installed in the exhaust system needs to move the working gas to obtain the control result. This is to compensate for the delay that occurs due to the response delay of the A/F sensor itself. In addition, in the system of this embodiment, if the rich signal continues to be output even after the forced lean state i ends and time _T4 has elapsed, the rich open state is set as shown in k in Fig. 4, and the A/F This open state is maintained until it is determined that the sensor is inactive or has failed and the next lean signal is output.

As mentioned above, in the past, in areas with low air density such as highlands, if the rich signal continues due to deceleration, etc., and the feedback control changes to open control, even after the deceleration ends, the feedback correction coefficient is used. Because correction to the lean side is not performed, the rich signal continues and feedback control cannot be restarted, or even if a lean signal is output and feedback control is restarted, the air-fuel ratio will not be sufficiently corrected by the feedback correction coefficient. However, according to the air-fuel ratio feedback control method of the present invention, such open-loop control is impossible. Feedback control can be reliably restarted from the control state, and therefore it is possible to perform smooth air-fuel ratio feedback control of the internal combustion engine even at high altitudes. In addition, it is possible to determine whether the A/F sensor is malfunctioning, inactive, or normal, depending on whether or not a lean signal can be obtained by temporarily forcing control to the lean side. This brings great convenience to air-fuel ratio feedback control using an A/F sensor.

In this configuration, in order to simplify the program, the first
Although the timer circuit 23 shown in the figure is used, it is obvious that the time T can be counted in the program as a so-called software timer in the interrupt arithmetic processing at predetermined time intervals shown in FIG.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of one embodiment of the present invention, and FIG.
Figures A, B, and 3 are flowcharts of part of the arithmetic processing in the above embodiment, and Figure 4 is a waveform diagram illustrating the operation of the above embodiment. 10...Intake air amount sensor, 11...Water temperature sensor,
12... Crank angle sensor, 13... A/F sensor, 14... Fuel injection solenoid valve, 15... Electronic control unit, 16... Analog multiplexer,
17...A/D converter, 18...gate circuit, 20...count and latch circuit, 21...
Waveform processing circuit, 22... Input/output port, 25...
Input port, 23...Timer circuit, 24...Interrupt latch circuit, 26...Output port, 27...Output control circuit, 29...Power amplifier circuit, 30...Common bus, 31...Memory, 32... CPU, 33
...I/O, memory control device.

Claims (1)

[Claims] 1 The air-fuel ratio feedback coefficient involved in fuel injection amount calculation is changed in accordance with the output signal of the air-fuel ratio sensor that detects the air-fuel ratio state of the internal combustion engine, and the amount of fuel supplied to the engine is adjusted according to the feedback coefficient. In an air-fuel ratio feedback control method for an internal combustion engine, in which a rich signal indicating that the air-fuel ratio state is on the rich side is continuously output from the air-fuel ratio sensor for a predetermined period of time or more, the air-fuel ratio is The feedback control is stopped with the feedback coefficient set at a constant value, and when the feedback control stopping procedure is continued for a predetermined period of time, the air-fuel ratio state of the engine is temporarily forcedly controlled to the lean side, and When a lean signal is output from the air-fuel ratio sensor within a predetermined time after the forced control is performed, the feedback control is restarted from a value where the air-fuel ratio feedback coefficient is smaller than the constant value. air-fuel ratio feedback control method.