JPH0367051A - Control device of car - Google Patents

Abstract

PURPOSE:To avoid a control system to become too large and complicated by furnishing a study control means as a common means used to a plurality of main controls. CONSTITUTION:A main control means 2 performs a plurality of main controls such as knocking control and idling air-fuel ratio control, on the basis of main control parameter optimized by study control. Therein the study control means shall be a common means to be used to a plurality of main controls. This precludes the control system from becoming too large and complex.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention provides a control device for an automobile, more specifically, a control device for an automobile, which includes a main control means and a learning control means, and is based on main control parameters optimized by learning control. The present invention relates to a control device for an automobile in which a main control means performs a plurality of main controls such as knocking control and idle air-fuel ratio control.

(Prior art) In order to optimize control parameters according to driving conditions, surrounding environment, etc., automotive control devices use learning control in many controls such as knocking control and idle air-fuel ratio control. Conventionally, control has been performed to learn the environment and set control parameters to values most appropriate for the operating conditions and surrounding environment.

(Problem to be solved by the invention) However, in conventional control devices equipped with a learning control function, a learning control means was provided for each main control such as knocking control and idle air-fuel ratio control. There is a problem in that the system becomes bloated and complicated, making it difficult to develop a control device that performs more diverse controls.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a control device for an automobile equipped with a learning control means that does not cause the control system to become bulky or complicated.

(Means for Solving the Problems) In order to solve the above problems, the present invention includes a main control means and a learning control means, and the main control means is controlled based on main control parameters optimized by the learning control. In a control device for an automobile that performs multiple main controls such as knocking control and idle air-fuel ratio control, the learning control means is a common means used in common for the plurality of main controls. Provides automobile control devices.

In a preferred embodiment of the present invention, the main control means and the learning control means each include an independent CPU, and the main control means includes a dual port RAM that is accessible from both the main control means and the learning control means. Means for storing parameters is provided.

(Function) As described above, the present invention includes a main control means and a learning control means, and the main control means performs knocking control, idle air-fuel ratio control, etc. based on the main control parameters optimized by the learning control. In a control device for an automobile that performs a plurality of main controls, the learning control means is a common means commonly used in the plurality of main controls, so that the control system does not become bulky or complicated.

Further, the main control means and the learning control means are each equipped with an independent CPU, and furthermore, a storage means for main control parameters consisting of a dual port RAM that can be accessed from both the main control means and the learning control means is provided. Therefore, the learning control function is further strengthened, and on the other hand, there is no need to manage data bus timing as in a general multi-CPU system, so the control system does not become bulky or complicated.

(Example) Hereinafter, the present invention will be described in detail based on the accompanying drawings.

In FIG. 1, l is a control device 2 according to an embodiment of the present invention.
3 is a spark plug of engine l, 4 is an intake passage, and 5 is an exhaust passage. A fuel injector 6, a throttle valve 7, and an air flow meter 8 are arranged in the intake passage 3 in this order from the intake downstream to the intake upstream. Further, a throttle opening sensor 9 is attached to the throttle valve 7, a knock sensor 10 is attached to the cylinder portion of the engine l, and a crank angle sensor 1 is attached near the crankshaft.
1 is arranged.

The control device 2 includes an input/output interface that inputs and outputs various signals, a CPU, and a memory that stores a control program.

The control device 2 receives an air flow signal from the air flow meter 8, a throttle opening signal from the throttle opening sensor 9, a knocking signal from the knock sensor 10, and an engine rotation speed signal from the crank angle sensor 11.
The control device 2 outputs an ignition signal to the spark plug 3 and an injector signal to the fuel injector 6.

As shown in Figure 2, the control program consists of a main program consisting of main control programs such as a knocking control program and an idle air-fuel ratio control program, a data table storing data such as control parameters, and a learning control program. It is configured. The learning control program is required for multiple routines for setting control parameters for main control to optimal values according to driving conditions and surrounding environment, that is, for each main control such as knocking control and idle air-fuel ratio control. It is structured as a subroutine consisting of individual routines for optimizing individual control parameters. The main program forms the main control means,
The learning control program forms a learning control means.

The learning control program performs learning control based on the control information from the main program, sets the main control parameters to optimal values according to the driving conditions and surrounding environment, writes the optimized main control parameters to the data table, The main program performs knocking control based on the main control parameters.
Performs main control such as idle air-fuel ratio control.

Next, a specific example of control by the control device according to the present embodiment will be described based on FIG. 3, which shows a part of the flow of the main control program, and FIG. 4, which shows part of the flow of the learning control program.

In the main control, while the car is running, the CPU judges whether the engine is in a high load state based on the engine speed signal, air flow signal, throttle opening signal, etc.81). In order to perform knocking control, first, (1) a knocking signal is taken as a variable S for determining the running state.

A comparison formula with the limit value α is used.

(2) Take the ignition advance angle as a control parameter, and set the parameter correction value Z by learning control to +5°.

(2) Take address 4050 as the address M of the data table where the initial values of the control parameters are stored.

Then, the learning control subroutine program is called using the above S, DXZ, and M as arguments (S3).

In learning control, the CPU judges whether or not the argument is giving a predetermined instruction (5ll), and if so,
The average value S of the non-king signal S of a predetermined n rotation needle is calculated based on the engine rotation speed signal and the knocking signal (S1
2) Compare the average value S and the predetermined limit value α513
), if s is larger than α, it is determined that there is a high possibility that knocking will occur, and the initial value of the ignition advance angle stored at address M of the data table, that is, address 4050 is read, and the initial value, e.g. 20' and the correction value Z, that is, 5°, to optimize the ignition advance angle, which is a parameter for knocking control, for the current driving condition, and write the optimal value, 25°, to address M of the data table, that is, address 4050 ( 314-516). If S is less than or equal to α, then CP
U determines that the possibility of knocking occurring is relatively low, and maintains the ignition advance angle, which is a parameter for knocking control, at the initial value (S13). As described above, the learning control for optimizing the knocking control parameters is completed and the main control is returned to. In the main control, the knocking initiation parameter, that is, the ignition advance angle, which has been optimized or maintained at the initial value, is read from address 4050 of the data table and nonking control is executed (S4).

In addition, in the main control, while the vehicle is running, the CPU appropriately determines whether the engine is in the idle state based on the engine speed signal, air flow signal, throttle opening signal, etc. In order to control the idle air-fuel ratio, first, 1) the engine speed signal is taken as the variable S for determining the driving state;

and a predetermined limit value β (or not is the average value of the engine rotation speed signal).

(2) The fuel increase coefficient is taken as a control parameter, and the parameter correction value Z by learning control is set to +0.1.

(2) Take address 4060 as the address M of the data table where the initial values of the control parameters are stored.

A decision is made (S6), and the learning control subroutine program is called using the above S, D, Z, and M as arguments (S7).

In learning control, the CPU judges whether or not the argument instructs execution of a predetermined operation (S17), and if it instructs the execution of a predetermined operation, the CPU executes the engine rotation for a predetermined n rotation needle based on the engine rotation speed signal. Find the variance S2 of the number signal S (818,
519), compare the variance S2 and the predetermined limit value β52
0), if S2 is larger than β, it is determined that the rotational state is unstable, and the initial value of the fuel increase coefficient stored at address M of the data table, that is, address 4060, is read and the initial value is , for example, by adding a correction value Z, that is, 0.1 to 0.1, to optimize the fuel increase coefficient, which is a parameter for idle air-fuel ratio control, for the current engine state,
Set the optimum value 0.2 to address M of the data table, that is, 40
Write to address 60 (S14-816). If S2 is less than or equal to β, the CPU determines that the rotational state of the engine is stable, and maintains the fuel increase coefficient, which is an idle air-fuel ratio control parameter, at the initial value (820). As described above, the learning control for optimizing the idle air-fuel ratio control parameters is completed, and the process returns to the main control. In the main control, idle air-fuel ratio control parameters that have been optimized or maintained at initial values, that is, fuel increase coefficients, are read from address 4060 of the data table and idle air-fuel ratio control is executed (S8).

As can be seen from the above description, in the control device according to this embodiment, the learning control program serving as the learning control means is
Individual main control parameters required for each main control such as knocking control and idle air-fuel ratio control, such as ignition advance angle and fuel increase coefficient, are optimized according to driving conditions and surrounding environment. Since the subroutine is a collection of routines, there is no need to incorporate a learning control routine corresponding to each main control into the main control program, which is the main control means for executing non-king control, idle air-fuel ratio control, etc. That is, the main control program, and by extension the entire control system, are prevented from becoming bulky and complicated.

Next, a control device according to another embodiment of the present invention will be described.

The control device 2' according to this embodiment, as shown in FIG.
The main control CPU performs main control such as knocking control and idle air-fuel ratio control, the learning control CPU performs learning control, and the dual port RAM allows data to be accessed from either of the two ports. It is equipped with a read/write RAM (hereinafter referred to as DPR), and both CPUs and DPR each have a control signal line through which read/write signals flow, and control information for control parameters for main control and learning control. They are connected by a data bus flowing therethrough and an address bus flowing an address signal indicating an address where a control parameter is stored. Furthermore, although not shown in FIG.
Like the control device 2 according to the first embodiment, the control device 2 includes an input/output interface for inputting and outputting various signals, and a memory storing a control program. The DPR stores control information for performing learning control and control parameters for main control.

The control device 2' is built into the same automobile engine as in the first embodiment, and the main control CPU of the control device 2' includes:
The air flow signal from the air flow meter 8, the throttle opening signal from the throttle opening sensor 9, the knocking signal from the knock sensor 10, and the engine speed signal from the crank angle sensor 11 are inputted to the main control CPU of the control device 2'. From there, an ignition signal is output to the spark plug 3 and an injector signal is output to the fuel injector 6. Further, signals from various sensors input to the main control CPU are transmitted to the learning control CPU via the main control CPU.

As shown in FIG. 6, the main control CPU writes control information for operating the learning CPU into the DPR based on the main control program, and the learning control CPU reads the control information from the DPR.
reads the control information, performs learning control based on the learning control program, sets the main control parameters to optimal values according to the driving conditions and surrounding environment, writes the optimized main control parameters to the DPR, and controls the main control CPU. reads the main control parameters from the DPR and performs main controls such as knocking control and idle air-fuel ratio control based on the main control program.

Next, a specific example of control by the control device according to the present embodiment will be described based on FIG. 7, which shows a part of the flow of the main control program, and FIG. 8, which shows a part of the flow of the learning control program.

In the main control, while the car is running, the main control CPU judges whether or not the engine is in a high load state based on the engine speed signal, air flow signal, throttle opening signal, etc.551). In order to perform knocking control in this case, first, (1) a knocking signal is taken as a variable S for determining the running state.

(2) Take the ignition advance angle as a control parameter, and set the parameter correction value Z by learning control to +5°.

(2) Take address 4050 as the address M of the DPR where the initial values of the control parameters are stored. decided (S52),
Write the above arguments S, D, Z, M to DPR (S 5
3).

In learning control, the learning control CPU reads arguments S, D, and ZSM from DPR (861), and judges whether or not execution is instructed (562). If so,
The average value S1 of the knocking signal S for a predetermined n rotations is calculated based on the engine rotation speed signal and the knocking signal (S6
3) Compare the average value S1 and the predetermined limit value α S6
4) If Sl is larger than α, it is determined that there is a high possibility that knocking will occur, and the initial value of the ignition advance angle stored at address M of the DPR, that is, address 4050 is read, and the initial value, e.g. The ignition advance angle, which is a parameter for knocking control, is optimized for the current driving condition by adding the correction value Z, that is, 5 degrees, to 20 degrees, and the optimal value of 25 degrees is written to address M of the DPR, that is, address 4050 (865 ~5
67).

If Sl is less than or equal to α, the learning control CPU determines that the possibility of knocking occurring is relatively low, and maintains the ignition advance angle, which is a parameter for knocking control, at the initial value (
864). As described above, the learning control for optimizing the knocking control parameters is completed and the main control is returned to. In the main control, the main control CPU reads the main control parameters that have been optimized or maintained at initial values, that is, the ignition advance angle, from address 4050 of the DPR and executes non-king control (854).

In addition, in the main control, while the car is running, the main CPU appropriately determines whether the engine is in an idle state based on an engine speed signal, an air flow signal, a throttle opening signal, etc. In this case, in order to control the idle air-fuel ratio, first, (1) take the engine rotational speed signal as the variable S for determining the driving state;

is the average value of the engine speed signal).

(2) The fuel increase coefficient is taken as a control parameter, and the parameter correction value Z by learning control is set to +0.1.

(2) Take address 4060 as the address M of the DPR where the initial values of the control parameters are stored.

(856), and the above arguments S, D, and Z.

Write M to DPR (857).

In the learning control, the learning control CPU reads the arguments S, D, Z, and M from the DPR (S61), and judges whether or not execution of the comparison expression is instructed (568), and if so, the engine Calculate the variance S2 of the engine speed signal S for a predetermined n rotations based on the speed signal (S69.5
70), compare the variance S2 and the predetermined limit value β571
), if S2 is larger than β, it is determined that the rotational state is unstable, and the address M of the DPR, that is, 406
Load the initial value of the fuel increase coefficient stored at address 0,
The fuel increase coefficient, which is a parameter for controlling the idle air-fuel ratio, is optimized for the current engine condition by adding a correction value Z, that is, 0.1, to the initial value, for example, 0.1, and then the optimum value is set to 0.
2 is written to address M of DPR, that is, address 4060 (865 to 567). If S2 is less than or equal to β, the learning control CPU determines that the rotational state of the engine is stable, and maintains the fuel increase coefficient, which is a parameter for idle air-fuel ratio control, at the initial value (S71). As described above, the learning control for optimizing the idle air-fuel ratio control parameters is completed, and the process returns to the main control. In main control, main control C
The PU optimizes or maintains the main control parameters, that is, the fuel increase coefficient, at the DPR's 406
Read from address 0 and execute idle air-fuel ratio control (3
58).

As can be seen from the above explanation, in the control device according to this embodiment, the main control and the learning control are performed using independent C
Since this is performed by the PU, that is, the main control CPU and the learning control CPU, the learning control function can be further strengthened. On the other hand, the data file is stored in the DPR, and the main control C
Since the PU and the learning control CPU each write or read the necessary data to the DPR via independent data buses, even when one CPU writes or reads data to the DPR, the other CPU data can be written or read. In other words, there is no need to manage data bus timing, that is, data read/write timing between CPUs, as in a general multi-CPU system, and the entire control system is prevented from becoming bulky and complicated.

Although the present invention has been described in detail above, it goes without saying that the present invention is not limited to the above-mentioned embodiments, and that various modifications can be made within the scope of the invention described in the claims.

(Effects of the Invention) As described above, the present invention includes a main control means and a learning control means, and the main control means performs knocking control and idle air-fuel ratio control based on main control parameters optimized by the learning control. In an automobile control device that now performs multiple main controls such as, the learning control means is a common means that is commonly used for multiple main controls, so the control system does not become bulky or complicated. .

Further, the main control means and the learning control means are each equipped with an independent CPU, and further, the main control parameter storage means is composed of a dual port RAM that can be accessed from both the main control means and the learning control means. Since this is provided, the learning control function is further strengthened, and on the other hand, there is no need to manage data bus timing as in a general multi-CPU system, so the control system does not become bulky or complicated.

Therefore, the present invention provides a control device for a vehicle that is equipped with a learning control means that does not cause the control system to become bulky or complicated.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram of an engine equipped with an automobile control device according to an embodiment of the present invention. FIG. 2 is a structural diagram of a control program for an automobile control device according to an embodiment of the present invention. FIG. 3 is a diagram showing a part of the flowchart of the main program in the program of FIG. 2, and FIG. 4 is a diagram showing a part of the flowchart of the learning control program in the program of FIG. FIG. 5 is a diagram showing a part of the system structure of a control device for an automobile according to another embodiment of the present invention. FIG. 6 is a diagram showing the control operation by the control device of FIG. 5. FIG. 7 is a diagram showing a part of the flowchart of the main program in the control system of FIG. 5, and FIG. 8 is a diagram showing part of the flowchart of the learning control program in the system of FIG. . 1...Engine, 2...Control device. Figure 1 Figure 2 Main test parameters Figure 4

Claims (2)

[Claims] (1) A vehicle that is equipped with a main control means and a learning control means, and the main control means performs multiple main controls such as knocking control and idle air-fuel ratio control based on main control parameters optimized by learning control. 2. A control device for an automobile, wherein the learning control means is a common means commonly used in a plurality of main controls. (2) The main control means and the learning control means are each equipped with an independent CPU, and furthermore, the main control parameter storage means consisting of a dual portram that can be accessed from both the main control means and the learning control means is provided. A control device for an automobile according to claim 1.