JP2000347882A - Processor for control program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a processor for control program which can be operated up to the limit of its processing capacity. SOLUTION: Processes A and B of the control program as base processes are so set that the priority of the process A is higher than the priority of the process B. In a process load measuring process (S201), the round time needed for single repetition of the base processes is measured and then the process load on the processor (CPU) corresponding to the round time is measured. In a thinning-out learning process (S202), conditions of a thinning-out process (thinning-out decision rotational frequency NELRN and thinning-out frequency N) are learnt according to the measured process load and then the conditions of the thinning-out process are changed and set again according to the process load. In thinning-out processes (S204 to S206), the process B (S208) is performed without being thinned out when the conditions of the thinning-out process set again in the thinning-out learning process are met (YES at S204 and YES at S206).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a control program processing device, and more particularly to a control program processing device for sequentially processing a plurality of control programs having different priorities in descending order of priority.

[0002]

2. Description of the Related Art For example, in engine control of a vehicle, a CPU in an electronic control unit (ECU) constantly processes all control programs in response to sensor signals from various parts of the vehicle or timer signals of a predetermined period. In such a case, the processing capacity of the CPU may be exceeded when the engine rotates at a high speed.

Therefore, as disclosed in Japanese Patent Application Laid-Open No. 60-3462, detected value signals of various operating parameters of the engine are input to a microcomputer, and the detected values of these parameters are synchronized with the rotation of the engine. When performing a predetermined process of the signal and calculating the basic amount of the control amount of the engine operation and the correction amount thereof based on the parameter detection value, the engine speed is detected and the detected value of the engine speed is calculated. It is determined whether or not the engine speed is equal to or higher than a predetermined engine speed, and at least part of the predetermined processing and the predetermined calculation is not performed when it is determined that the engine speed is equal to or higher than the predetermined engine speed. A method of processing and calculating an operating parameter signal of an internal combustion engine in such a manner has been proposed.

That is, according to the technique described in the publication, priorities of processing are set for each of a plurality of control processing programs executed by a CPU in an ECU, and when the engine speed is equal to or higher than a predetermined speed, the priorities are set. This is to prevent the processing of the control program having a low level from being skipped. The predetermined engine speed is set to a fixed value corresponding to the maximum processing load (processing load limit) of the CPU. This is because if the predetermined number of revolutions is set corresponding to a value exceeding the processing load limit of the CPU, the processing capacity of the CPU may be exceeded when the engine is running at a high speed.

[0005]

In the engine control, even if the engine speed is the same, the processing load on the CPU in the ECU varies depending on the control state (operating state) of the engine, and the processing load on the CPU is reduced. CPU
The processing capability limit value may also be increased, and it may not be necessary to skip control programs with lower priorities.

However, according to the technology described in the above publication, the predetermined rotational speed is set to a fixed value corresponding to the maximum value of the processing load of the CPU, so that the load on the engine is reduced depending on the control state of the engine. In such a case, there is a case where the processing capacity of the CPU has a margin and the control program with a low priority is not thinned out even though the control program with a low priority can be processed. In other words, according to the technology described in the publication, the predetermined number of revolutions for thinning out control programs with low priorities is set to a fixed value regardless of the actual processing load of the CPU.
However, there has been a problem that the control program cannot be operated up to the limit of the processing capacity.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide a control program processing apparatus capable of operating up to the limit of processing capacity.

[0008]

According to the first aspect of the present invention, there is provided a method for setting a priority of processing for each of a plurality of control programs, and starting with a control program having a higher priority. A processing device for a control program that performs sequential processing, and includes an execution unit, a thinning unit, a measurement unit,
It has learning means. The execution means executes the processing sequentially from the control program having the higher priority. The thinning means is
When the set thinning condition is satisfied, the processing of the control program with a lower priority order among the plurality of control programs is thinned out so that the execution means does not execute the processing of the thinned out control program. The measuring means measures the processing load of the control program in the processing device. The learning means learns the thinning conditions in the thinning means based on the processing load measured by the measuring means, and changes and resets the thinning conditions in accordance with the processing load.

Therefore, according to the present invention, the processing load of the actual processing device is measured, the conditions for thinning out the control program are learned based on the processing load, and the conditions for thinning out the control program are determined according to the magnitude of the processing load. By making it variable, the control program can be operated up to the limit of the processing capacity of the processing device.

According to a second aspect of the present invention, in the control program processing device according to the first aspect, when the processing of the plurality of control programs is performed repeatedly, The processing load corresponding to the circulation time may be measured by measuring the circulation time required for the repetition.

In this way, the processing load on the processing device can be measured simply and easily. Claim 3
In the control program processing device according to claim 1 or 2, the learning means, based on the processing load measured by the measuring means, when the processing load is small, The thinning condition in the thinning means may be changed to a loose condition and reset, and when the processing load is large, the thinning condition may be changed to a strict condition and reset.

In this way, when the processing load of the processing device is small and the processing capability of the processing device has a margin, it is possible to execute a control program with a low priority without thinning out. Also, the processing load of the processing device is large,
When the processing capacity of the processing device is insufficient, the processing load on the processing device is reduced by thinning out the control programs with low priorities, so that the control program with high priority can be reliably executed. .

According to a fourth aspect of the present invention, in the control program processing device according to any one of the first to third aspects, the plurality of control programs are engine control programs. The thinning conditions in the thinning means are set based on the engine speed,
When the processing load measured by the measurement unit is small, the execution unit executes the processing of the control program with a low priority without thinning out to a region where the engine speed is large. The processing of a control program with a higher priority may be executed by thinning out the processing of a control program with a lower priority.

In engine control, even if the engine speed is the same, if the processing load on the CPU in the ECU fluctuates due to the control state (operating state) of the engine and the processing load on the CPU decreases, In some cases, the limit value of the processing capacity also increases, so that it is not necessary to thin out control programs with low priorities. Therefore, according to the invention of claim 4, in such a case, it is possible to execute a control program with a low priority without thinning out.
The effect of the invention described in any one of claims 1 to 3 can be sufficiently brought out.

In the embodiments of the invention described below, the "executing means" described in claims or means for solving the problems corresponds to S203, S2 in the ECU 1.
08, and the “thinning-out means” is also the ECU
1, the "measurement means" corresponds to the processing of S201 in the ECU 1, and the "learning means" also corresponds to S202 in the ECU 1.
Similarly, the “thinning condition” corresponds to the thinning determination rotation speed NELRN and the thinning frequency N.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is embodied in an electronic control unit for a vehicle engine will be described below with reference to the drawings. FIG. 2 shows an electronic control unit (ECU) 1 according to the embodiment.
FIG. 2 is a block diagram showing an input / output system of FIG.

The air sucked from the air filter F is mixed with the fuel injected from the fuel injection valve 3 disposed in the intake pipe 2 to form an air-fuel mixture, and the air-fuel mixture is disposed in the intake pipe 2. From the throttle control valve 4, the air flows into each cylinder (not shown) of the engine E of the vehicle through the intake manifold 5. And the exhaust from each cylinder of the engine E is
The gas is exhausted from the exhaust pipe 6 through the silencer 7.

In the intake pipe 2, an intake pressure sensor 8 for detecting the pressure of the intake air and an intake temperature sensor 9 for detecting the temperature of the intake air are arranged. The water-cooled engine E is provided with a cooling water temperature sensor 10 for detecting a cooling water temperature.
An air-fuel ratio sensor 11 for detecting an air-fuel ratio is provided in the exhaust pipe 6. The throttle control valve 4 has a throttle opening sensor 1 for detecting its opening (throttle opening).
2 are connected.

A distributor 15 is provided between the igniter 13 and the ignition coil 14 for sequentially distributing the high voltage generated by the igniter 13 to the ignition plug (not shown) of each cylinder. Distributor 15
A rotation angle sensor 16 that generates a high-level signal at every predetermined crank angle of the engine E is built in.

The power from the vehicle-mounted battery BATT is directly supplied to the ECU 1 and also supplied through an ignition switch IG. The ECU 1 receives analog signals from the sensors 8, 9, 10, 11, 12, 16 and the like, and digital signals from various in-vehicle devices (not shown) such as a wiper and a power window, based on these signals. An arithmetic process for controlling the engine E is performed, a basic injection amount of fuel injection from the fuel injection valve 3 and a correction injection amount thereof are calculated, a control signal for the fuel injection valve 3 is generated and output, and the igniter 13 is controlled. And outputs the control signal.

FIG. 1 is a block diagram showing the configuration of the ECU 1. The ECU 1 includes a microcomputer 21, input interface circuits 22 to 24, an output interface circuit 25, and power supply circuits 26 and 27.
The microcomputer 21 includes a CPU 31 and a RAM 3
2, a ROM 33, a counter 34, a timer 35, an A / D converter 36, an input port 37, and an output port 38 are connected via a data bus 39.

The RAM 32 functions as a work memory for the microcomputer 21. Power is directly supplied from the vehicle battery BATT via the power supply circuit 26 to provide backup. The ROM 33 stores a program described later. Power is supplied from a vehicle-mounted battery BATT via an ignition switch IG and a power supply circuit 27 to each component of the microcomputer 21 except the RAM 32.

The analog signal from the rotation angle sensor 16 is
The data is input to the input port 37 of the microcomputer 21 via the input interface circuit 22 and the input port 3
7 to the CPU 31 via the data bus 39. Then, the CPU 31 calculates the rotational speed (engine rotational speed) NE of the engine E based on the analog signal from the rotational angle sensor 16. The high-level signal from the rotation angle sensor 16 has an effect as an interrupt signal for determining the execution timing of an external input interrupt process described later.

An analog signal from each of the sensors 8 to 12 is input to an A / D converter 36 of the microcomputer 21 via an input interface circuit 23,
After being converted into a digital signal by the D converter 36, the digital signal is transferred to the CPU 31 via the data bus 39.

Digital signals from various in-vehicle devices (not shown) are input to the microcomputer 21 via the input interface circuit 24, and are transferred from the input port 37 to the CPU 31 via the data bus 39. CPU3
The calculated basic injection amount of fuel injection from the fuel injection valve 3 and its corrected injection amount are transferred from the data bus 39 to the output interface circuit 25 via the output port 38. And a control signal is generated, and the control signal is output to the fuel injection valve 3. The control amount of the igniter 13 calculated by the CPU 31 is transferred from the data bus 39 to the output interface circuit 25 via the output port 38, converted into a control signal by the output interface circuit 25, and output to the igniter 13.

Next, the operation of the ECU 1 of the present embodiment configured as described above will be described. When the ECU 1 is started, according to the program recorded in the ROM 33,
The computer executes the following steps by various types of arithmetic processing. In addition, a computer-readable recording medium (semiconductor memory, hard disk, floppy disk, data card (IC
Card, magnetic card, etc.), optical disk (CD-ROM,
DVD, etc.), a magneto-optical disk (MD, etc.), a phase change disk, a magnetic tape, etc.), and the program may be loaded into the ECU 1 as needed and started up to be used.

The control program of the ECU 1 includes control programs for an initialization process, a base process, and an interrupt process. FIG. 3 is a flowchart illustrating a control program for the initialization process.

When the control program for the initialization process is started, in step (hereinafter referred to as "S") 101,
Operating environment setting of microcomputer 21 and RAM
After the initialization processing such as the initialization of 33 is performed, the processing shifts to the base processing. FIG. 4 is a flowchart for explaining a control program of the base processing.

When the control program for the base process is started, first, a processing load measurement process is performed in S201, and then a thinning learning process is performed in S202, and a thinning determination rotation speed NELRN and a thinning frequency N are set. Subsequently, the control program of the process A is performed in S203.

Then, at S204, the engine speed N
It is determined whether E is equal to or greater than the thinning-out determination rotational speed NELRN set in the thinning-out learning process of SS02, and NE ≧ NE
In the case of LRN (S204: YES), in S205, the count value CNT of the thinning counter provided in the counter 34 is incremented, and in S206, the count value CNT is thinned out in the thinning learning process of SS02. It is determined whether or not the number is less than N. If CNT <N (S206: YES), the process returns to S201.

When it is determined in S204 that NE <NELRN (S204: NO), and in S206, C
If it is determined that NT ≧ N (S206: NO), S2
At 07, the count value is cleared. At S208, the control program of the process B is performed, and then the process returns to S201.

Here, the process A is a process for calculating the basic injection amount of fuel injection from the fuel injection valve 3. Processing B
Are intake pressure sensor 8, intake temperature sensor 9, cooling water temperature sensor 10, air-fuel ratio sensor 11, throttle opening sensor 12
This is a process for calculating a correction injection amount of fuel injection from the fuel injection valve 3 obtained based on an analog signal from the above. Therefore, the process A has a higher priority than the process B.

As shown in FIG. 4, the base processing is repeatedly performed, but the processing load on the CPU 31 increases, and the processing A
If it becomes difficult to perform both the processing B and the processing B, the processing load on the CPU 31 is reduced by thinning out the execution of the processing B having a lower priority than the processing A.

The engine speed NE is the thinning-out determination speed N
If it is less than ELRN (S204: NO), the process moves from S204 to S207 to S208 every time the base process is repeated, and the process B is executed. When the engine speed NE is equal to or greater than the thinning-out determination speed NELRN (S204: Y
ES), when the count value CNT is equal to or greater than the number N of thinning-out times (S206: NO), the process B is executed only once for N repetitions of the base process, and the process B is not executed. It will be thinned out.

Here, the count value CNT is determined when the engine speed NE is equal to or greater than the thinning-out determination speed NELRN (S
204: YES), is incremented each time the base process is repeated, and cleared when process B is executed. If the interrupt processing control program is started while the base processing is being repeated, the base processing is interrupted, the interrupt processing is performed, the processing returns to the interrupted base processing after the end of the interrupt processing, and the base processing is repeated again. It is.

Control programs for interrupt processing include those started by an external input and those started by the timer 35 at regular intervals. FIG. 5 is a flowchart for explaining a control program of an external input interrupt process started by an external input.

As described above, the high-level signal from the rotation angle sensor 16 which is an external input has an effect as an interrupt signal for determining the execution timing of the external input interrupt processing. Therefore, in the present embodiment, the control program for the external input interrupt process is started in synchronization with the engine speed NE. Then, when the control program of the external input interrupt process is started, the control program of the process C is executed in S301, and then the process returns to the interrupted base process.

Here, the process C is a process of calculating the engine speed NE obtained based on the period measurement of the analog signal from the rotation angle sensor 16 and the control amount of the igniter 13 performed based on the engine speed NE. And the like. FIG. 7 is a flowchart for explaining a control program for a fixed-time interrupt process started at regular intervals.

The control program for the fixed time interrupt processing is started by the timer 35 at regular intervals (for example, 1 ms). Then, when the control program for the fixed time interruption process is started, the count value TIMER of the processing load measurement counter provided in the counter 34 is incremented in S501.

FIG. 6 is a flowchart for explaining the details of the processing load measurement processing in S201 shown in FIG.
First, in step S401, a count obtained by the fixed-time interrupt processing shown in FIG. 7 is set as a round time (base round time) TR required to repeat the base processing once (that is, the base processing makes one round). The value TIMER is set,
Next, after the count value TIMER is cleared in S402, the process proceeds to S202 shown in FIG.

Thus, every time the base processing is repeated, S
By monitoring the count value TIMER at 201, the base circulation time TR, which is the time required to repeat the base processing once, can be measured. Where C
As the processing load on the PU 31 increases, the time required for the base processing increases, so the base rotation time TR also increases. Therefore, the base rotation time TR becomes a value corresponding to the processing load of the CPU 31. Therefore, the base circuit time TR
, The processing load on the CPU 31 can be easily and easily measured.

FIG. 8 is an example of a timing chart of the external input interrupt processing, the base processing, the fixed time interrupt processing, and the count value TIMER. Incidentally, in FIG. 8, although the fixed time interruption processing is actually included in the base circuit time TR,
It is omitted to avoid complicating the drawings.

Each time the base processing is repeated, the base circulation time TR is measured by the processing load measurement processing. When the external input interrupt processing is started during the execution of the processing A and the processing B of the base processing, the processing A and B are interrupted, the processing C of the external input interrupt processing is performed, and the processing is interrupted after the processing C is completed. The process returns to each of the processes A and B. When the base process including the external input interrupt process and the fixed time interrupt process is being executed, the count value TIMER is incremented each time the fixed time interrupt process is started (for example, every 1 ms). Count value T monitored each time
IMER is the base rotation time TR.

FIG. 9 is a flowchart for explaining the details of the thinning-out learning process in S202 shown in FIG. First, in S601, it is determined whether or not the base rotation time TR measured in S201 is equal to or longer than a predetermined time interval TA.
When R ≧ TA (S601: YES), in S602
It is determined whether or not the engine rotational speed NE is equal to or greater than a thinning-out determination rotational speed NELRN set before that, and NE ≧ N
In the case of ELRN (S602: YES), in S603
The thinning number N set before that is equal to the predetermined number ΔN
In S604, it is determined whether or not the number of thinnings N after the addition exceeds the maximum number of thinnings Nmax.
If> Nmax (S604: YES), the number of thinnings N is set to the maximum number of thinnings Nmax in S605, and the process proceeds to S203 shown in FIG. 4. If N ≦ Nmax (S604: NO), S203 is left as it is. Move to.

If it is determined in S602 that NE <NELRN (S602: NO), in S606, the thinning-out determination rotation speed NELRN set before that is reduced by a predetermined rotation speed ΔNE, and in S607. It is determined whether or not the thinning-out determination rotational speed NELRN after the subtraction is less than the rotational speed (NE-α) obtained by subtracting the predetermined rotational speed α from the engine rotational speed NE. If NELRN <NE-α (S607:
YES) in S608, the thinning-out determination rotation speed NELRN
Is set to the rotation speed (NE-α), and NELRN ≧ NE−
In the case of α (S607: NO), the thinning-out determination rotation speed NELRN subtracted in S606 is set. And S
At 609, the thinning-out determination rotational speed NELRN set at S606 or S608 is the minimum thinning-out determination rotational speed NE.
It is determined whether it is less than LRNmin, and NELRN <NE
In the case of LRNmin (S609: YES), in S610, the thinning-out determination rotational speed NELRN is set to the minimum thinning-out determination rotational speed NELRNmin, and then the process proceeds to S203.
(NO) shifts to S203 as it is.

Here, the predetermined time interval TA is set to a time interval (for example, 16 ms) at which the process A must be executed in accordance with the control content. As described above, when the base rotation time TR is equal to or longer than the predetermined time interval TA (S60).
1: YES) is to determine whether or not the process B is currently being thinned out by determining whether or not the engine speed NE is equal to or greater than the thinning determination rotation speed NELRN (S602: YES). ) Increases the number N of thinnings by adding a predetermined number ΔN to the number N of thinnings (S603). As shown in FIG. 4, the process B is executed only once for the repetition of the base process thinning frequency N, and the process B is thinned. Therefore, as the number N of thinning increases, the process B is controlled to be thinned.

If the process B is not currently being thinned (S602: NO), the predetermined rotation speed ΔNE is subtracted from the thinning determination rotation speed NELRN (S60).
6) Decrease the thinning-out determination rotational speed NELRN. FIG.
As shown in (2), as the thinning-out determination rotational speed NELRN decreases, the process B is controlled to be thinned out.

Therefore, the state where the processing B is thinned out (S2)
04: YES and S206: YES), until the base circling time TR becomes shorter than the predetermined time interval TA (S
601: NO), the number N of thinnings increases (S60)
3) The process B is controlled in the direction of thinning out.

The predetermined number of rotations ΔN and the predetermined number of rotations ΔNE are set experimentally so that the number of thinnings N and the number of rotations NELRN for judging thinning do not suddenly change too much. Also, the number N of thinnings is set so as not to exceed the maximum number Nmax of thinnings set experimentally (S604, S605).

Then, the thinning-out determination rotational speed NELRN is
The engine speed is set not to be less than the engine speed (NE-α) obtained by subtracting the experimentally set predetermined engine speed α from the current engine speed NE (S607, S608).
Experimentally set minimum thinning-out determination rotation speed NELRNm
in (S609, S6)
10).

As described above, the upper limit of the thinning frequency N and the lower limit of the thinning determination rotational speed NELRN are set because the thinning frequency N is excessively increased or the thinning determination rotational speed NELR is set.
This is to prevent the processing B from being constantly thinned out when N is excessively reduced.

When it is determined in S601 that TR <TA (S601: NO), in S611, the base rotation time TR is a time interval (TA-γ) obtained by subtracting the predetermined time γ from the predetermined time interval TA. Is less than or equal to TR
If ≧ TA−γ (S611: NO), S20 is used as it is.
3 and TR <TA-γ (S611: YE
In S612, it is determined in S612 whether or not the engine speed NE is equal to or greater than a thinning-out determination speed NELRN set before that. If NE ≧ NELRN, it is determined (S612: YE).
In S613, it is determined in S613 whether or not the number N of thinnings set before is one, and if N ≠ 1 (S61).
3: NO) in S614, the number N of thinnings set earlier is subtracted by a predetermined number ΔN ', and in S615
It is determined whether or not the number of thinnings N after the subtraction is less than one. If N <1 (S615: YES), after the number of thinnings N is set to one in S616, FIG. The process proceeds to S203 shown, and if N ≧ 1 (S615: NO), the process directly proceeds to S203.

When it is determined in S612 that NE <NELRN (S612: NO), and in S613, N <NELRN
= 1 (S613: YES), S617
, The thinning-out determination rotational speed NE set before that
LRN is added by a predetermined rotation speed ΔNE ′, and in S618, the thinning-out determination rotation speed NELRN after the addition is a rotation speed (NE + β) obtained by adding a predetermined rotation speed β to the engine rotation speed NE.
Is determined, if NELRN> NE + β (S618: YES), the thinning determination rotation speed NELRN is set to the rotation speed (NE + β) in S619, and NE
If LRN ≦ NE + β (S618: NO), S617
The thinning determination rotation speed NELRN added by is set. Then, in S620, it is determined whether or not the thinning-out determination rotation speed NELRN set in S617 or S619 exceeds the maximum thinning-out determination rotation speed NELRNmax. If NELRN> NELRNmax is satisfied (S620:
YES) in S621, the thinning-out determination rotation speed NELRN
Is set to the maximum thinning-out determination rotation speed NELRNmax, and the process proceeds to S203, where NELRN ≦ NELRNmax.
In the case of (S620: NO), the process directly proceeds to S203.

As described above, the base rotation time TR is the time interval (T) obtained by subtracting the predetermined time interval γ from the predetermined time interval TA.
If (A-γ) is smaller than (S611: YES), it is determined whether the engine speed NE is equal to or greater than the thinning determination rotation speed NELRN (S612), thereby determining whether the process B is currently being thinned. When thinning is being performed (S612: YE
S) reduces the number N of thinnings by subtracting the predetermined number ΔN ′ from the number N of thinnings (S614). As shown in FIG. 4, as the number N of thinnings decreases, the processing B
Is controlled so as not to thin out. And
If the number N of thinning becomes one, the process B is executed once each time the base process is repeated once, which is the same as the case where the process B is not thinned.

If the process B is not currently being thinned (S612: NO) and if the number N of times of thinning is one (S613: YES), the thinning determination rotation speed NELRN is used.
Is added to the predetermined rotation speed ΔNE ′ (S61).
7) Increase the thinning determination rotation speed NELRN. FIG.
As shown in (2), as the thinning-out determination rotational speed NELRN increases, the process B is controlled so as not to thin out.

Therefore, the state where the processing B is not thinned out (S
If the answer is 204: NO or S206: NO), the base rotation time TR becomes equal to or longer than the predetermined time interval TA (S6).
01: YES), the number N of thinnings decreases (S61).
4) If the number N of thinnings is 1, it is the same as not thinning out the process B, so that thinning is not necessary, and the thinning determination rotation speed NELRN increases (S617), so that the process B is not thinned out. To be.

Note that the predetermined time interval γ is determined in steps S602 to S602.
610 (the direction of thinning out process B) and S61
This is to provide a hysteresis characteristic in order to prevent the occurrence of a hunting state in which each of the processes 2 to S621 (the direction in which the process B is not thinned out) is switched each time the base process is repeated.

The predetermined number of times ΔN ′ and the predetermined number of rotations ΔNE ′ are set experimentally so that the number of thinnings N and the number of rotations for thinning determination NELRN do not suddenly change too much. Also, the number N of thinning is 1
It is set so as not to be less than the number of times (ie, 0 times) (S615, S616).

The thinning-out determination rotational speed NELRN is
A rotational speed (N) obtained by adding a predetermined experimental speed β (for example, 200 rpm) experimentally set to the current engine rotational speed NE.
(E + β) (S618, S6)
19), the rotation speed is set so as not to exceed the experimentally set maximum thinning determination rotation speed NELRNmax (for example, 5000 rpm) (S620, S621).

As described above, the lower limit (= 1)
Times) and the upper limit of the thinning-out determination rotational speed NELRN is set when the thinning-out frequency N becomes 0 or the thinning-out determination rotational speed NELRN becomes too large, so that the thinning out of the processing B is not performed at all. This is to prevent that from happening.

FIG. 10 shows the engine speed NE, the thinning determination rotation speed NELRN, the thinning frequency N, and the base rotation time T.
5 is an example of a timing chart of R. From time T1 to T3, the processing load on the CPU 31 increases due to any change in the control state (operating state) of the engine, and the base circuit time TR also increases accordingly.

At time T2, when the base rotation time TR exceeds a predetermined time interval TA (a time interval during which the process A must be executed in accordance with the control content), the thinning-out judging rotation speed NELRN starts to decrease, and at time T3 , The thinning of the process B is started. Subsequently, from time T3 to T4, the number of thinnings N is added, and the processing load on the CPU 31 is reduced by thinning out the processing B. Then, at time T4, when the base circling time TR falls below the predetermined time interval TA, the increase in the number N of thinning stops. At this time, the thinning determination rotation speed NELRN does not fall below the rotation speed (NE-α).

Thereafter, from time T5 to T7, the processing load on the CPU 31 decreases due to some change in the control state of the engine, and the base circuit time TR
Also decrease. Then, at time T6, when the base circling time TR falls below the time interval (TA-γ), the number N of thinnings starts to decrease. Subsequently, at time T7, when the number of thinnings N becomes one, since the thinning of the process B becomes unnecessary, the thinning of the process B is stopped and the process B is executed, and the thinning determination rotation speed NELRN increases. Begin to.

Thereafter, at time T8, when the thinning-out determination rotation speed NELRN increases to the rotation speed (NE + β),
After that, the increase in the thinning determination rotation speed NELRN stops. As described in detail above, in the present embodiment, the actual processing load (base rotation time TR) of the CPU 31 is measured. When the processing load of the CPU 31 is small (the processing load is light) and the processing capacity of the CPU 31 has an allowance, the thinning determination rotation speed NELRN is set to a large value, and the number N of thinning is set to a small value. The processing B having a low priority in the base processing is executed without thinning out the area up to the area where the engine speed NE is high (the high rotation area of the engine E). When the processing load of the CPU 31 is large (the processing load is heavy) and the processing capacity of the CPU 31 has no margin, the thinning determination rotation speed NELRN is set to a small value and the number N of thinning is set to a large value. In the base processing, low-priority processing B is prevented from being thinned out, and high-priority processing A in the base processing is reliably executed at a desired processing interval.

For example, in the control of the engine E, even if the engine speed NE is the same, the processing load on the CPU 31 varies depending on the control state (operating state) of the engine E, and the processing load on the CPU 31 decreases. , CPU
In some cases, the limit value of the processing capability of 31 may be increased, and it may not be necessary to skip the process B, which is a control program with a low priority. In the present embodiment, even in such a case, the processing B can be executed without thinning out.

Therefore, according to the present embodiment, the actual CP
The processing load of U31 (base rotation time TR) is measured, and based on the processing load, the conditions for thinning out the control program (the thinning-out determination rotational speed NELRN and the number of thinning-out times N) are learned. By varying the conditions for the thinning, the control program can be operated up to the limit of the processing capacity of the CPU 31.

It should be noted that the present invention is not limited to the above-described embodiment, and may be modified as follows. Even in such a case, the same operation or effect as or more than that of the above-described embodiment can be obtained. . [1] As described above, when the control state (operating state) of the engine E changes, the processing load on the CPU 31 also changes, and accordingly, the base circuit time TR also changes. But C
Even if the processing load of the PU 31 changes significantly,
Every time the base process is repeated, the thinning-out determination rotation speed NELR
N changes only by the predetermined rotation speeds ΔNE, ΔNE ′, and the number N of thinnings changes only by the predetermined times ΔN, ΔN ′. Therefore, following a large change in the processing load of the CPU 31, a large delay may occur until the thinning determination rotation speed NELRN and the thinning frequency N are changed to optimal values.

Therefore, as illustrated in FIG. 11, the control state of the engine E is divided into a plurality of regions based on the engine speed NE and the intake pressure detected by the intake pressure sensor 8. The final setting values of the thinning-out determination rotation speed NELRN and the number of thinning-out times N when in the region are set as initial values when the vehicle enters the region next.

That is, as shown in FIG. 11, when the engine speed NE is 3000 rpm and 6000 rpm, the intake pressure is 400 mmHg and 700 mmHg, the control state of the engine E is divided into nine regions. And, for example,
For the region where the engine speed NE is less than 3000 rpm and the intake pressure is less than 400 mmHg, the final set values of the thinning-out determination speed NELRN and the number of thinning times N when the region was previously in the region, Are set as initial values (NELRN (0), N (O)). Similarly, when the engine speed NE is 6000 rpm
As described above, for the region where the intake pressure is 700 mmHg or more,
The thinning-out determination rotation speed N when in the area before that
The final set values of the ELRN and the number of thinnings N are set to the initial values (NELRN (8), N
(8)).

Here, for example, when the engine speed NE is 1
000 rpm and the intake pressure is 100 mmHg,
Engine speed NE is 7000 rpm and intake pressure is 80
When the control state of the engine E suddenly changes to the state of 0 mmHg, the initial value of the thinning determination rotation speed NELRN is set to NE.
LRN (8) is set and the initial value of the number N of thinning is set to N (8), and the thinning learning process shown in FIG. 9 is performed.

In this way, even if the control state of the engine E changes suddenly and the processing load on the CPU 31 changes significantly, the thinning determination rotational speed NELRN and the thinning determination rotation number NELRN follow the large change in the processing load. It is possible to reduce the delay until the number of thinnings N changes to an optimal value.

[2] In the above [1], the intake pressure sensor 8
The control state of the engine E is divided into a plurality of regions based on the detected intake pressure and the engine speed NE, but instead of the intake pressure, based on the throttle opening detected by the throttle opening sensor 12 and the engine speed NE. The control state of the engine E may be divided into a plurality of regions.

In this case, the same operation and effect as the above [1] can be obtained. [3] In the above embodiment, the base processing is configured to include two control programs of processing A and processing B, but the base processing is configured to include three or more control programs of which priorities are set. You may do so.

In this case, the priorities are set for each control program, and the thinning conditions (thinning determination rotational speed NELRN, thinning frequency N) are set in a plurality of stages according to the number of control programs. Good. For example, when there are three control programs, the thinning conditions may be set in two stages, and when there are four control programs, the thinning conditions may be set in three stages.

[4] The present invention is not limited to the engine control device (ECU 1) for controlling the engine of the vehicle, but may be applied to various processing devices for processing a plurality of various control programs having different priorities.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment embodying the present invention.

FIG. 2 is a block diagram showing an input / output system according to one embodiment.

FIG. 3 is a flowchart for explaining the operation of the embodiment;

FIG. 4 is a flowchart for explaining the operation of the embodiment;

FIG. 5 is a flowchart for explaining the operation of the embodiment;

FIG. 6 is a flowchart for explaining the operation of the embodiment.

FIG. 7 is a flowchart for explaining the operation of the embodiment;

FIG. 8 is a timing chart for explaining the operation of the embodiment.

FIG. 9 is a flowchart for explaining the operation of the embodiment.

FIG. 10 is a timing chart for explaining the operation of the embodiment.

FIG. 11 is an explanatory diagram for explaining the operation of another embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Electronic control unit (ECU) 21 ... Microcomputer 8 ... Intake pressure sensor 9 ... Intake air temperature sensor 10 ... Cooling water temperature sensor 11 ... Air-fuel ratio sensor 12 ... Throttle opening degree sensor 16 ... Rotation angle sensor 22-24 ... Input interface Face circuit 25 ... Output interface circuit 26,27 ... Power supply circuit 31 ... CPU 32 ... RAM 33 ... ROM 3
4: Counter 35: Timer 36: A / D converter 37: Input port 38: Output port 39: Data bus

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] August 13, 1999 (1999.8.1)
3)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0030

[Correction method] Change

[Correction contents]

Then, at S204, the engine speed N
It is determined whether E is equal to or greater than the thinning-out determination rotation speed NELRN set in the thinning-out learning process in S202 , and NE ≧ NE
In the case of LRN (S204: YES), in S205, the count value CNT of the thinning counter provided in the counter 34 is incremented, and in S206, the count value CNT is thinned out in the thinning learning process set in S202. It is determined whether or not the number is less than N. If CNT <N (S206: YES), the process returns to S201.

Continued on front page F-term (reference) 3G084 BA13 BA17 CA09 DA00 EB05 EB08 EB18 EC01 FA02 FA10 FA11 FA20 FA29 FA33 FA38 5B098 GC03 GC04 GC10

Claims (4)

[Claims] 1. A control program processing apparatus for setting a priority order of processing for each of a plurality of control programs and sequentially executing processing from a control program with a higher priority, wherein the processing is performed sequentially from a control program with a higher priority. Executing the thinning process of the control program having a lower priority order among the plurality of control programs when the set thinning condition is satisfied. A thinning means for preventing execution, a measuring means for measuring a processing load of a control program in the processing device, and learning conditions for the thinning in the thinning means based on the processing load measured by the measuring means. Learning means for changing and setting the thinning conditions in accordance with the processing load. Control program of the processor. 2. The control program processing device according to claim 1, wherein the measuring unit measures a circulation time required for one repetition when the processes of the plurality of control programs are repeatedly performed. A processing program for measuring the processing load corresponding to the circulation time. 3. The processing device for a control program according to claim 1, wherein said learning means is based on a processing load measured by said measuring means and said thinning means when said processing load is small. Wherein the thinning condition is changed to a loose condition and reset, and if the processing load is large, the thinning condition is changed to a strict condition and reset. 4. The control program processing device according to claim 1, wherein the plurality of control programs are engine control programs, and a condition of the thinning in the thinning means is an engine speed. When the processing load measured by the measurement unit is small, the execution unit executes the processing of the control program with low priority up to a region with a high engine speed without thinning out, and the processing load is A control program processing device characterized in that, when it is larger, the processing of a control program with a higher priority is executed by thinning out the processing of a control program with a lower priority.