JPH05248286A - Control device for engine - Google Patents

Abstract

PURPOSE:To prevent unstabilization of control by limiting changing a control constant when a change of operative condition of an engine exceeds a predetermined value, in the case of feedback-controlling a condition quantity relating to the engine so as to approach a target vague based on the predetermined control constant. CONSTITUTION:In an engine 1, oxygen concentration in exhaust gas is detected by an air-fuel ratio sensor 3 in an exhaust passage 2, thus to control a fuel injection valve 5 by a controller 4 so that air-fuel ratio obtains theoretical air-fuel ratio. A bypass valve 9 is arranged in a bypass passage 8 of a throttle valve 7 in an intake air passage 6, thus to control the bypass valve 9 by the controller 4 so as to maintain an idle rotational speed always optimum. Here in the controller 4, whether a change is provided or not in an operative condition of the engine 1 is detected. When the change of the operative condition exceeds a predetermined value, in response to this change, a change of control constant is limited. In this way, control is prevented from being not stabilized due to diffusion of adaptive control.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an engine control device, and more particularly to an engine control device using adaptive control.

[0002]

2. Description of the Related Art Generally, a vehicular engine is provided with a catalytic converter composed of a three-way catalyst in the exhaust passage in order to remove harmful components such as CO, HC and NO _x contained in the exhaust gas. Three-way catalyst for CO, HC, and NO _x
In order to obtain a high sterilization rate for the components at the same time, it is necessary to accurately control the air-fuel ratio (A / F) within a narrow region near the stoichiometric air-fuel ratio. Therefore, as shown in FIG. 22, the oxygen concentration in the exhaust gas is detected by the air-fuel ratio sensor (O ₂ sensor) 3 provided in the exhaust passage 2 of the engine 1 so that the air-fuel ratio becomes the stoichiometric air-fuel ratio. The controller 4 including a microcomputer feedback-controls the fuel injection amount from the fuel injection valve 5 (hereinafter, this air-fuel ratio control is referred to as "A / F control").

In order to always control the idle speed of the engine optimally, a bypass passage 8 for bypassing the throttle valve 7 is provided in the intake passage 6 of the engine 1 and the passage 8 is provided in the bypass passage 8. A bypass valve 9 composed of a duty solenoid valve for adjusting the amount of passing air is provided, and the controller 4 controls the bypass valve 9 based on the output of the engine speed sensor 10 so that the idle speed _Ne becomes a predetermined speed. Is feedback-controlled (hereinafter, this idle speed control is referred to as "ISC"). Also, intake negative pressure (Boost)
Intake negative pressure sensor 11 and engine water temperature T _W
Various sensors such as the water temperature sensor 12 for detecting the water temperature are also used for control.

By the way, PID control has been generally used for the above-mentioned feedback control, but in recent years, more advanced feedback control is possible with the progress of the control theory and the improvement of the capability of the control controller. Has become.

As a typical example thereof, the state quantity (generated torque etc.) of the controlled object (engine) is detected or estimated, and the input (engine speed, A / F etc.) is optimized so that the output (engine speed, A / F etc.) becomes optimum. A control method for setting the bypass valve opening degree, A / F, etc.) is used and is called state feedback control.

FIG. 23 shows a block diagram of the state feedback control. In the ISC, the input r is the target engine speed, the engine input u is the bypass valve opening, and the output y is the engine speed. In the A / F control, the input r is the target A / F, the engine input u is the fuel injection amount, and the output y is the A / F.

In this state feedback control, a dynamic characteristic model of the engine is set in advance, and a state observer called an observer detects or estimates the state quantity of the engine (for example, internally generated torque), and according to the order of the engine model. The feedback gain K (vector quantity) is determined for a plurality of state quantities x (vector quantity), and feedback control is performed.

Such state feedback control is performed by P
The controllability is far better than that of the ID control, but when this is applied to the feedback control of the engine, there are also drawbacks as described below.

(1) Since it is generally difficult to detect the state quantity of the controlled object, this is estimated by an observer.
In order to estimate the state quantity, it is necessary to know the dynamic characteristic model of the controlled object, but the controlled object such as an engine not only has a large change in characteristics due to operating conditions and changes over time, but also for each controlled object. Due to the variation, it is impossible to determine the dynamic characteristic model in advance.

(2) The gain when setting the input from the state quantity cannot be set to the optimum value because the dynamic characteristic cannot be determined.

Therefore, adaptive control as shown in the block diagram of FIG. 24 is effective as a feedback control law of the engine.

FIG. 25 shows a schematic flow when the above adaptive control is applied to engine feedback control.
Steps (a) to (e) are an adaptive control system.

(A) Engine state (rotation speed, intake negative pressure)
Set the tuning gain and the vibration input to change the input.

(B) The engine model is identified by engine model tuning.

(C) The feedback gain is determined based on the identified model.

(D) An observer estimates the state quantity based on the identified model.

(E) State feedback control is performed by multiplying the estimated state quantity by the feedback gain.

By repeatedly learning the above process, the estimated gain of the engine model is gradually optimized as shown in FIG. In other words, if adaptive control is used, even if the characteristics of the controlled object change, it follows this, so that it is possible to gradually approach the optimum control state. Incidentally, changing the input by adding input (r) vibration to the input (r _e) is to promote the identification.

27 to 29 are flowcharts showing a specific example of adaptive control.

[0020] First, in step S1 of FIG. 27, the engine speed (N _e), from the table of the intake negative pressure (Boost), reads the vibration data (r _e) and tuning gain (K _g), the next step The output y is detected in S2. This output y is the air-fuel ratio A in A / F control.
/ F, and the engine speed (N _e ) in ISC
Is. Then, in step S3, the error E _s is obtained. In addition, r'is a previous control target and a is a constant. In the next step S4, it is checked whether the error E _s is less than or equal to a constant value E _so ,
If E _s ≦ E _{so is} not satisfied, the input r is added with the vibration input r _e · (−1) ⁿ in step S5. n represents a control cycle.

Next, moving to FIG. 28, steps S6 to S11.
Identify the engine model. Here, u ₁ is the value of u one cycle before, u ₂ is the value of u one cycle before u ₁ , and u ₃ is the value of u one cycle before u ₂ (same below). Further, g _{1 to} g ₄ are control constants representing the engine model, and here, four models are shown, but the number may be changed according to the model. At that time, the numbers of du, u, x, and K may be changed according to the number of g. Y _m represents the engine model output.

Next, proceeding to step S12 in FIG. 29, the gain is set from the gain setting equation for the finite settling response. The following steps S13 to S15 show the state estimation operation by the observer, and step S16 is an integral term. F is an observer gain. Then, state feedback is performed in step S17, and u is output in the next step S18. Here, u is the fuel injection amount in the A / F control and the bypass valve opening in the ISC, as described above. Next, the input / output values are updated in step S19, and the previous control target r'is updated in step S20.
End the cycle.

When the determination in step S4 is "YES", the process directly proceeds to step S13 in FIG. That is, if the error E _s is equal to or less than the constant value E _so , the model error is considered to be small, and the model identification step is omitted.

[0024]

However, when the adaptive control as described above is applied to engine control, there are problems as described below.

(1) Depending on engine conditions such as engine speed and water temperature, the model may diverge instead of converge, and control may become extremely unstable.

(2) If the output value (engine speed in the case of ISC) suddenly changes due to disturbance during adaptive control, identification becomes unstable and control becomes unstable.

(3) When the adaptive control program is incorporated in the engine control unit, the program becomes long because of the large number of processing steps. Therefore, the ROM capacity may increase or the processing capacity of the CPU may be exceeded.

(4) Since the engine model is identified and the gain is set, there are many adaptive control steps and it takes time to converge. Therefore, unstable control is continued during that time.

In view of the above problems, it is a first object of the present invention to provide an engine control device which prevents the adaptive control from becoming unstable due to divergence.

A second object of the present invention is to provide an engine control device which prevents control from becoming unstable when the model suddenly changes due to disturbance.

A third object of the present invention is to provide an engine control device that suppresses an increase in ROM capacity and performs adaptive control without exceeding the processing capacity of the CPU.

Further, a fourth object of the present invention is to provide an engine control device which speeds up control convergence.

[0033]

An engine control apparatus according to the present invention comprises a control means for performing feedback control so that a state quantity related to an engine approaches a target value based on a predetermined control constant, and a state associated with the feedback control. In an engine control device including a control constant changing unit that changes the limiting constant so that the degree of change in the amount matches the required degree of change, a unit that detects the presence or absence of a change in the engine operating state, and the detecting unit When it is detected that the change of the engine operating state exceeds a predetermined value, in response to this, a limiting means for limiting the change of the control constant is provided.

Further, the engine control apparatus according to the present invention includes a control means for performing feedback control so that the state quantity related to the engine approaches a target value based on a predetermined control constant, and the degree of change of the state quantity associated with the feedback control. A control constant changing means for changing the control constant so as to match the required degree of change, in an engine control device, an external load detecting means for detecting an operating state of an external load with respect to the engine; Means for decreasing the tuning gain of the control constant and then gradually increasing the tuning gain.

[0035]

According to the present invention, the adaptive control diverges because the limiting means for limiting the change of the control constant is provided when it is detected that the change in the operating state of the engine exceeds a predetermined value. However, it is possible to prevent the control from becoming unstable.

Further, according to the present invention, when the external load is actuated, the tuning gain of the control constant is made small, and means for gradually increasing the tuning gain is provided thereafter. It can be prevented from becoming stable.

[0037]

Embodiments of the present invention will be described below with reference to the drawings.

1 and 2 are a block diagram showing an engine control apparatus according to a first embodiment of the present invention and a flow chart of its control routine. In this embodiment, operating conditions (engine speed, engine water temperature) are shown. , Intake negative pressure, etc.)
The control constants (g ₁ to g _{4 in} FIG. 28) identified for each are stored in a table, and this is used as the initial value of the next adaptive control, whereby the control convergence is accelerated.

Step S in the flowchart of FIG.
21 to S26 correspond to step S5 of FIG.
This is a process inserted between step S6 of 8 and step S6. First, in step S21, the engine speed (N _e ), the intake negative pressure (Boost), and the water temperature (T _w ) are input, and in the next step S22, N _e and Boots read in step S21.
A cell number (N _c ) corresponding to t and T _w is input, and it is checked in step S23 whether or not the cell number is the same as the previous time. If this determination is “YES”, the process proceeds to step S6, but the determination in step S23 is If "NO", the step S24
Then, g _1c to g _4c of the cell with the cell number N _c ′ in the previous table
Are defined as control constants g _{1 to} g ₄ . And the next step S2
5, reads g _1c to g _4c of the cell with the cell number N _c of the table, and updates the cell number N _c 'of the previous in the next step S26.

Next, the flow chart of FIG. 3 shows the operation of the engine control apparatus according to the second embodiment of the present invention, and if “YES” is determined in the step S23 of the flow chart of FIG. In contrast to the process directly proceeding to step S6, in the present embodiment, the process proceeds to step S6 through steps S27 to S31 and steps S32 to S35.

That is, in step S27, it is checked whether or not the flag F is set. If F = 1, step S
28 to S31, the control constants g _{1 to} g ₄ during adaptive control are
Control constants under each condition (engine speed, intake negative pressure, water temperature) are compared with initial values g _{1c to} g _4c, and their difference is a constant ε.
If less proceeds to step S6, but control constant initial value g _1c ~ control constants g ₁ to g ₄ in the adaptive control in each condition
If it is far from g _4c , step S32
Advances to ～S35, with a gain K _g and vibrating input r _e stops adaptive control by zero, a normal state feedback control using an initial value (table value) as the control constants. By executing such processing, it is possible to prevent the adaptive control from becoming unstable due to divergence.

[0042] In the flowchart of FIG. 3, although the detection of the difference between the initial value and the control constants g ₁ to g ₄ in step S28~S31 is performed for each constant g ₁ to g _4, instead of this As shown in steps S36 and S37 of FIG. 4, it is also possible to evaluate by the sum of squares S _e of all differences (which may be the sum of absolute values).

Next, FIGS. 5 and 6 are a block diagram showing an engine control apparatus according to a third embodiment of the present invention and a flow chart of its control routine. In this embodiment, a cooler switch signal and a field ON signal indicating that the cooler 13 and the alternator 14 driven by the engine 1 are operated are input to detect the load application,
The identification gain (tuning gain) K _g is adjusted. In this case, as shown in the flowchart of FIG. 6, the identification gain K _g is initially made small and gradually increased. By performing such control, it is possible to prevent destabilization of control by high gain adaptive control when the model suddenly changes due to load application.

Step S in the flowchart of FIG.
41 to S50 also correspond to step S5 of FIG. 27 and FIG.
This is a process inserted between step S6 of 8 and step S6.

[0045] First examining whether the flag F _L = 1 in step S41, the flag F _L is at the initial value setting has a F _L = 0. If _FL = 0, step S4
2, S43 in the investigation whether磁信No. or cooler switch signal field each is generated, and when at least one of the signals is generated, flagged F _L at step S44, S45, a table value T _L zero To do. Then, in step S46, a gain coefficient C _kg (C _kg ≤1) that gradually increases corresponding to T _L is read from the table, and in step S47, the tuning gain K _g is multiplied by C _kg to obtain K.
Correct _g .

Next, if the determination in step S41 is "YES", the table value T _L is incremented in step S48, and it is checked in the next step S49 whether the table value T _L has reached the maximum value T _Lmax. this determination proceeds to step S46, S47 while "NO" is corrected to gradually increase the tuning gain K _g, but if the determination in step S49 is "YES", the flag F _L at step S50 To step S6.

7 to 9 are views showing a fourth embodiment of the engine control apparatus according to the present invention. In this example,
The software is shared by using the adaptive control part of the ISC and the A / F control as a subroutine to reduce the number of program steps.

FIG. 7 is an image diagram showing the features of this embodiment.

8 and 9 show the main flow of this embodiment, and steps S52 to S58 of FIG. 8 are A / F.
For control, steps S59 to S65 in FIG. 9 indicate ISC, respectively.

First, in step S51 of FIG. 8, from the table of the engine speed (N _e ) and the intake negative pressure (Boost), the excitation data r _ea of the A / F control, the tuning gain K _ga, and the ISC are added. The vibration data r _es and the tuning gain K _gs are read, and in step S52,
Let r _ea and K _{ga be} the initial values of the excitation input r _e and the tuning gain K _g of A / F control, respectively. Next, in step S53, the target air-fuel ratio is set to the input r, and in step S54
The air-fuel ratio A / F is set as the output y.

In the next step S55, parameters for A / F control are set in the register, and in step S56.
Then, the adaptive control subroutine according to steps S3 to S20 of FIGS. 27 to 29 is executed, and step S57 is executed.
Then, the engine input u is used as the fuel injection amount. Then, in step S58, the parameters are set in the register and the A / F control is ended.

Next, in the ISC shown in FIG. 9, step S5
The target engine speed is set to the input r in step 9, and step S
At 60, _es and K _gs are respectively set as the initial values of the excitation input r _{e of the} ISC and the tuning gain K _g . Next, in step S61, the engine speed is set to the output y.

In the next step S62, the parameters for ISC are set in the register, in step S63, the adaptive control subroutine according to steps S3 to S20 of FIGS. 27 to 29 is executed, and in step S64, the engine input is performed. Let u be the bypass valve opening. Then, in step S65, the parameters are set in the register and ISC is performed.
To finish.

Next, a fifth embodiment of the present invention will be described. In the above-described fourth embodiment, the adaptive control is individually performed with respect to the A / F control and the ISC, but in the present embodiment, instead of separately performing the adaptive control, as shown in FIG. And the control constants for A / F control are arranged in the same manner on the RAM, the subprogram is stored in the memory (RA
M) By simply shifting the access address, data transfer processing becomes unnecessary, and the total number of steps and processing time can be shortened.

Steps S71 to S87 of FIG. 11 are the main control flow in this embodiment. In the main flow of FIG. 11, steps S76 to S81 are an A / F control routine, and steps S82 to S87 are an ISC control routine. Steps S79 and S85 of FIG.
The flow of the common adaptive control subroutine in FIG.
13 to FIG. 13, steps S96 to S100 of FIG. 12 are model identification, step S101 is gain setting, and step S102.
S104 is state estimation by an observer, step S105 of FIG. 13 is an integral term, and step S106 is state feedback.

The above-mentioned fourth and fifth embodiments are examples in which the A / F control and the ISC are made common in software, but the sixth embodiment of the present invention described below is This is an example of a case in which the A / F control and the ISC are commonly used in hardware. That is, in this embodiment,
As shown in the system diagram of FIG. 14, only the adaptive control part is processed by another CPU 2, and input / output (FIGS. 27 to 29) is performed.
R _e , K _g , r, y, u) of two CPs
U uses a dual port RAM that can share the RAM without an arbitration circuit. FIG. 15 is an image diagram showing memory addresses of the dual port RAM. By using this dual-port RAM, the CPU 1 that performs the main control and the CPU 2 that performs the adaptive control can operate independently of each other regardless of the status (timing) of the CPU of the other party.
Since there is no need for a logic for synchronizing, the number of program steps can be reduced and the design becomes easy.

Steps S111 to S123 in FIG. 16 are C
17 shows a flow of a control routine executed by PU1, and FIG.
Steps S131 to S134 are the main flow of the adaptive control routine executed by the CPU 2. A / F control is performed in steps S116 to S119 of FIG. 16, step S120.
~ S123 is ISC. The flow of the adaptive control subroutine is the same as steps S91 to S108 in FIGS. 12 and 13 described above.

FIG. 18 is a block diagram showing an engine controller according to the seventh embodiment of the present invention. The feature of this embodiment is that the internal state quantity of the engine such as the torque generated by the engine is directly measured by a sensor, and the output of this sensor is incorporated into the adaptive control system.

The torque generated by the engine can be detected by a shaft torque sensor or an average effective pressure calculated by combustion pressure.

According to this embodiment, the convergence speed of state estimation is improved, and the convergence of the entire adaptive control is accelerated. Therefore,
Highly accurate feedback control becomes possible from an early stage.

Flow charts of the control routine in this embodiment are shown in FIGS. 19 to 21, the step S142 for reading the torque T is inserted between the steps S2 and S3 in FIG. 27, and of the steps S13 to S15 showing the state estimation operation by the observer in FIG. 29. After step S14, step S142 for adding a state quantity representing torque is newly provided, and step S143 is provided instead of step S15, and the flow is the same as the flow of FIGS. 27 to 29.

[Brief description of drawings]

FIG. 1 is a block diagram showing an engine control device according to a first embodiment of the present invention.

FIG. 2 is a flowchart of the control routine.

FIG. 3 is a flowchart of a control routine in the engine control device according to the second embodiment of the present invention.

FIG. 4 is a flowchart showing a modified part of the flowchart of FIG.

FIG. 5 is a block diagram showing an engine control device according to a third embodiment of the present invention.

FIG. 6 is a flowchart of the control routine.

FIG. 7 is an image diagram of a RAM address map of an engine control device according to a fourth embodiment of the present invention.

FIG. 8 is a flowchart showing the first half of the main flow of the control routine.

FIG. 9 is a flowchart showing the latter half of the main flow of the control routine.

FIG. 10 is an image diagram of a RAM address map of an engine control device according to a fifth embodiment of the present invention.

FIG. 11 is a flowchart showing a main flow of the adaptive control routine.

FIG. 12 is a first half of a flowchart of the adaptive control subroutine.

FIG. 13 is the latter half of the flowchart of the adaptive control subroutine.

FIG. 14 is a system diagram of an engine control device according to a sixth embodiment of the present invention.

FIG. 15 is an image diagram showing memory addresses of the dual port RAM.

FIG. 16 is a flowchart showing a control routine executed by the CPU 1.

FIG. 17 is a flowchart showing a main flow of a control routine executed by the CPU 2.

FIG. 18 is a block diagram showing an engine control device according to a seventh embodiment of the present invention.

FIG. 19 is a flowchart showing a part of the control routine.

FIG. 20 is a flowchart showing a part of a control routine continued from FIG.

FIG. 21 is a flowchart of a control routine continued from FIG.

FIG. 22 is a configuration diagram of an engine control system.

FIG. 23 is a block diagram used for explaining state feedback control.

FIG. 24 is a block diagram for explaining adaptive control.

FIG. 25 is a flowchart for explaining adaptive control.

FIG. 26 is a timing chart showing a convergence state of control by adaptive control.

FIG. 27 is a part of a flowchart showing a specific example of an adaptive control routine.

FIG. 28 is a part of the flowchart of the adaptive control routine following FIG. 27.

FIG. 29 is a flowchart of the adaptive control routine following FIG. 28.

[Explanation of symbols]

 1 engine 2 exhaust passage 3 air-fuel ratio sensor 4 controller 5 fuel injection valve 6 intake passage 7 throttle valve 8 bypass passage 9 bypass valve 10 engine speed sensor 11 intake negative pressure sensor 12 water temperature sensor

Claims (5)

[Claims] 1. A control means for performing feedback control so that a state quantity related to an engine approaches a target value based on a predetermined control constant, and a degree of change of the state quantity associated with the feedback control matches a required degree of change. In an engine control device provided with a control constant changing means for changing the control constant, a detecting means for detecting whether or not there is a change in an engine operating state, and a change in the engine operating state exceeding a predetermined value by the detecting means. The engine control device is provided with a limiting means for limiting the change of the control constant in response to the detection of the. 2. The engine control device according to claim 1, further comprising tuning means for operating the control constant changing means for each operating state of the engine. 3. The engine control device according to claim 1, wherein the control means is an air-fuel ratio feedback control means. 4. The engine control device according to claim 1, wherein the control means comprises an idle speed feedback control means. 5. A control means for performing feedback control so that a state quantity relating to the engine approaches a target value based on a predetermined control constant, and a degree of change of the state quantity associated with the feedback control matches a required degree of change. In an engine control device including a control constant changing unit that changes the control constant, an external load detecting unit that detects an operating state of an external load with respect to the engine, and a tuning gain of the control constant when the external load is operating are reduced. An engine control device comprising means for gradually increasing the tuning gain thereafter.