 |
Description  |
|
|
BACKGROUND OF THE INVENTION
This invention relates to a control system for the internal combustion
engine or more in particular to an electronically controlled fuel supply
system for controlling the amount of fuel supply by measuring the flow
rate of air sucked into the engine.
Generally, in an electronically controlled fuel supply system, fuel is
supplied into the air path leading to the engine by a fuel injector, which
includes an injection valve opened in synchronism with the engine rotation
and kept open for a predetermined period of time. This valve-open period
is the fuel injection time regulated by the electronically controlled fuel
supply system in such a manner as to attain a predetermined air-fuel ratio
for the amount of air sucked in. Theoretically, the air-fuel ratio is
controlled to obtain the theoretical air-fuel ratio at which oxygen held
in the sucked air and injected fuel is used for combustion in proper
quantities. The precision of air-fuel ratio depends on the detection
accuracy of a sucked air flow rate sensor and the response accuracy of the
fuel injection valve. These accuracies of detection and response change
with time after a protracted use of the engine. Therefore, accurate
control is impossible merely by controlling the fuel injection time
against the sucked air flow rate in such a manner as to attain a certain
fixed air-fuel ratio.
In order to eliminate the above-mentioned disadvantage, a system has been
commercialized by which the oxygen concentration in the exhaust gas from
the engine is detected and the fuel injection time against the sucked air
flow rate is controlled in such a manner that the detected oxygen
concentration corresponds to the theoretical air-fuel ratio. This system
takes advantage of the fact that the oxygen concentration of the exhaust
gas is sharply reduced with the increase in the air-fuel ratio in the
neighbourhood of the theoretical air-fuel ratio. The oxygen concentration
in the exhaust gas is generally detected by an oxygen sensor including an
air-permeable zirconia solid electrolyte. This control system generally
employs a closed loop and therefore is effective in control to hold the
engine operating condition at the theoretical air-fuel ratio. When the
engine is required to be operated in the condition displaced from the
theoretical air-fuel ratio as in the automobile warm-up, acceleration,
running up or down a comparatively steep slope or along a freeway,
however, such a control system cannot maintain the proper operating
condition, since it is necessary to operate the engine with the mixture
gas whose the fuel concentration is thicker or thinner than the
theoretical air-fuel ratio. Further, when the air flow rate is sharply
changed, such a control system cannot maintain the proper operating
condition, since the time lag in which the air-fuel ratio information is
transferred from oxygen sensor to injection valve is too long. In such
special operating conditions, the closed loop control based on the data of
the theoretical air-fuel ratio from the oxygen sensor is impossible, so
that an open loop control is employed in which control is based on a
predetermined air-fuel ratio determined as suitable for each of the
above-mentioned special operating conditions. In such an open loop
control, it is quite impossible to correct the change with time of the
functions or response of the air flow rate sensor or injector. In such
special operating conditions, therefore, neither fuel consumption is saved
nor exhaust gas is purified properly and smoothly.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a control system for the
internal combustion engine in which the air-fuel ratio of the mixture gas
sucked into the engine is controlled on the basis of the data of the
theoretical air-fuel ratio detected by the oxygen sensor under the normal
operating conditions, the data of the air-fuel ratio for the normal
operating conditions is stored in a memory, and the air-fuel ratio is
controlled under the special operating conditions by correcting the
air-fuel ratio on the basis of the stored data.
Another object of the invention is to provide a control system for the
internal combustion engine in which the air-fuel ratio of the mixture gas
in the control system according to the first-above written object is
controlled in accordance with the ranges of magnitude of the sucked air
flow rate under the special operating conditions.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a diagram for explaining the general configuration of the control
system for the internal combustion engine according to the present
invention.
FIG. 2 is a diagram for explaining the relation in timing between fuel
injection and the operating condition of each engine cylinder as related
to the rotational angle of the crank shaft.
FIG. 3 is a block diagram showing in detail the control circuit shown in
FIG. 1.
FIG. 4 is a flow chart for the control system according to the present
invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
A configuration of the essential parts of the electronic engine control
system is shown in FIG. 1. The air taken in through an air cleaner 12 is
applied to an air flow meter 14, where the flow rate thereof is measured.
An output QA representing the air flow rate is supplied from the flow
meter 14 to a control circuit 10. The air flow meter 14 includes a sucked
air temperature sensor 16 for detecting the temperature of the sucked air.
The sensor 16 produces an output TA representing the temperature of the
sucked air, which is applied to the control circuit 10.
The air that has passed the air flow meter 14 is then passed through the
throttle chamber 18. The amount of air passing the throttle chamber 18 is
controlled by changing the opening of the throttle valve 20 in the
throttle chamber 18 and mechanically interlocked with the accelerating
pedal 22. The valve position detector 24 detects the opening of the
throttle valve 20 by detecting the position of the throttle valve 20. The
signal QTH representing the position of the throttle valve 20 is applied
from the throttle position detector 24 to the control circuit 10. The air
that has passed the throttle chamber 18 is sucked into the combustion
chamber 34 through the intake manifold 26 and the suction valve 32. Thus
the amount of air sucked into the combustion chamber 34 is regulated by
the accelerating pedal 22.
The throttle chamber 18 includes a bypass 42 for idling and an idle adjust
screw 44 for regulating the amount of air flowing through the bypass 42.
While the engine is idling, the throttle valve 20 is closed up. The sucked
air from the air flow meter 14 flows through the bypass 42 and is sucked
into the combustion chamber 34. The amount of sucked air when the engine
is idling, therefore, may be changed by operation of the idle adjust screw
44. The energy generated at the combustion chamber 34 is substantially
determined by the amount of air flowing in from the bypass 42, and
therefore, the engine rotational speed under the idling state is adjusted
at proper value by changing the amount of sucked air into the engine by
regulating the idle adjust screw 44.
The throttle chamber 18 further includes another bypass 46 and the air
regulator 48. The air regulator 48 regulates the amount of air passing
through the path 46 in response to the output signal NIDL of the control
circuit 10 so as to regulate the amount of air supplied to the engine in
accordance with the control response of the fuel injection when the engine
is under warm-up state or when the throttle valve 20 undergoes a sudden
change or especially when it is closed suddenly. Also, the air flow rate
at the time of idling may be changed if required.
Next, the fuel supply system will be described. The fuel stored in the fuel
tank 50 is sucked by the fuel pump 52 and supplied under pressure to the
fuel damper 54. The fuel damper 54 absorbs the pressure pulsation of fuel
from the fuel pump 52 and supplies fuel of predetermined pressure to the
fuel pressure regulator 62 through the fuel filter 56. The fuel from the
fuel pressure regulator 62 is supplied under pressure to the fuel injector
66 through the fuel pipe 60. In response to the output INJ of the control
circuit 10, the injection valve of the fuel injector 66 opens and fuel is
injected.
The amount of fuel injected from the fuel injector 66 is determined by the
valve open time of the injector 66 and the difference between the pressure
of the fuel supplied to the injector and the pressure in the intake
manifold 26 into which fuel is injected. The amount of fuel injected from
the fuel injector 66, however, preferably depends solely on the valve open
time determined by the signal produced by the control circuit 10. Thus the
pressure of the fuel supplied by the fuel pressure regulator 62 to the
fuel injector 66 is controlled in such a manner that the pressure
difference between the fuel to the fuel injector 66 and the intake
manifold 26 is kept constant. The fuel pressure regulator 62 includes a
diaphragm 62A operated in response to the pressure difference on both
sides thereof and a needle adjust valve with the valve body 62B fixed to
the diaphragm 62A for adjusting the flow rate of fuel returned to the fuel
return pipe 58. One of the chambers of the fuel pressure regulator 62 is
supplied from the fuel pump 52 with fuel of a pressure slightly higher
than proper fuel pressure, while the other chamber thereof is impressed
with the intake manifold pressure through the conduction pipe 64. When the
fuel pressure in the fuel pipe 60 exceeds a predetermined level as
compared with the intake manifold pressure, the fuel pipe 60 communicates
with the fuel return pipe 58, so that fuel corresponding to the excess
pressure is returned to the fuel tank 50 through the fuel return pipe 58.
In this way, the difference between the fuel pressure in the fuel pipe 60
and the manifold pressure in the intake manifold is kept constant.
The fuel tank system 50 further includes a pipe 58 for absorbing the
gasified fuel and a canister 70. While the engine is running, air is
sucked from the atmospheric opening and the fuel gas thus absorbed is
introduced to the intake manifold 26 by the pipe 72. When the engine is
stationary, on the other hand, the fuel gas is discharged into atmosphere
through activated carbon.
As explained above, the fuel is injected from the fuel injector 66 and the
suction valve 32 is opened in synchronism with the motion of the piston
74, thus introducing the air-fuel mixture gas to the combustion chamber
34. By compression and the resulting combustion of this mixture gas by the
spark energy from the ignition plug 36, the combustion energy of the
mixture gas is converted into kinetic energy for operating the piston.
The combusted mixture gas is discharged as an exhaust gas into the
atmosphere through an exhaust valve (not shown), the exhaust tube 76, the
catalyst converter 82 and the muffler 86. The exhaust tube 76 includes the
exhaust gas recycle tube 78 through which part of the exhaust gas is led
to the intake manifold 26. In other words, part of the exhaust gas is
returned to the suction side of the engine. The amount of exhaust gas thus
returned is determined by the degree of opening of the valve of the
exhaust gas recycle apparatus 28. The degree of valve opening is
controlled by the output EGR of the control circuit 10. The valve position
of the exhaust gas recycle apparatus 28 is converted into an electrical
signal and in the form of signal QE, applied to the control circuit 10.
The amount of nitrogen oxide contained in the exhaust gas increases in
proportion to the combustion temperature in the cylinders. Therefore, the
amount of oxygen is required to be reduced if the combustion temperature
is to be reduced. For this purpose, water, methanol or carbon dioxide is
mixed with the sucked air. The exhaust gas recycle apparatus 28 so
operates that the exhaust gas most of which comprises carbon dioxide is
mixed with the sucked air, thus reducing the combustion temperature in the
combustion chamber.
The exhaust tube 76 includes a .lambda. sensor 80 for detecting the mixing
ratio of mixture gas sucked into the combustion chamber 34. This .lambda.
sensor 80 generally takes the form of oxygen sensor (O.sub.2 sensor) and,
detecting the oxygen concentration in the exhaust gas, generates a voltage
V.lambda. corresponding to the oxygen concentration. The output V.lambda.
of the .lambda. sensor 80 is applied to the control circuit 10. The
catalyst converter 82 includes an exhaust gas temperature sensor 84, so
that the output TE corresponding to the exhaust gas temperature is applied
to the control circuit 10.
The control circuit 10 has a negative power terminal 88 and a positive
power terminal 90. From the control circuit 10, the signal IGN for
controlling spark generation of the ignition plug 36 as mentioned above is
applied to the primary winding of the ignition coil 40. A high voltage
thus produced at the secondary winding is applied through the distributor
38 to the ignition plug 36, thereby generating a spark for combustion in
the combustion chamber 34. More specifically, the ignition coil 40 has a
positive power terminal 92, and the control circuit 10 has a power
transistor for controlling the primary winding current of the ignition
coil 40. A series circuit including the primary winding of the ignition
coil 40 and the power transistor is formed between the positive power
terminal 92 of the ignition coil 40 and the negative power terminal 88 of
the control circuit 10. By the turning on of the power transistor,
electromagnetic energy is stored in the ignition coil 40, while by the
turning off of the power transistor, the electromagnetic energy is applied
to the ignition plug 36 as energy of high voltage.
The engine 30 has a water temperature sensor 96 for detecting the
temperature of the engine cooling water 94. The water temperature sensor
96 applies the signal TW associated with the detected temperature to the
control circuit 10. Further, the engine 30 has an angle sensor 98 for
detecting the rotational angle of the engine. The sensor 98 produces a
reference signal PR every 120 degrees of engine rotation in synchronism
with engine rotation, and an angular signal PC at each predetermined
angle, for instance, 0.5 degrees, of engine rotation. These signals are
applied to the control circuit 10. The number of revolutions of the
crankshaft is easily determined from the reference signal PR.
In FIG. 1, the air flow meter 14 may be replaced with a negative pressure
sensor. In the drawing, such a negative pressure sensor is shown by a
dotted line, and applies to the control circuit 10 a voltage VD
corresponding to the negative pressure of the intake manifold 26.
Specifically, the negative pressure sensor 10 may take the form of a
semiconductor negative sensor. The boost pressure of the intake manifold
is caused to act on one side of the negative pressure sensor and the
atmospheric pressure or a fixed pressure on the other side thereof. Such a
pressure may be vacuum. In such a construction, the voltage VD
proportional to the manifold pressure is generated by the piezo resistance
effect or like and applied to the control circuit 10.
The diagram of FIG. 2 is for explaining the ignition timing as relative to
the crank angle of the six-cylinder engine and the fuel injection timing.
(A) shows the crank angle. A reference signal PR is produced from the
angle sensor 98 each 120 degrees of crank angle. In response to this
signal, the control circuit 10 produces a signal at the crank angles of
0.degree., 120.degree., 240.degree., 360.degree., 480.degree., 600.degree.
and 720.degree..
In the drawing under consideration, (B), (C), (D), (E), (F) and (G) show
the operation of the first, fifth, third, sixth, second and fourth
cylinders respectively. J1 to J6 show the opening positions of the suction
valves of the respective cylinders. The valve opening positions of the
cylinders are displaced by 120 degrees of the crank angle T as shown in
FIG. 2. Somewhat depending on the engine structure, the valve opening
positions and the valve opening width are substantially reflected in the
drawing.
In the drawing, A1 to A5 show the valve open timing of the fuel injector
66, i.e., the fuel injection timing. The length JD of the respective
injection time A1 to A5 represent the valve open time of the fuel injector
66. This time length JD may be considered to represent the amount of fuel
injection from the fuel injector 66. The fuel injector 66 is provided for
each cylinder and connected in parallel to the drive circuit in the
control circuit 10. In response to the signal INJ from the control circuit
10, the fuel injector for each cylinder opens the valve thereof
simultaneously for fuel injection. The signal INJ is for determining the
pulse width of the fuel injection.
Explanation will be made below of the first cylinder shown in (B) of FIG.
2. In synchronism with the reference signal INTIS generated at the crank
angle of 360 degrees, the output signal INJ from the control circuit 10 is
applied to the fuel injector 66 provided at the intake port or manifold of
each cylinder. As a result, fuel is injected as shown by A2 for the time
period JD calculated by the control circuit 10. Since the suction valve of
the first cylinder is closed, however, the injected fuel is held in the
neighbourhood of the intake port of the first cylinder but not sucked into
the cylinder. Next, in response to the reference signal INTIS generated at
the crank angle of 720 degrees, a signal is applied from the control
circuit 10 again to the respective fuel injectors 66, thus injecting fuel
as shown by A3. Almost simultaneously with this injection, the suction
valve of the first cylinder is opened so that both the fuel injected in A2
and fuel injected in A3 are sucked into the combustion chamber. The same
applies to the other cylinders. Thus, in the fifth cylinder shown by (C),
the fuel injected in A2 and A3 is sucked into the combustion chamber at
the valve open position J5 of the intake valve. In the third cylinder
shown by (D), part of the fuel injected in A2, fuel injected in A3 and
part of the fuel injected in A2 are sucked into the combustion chamber at
the open position J3 of the suction valve. The part of the fuel injected
in A2 combined with the fuel part injected in A4 makes up the amount of
fuel for one injection. Thus the amount of fuel for two injections is
taken in each suction stroke of the third cylinder. Similarly, in the
sixth, second and fourth cylinders shown in (E), (F) and (G), the amount
of fuel corresponding to two injections of the fuel injector are sucked in
a single suction stroke. As apparent from the foregoing explanation, the
amount of fuel injected which is designated by the fuel injection signal
INJ from the control circuit 10 is one half that of fuel required for
suction, so that the amount of fuel commensurate with the air sucked into
the combustion chamber 34 is obtained by two injections by the fuel
injector 66.
In FIG. 2, G1 to G6 show ignition timings for the first to sixth cylinders.
By turning off the power transistor in the control circuit 10, the current
in the primary side of the ignition coil 40 is cut off, thus generating a
high voltage in the secondary coil thereof. This high voltage is generated
in timing with the ignition timings G1, G5, G3, G2 and G4, and distributed
to the ignition plug 36 of each cylinder through the distributor 38. Thus
the ignitions plugs of the first, fifth, third, sixth, second and fourth
cylinders are started in that order, thus combusting the mixture gas of
fuel and air.
The detailed circuit configuration of the control circuit 10 of FIG. 1 is
shown in FIG. 3. The positive power terminal 90 of the control circuit 10
is connected to the positive terminal 110 of the battery, so that a
voltage VB is supplied to the control circuit 10. The source voltage VB is
kept at constant voltage PVCC, say, 5 V, by the constant voltage circuit
112. This constant voltage PVCC is applied to the central processor
(hereinafter referred to as CPU) 114, the random access memory
(hereinafter referred to as RAM) 116, and the read only memory
(hereinafter referred to as ROM) 118. Further, the output PVCC of the
constant voltage circuit 112 is applied to the input/output circuit 120.
The I/O circuit 120 includes a multiplexer 122, an analog-to-digital
converter 124, a register 125, a pulse output circuit 126, a pulse input
circuit 128 and discrete I/O circuit 130.
The multiplexer 122 is impressed with an analog signal. In response to a
command from CPU, one input signal is selected and applied to the A/D
converter 124. The analog input signals QA applied to the multiplexer 122
via the filters 134, 136, 138, 140 and 144 include the analog signal TW
representing the temperature of the engine cooling water, the analog
signal TA indicating the temperature of the sucked air, the analog signal
TE showing the temperature of exhaust gas, the analog signal QTH
indicating the throttle opening, the analog signal QE showing the valve
open state of the exhaust gas recycle apparatus 28, the analog signal
V.lambda. representing the oxygen concentration of exhaust gas, i.e., the
excess air in the sucked mixture gas and the analog signal QA showing the
amount of air sucked in, which are produced from the water temperature
sensor 96, the sucked air temperature sensor 16, the exhaust gas
temperature sensor 84, the throttle position detector 24, the exhaust gas
recycle apparatus 28, the .lambda. sensor 80 and the air flow meter 14
shown in FIG. 1, respectively. The output V.lambda. of the .lambda. sensor
80, however, is low in voltage level and therefore applied to the
multiplexer through the amplifier 142 having a filter circuit.
Also, the analog signal VPA representing the atmospheric pressure, which is
produced from the atmospheric pressure sensor 146, is applied to the
multiplexer 122. The voltage VB is applied to the series circuit including
the resistors 150, 152 and 154 through the resistor 160, from the positive
power terminal 90. The voltage across the series circuit of the resistors
is maintained constant by the zener diode 148. The voltages VH and VL at
the junctions points 156 and 158 between the resistors 150 and 152 and
between the resistors 152 and 154 respectively are applied to the
multiplexer 122.
The above-mentioned CPU 114, RAM 116, ROM 118 and I/O circuit 120 are
connected with each other by the data bus 162, address bus 164 and control
bus 166. Further, clock signals E are applied from CPU to RAM, ROM and I/O
circuit 120, so that data is transmitted via the data bus 162 in
synchronism with the clock signal E.
The multiplexer 122 of the I/O circuit 120 is impressed with the water
temperature signal TW, sucked air temperature signal TA, exhaust gas
temperature signal TE, throttle opening signal QTH, exhaust gas recycle
rate signal QE, .lambda. sensor output V.lambda., atmospheric pressure
signal PVA, reference voltages VH and VL and the sucked air amount signal
QA or negative pressure signal VD. The addresses of these inputs are
designated by the CPU 114 through the address bus according to the command
program stored in ROM 118, so that the analog input of the addresses
designated is taken in. This analog input is applied from the multiplexer
122 to the analog-to-digital converter 124. The digitally converted value
is held in the register 125 corresponding to the respective inputs, and
then applied to CPU 114 or RAM 116 in response to the command sent from
CPU 114 via the control bus 166, as required.
The reference pulse P and the angle signal PC in the form of pulse train
are applied from the angle sensor 98 to the pulse input circuit 128
through the filter 168. Further, the pulses PS of the frequency
corresponding to the vehicle speed are applied from the vehicle speed
sensor 170 to the pulse input circuit 128 through the filter 172.
The signal processed by CPU 114 is held in the pulse output circuit 126
having the functions of a register. One of the outputs of the pulse output
circuit 126 is applied to the power amplifier circuit 186, on the basis of
which the fuel injector is controlled.
Reference numerals 188, 194 and 198 show power amplifier circuits for
controlling the current in the primary side of the ignition coil 40, the
opening of the exhaust gas recycle apparatus 28 and the opening of the air
regulator 48 in response to the output pulses from the pulse output
circuit 126. The discrete I/O circuit 130 receives and holds, via the
filters 180, 182 and 184, signals from the switch 174 for detecting the
closed-up state of the throttle valve 20, the starter switch 176 and the
gear switch 178 indicating that the transmission gear is in "top speed"
position, respectively. Further, the processed signal from the central
processing unit CPU 114 is held. The signal associated with the discrete
I/O circuit 130 is one capable of indicating the content thereof by one
bit. In response to the signal from the central processing unit CPU 114,
signals are applied from the discrete I/O circuit to the power amplifier
circuits 196, 200, 202 and 204 for such operations as closing the exhaust
gas recycle apparatus 28 to stop exhaust gas recycle, control of the fuel
pump, indication of an abnormal temperature of the catalyst on the lamp
208 and indication of engine overheated condition on the lamp 210.
The air-fuel ratio for the internal combustion engine is given as
R.varies.QA/nTi (1)
where QA is a signal representing the amount of sucked air detected by the
air flow sensor 14 in FIG. 1, n the number of engine revolutions
determined by dividing the pulses obtained from the angle sensor 98, and
Ti the injection pulse width corresponding to the open time of the
injection valve of the fuel injector 66. From equation (1), the injection
pulse width Ti is expressed as
Ti.varies.QA/nR (2)
Under the normal operating condition of the internal combustion engine, the
injection pulse width Ti is subjected to closed-loop control on the basis
of the sucked air mount QA and number of revolutions n in such a manner
that the theoretical air-fuel ratio Ro is attained, utilizing the fact
that the output of the .lambda. sensor 80 suddenly changes in the
neighborhood of the theoretical air-fuel ratio. The sucked air mount QA is
divided into five ranges from zero to maximum. The number of ranges into
which QA is divided may alternatively be eight or more, as desired. The
value of the right side Ro for a brand new vehicle which has not yet been
driven in the streets after being manufactured in the factory is expressed
as
Ro=QA/nTi (3)
The values QA, n and Ti change with time. The value Rti of the right side
of equation (1) after such changes with time is give as
Rti=QA/nTi (4)
The ratio Ki between Rti and Ro for each range of the sucked air amount is
Ki=Rti/Ro (5)
For a brand new vehicle, Rti=Ro, and therefore Ki=1. The value Ki is a
correction factor for the change with time of the performance of the air
flow meter 14 and injector 66.
A flow chart for explaining the operation of an embodiment of the present
invention is shown in FIG. 4. RAM 116 includes a nonvolatile memory which
keeps information in store even when power is thrown off. The step 1 of
FIG. 4 is concerned with a brand new vehicle in the initial state having
not yet experienced the driving in the streets. Under this condition, "1"
is written in the memory sections K.sub.1, K.sub.2, K.sub.3, K.sub.4 and
K.sub.5 of the nonvolatile memory of RAM 116. At the same time, the value
Ro is calculated by CPU 114 according to the program of ROM 118 and stored
in RAM 116. In step 2, the continuous number N in RAM 116 stored in RAM
116 is set to "0".
The mixture gas of a certain air-fuel ratio comprised of sucked air and the
fuel injected from the fuel injector 66 is combusted in the combustion
chamber 34 and discharged into the exhaust gas tube 76. It requires
approximately 100 msec on the average for the air at the injector 66 to
reach the .lambda. sensor 80. If normal operation continues during this
period, the air-fuel ratio at the position of the injector 66 is
considered identical to that before combustion of the exhaust gas at the
position of the .lambda. sensor 80. This air-fuel ratio is the theoretical
one based on the output of the .lambda. sensor 80. If normal operation is
changed to a special operating condition or a special operating condition
continues during the 100-msec period, on the other hand, the output of the
.lambda. sensor 80 indicates a value different from the theoretical
air-fuel ratio and therefore the air-fuel ratio of the mixture gas at the
position of the injector 66 is not the theoretical value.
Step 3 in FIG. 4 is such that the number of revolutions n, the injection
pulse width and the sucked air amount QA for N=0 are stored in the
corresponding memory sections of the register 125. In step 4, whether the
engine is in normal operating condition or not is determined. Under the
normal operating condition, the output of the .lambda. sensor changes
suddenly in the neighbourhood of the theoretical air-fuel ratio. When it
is decided that the output of the .lambda. sensor is within the range
V.sub.1 to V.sub.2 corresponding to the theoretical air-fuel ratio, the
engine is considered in the normal operating condition and advance is made
to the next step 5.
It takes 100 msec for the gas to proceed from the injection position of the
injector 66 to the .lambda. sensor 80. After confirming that the normal
operating condition continues for at least 100 msec, the values QA, Ti and
n stored in step 3 may be processed as values giving the theoretical
air-fuel ratio subjected to closed loop control. Assume that the
continuous number N has the maximum value Nmax of 10. If 10 msec is set
for one continuous number N, it takes 100 msec before N reaches 10. The
period of N may be conveniently in proportion to the flow velocity in
synchronism with engine rotation instead of being fixed. In step 5 in FIG.
4, whether or not the continuous number N has reached maximum Nmax is
determined. If N is smaller than Nmax, "1" is added to N in step 6 and a
delay time to attain 10 msec for one continuous number N is given in step
7, from which the process is returned to step 4. After that, a similar
process from steps 4 to 7 is repeated. When N reaches Nmax in step 5,
advance is made to step 8 where the values QA, Ti and n stored in the
register 125 in step 3 are read out and the value Rti of equation (4) is
calculated according to the program stored in ROM 118. In step 9, the
value Ro is read out of the RAM 116 and value Ki is calculated according
to equation (5). This value Ki is rewritten to the correction factor Ki
for the range corresponding to the sucked air amount in step 10. The
process is then returned to step 2 for repetition of a similar operation.
In one operation from steps 2 to 10, the correction factor for only one of
the five ranges of the sucked air amount is rewritten. Although the
performance of the air flow meter 14 or injector 66 changes with time on
the order of day or month, the correction factor table for the nonvolatile
RAM is rewritten at time intervals on the order of second. Thus the table
is rewritten sufficiently prior to the change with time of such devices.
Instead of the sucked air amount QA, the negative pressure VD of the
negative pressure sensor 100 may be used with equal effect.
If "No" is the answer at step 4, the engine is in a special operating
condition, under which the air-fuel ratio is controlled at a value
different from the theoretical ratio. In warm-up, acceleration or driving
up a slope, for example, the air-fuel ratio is reduced below the
theoretical value; while in deceleration or driving down a slope, the
air-fuel ratio is controlled at a larger value than the theoretical one.
The 256 values (16.times.16) of air-fuel ratio .lambda. corresponding to
the number of revolutions n and the sucked air amount divided into 16
ranges are tabulated and stored in ROM 118. Under a special operating
condition, the air-fuel ratio suitable for that operation is retrieved
from the table of ROM 118 and the injection pulse width Ti based on the
air-fuel ratio .lambda. is determined from equation below.
Ti=QA/.lambda.n (6)
thus setting the injection time Ti. It should be noted that the value
.lambda. for the conventional systems is already set at the time of
assembly of ROM 118 and therefore not corrected for any change with time
of the sensors or other operating devices and that in the conventional
systems, the air-fuel ratio is not controlled on the basis of the .lambda.
sensor under the special operating conditions where the system is
subjected to open-loop control. According to the present invention, in
spite of the open-loop control under the special operating conditions, the
injection pulse width Ti is calculated from the equation (7) below
T.sub.1 =QA/Ki.lambda.n (7)
taking into consideration the correction factor Ki for the air-fuel ratio.
It is thus possible to drive the vehicle under the special operating
conditions at the air-fuel ratio intended in design stage, which has been
corrected against the change with time of the performance of the air flow
meter 14 and injector 66.
In other words, step 11 is one in which the air-fuel ratio associated with
the special operating condition at the particular time is retrieved | | |
