JPH06100143B2 - Engine operation pattern prediction device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic control device for controlling fuel supply amount, injection timing, intake air amount, ignition timing, etc. of an engine for an automobile, for example, an internal combustion engine, and more particularly to an engine. The present invention relates to a technique for performing precise control by predicting the operating state of.

(Prior Art) As a conventional total engine electronic control unit, there is one as shown in FIGS. 1 and 2, for example.

The above device is described in SAE papers 800056 and 800825. FIG. 1 is a hardware configuration diagram, and FIG. 2 is a correspondence diagram of control system sensors, signals, and actuators.

In FIG. 1, 1 is an air flow sensor that detects the intake air amount, 2 is a throttle position switch that detects when the throttle valve is fully closed, 3 is an AAC valve that adjusts the intake air amount to control the idle speed, and 4 is An EGR control valve that controls the exhaust gas recirculation amount, 5 is a negative pressure converter (a composite of a constant pressure valve and an on / off solenoid valve) that controls the opening of the AAC valve 3 and the EGR control valve 4, 6 is a fuel injection valve, 7 is an oxygen sensor, 8 is an ignition coil, 9 is a distributor, 10 is a three-way catalyst, 11 is an exhaust temperature sensor, 12 is a neutral switch that detects the neutral position of the transmission, and 13 is a crank that detects the rotation angle of the crankshaft. An axis sensor, 14 is a cooling water temperature sensor, 15 is a fuel pump, 16 is a fuel pump relay, 17 is a vehicle speed sensor, and 18 is an air conditioner switch.

In the above-mentioned device, input signals from various sensors as shown in FIG. 2 are input to a microcomputer (not shown) to comprehensively control various actuators as shown.

The control contents of the above-mentioned device are as follows.

(1) Fuel supply amount control (EGI) according to the intake air amount.

(2) Control of air supply (ISC) that keeps the engine speed constant during idling.

(3) Ignition timing control (IGN) according to engine speed and engine load.

(4) Exhaust gas recirculation amount control (EGR) according to engine speed and engine load.

The above control contents are changed according to various operating conditions.

However, in the conventional device as described above, the control is performed only on the basis of the result of the actual operation of the engine, so that problems such as slow response of the control system and engine stall can be avoided early. There are problems such as not being able to perform control like that, and being unable to prevent recurrence of the same defect.

(Object of the Invention) The present invention has been made in order to solve the problems of the prior art as described above, and uses the change amount of the operation pattern after the engine and the auxiliary machinery are in a predetermined state as pattern data. It is stored in memory, and when a predetermined state occurs, based on the pattern data of the change, the subsequent operation pattern can be predicted, and the result can be reflected in engine control and diagnosis. An object is to provide a motion pattern prediction device.

(Outline of the Invention) FIG. 3 is a block diagram showing the overall configuration of the present invention.

First, in FIG. 3 (A), 20 is an operating state of an engine and auxiliary machinery (transmission, air conditioner, etc.) (in the case of a vehicle, the state of the vehicle, for example, the vehicle speed is also included. ) Various sensor groups for detecting ()
(200, 210, 220, 230, 250, 270, 280, etc. in the figure), 21 is a calculation means for performing various calculations for control and diagnosis, 22 is various actuators and display means that operate according to the calculation result of the calculation means 21 (35, 38, 41 in FIG. 42, 60, 70, 80 etc.).

Reference numeral 23 is an operation change pattern creating means, which is pattern data of a change in engine operation corresponding to a predetermined state of the engine or accessories (a value at which the engine state changes after a certain state is reached). When the output of the sensor group 20 indicates a predetermined state, the pattern data corresponding to the change is output.

Reference numeral 24 is a synthesizing means for the predicted operation pattern, and synthesizes the output of the sensor group 20 and the pattern data of the above change.

The calculation means 21 performs a calculation based on the output of the sensor group 20 and the output of the predicted operation pattern synthesis means 24 to control the actuator or the display means 22 or to perform a display operation. It is also possible to perform fault detection and self-diagnosis of control status.

As described above, by predicting the engine state after a certain point in time based on previously stored data, and performing control according to the predicted result, the control response is improved and problems such as engine stall are avoided in advance. It becomes possible to do.

Next, in FIG. 3 (B), 25 is an actual operation pattern measuring means, which samples the operation signal of the engine given from the sensor group 20 at a predetermined cycle, and the sampling data for a predetermined period is time-series pattern data. Stored as actual operation pattern data.

The predicted motion pattern synthesizing means 26 includes the pattern data of the change given from the motion change pattern creating means 23,
The above-mentioned actual operation pattern data is combined.

The other parts are the same as those in FIG.

The predicted motion pattern synthesizing means 26 extrapolates the actual motion pattern data to obtain data (temporary predicted pattern data) on the assumption that a predetermined state has not occurred, and changes patterns in the data.・ Data is added and combined.

With the above configuration, it is possible to accurately predict the engine operation.

Next, in FIG. 3 (C), 27 is an actual state determination means, which determines from the output of the sensor group 20 that the engine state has reached a predetermined state (a plurality of states are possible).

28 is an operation pattern storage means for each state, and the actual state determination means 27 has a predetermined engine state (for example, a stall state)
Actual operation pattern measuring means when it is determined that
The value stored by 25 is stored as operation pattern data corresponding to the predetermined engine state.

Reference numeral 29 denotes an operation change amount pattern creating means, which indicates an amount of change from the predetermined value of the operation pattern data stored in the state operation pattern storage means 28 when the actual state determination means 27 determines that the engine state is a predetermined engine state. It is stored as variation pattern data, and when the output of the sensor group 20 or the actual state determination means 27 indicates a predetermined state, the variation pattern data corresponding to it is output.

Reference numeral 31 is a predictive motion pattern synthesizing means, which synthesizes the real motion pattern data of the real motion pattern measuring means 25 and the motion pattern data of the motion change pattern creating means 29.

Other operations are the same as those in FIG.

With the above-described configuration, accurate prediction can be performed by storing and using the data of the result of actual operation even with respect to changes over time of the engine and changes in the engine operation pattern due to control. There is an advantage called.

(Examples of the Invention) Hereinafter, the present invention will be described in detail based on examples.

First, an outline of the system of the electronic control unit of the present invention will be explained based on FIG.

FIG. 4 shows a case where the present invention is applied to a 4-cycle 6-cylinder engine, and the control target is as follows.

(1) Injector 35 provided in each cylinder of the engine
Fuel injection (EGI) control (EGI OUT110) is performed by controlling the valve opening start time and valve opening time.

(2) Ignition (IGN) control (IGN OUT120) that controls the ignition timing and energization time by controlling the energization / interruption of the primary coil of the ignition coil 38.

(3) Exhaust gas recirculation (EGR) control (EGR OUT130) performed by negatively controlling the lift amount of the EGR valve 30 using the VCM valve 40.

(4) By controlling the lift amount of the AAC valve 50 by using the VCM valve 40 to control the negative pressure, the idle rotation performed by controlling the amount of air that bypasses the throttle valve 510 (IS
C) Control (ISC OUT150).

The above is the main control target, but there are the following additional control or information output. (5)
ON / OFF control of the fuel pump 530 by controlling the fuel pump relay 60 (F / P OUT160), (6) Output of fuel consumption data to the fuel consumption meter 70 (FCM OUT170), (7) System self-diagnosis Data exchange (CHECK) with checker 2000 or vehicle information provision device 2500, (8) Output of alarm by alarm result to alarm lamp 80 (ALARM OU)
(T180), (9) Display (MONIT) of self-diagnosis results on the display 1900. (10) Air conditioner (air conditioner)
Control (AIR CON) by air-conditioner control signal (A / C OUT190) to air-conditioner relay 90 that turns on and off.

In order to perform the above control and output, the following control information is obtained from the engine and each part of the vehicle.

(1) From the crank angle sensor 200 built in the distributor 520, the REF signal 201 that rises every 120 ° at the crankshaft rotation angle (twice the distributor rotation angle).
And POS signal 202 where rising and falling alternately occur every 1 °
To get

An engine speed signal 203 is obtained by counting the POS signal 202 for a predetermined time.

(2) The engine intake air amount Q _a is an air flow meter
It detects by 210. Note intake air amount Q _a is made inversely related to the air flow meter output voltage signal (AFM) 211.

(3) The output voltage of the O ₂ sensor 220 changes according to the oxygen concentration in the exhaust gas, and a signal (O ₂ ) 221 corresponding to the air-fuel ratio is obtained.

(4) The water temperature sensor 230 obtains a voltage signal (T _W ) 231 representing the temperature of the engine.

(5) The on-vehicle battery 240 supplies electricity to each part of the control system. Control to the control unit 1000
Main power supply 241 via unit relay 540 and battery 24
The auxiliary power supply 242 which is directly input from 0 is supplied. The voltage signal (V _B ) 241 of the main power source is also used as information for control. The ON terminal 262 of the ignition switch 260 is
Since the battery voltage is applied not only in the ON position but also in the START position, the control unit 1000 can be used during cranking.
The main power supply 241 is supplied to the.

(6) The vehicle speed sensor 250 obtains a signal (VSP) 251 having a pulse density proportional to the vehicle speed.

(7) The ignition switch 260 is a switch operated by the driver to start and operate the engine.
The voltage signal (START) 261 at the T terminal makes it possible to know whether cranking is in progress.

(8) Throttle valve sensor 270 is a throttle opening signal (TVO) 271 proportional to the opening of the throttle valve.
Is output.

(9) The air conditioner switch 280 is a switch that closes when the air conditioner is operated, and detects whether the air conditioner is operating by the terminal voltage signal (A / C) 281.

(10) The neutral switch 290 is a switch that closes when the gear position of the transmission is in neutral or in the parking position, and its opening / closing signal (NEUT).
291 detects the gear position of the transmission.

The signals described above are input to and output from the control unit 1000. In addition to the input / output to / from the control unit 1000, a checker 2000 for diagnosing the control system and displaying the result is connected via a check connector 2010. Further, the vehicle information providing device 2500 is connected via a data transfer connector 2510. The control unit 1000 has a microcomputer, determines the control state of each controlled object based on the above control information (input signal), outputs a control signal (output signal), and optimally controls the engine. Output information related to.

The functions of 21, 23, 24, 25, 26, 27, 28, 29, 31, etc. in FIG.
Included in control unit 1000 above.

Next, a control that comprehensively performs the control described above.
The circuit configuration of the unit 1000 will be described with reference to FIG.

In FIG. 5, 1100 is a signal shaping circuit, which inputs various input signals from the engine and various parts of the vehicle, removes noise from these various input signals, absorbs surges, and causes malfunction of the control unit 1000 due to noise. In addition to preventing damage due to surges, various input signals are amplified or converted so that the next input interface circuit 1200 can be properly operated. 1200 is an input interface circuit, which performs analog-digital (AD) conversion of various input signals shaped by the signal shaping circuit 1100,
It counts the number of pulses during a predetermined time, converts it into a digital code signal so that the next central processing unit (CPU) 1300 can read it as input data, and stores it as an input data in a register inside. 1300 is a central processing unit (CPU), which is an oscillation signal 1311 of a crystal oscillator 1310.
It operates in synchronization with a clock signal based on
Connected to each part via 20 and mask ROM 141 of memory 1400
0 and the program stored in the PROM 1420 are executed, various input data are read from each register in the input interface circuit 1200, various output data are calculated by arithmetic processing, and predetermined timing is set in the register in the output interface circuit 1500. To send output data. memory
1400 is a data storage device, such as mask ROM1410, PROM1420, R
It has an AM 1430 and a memory holding memory 1440. The mask ROM 1410 permanently stores the program executed by the CPU 1300 and the data used when executing the program when the IC is manufactured, and the PROM 1420 has the same program and data as the mask ROM 1410, which is highly likely to be changed according to the vehicle model and engine type. Is permanently written and stored before being incorporated into the control unit 1000. Further, the RAM 1430 is a readable / writable memory that stores data that is temporarily stored in the output interface circuit 1500 before being sent to the output interface circuit 1500 as data during calculation processing or result data. Turns off the main power supply 241
It will not be retained when it runs out. The memory holding memory 1440 stores the result data of the arithmetic processing and the intermediate data in the ignition
The memory is retained even when the switch 260 is turned off, that is, when the vehicle is not driven.

1350 is a calculation timer circuit, which enhances the function of the CPU 1300, a multiplication circuit for speeding up calculation processing,
An interval timer that sends an interrupt signal to the CPU 1300 at every predetermined time cycle, a free-running timer for the CPU 1300 to know the elapsed time from a certain event to the next event and the event occurrence time.
It has a counter and the like. An output interface circuit 1500 receives output data from the CPU 1300 in an internal register and converts it into a pulse signal having a predetermined timing and time width, or a predetermined cycle and duty ratio, or "1" or "0". Driving circuit by converting to switching signal
Send to 600. The drive circuit 1600 is a power amplifier circuit,
In response to a signal from the output interface circuit 1500, a transistor or the like performs voltage / current amplification to drive various actuators, performs display, or is connected to the control unit 1000 via the connector 2010 to control the system. An output signal is sent to the checker 2000 for making a diagnosis and displaying the result.

1700 is a backup circuit, which detects a failure by monitoring the signal of the drive circuit 1600, and when the CPU 1300, memory 1400, etc. fails and does not operate normally, the signal shaping circuit 1100
In response to a part of the signal from the above, the engine outputs the minimum necessary control output for driving the vehicle and also the switching signal 1701 for notifying the occurrence of the failure. 1750 is a switching circuit, and a switching signal from the backup circuit 1700
The signal from the output interface circuit 1500 is blocked by 1701 and the signal from the backup circuit 1700 is passed.

1800 is a power supply circuit, which has a stable power supply voltage of 181
It supplies 0, 1820, 1860, 1880, 1890 and outputs a RESET signal 1840, a HALT signal 1850, a battery voltage signal 1830, etc. for controlling the operation of the CPU 1300.

Next, FIG. 6 is a block diagram showing an embodiment of a control system to which the present invention is applied and a signal flow.

In an actual system, each block shown in FIG.
It is realized by the hardware shown in the figure and a program executed by the CPU 1300, but is shown in the form of a block diagram to make the system easy to understand.

The overall configuration and schematic operation will be described below.

First, the actual operation pattern measuring means 3100 uses various input signals 203,
211, 271, etc. are input, sampled at predetermined intervals for a predetermined period, and sequentially stored, and the operation data of the engine speed, intake air amount, throttle opening, etc. is displayed in pattern data. It is stored in the form as actual operation pattern data 3101.

On the other hand, various input signals 281, 291 and the like are input to the operation change pattern creating means 3200, and operation pattern data predicted as changes in the engine rotation speed, intake air amount, etc. are selected according to the movement of the signals, Movement change pattern
Output as data 3201.

The predicted motion pattern synthesizing means 3300 inputs the actual motion pattern data 3101 and the motion variation pattern data 3201 and synthesizes and processes both to generate future rotation, intake air amount,
Predicted operation pattern data 3301 that predicts what the operation pattern of the throttle, etc. will be, is created. Predicted motion pattern data 3301 when the motion change pattern is zero
Is the same as the actual operation pattern data 3101.

The actual state discrimination means 3400 uses various input signals 203, 211, 231, 271,
From 281, 291 and the like, the engine state, for example, an unsteady state such as engine stall (engine stall), acceleration, deceleration, gear change, etc., is determined, and actual state data 3401 is output.

The operation pattern storage means 3500 for each state stores the actual state data 3401.
According to the above, the actual operation pattern data 3101 such as rotation, intake air amount, throttle opening, etc. when the engine state occurs are stored as state-specific operation pattern data 3501 separately for each engine state.

As described above, the state-specific operation pattern data 3501 includes not only the actual operation pattern data 3101 generated while the engine is in use and stored, but also data stored in advance when the control device is manufactured.

The engine state estimation means 3600 uses the predicted operation pattern
The data 3301 is compared and collated with the operation pattern data 3501 for each state, and if they match or approximately match, the engine state is estimated to be the engine state corresponding to the matched operation pattern data for each state. The state estimation data 3601 is output.

The control output computing means 3700 uses EGI, IG
Control outputs (110, 120, 130, 150, 190, etc.) of N, EGR, ISC, etc. are calculated and output, but the calculation method or correction data is changed according to the engine state estimation data 3601.

In the embodiment of FIG. 6, the state-specific operation pattern storage means 3500, the engine state estimation means 3600 and the like are included, but the present invention is at least a minimum operation change pattern generation means 3200 and a predicted operation. The pattern synthesizing means 3300 can be mainly configured.

For example, the sensor signal (a signal indicating the current engine state) and the motion change pattern data 3201 may be combined to form the predicted motion pattern data 3301. In addition, the output of the predicted motion pattern synthesizing means 3300
The engine state may be estimated by giving it to the 3600, or the engine may be controlled based on the output of the predictive operation pattern synthesizing means 3300.

Next, the operation of FIG. 6 will be described in detail based on an example.

This example is an example in which a state in which engine stalling is likely to occur is predicted from changes in engine speed, and control is performed so as to avoid it.

FIG. 7 is a diagram showing a pattern of engine rotation before and after engine stall.

In FIG. 7, section A is a deceleration section. When the clutch is disengaged at the end of deceleration, the load decreases, so the engine speed increases once and then begins to decrease again. At this time, for example, the parts of the fuel supply system of the engine are changing over time, the fuel supply amount (mixing ratio) is not appropriate, the timing of disengaging the clutch is late, and the drop in rotation speed is too large.
When the ignition timing is not appropriate or the vicinity of the throttle valve is dirty and the air-fuel mixture is not supplied stably, the engine does not stably converge to the idle state and causes a hunting phenomenon, and gradually The rotation speed may decrease (section B), and finally engine stall (section C) may occur.

Here, for example, when the rotation operation pattern in the section D is measured, it is judged whether or not the pattern is as shown in the figure (pattern recognition), and it is judged that the pattern is likely to stall, that is, the stall If it is estimated that the engine is not stalled by increasing the amount of air-fuel mixture and increasing the torque generated by the engine at the end of section D, for example.

FIG. 8 is a flowchart of a program 3150 executed by the CPU 1300 as the actual operation pattern measuring means 3100.

This program is a regular interrupt program activated by an interrupt signal sent from the interval timer at regular intervals. First, in 3151, it is judged whether or not it is a measurement section. The measurement section is, for example, the section D in FIG. This determination is performed by determining deceleration based on the throttle opening and the vehicle speed, determining the start of the section based on whether the engine speed reaches a predetermined value, and determining the end of the section based on whether a predetermined time has elapsed. . If it is within the measurement section, the measured data is sequentially sampled at 3152 and stored in the RAM 1430. As a result, the actual engine rotation pattern is measured and stored as the actual operation pattern data 3101.

Next, the operation of the operation change pattern creating means 3200 will be described by taking the operation when the air conditioner is on / off as an example.

FIG. 9 is a diagram showing changes in the engine rotation speed when the air conditioner is turned on and off.

It is experimentally known that the engine speed changes as shown in the figure depending on whether the air conditioner is on or off. This rotation change pattern is stored in advance as data,
For example, when the air conditioner switch is turned on,
This is detected, and it is predicted that the rotation speed changes as shown in the section A, starting from that time.

FIG. 10 is a flowchart of the program 3250 for calculating the movement change pattern data 3201. The microcomputer previously stores the pattern data of the change in the rotation speed when the air conditioner is on (a value where the starting point of the section A in FIG. 9 is zero) and the change in the rotation speed when the air conditioner is off (the ninth value). (Value with the starting point of section B in the figure set to zero)
Data and are stored. The program started by the periodic interrupt judges from the data of the air conditioner switch whether or not the air conditioner is turned on in 3251. If YES, 3
In 252, pre-stored pattern data for the change when the air conditioner is turned on is selected and output as operation change pattern data 3201. Similarly, 32 when off
53 and 3254 select the OFF data and output it.

Next, the operation of the predictive operation pattern synthesizing unit 3300 will be described by taking the case where the air conditioner is turned on during deceleration as an example.

FIG. 11 is a diagram showing changes in the engine rotation speed when the air conditioner is turned on during deceleration.

In FIG. 11, when the air conditioner is turned on at the time when the rotation speed is changing like the line a during deceleration, the rotation speed changes like the b. In this case, the rotation speed is sufficiently high and there is no concern about engine stall. on the other hand,
If the air conditioner is turned on at that point, it will change like c, and the rotation speed will drop significantly and there is a strong possibility that engine stall will occur.

FIG. 12 is a flowchart of the program 3350 for creating the predicted motion pattern data 3301 for predicting the rotation as described above. First, 3351 is executed, for example, at the time point of FIG. 11, and extrapolation is performed from the actual operation pattern data 3101 at that time point to calculate the subsequent rotation change pattern. That is, the extension of the line a is estimated. This estimated data is called temporary predicted pattern data.

At 3352, the movement change pattern data 3201 is added to the extrapolation data. As a result, predicted motion pattern data 3301 such as b and c is created. If the change is gradual, it may be possible to simply connect the motion change pattern data 3201 to the rotation speed at the time without extrapolation. Further, when there is no operation such as air conditioner on / off, the operation change pattern data 3201 is zero, so the predicted operation pattern data 3301 becomes the actual operation pattern data 3101 itself. This corresponds to the case of deceleration hunting in FIG.

Next, FIG. 13 shows a program 34 for creating the actual state data 3401.
It is a flowchart of 50.

First, at 3451, the engine speed is checked, and if it is 20 rpm or less, at 3452, data 3401 is changed to data indicating engine stall. Otherwise, at 3453, the data 3401 is changed to data indicating that the engine is rotating. In addition to the engine speed, the intake air amount, oil amount, etc. can be used for the engine stall determination. It is also possible to determine the actual state of the engine, such as acceleration and deceleration, from movements such as the throttle and intake air amount.

Next, FIG. 14 is a flowchart of the program 3550 for creating the operation pattern data 3501 for each state.

First, the actual state data 3401 is checked at 3551, and in the case of stalling, at 3552, the actual operation pattern data 31 immediately before that is checked.
01 (pattern data of section D in FIG. 7) is stored as operation pattern data at the time of engine stall. This data is stored in the memory holding memory 1440 (see FIG. 5) so that the memory is held even when the ignition switch 260 is turned off and the main power is turned off.

As a result, the operation pattern of the engine rotation when the engine actually stalls during use of the engine is stored.

In addition to the above-mentioned data, the operation pattern at the time of engine stall that occurred in a development experiment or the like is stored in advance as another operation pattern data 3501 at the time of engine stall.
Specifically, it is stored in the mask ROM 1410, PROM 1420, or the like when the control device is manufactured. It is also possible to store operation pattern data according to the state of the engine such as acceleration and deceleration by adding a similar program.

Next, the operation of the engine state estimating means 3600 will be described.

Figure 15 shows predicted motion pattern data 3301 at the time of engine stall.
FIG. 5 is a diagram showing a relationship between state-based operation pattern data 3501.

In FIG. 15, the line a is the data stored in the operation pattern storage means 3500 for each state, and is the operation pattern data 3501 when the air conditioner is turned on and the engine stalls during deceleration. Actually, the section B in the figure is stored, but the state of engine rotation in the sections A and C before and after that is also shown for the sake of clarity.

Line b is the predicted operation pattern data 3301 created by the above-described predicted operation pattern synthesizing means 3300 when the air conditioner is turned on at that time. Similar to the operation pattern data 3501 for each state, the state of rotation before and after that is also illustrated.

The engine state estimating means 3600 finds the area of the hatched portion (the portion surrounded by the lines a and b of the section B) in the figure, and depending on whether the size is larger or smaller than a predetermined value, if left as it is, the engine will stall. If it is smaller than a predetermined value, the engine state is predicted and determined to reach engine stalling.

Next, FIG. 16 is a flowchart of a program 3650 in which the engine state estimation means 3600 calculates the engine state estimation data 3601.

First, in 3651, the first of the predicted motion pattern data 3301 and the plurality of state-based motion pattern data 3501 stored
Pattern data 3501-1 (for example, this is the pattern data when the air conditioner was turned on during deceleration and was stalled) (3301-3501-1) at each time point, sequentially over the entire section B Calculate and integrate. As a result, the difference area data between the lines b and a in FIG. 15 is calculated with a sign. This difference area data is compared with a predetermined value at 3652. If it is larger than the predetermined value, the line b (predicted rotation operation pattern) is relatively above the line a (rotation operation pattern when the engine actually stalls), and there is no risk of engine stall. If it is smaller than the predetermined value, the line b is relatively close to the line a or is below the line a, and there is a strong possibility of stalling, so it is determined that stalling will occur, and at 3653, the engine state estimation data 3601 Let be the data representing the engine stall.

In 3654 and 3655 or less, the same processing is performed on the second and third (for example, pattern data at the time of engine stall due to deceleration hunting in FIG. 7 etc.) engine state-specific operation pattern data.

As a result, it is presumed that the engine stalls if it coincides with the rotational motion pattern that has reached the engine stall up to now, or if the rotational speed becomes relatively lower than that.

It should be noted that in this example, since only the engine stall is determined, the difference is obtained with a sign, and the determination is made only whether the engine stall or not.
When identifying engine states other than engine stalling, such as acceleration and deceleration, the absolute values of the differences are integrated to calculate the areas themselves and compared, and the engine state estimation data can be set to different values according to the results. For example, it is possible to identify which of a plurality of motion patterns is matched or approximately matched. Even in the case of stalling, there are multiple patterns, so it is possible to identify which state of stalling.

Next, FIG. 17 is a flowchart showing a part of the program 3750 executed as the control output calculation means 3700.

This program is rotation synchronization, that is, the crank angle sensor 200
Is a program started by an interrupt signal by a signal (REF signal 201) for every 120 ° crank angle.

First, at 3751, the engine state estimation data 3601 is checked to determine whether it is estimated to be stalled. If it is determined that the engine has not reached the stalling, the normal control is performed at 3952. The details of normal control are well known in the above-mentioned SAE papers 800056, 800825, etc., and therefore omitted. If the engine is estimated to be stalled, the amount of air due to ISC (that is, the amount of the air-fuel mixture) is increased at 3753, and the rotational torque is controlled to increase. Specifically, there is a method of increasing the control target speed of ISC and indirectly increasing the air amount in feedback control, and ISC
The control output of can be applied to the method of directly increasing the air amount by feedforward control, but the latter has better response. However, in the latter method, it is not possible to strictly control how much the rotation speed increases, so if you want to keep the increase width constant, you can control by short feedforward and gradually shift to feedback. Should be taken.

In 3754, the ignition timing is corrected in the direction in which the torque increases, that is, the direction in which the torque advances.

In some cases, the engine may be stalled easily due to lack of ignition energy. Therefore, it is also effective to increase the ignition energy, that is, to control the energization time to the ignition coil.

With the 3755, the EGR is reduced and the combustion condition is improved. In addition, in order to prevent engine stalling, control such as controlling the mixture ratio of the air-fuel mixture in the direction of increasing torque (EGI control) or reducing the load on the engine (for example, turning off the air conditioner) is also effective.

Although the above description has been given of the example of predicting and avoiding engine stall, it can also be applied to detection of acceleration / deceleration or gear change.

It is also possible to use operation data other than the rotation speed, for example, an intake air amount signal (AFM signal 211), a throttle opening signal (TVO signal 271) and the like. These will be described below.

For example, regarding a gear change, the presence or absence of a gear change can be detected by detecting the operation of a clutch or the like, but it is not possible to distinguish which gear change from what speed. In order to make a distinction, a complicated sensor for identifying all gear positions is required, which is expensive and complicated. However, by applying the method of the present invention, the change pattern of the intake air amount, the throttle opening, the engine rotation, etc. is detected and collated with the pattern data that is experimentally measured in advance and stored at the time of manufacture. You can accurately and quickly estimate whether a gear change has occurred. Then, by adding the result to the control, the control suitable for each engine state can be performed. As an example, a method of controlling the fuel supply pattern according to the gear change pattern so as to minimize the emission of harmful components of the exhaust gas can be considered.

Further, with respect to acceleration and deceleration, it is possible to not only judge whether acceleration or deceleration but also distinguish acceleration / deceleration on a slope (individually up and down) or to estimate the degree of acceleration / deceleration.

This enables fine control of fuel, ignition, and EGR so as to reduce the emission of harmful components of exhaust gas.

Next, FIG. 18 shows another embodiment of the present invention.

In this embodiment, the movement change pattern creating means 3800
Is the actual state data 3401 and the operation pattern data 35 for each state.
01 and the actual state data 3401 is determined to be a predetermined engine state.
The change amount from the predetermined value is stored as motion change amount pattern data. Then, when the input signal from the sensor or the actual state data 3401 indicates a predetermined state, the operation change amount pattern data 3201 corresponding thereto is output.

Other operations are similar to those of the embodiment shown in FIG.

Next, FIG. 19 is a flow chart of a program 3260 for creating the motion variation pattern data 3201 stored in the motion variation pattern creating means 3800.

In FIG. 19, at 3261, it is checked from the actual state data 3401 whether the air conditioner is on. If the air conditioner is on, go to 3262.

In 3262, the operation pattern data 3501-3 when the air conditioner is turned on is read from the operation pattern data 3501 for each state, and the change pattern data is calculated with the value at the moment the air conditioner is turned on as zero. . That is, the value at the moment when the air conditioner is turned on is subtracted from the entire pattern data.
The change pattern data is stored as operation change pattern data 3201-1 when the air conditioner is turned on. Less than
3263, 3264 change data when the air conditioner is off, 3265, 3
Data of the change when the gear position of the automatic transmission changes from N (neutral) to D (drive) in 266, 3267, 3268
The data of the amount of change when moving from D to N is calculated and stored respectively.

As a result, the motion change pattern is stored as the pattern when it actually enters that state, so that it is possible to predict motion that is close to the actual engine motion with respect to changes over time in the engine and changes in motion due to control. Become.

The above description of the program has been explained so that it is executed continuously with a timed interrupt. However, when actually executing the program, a program (OS) that controls the execution schedule before each program Insert
Effectively use data by controlling the execution interval and execution order.

As an example, the movement of pattern data when the air conditioner is turned on will be described.

The actual operation pattern data 3101 _n measured when the air conditioner is turned on for the nth time is used to create the predicted operation pattern data 3301 _n , but the operation change pattern data n used at that time is the same as the previous operation (n -1st time) Actual operation pattern data 3101 _n-1 is the operation pattern data for each state 35
01 _n−1 , and the change amount thereof is also stored in the motion change amount pattern creating means 3800.

Similarly, the data compared by the engine state estimating means 3600 is the predicted motion pattern data 3301 _{n for the} nth time and the motion pattern data for each state 3501 _n-1 stored one time before.
Is. After the comparison and collation is completed, the n-th actual operation pattern data 3101 _n is newly stored as 3501 _n .

Note that the control method may be a method of directly calculating the control amount from the predicted operation pattern data 3301 other than the method of estimating the state as described above.

For example, it is also possible to calculate the fuel supply amount from the predicted operation data of the intake air amount and the engine rotation, supply the fuel expected to be necessary early, and control so as to reduce the influence of the operation delay of the fuel supply system.

(Effects of the Invention) As described above, in the present invention, it is possible to predict an operation pattern after the engine reaches a predetermined state. Therefore, by performing an early control based on the prediction result, engine stall, etc. The problem of can be avoided. As a result, the range of use of the engine (for example, idling speed) can be expanded to the limit and the performance can be fully utilized (for example, idling speed can be reduced to reduce fuel consumption).

Further, in the case of extrapolating the actual motion pattern, more accurate motion prediction can be performed.

Also, in the case of sequentially storing operation change pattern data, accurate prediction can be made even with respect to changes over time of the engine and changes in the operation pattern of the engine due to control by using the data of the results of actual operation. The effect is that it can be done.

[Brief description of drawings]

FIG. 1 is an example of a conventional device, FIG. 2 is a corresponding diagram of a control system of the device of FIG. 1, FIG. 3 is a block diagram showing the entire configuration of the present invention, and FIG. 4 is an electronic control of the present invention. FIG. 5 is a block diagram showing an embodiment of a control system to which the present invention is applied. FIG. 7 is a diagram showing the engine rotation before and after engine stall. FIG. 8 shows a pattern, FIG. 8 is a flowchart of a program executed by the CPU as an actual operation pattern measuring means,
FIG. 9 is a diagram showing a change in engine speed when the air conditioner is turned on and off, and FIG. 10 is a flow chart of a program for calculating operation change pattern data 3201.
The figure shows the change in engine speed when the air conditioner is turned on during deceleration. Figure 12 is the flowchart of the program that creates the predicted motion pattern data 3301.
The figure is a flow chart of the program that creates the actual state data 3401. Fig. 14 is the flow chart of the program that creates the state-specific operation pattern data 3501. Fig. 15 is the predicted operation pattern data 3301 at the time of stall and the operation pattern by state. FIG. 16 is a diagram showing the relationship with the data 3501, FIG. 16 is a flow chart of a program 3650 for calculating the engine state estimation data 3601, FIG. 17 is a flow chart of a program executed as the control output calculation means 3700, and FIG. Another embodiment and FIG. 19 are flowcharts of a program for creating operation change pattern data 3201. Explanation of code 20 …… Sensor group 21 …… Computing means 22 …… Actuator or display means 23,29 …… Motion change pattern creating means 24,26,31 …… Predictive motion pattern synthesizing means 25 …… Actual motion pattern… Measuring means 27 …… actual state discrimination means 28 …… state-specific operation pattern storage means

Claims (3)

[Claims] 1. A sensor for detecting an operating state of an engine or an operating state of an engine and an auxiliary machine, and at least one of the engine and the auxiliary machine, which are known in advance to cause a change in engine operation. A first means for pre-storing the time-series pattern data of the change occurring in the engine operation thereafter, that is, the continuous time-series pattern data of the predetermined period; Detecting that the output of the sensor has reached the predetermined state and reading the variation pattern data corresponding to the predetermined state from the first means and outputting the same; and an engine provided from the sensor. Third means for predicting a future operation pattern of the engine by synthesizing the current operation data of the engine and the variation pattern data given from the second means. An engine operation pattern prediction device characterized by the above. 2. A predetermined state in which at least one of the sensor for detecting the operating state of the engine or the operating state of the engine and the auxiliary machinery and at least one of the engine and the auxiliary machinery is known in advance to cause a change in the engine operation. A first means for pre-storing the time-series pattern data of the change occurring in the engine operation thereafter, that is, the continuous time-series pattern data of the predetermined period; Second output means for detecting the change in the predetermined state from the output of the first output means and reading out the change pattern data corresponding to the predetermined state from the first means, and sampling the sensor output. Means for storing sampling data of a predetermined period corresponding to the actual operation of the engine as continuous time-series actual operation pattern data; Fifth means for predicting time-series pattern data of temporary prediction of future motion patterns by extrapolating to turn data, that is, temporary prediction pattern data, and temporary prediction pattern data given from the fifth means And a sixth means for predicting a future operation pattern of the engine by synthesizing the variation pattern data given from the second means, and an operation pattern prediction device for the engine. 3. A sensor for detecting an operating state of an engine or an operating state of an engine and auxiliary equipment, and sampling data of the sensor output for sampling data of a predetermined period corresponding to an actual operation of the engine continuously. A fourth means for storing as actual movement pattern data of a series; and a fourth means for extrapolating to the above-mentioned actual movement pattern data to predict temporal series pattern data of temporary prediction of a future movement pattern, that is, temporary prediction pattern data. 5 means, 7th means for detecting from the output of the sensor that the state of the engine has reached a pre-determined predetermined state, and 7th means is determined to be the predetermined engine state The value stored by the fourth means at the time is set as a continuous time-series operation pattern corresponding to the predetermined state.
Eighth means for storing as data, and a change amount of the operation pattern data stored by the eighth means from a predetermined value is stored as continuous time-series operation change pattern data for a predetermined period. And then thereafter, when the sensor output indicates the predetermined state, the ninth means for outputting the motion change amount pattern data corresponding thereto and the provisional predicted pattern data given from the fifth means. And the motion change pattern given from the ninth means
An engine operation pattern prediction device comprising: a tenth means for predicting a future operation pattern of the engine by synthesizing the data with the engine.