JPH01273849A - Air-fuel ratio controller for engine - Google Patents

Abstract

PURPOSE:To prevent the deterioration of the learning precision by suspending the learning of the correction value for compensating the deviation between the banks when the learning condition in one system is not established and by continuing the feedback control in the pertinent system, in the device equipped with a plurality of feedback control systems. CONSTITUTION:The air-fuel ratio is feedback-controlled by each feedback control means B on the basis of the output of an air-fuel ratio detecting means A which detects the air-fuel ratio in the exhaust gas in each exhaust system of at least two cylinder groups. The first judging means C for judging if the feedback control execution condition is established or not is installed. Further, a learning means D for learning the correction value for compensating the deviation between the intake and exhaust systems for each cylinder group and the second judging means E which judges if the prescribed learning conditions are established or not are installed. Therefore, when the learning condition of one cylinder group is not established, a control means F continues the feedback control of the cylinder group until the feedback condition of other cylinder group is not established.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an air-fuel ratio control device for an engine having at least two or more air-fuel ratio feedback control systems, and particularly relates to an air-fuel ratio control device for an engine having at least two or more air-fuel ratio feedback control systems. Concerning improvement of combustion conditions.

(Prior art) The technology of feedback controlling the air-fuel ratio of the air-fuel mixture supplied to the engine based on the output of an air-fuel ratio sensor installed in the exhaust system is well known in order to improve emission characteristics. .

By the way, a case is assumed in which the vehicle is stopped from a running state and maintained in an idling state while performing air-fuel ratio feedback control. At this time, the engine temperature, the temperature of the exhaust gas purification device, etc. naturally decrease. For example, an air-fuel ratio sensor such as an oxygen sensor has an active region, and outside this active region, the sensor output becomes unreliable, so feedback control actually worsens emissions. Furthermore, since combustibility deteriorates as the engine temperature decreases, it is not preferable to maintain the air-fuel ratio of the air-fuel mixture at the stoichiometric air-fuel ratio in order to perform feedback control. Therefore, in a state where the conditions for performing such feedback control are not satisfied, it is conventional practice to stop feedback control and perform oven loop control.

On the other hand, especially in the case of type ■ engines, the exhaust system is divided into two banks (left and right), and an oxygen sensor is installed in the exhaust system of each bank, and the output of each of these two sensors is Based on this, there is a technique (Japanese Patent Application Laid-Open No. 61-118538) that performs feedback control of the air-fuel ratio for both banks independently.

JP-A No. 61-118538 discloses that if feedback control → oven loop control is performed independently between each bank, control modes may differ, and in such cases, engine drivability deteriorates. This was done in view of certain points. A case where the control mode must be different is, for example, a case where the temperature conditions between the left and right banks become different due to wind direction or the like. and,
The technology of JP-A-Sho mentioned above is such that when the left and right banks shift to a state where the control mode is different, the control mode of the bank to which the control mode has shifted first is forced to control the other bank. Either the banks are changed at the same time, or the control of the bank to which the control mode is to be transferred is forcibly continued until the control of the other bank is changed.

(Problem to be Solved by the Invention) However, when attempting to perform feedback control independently in the two systems, there are differences in the resistance of the intake system, differences in the resistance of the exhaust system, and air-fuel ratio sensors between the two systems. Since there are differences in response, etc., it is necessary to make corrections to compensate for the inherent differences between these two systems. and,
Correction to compensate for the difference in resistance of the intake system and the difference in resistance of the exhaust system is necessary even when feedback control is stopped and transition is made to oven loop control.

These corrections are performed based on learning during the feedback control period. Therefore, although the accuracy of learning depends on the balance between the timing at which feed pack control is stopped and the timing at which learning is performed, it is necessary to If you ignore learning and simply match the control between the two systems, the accuracy of learning will be lost, and then,
Air-fuel ratio control based on this learning will also not be appropriate.

Therefore, the present invention was proposed in order to eliminate the drawbacks of the above-mentioned conventional example, and its purpose is to prevent the above-mentioned learning accuracy from decreasing in an air-fuel ratio control device having two or more systems of feedback control. This paper proposes an air-fuel ratio control device for an engine.

(Means and effects for solving the problems) The configuration of the present invention for achieving the above-mentioned problems is as shown in FIG. The air-fuel ratio of the air-fuel mixture supplied to the cylinders of each group is set to a target value based on the output of the air-fuel ratio sensor arranged in the exhaust system of each group and the sensor of that group for each cylinder group. a feedback control execution means for performing feedback control; a first determination means for determining whether the execution condition for the feedback control is satisfied; and a first determination means for determining whether or not the execution condition for the feedback control is satisfied, a learning means for learning a correction value for compensating for deviations between systems; a second determining means for determining whether the predetermined learning condition is satisfied; and a learning means for determining whether the predetermined learning condition is satisfied for any cylinder group. , the learning for that group is stopped, and at least until the feedback control execution conditions of other cylinder groups are not satisfied, the above-mentioned The present invention is characterized by comprising a control means for controlling to continue feedback control of a cylinder group for which a learning condition is not satisfied.

(Example) An example in which the present invention is applied to a 6-cylinder ■type engine will be described below with reference to the accompanying drawings.

<Overview of Example System> As shown in FIG. 2, the engine 10 includes an air filter 12 that filters intake air.
The air filtered by 2 passes through the intake pipe 14, passes through the intake valves 16a and 16b, and enters the cylinders 1 arranged in a figure 7 shape.
8a, 18b. In addition, in this FIG. 2, other four cylinders are arranged on the opposite side of the drawing, but are not shown. Also, each of the six cylinders, the first cylinder ~
If it is the 6th cylinder, the 1st cylinder is 18a and the 2nd cylinder is 18a.
8b, the cylinder 18a and the two cylinders on the opposite side form a left bank cylinder, and the cylinder 18b and the two cylinders on the opposite side form a right bank cylinder.

The upstream side of the intake pipe 14 has a flow rate (
An air flow meter 20 that measures the temperature (T8) of the intake air and an intake air temperature sensor 22 that measures the temperature (T8) of the intake air are attached.

The middle part of this intake pipe 14 is connected to each left and right bank.
The passage is divided into two passages, and each divided passage is provided with a throttle valve 24a, 24b. An opening sensor 27 is attached to one throttle valve 24a to detect its opening (TVO). In addition, the intake pipes 1 on the upstream and downstream sides of the throttle valves 24a and 24b
A bypass pipe 26 is connected to the intake pipe 14 in a state where the portion 4 is bypassed. An idle speed control valve 28 is attached to a midway portion of the bypass pipe 26. When there is A/C operation, power steering operation, heater operation, etc., the on/off operation of this ISO valve 28 is controlled according to the signal ISO according to the load, and the air flow rate (bypass air volume) will be controlled.

Further, the downstream side of the intake pipe 14 is branched and connected to cylinders 18a, 18b, etc. belonging to the left and right banks, and each branch connection pipe has a fuel injection valve 34a, 34b that supplies fuel into the corresponding cylinder. It is arranged. For each of the fuel injection valves 34a and 34b, a fuel injection pulse (τ6, τB) width is determined by an engine control unit (hereinafter referred to as single EcU) 36, which will be described later. In addition, the subscript “
"A" refers to a signal related to the left bank system, and "B" refers to a signal related to the right bank system.

Inside the left and right cylinders 18a and 18b, a piston 32a is provided.
, 32b are slidably disposed.

By sliding and reciprocating within the cylinders 18a and 18b, each piston is connected to the crankshaft (
(not shown) is rotated. Also, although not shown, the cylinder 18a.

A spark plug is provided at the top of 18b to combust the fuel injected here. Further, each cylinder 18a, 18b is provided with a cooling water passage 40a, 40b for cooling each cylinder, and the temperature of the cooling water passing through the cooling water passage is adjusted to the engine temperature (T,
) A water temperature sensor 42 is attached to detect the water temperature.

On the other hand, the fuel burned in each cylinder 18a, 18b is passed through the corresponding exhaust valve 44a, 44b as exhaust gas to the exhaust manifold 46a, 46b, and then to the exhaust pipe 48.
a, 48b. This exhaust pipe 48a, 4
Catalytic converters 50a and 50b for purifying exhaust gas are provided in the middle of 8b.

As in the case of the intake pipe, in order to suppress mutual interference of exhaust gas, the exhaust manifold 46a from the left bank cylinder 18a and the exhaust manifold 46C from the remaining two left bank cylinders (from the other one cylinder) (exhaust manifold not shown) are combined into one exhaust pipe 48a,
The exhaust manifold 46b of the right bank cylinder 18b and the exhaust pipe 46d from the remaining right bank cylinders (the exhaust manifold of the other cylinder is not shown) are combined into one exhaust pipe 48b. 02 sensors 52a, 52b are attached to the exhaust pipes 48a, 48b to measure the oxygen concentration (o2) remaining in the exhaust gas passing through the exhaust gas at a position as close as possible to the above-mentioned merging point. The oxygen concentration output signals from these sensors 52a and 52b are respectively
Let EA be Ea.

53a and 53b are catalytic converters 5°a and 50b, respectively.
temperature sensor (output is TAC, Tac). Based on the output EA from the oxygen sensor 52a, air-fuel ratio control for the left bank (cylinder 18a, etc.) is performed, and the oxygen sensor 5
Based on the output Ea from 2b, the right bank (cylinder 1
8b, etc.) is performed independently between each bank.

25 is a fuel tank. The fuel gas components evaporated within this tank are trapped in the canister 19. This trapped gas is controlled by the signal PC from the ECU 36.
By opening the solenoid valve 17, the pipe 15a
, 15b, and is supplied to the left and right banks. As will be described later, the E CtJ 36 can detect the start and end of vaporized fuel supply by monitoring this signal PC.

FIG. 3 shows the signals input to the ECU 36 and the signals input to the ECU 36.
This is a collection of signals output from. Although not shown in FIG. 1, as shown in FIG. 3, a signal GP, which is a signal from a transmission (not shown) and indicates the current gear position;
A signal I that is a signal from a switch provided in the throttle opening sensor 27 and indicates that it is in an idle state.
DLE, and a signal N indicating the engine speed from the distributor etc. are also input to the ECU 36.

Although not shown in FIG. 2, the ECU 36 has a battery-backed R for storing learned values.
AM is provided.

Principle of Operation> The principle of the air-fuel ratio subscript of this embodiment will be explained with reference to FIG. In the figure, CAFB is a correction coefficient for feedback control to the target air-fuel ratio (stoichiometric air-fuel ratio) in the left bank system. Flag F ALN is a flag indicating that learning should be performed in the left bank system, Flag F
A8TP is a flag indicating that feedback control should be performed in the left bank system. CIIFII
, FIILN, FBIITP are the above correction coefficients and flags for the right bank system.

FIG. 4 shows a case where the learning condition for the right bank system becomes unsatisfied first, and then the learning condition for the left bank system becomes unsatisfactory. Here, the learning condition refers to the case where the logical formula % formula %) * (GP・0) * Sensor normal force (continues for 10 seconds or more, and if this is satisfied, the flag is If F ALN or F BLN is set and is not established, those flags are reset.

More specifically, whether or not it is in the F/B zone is determined based on, for example, engine rotation speed N, vehicle speed, and catalytic converter temperature (T
AC, Tac), etc. from a predetermined map. As will be described later, in this embodiment, being in the feedback zone is different from the feedback control execution conditions. The learning condition is satisfied when the engine is in the idle state, the engine water temperature Tw is 75°C or higher, and the transmission is in the neutral position ( GP=O), and the air flow meter 20, air-fuel ratio sensor (52a, 52b),
This refers to a case where the water temperature sensor 42, intake temperature sensor 22, etc. remain normal for 10 seconds or more. For example, whether or not an air-fuel ratio sensor is normal can be determined by whether a predetermined output is obtained from the sensor when a current is passed through the sensor.

According to FIG. 4, the learning condition is not satisfied (FBLN=O) in the right bank before the left bank. In this case, learning for obtaining correction coefficients for compensating for deviations between banks is immediately stopped. This stopping is performed in order not to reduce the accuracy of the learned values up to that point. However, as shown in FIG. 4, the feedback control in the left bank continues even if the learning condition is not satisfied in the right bank system. How long it will continue is until the learning condition is not satisfied in the left bank system.

The reason why the learning of the left bank system is not stopped immediately when the learning condition of the right bank system becomes unsatisfied is because the learning of the left bank system is not stopped immediately.
Since feedback control is continuing, learning is continued during that time to improve learning accuracy.

Also, one reason why feedback control of the right bank is continued even after learning is stopped in the right bank system is that during that time, there is a difference in the control state between the left and right banks (feedback control for the left bank, feedback control for the right bank). This is to prevent the occurrence of vibrations caused by differences in engine output. Second, by forcibly continuing feedback control until at least the learning condition is no longer satisfied in the left bank system, deterioration of emissions during that time can be prevented.

Details of air-fuel ratio control> In this embodiment, as described later, during feedback control, the air-fuel ratio correction coefficient CAF by feedback control is
B + CBFII, air-fuel ratio sensor (52a, 52b
) is obtained based on integral control according to the outputs (EA, Es).

And during learning (FALN = 1, FBLN =
Only 1) obtains learning correction coefficients CALN and C3LN by learning the value of CAFII + CBre. In addition, as will be described later, during learning, the correction coefficient CA
F! 1°cars is slightly corrected.

The final fuel injection amount (τ, 8) for both banks is determined by the following equation. Feedback control in progress (F ＾゛5
tp= 1 , F estp= 1 ) then, te,=
teo (1+CAFB +CALN +CO)te,=
hand. (i+c,.10CBLN+Co), and feedback control is stopped (FAIIITP"0, F
l58TP=O) is te, two τ. (1+CALN+Co)te,=tea
(1+CBLN+CO). That is, it is open loop control.

Note that C8 in the above equation is a coefficient for correction based on the intake air temperature Ta%, engine temperature Tw, etc.

Details of Air-Fuel Ratio Control> Details of the control program related to air-fuel ratio control in this embodiment will be described with reference to FIGS. 6A to 6C.

Now, step S2 in FIG. 6A. In step S4, counters CNTA, CNTe, flag FA8? P+
Initialize F a+srp. Here, the counter determines the number of times of learning, and in this embodiment, the number of times of learning is set to, for example, three times. Step S2. Step S4 is activated only when the engine is started.

In step S6, intake air ff1Q, and engine rotational speed N are input. In step S8, the basic fuel injection amount is determined based on the engine speed N and intake air ff1Q. Calculate. In step SIO, a signal IDLE and a gear position signal GP are input. In step S12, the intake air temperature T, engine temperature Tw, and catalytic converter temperature (TAc, Tac) are input. This is to determine whether the learning conditions are met. In step S13, a correction coefficient Co is calculated based on the intake air temperature T6, the engine temperature Tw, etc.

Steps S14 to S18 are based on the current output EA of the air-fuel ratio sensor 52a in the left bank system.
(2) Determine the correction coefficient CAFB for controlling (integral control). That is, if EA=1, CAFII=CAF[l-ΔIA, and if EA≠1, CAF[l=CAFB+ΔrA. ΔIA is a constant for integral control.

Care is also determined for the right bank system through steps S20 to S24.

Steps S26 to S40 are learning control and feedback control stop control for the left bank system,
Steps S42 to S56 are learning control and feedback control stop control for the right bank system.

In step S2B, it is checked whether the learning conditions for the left bank system are satisfied.

First, a case will be described in which the learning conditions are currently established for both the left and right banks.

In step S28, the feedback control stop flag F
Reset AIITP. This means that in this embodiment, feedback control is stopped only after learning has been stopped, which means that if the learning condition is satisfied, the feedback control execution condition is always satisfied. Because it does. In step S30, the learning value CA1. Update N.

CA = CALN +1. /2・(CAFB)Here, (CAys) is denoted by da CApB(i) (CAFB1) =□, and CAre(t) (i = 1~n 1
) is the correction coefficient for the past (n-1) times, and by adding the current CAre to this, (CAFE) becomes the moving average of n pieces of data.

In step S32, the correction coefficient C is calculated based on this moving average value.
AFfi is corrected according to the following formula.

CAra = CAFII l/2・(CAF
B) That is, each one of the moving average values of correction coefficients (CAFII)
/2 is reflected on one hand in the learning value CALN and on the other hand in the current CAFll. In step S34,
A counter CNTA indicating the number of times of learning is incremented. This counter CNTA is used to guarantee the minimum number of learning times that should be performed in the left bank system, and in this embodiment, this minimum number of times is three.

Regarding the right bank, when the learning conditions are satisfied,
Step 542->Step 544=>Step S46
Step 348g Step S50 is reached, where the learned value CBLH is updated, the correction coefficient C3,1 is corrected according to the learned value, and the counter CNTB is incremented.

In steps S60 to S70, the flag F is currently
Since ASTP and F as are not set, it is assumed that feedback control is in progress, and step S64 is executed.
．． In step S66, te=teo (1+CAFB +CALN +GO)
τ6=te. (1+CBpB +CIILN +CO) is calculated, and in step S72, fuel is injected from the injectors 34a and 34b according to this fuel injection amount. Then, the process returns to step S6.

Thus, as long as the learning condition is satisfied, step S
In steps 30 to S32 and steps 546 to S48, the learning values are updated, and the learning accuracy is improved. That is, as described above, each 1/2 of the moving average value (CAFB) of the correction coefficient is calculated as the learning value CALN.
If the learned value is too small, the moving average value (CAFB) is reflected in the current CAFB.
FB) increases, and the learning value CA1. This is because s is also updated to be thicker in step S30, and at the same time, the feedback correction coefficient CAFB is also corrected in step S32 so that its fluctuation range becomes smaller.

As shown in FIG. 4, assume that the learning condition for the right bank system is not satisfied first. In this case, step 542 =
=5 Step S52 is proceeding to check whether the counter CNT has reached three times or more. As mentioned above, since this counter guarantees the minimum number of learning times, generally CN Tn > 3. Proceeding to step S54, it is checked whether the learning conditions for the left bank system are satisfied.

Since this holds true for the left bank system, the process does not proceed to step S56. That is, the process does not proceed to step S56, so even if the learning condition no longer holds true, the flag Fll5TP will not be set unless the learning condition for the left bank system falls, so the feedback control will continue. This control of continuing feedback control of the other system as long as the learning condition of one system is satisfied is performed in the same way for the left bank system.

<Effects of the Example> In this way, even after the learning of one system is stopped, by continuing the feedback control of that system, the difference in the control state between the left and right banks (the left bank is under feedback control, the right bank is under feedback control). (Oven control) is prevented from occurring, and vibration generation is suppressed. Furthermore, deterioration of emissions during this period is prevented.

Furthermore, since the learning of the system in which the learning condition is not met first is immediately stopped, the accuracy of the learned values is prevented from deteriorating. On the other hand, since learning of the established system continues, its accuracy further improves.

In addition, since an appropriate minimum number of learning times is set, even if the learning condition becomes unsatisfied before learning reaches that number, learning continues (step S36 or step 552), so feedback control is performed. Learning values can be obtained with a degree of accuracy that can provide some assistance. Furthermore, since learning is stopped after about three times, there is no possibility of obtaining an extremely inaccurate learned value.

Modification of Air-Fuel Ratio Control> In the embodiment described above, the catalytic converters 50a and 50b were disposed in the left and right banks, respectively.

However, as is clear from the explanation of the operation of the above-mentioned embodiment, the present invention relates to learning accuracy between two banks. It is applicable as long as the air-fuel ratio control is performed independently based on the respective outputs.

That is, it is also applicable to a modification in which the exhaust pipes 48a and 48b are combined into one and only one catalytic converter is present.

Furthermore, the feedback control is not limited to integral control, but the present invention is also applicable to a so-called PI sub-control system that combines integral control and proportional control.

Note that step 838 in FIG. 6B. In step S54, it was checked whether the learning condition was not satisfied, but this was done in step S38.
．． When the feedback control stop flag F□TP+F88TP of the other system is checked in step S54, for example,
This is because, if the learning condition is not satisfied in both systems at the same time, a program deadlock may occur in which the feedback control continues indefinitely in both systems.

It goes without saying that this invention is not limited to the configurations of the embodiments and modified examples described above, and can be modified in various ways without departing from the gist of the invention.

(Effects of the Invention) As detailed above, according to the engine air-fuel ratio control device according to the present invention, when air-fuel ratio feedback control is performed independently between different banks, the deviation between the two banks can be reduced. Learning of correction values for compensation is stopped immediately for a certain system when the learning conditions for that system are not satisfied, but feedback control continues until the feedback control for the remaining systems is stopped, so that the learning conditions are first set. It is possible to prevent the learning accuracy of the system in which the equation is not established from decreasing, and improve the learning accuracy of the remaining systems.Furthermore, since feedback control is stopped simultaneously in all systems, output deviations based on differences in control modes can be prevented. This also prevents vibrations from the vehicle body.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the configuration of the present invention, Fig. 2 is a diagram showing the configuration of an embodiment in which the present invention is applied to a type II engine, Fig. 3 is a diagram showing signals input and output to the ECU, FIG. 4 is a diagram showing the operating principle of this embodiment, FIG. 5 is a conceptual diagram explaining the learning conditions, and FIGS. 6A to 6C are flowcharts showing the air-fuel ratio control program of the embodiment. In the figure, 10...Engine, 11a, 11b...Intake manifold, 12...Air filter, 13...Dashpot, 14...Intake pipe, 15a, 15b...
・Evaporative fuel supply pipe, 16a, 16b...Intake valve, 17
...Evaporative fuel supply control solenoid valve, 18a, 1
8b...Cylinder, 19...Canister, 20...
・Air flow meter, 22... Intake temperature sensor, 24a,
24b... Throttle valve, 25... Fuel tank, 2
6... ISC supply pipe, 27... Throttle opening sensor, 28... Idle speed control valve (I
SC valve), 32a, 32b... piston, 34a,
34b...Fuel injection valve, 36...Engine control unit (ECU), 40a, 40b-cooling water passage, 42...Water temperature sensor, 44a, 44b-
...Exhaust valve, 46a. 46b, 46c, 46d...exhaust manifold, 48a
, 48b...Exhaust pipe, 50a, 50b...Catalytic converter, 53a, 53b...Temperature sensor, 52a. 52b...0□ sensor. Fig. 1 Fig. 5 Fig. 6 C Fig. 7

Claims (1)

[Claims] (1) An air-fuel ratio sensor arranged in the exhaust system of each cylinder group divided into at least two or more groups of a multi-cylinder engine, and for each cylinder group, based on the output of the sensor of that group, Feedback control execution means that feedback controls the air-fuel ratio of the air-fuel mixture supplied to the cylinders of each group to a target value; and a first determination means that determines whether the execution condition of the feedback control is satisfied. learning means for learning correction values for compensating for deviations between the intake and exhaust systems for the two or more cylinder groups under predetermined learning conditions; and a learning means for determining whether the predetermined learning conditions are satisfied. When the learning condition for any cylinder group is not satisfied, learning for that group is stopped, and at least until the feedback control execution condition for the other cylinder group is not satisfied. , a control means for controlling to continue feedback control of a cylinder group for which the learning condition is not satisfied, regardless of whether the feedback control execution condition for the cylinder group is satisfied. Air-fuel ratio control device.