JPH0635506A - Variable structure control method - Google Patents

Abstract

PURPOSE:To vary the PI constant and the primary delay filter time constant of an equivalent disturbance observer and to attain a high function for the control response by multiplying the deviation value between the command input and the state input of a controlled system and the differential value of the deviation value by a weight coefficient and adding together these multiplied values. CONSTITUTION:For instance, the velocity deviation and its differential value are multiplied by a weight coefficient and these multiplied values are added together. Thus a parameter E is obtained. The function generators 4-6 calculates a proper P (proportion) constant, an I (integration) constant, and a primary delay filter time constant based on the parameter E respectively. Then each output of the generators 4-6 is sent to the corresponding parts of a PI controller 1 and an equivalent compensating part 3 with the velocity deviation defined as an input. Thus the output of generators 4-6 are changed into a proper proportion constant, an integration constant, and a filter time constant respectively. In such a way, the proportion time constant, the integration constant, and the primary delay filter time constant of an equivalent disturbance observer can be varied in accordance with the operating states. Thus it is possible to improve the robust properties and the stability against the setting fluctuation and the load fluctuation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to application of an equivalent disturbance observer which is usually used to realize robust control in an industrial application motor drive system, and is equivalent to an equivalent disturbance observer and a proportional integral control (PI control). ) Is applied to make the compensation amount variable according to the operating condition by the deviation amount or its differential value, thereby enhancing the adaptability of control and improving the steady-state and transient characteristics.

[Prior art]

It is often practiced to apply a proportional-plus-integral control (PI control) and an equivalent disturbance observer to a speed control system of an induction motor to achieve high functionality. In this case, the proportional constant (KP) and the integral constant (KI) were controlled as constant values. Further, as a constant that determines the control frequency band of the equivalent disturbance observer, there is a first-order filter time constant of the observer, which was also conventionally controlled by a constant value.

Therefore, in order to improve the control performance, if these constants are excessively compensated, the response of the control is improved. However, since the control compensation amount is superposed on the torque command of the electric motor, at the steady state. Since high frequency noise is generated and high frequency torque ripple and noise of the electric motor are generated, these time constants have to be appropriately adjusted and selected on the stable side. These were trade-off relationships.

An example of a speed control system of an induction motor is shown in FIG.
Let's explain the above problems based on. The symbols shown are ω _m ^* ; speed command KP; proportional constant KI; integration constant T ^* ; torque command Kt; torque generation count Jm; electric motor inertia Dm; electric motor damping ω _m ; electric motor speed.

In FIG. 6, 1 is a PI control device, 2 is a control target, and the deviation e between the set input ω _m ^* and the state quantity ω _m is given to the control target 2 via the PI control device 1 as a torque command T ^*. Is applied to stabilize the speed control system. Here, the PI constant was controlled at a constant value. Also, the equivalent disturbance compensator of the part 3 surrounded by the broken line is Go
It is composed of a minimum-dimensional torque observer using the pinath method and can be expressed as the following equation.

[Formula 1] T _L = 1 / (1 + T _f S) {Kt * T ^* -Jm S ω _m } (1)

T _f is a primary filter time constant, and this value is also controlled at a constant value regardless of such changes in operating conditions even if speed setting fluctuations, parameter fluctuations, or load fluctuations occur with respect to the controlled object. Therefore, various problems have occurred in control.

[0007]

Since the control is performed with a constant value irrespective of such a change in the driving situation, the influence thereof may deteriorate the control characteristics. For example, if the PI constant is set to a constant value, overshoot or windup phenomenon occurs when the setting changes. When the observer time constant is set to a constant value, the robustness of the entire control is significantly improved, but high-frequency noise tends to be superimposed on the compensation output, which constantly causes high-frequency torque ripple and increases motor noise. There was an occasion.

If the PI constant is large and the observer time constant is fast in order to improve the characteristics, the above problems occur, and these constants are rather small in proportion constant KP and integral constant KI so that they are stable in practical use. In addition, the observer time constant T _f had to be large.

[0009]

In order to solve the above problem, the adaptiveness of control is enhanced by varying the compensation amount according to the operating condition as described below, and the steady and transient characteristics are improved. The control means to be improved will be described.

In the speed control system, the speed deviation e is a control variable for judging the operating condition. Using this value as a parameter, the above PI control constant and load torque observer 1
The next delay time constant is variably controlled. When the value of the speed deviation e is large, the control compensation amount is increased to improve the responsiveness, and when it is small, the compensation amount is decreased to form a variable structure to improve the control stability. Therefore, as shown in the equation (2), the speed deviation e and its differential value de / dt are added to the weighting factors K ₁ and K ₂
E is obtained as a value for evaluating the driving situation by multiplying and adding

[Equation 2] E = K ₁ e + k ₂ de / dt (2) By varying the constants such as P (proportional) and I (integral) according to the magnitude of E, constants according to the operating conditions It is intended to provide a variable structure control method for selecting. These PI constants and filter time constants may be given by a function having E as a parameter so as to be variable in accordance with the operating conditions. In order to make the PI constants and filter time constants variable in accordance with the operating conditions in advance. The defined data table may be referred to and selected, and the PI constant and the filter time constant may be given by the host computer according to the operating condition.

As described above, the operating state is judged by the speed deviation e and the differential value de / dt of the speed deviation, and the proportional constant KP, the integral constant KI and the filter time constant are determined by these values.
Select T _f to an optimum value.

[0012]

[Operation] With the above configuration, by selecting the proportional constant KP, the integration constant KI, and the filter time constant T _f to the optimum values according to the operating conditions, high frequency torque ripple is not generated and the motor noise is reduced. It is possible to maintain good control characteristics without increasing overshooting and without causing an overshoot or a wind-up phenomenon when the setting changes.

[0013]

1 and 3 show an embodiment of a variable structure control method according to the present invention. In the illustrated control, E obtained by the equation (2) is used as a parameter. In this equation (2), the velocity deviation e and its differential value de / dt are K ₁ and K ₂ respectively.
At this time, if the evaluation of the deviation value e is to be increased, the value of K ₁ may be set larger than the value of K ₂ .
In the embodiment shown in FIG. 1, function generators 4, 5 and 6 are provided which can obtain an appropriate proportional constant KP, integral constant KI and filter time constant T _f in accordance with the parameter E.
The velocity deviation e is input and their outputs are sent to the corresponding parts of the PI controller 1'and the equivalent disturbance compensator 3 ', respectively.
Appropriate proportional constant KP, integral constant KI and filter time constant T _f
Change to. FIG. 3 is a diagram showing the characteristics of these function generators, where (a) is the characteristic of the proportional constant KP for the parameter E in the function generator 4, and (b) is the integration constant KI for the parameter E in the function generator 5. , (C) is the filter time constant T for the parameter E in the function generator 6.
_Shows the characteristics of _f . Here, in the actual system, the control upper limit value is defined by the acceleration / deceleration limiter, torque limiter, current limiter, etc. Therefore, the upper limit value may be provided in the variable output of each function generator.

Further, instead of using the function generator, the function characteristic curve can be converted into a data table and the data table can be used. That is, as shown in FIG. 4, a data table (a) recording appropriate values of the proportional constant KP, the integral constant KI, and the filter time constant _Tf with respect to the value of the speed deviation e, or the proportional constant with respect to the value of the parameter E. The data table (b) in which appropriate values of KP, integration constant KI, and filter time constant T _f are recorded is used instead of the function generators 4, 5, and 6, and the proportional constant KP and integration constant KI are used according to the operating conditions. Also, the filter time constant T _f can be selected as an optimum value.

Further, when the control system has a host computer, a method for obtaining an output by a function operation in the computer 7 as shown in FIG. 5 is also possible.

In the above description, the proportional constant KP and the integral constant KI
And the method to make the filter time constant T _f variable at the same time was described, but the combination such as controlling only the proportional constant KP and the integration constant KI or controlling only the filter time constant T _f can be freely selected. Of course, of course.

[0017] When for example K ₂ 0 of the weight counter K _1, K ₂ in the formula (2), but not the evaluation of the differential value de / dt of the speed deviation, the evaluation of only the speed deviation e, PI The simplest method of changing the constant and the filter time constant according to the operating condition can be realized as shown in FIG. In FIG. 2, (a) is the characteristic of the proportional constant KP with respect to the speed deviation e in the function generator 4, (b) is the characteristic of the integration constant KI with respect to the speed deviation e in the function generator 5, and (c) is the function generator 6 The characteristic of the filter time constant T _{f with} respect to the speed deviation e in FIG.

[0018]

According to the variable structure control method as described above, the operating condition can be instantaneously and instantaneously observed, and the control can be performed with the control constants according to the operating condition, whereby the robustness and the stability against the setting fluctuation and the load fluctuation can be obtained. It is possible to improve. The amount of overshoot can be greatly reduced, and the amount of impact drop can be significantly reduced.

In particular, since the control constant is made large during the transition, the response becomes fast, and on the contrary, at the time of settling or in the steady state, the response is made small to make the control system robust and stable. Furthermore, the occurrence of steady torque ripples and speed ripples, which occurs when the control is performed in a constant manner as in the past, is greatly suppressed. It can also be expected to reduce noise.

According to the present invention, the control constant can be easily changed to a parameter adapted to the operating condition, and the variable selection of the constant greatly improves the degree of freedom in selection. It can also be applied to position control and torque control.

[Brief description of drawings]

FIG. 1 is a basic block diagram showing a variable structure control method for a system to which a velocity feedback control and an equivalent disturbance observer are applied to facilitate understanding of the technical idea of the present invention.

FIG. 2 of the present invention, the differential value de / dt of the speed deviation is not evaluated, but only the speed deviation e is evaluated, and the PI constant and the filter time constant are made variable according to the driving situation. Shows the characteristic curve of various methods, (a) is the characteristic of proportional constant KP for speed deviation e, (b) is the characteristic of integration constant KI for speed deviation e, (c) is the filter time constant T _f for speed deviation e. It shows the characteristics.

FIG. 3 is an example of a function characteristic curve of a variable structure control method, wherein a PI constant and a filter time constant are given by a function having E as a parameter so as to be variable according to an operating condition, among the present invention. (A) shows the characteristic of the proportional constant KP with respect to the parameter E, (b) shows the characteristic of the integration constant KI with respect to the parameter E, and (c) shows the characteristic of the filter time constant T _{f with} respect to the parameter E.

FIG. 4 shows an example in which a function characteristic curve of a variable structure control method is used as a data table, and (a) records appropriate values of a proportional constant KP, an integration constant KI, and a filter time constant T _f with respect to a value of a speed deviation e. Data table, (b) is parameter E
It is a data table which recorded appropriate values of a proportional constant KP, an integral constant KI, and a filter time constant _Tf with respect to the value of.

FIG. 5 is a block diagram showing an example of a method of obtaining an output by a function operation in the computer when there is a host computer in the control system.

FIG. 6 shows a proportional-integral control (PI
FIG. 8 is a block diagram showing a conventional control system that has been made highly functional by applying control) and an equivalent disturbance observer.

[Explanation of symbols]

 1, 1'Velocity PI control unit 2 Control object 3 3'Equivalent disturbance compensation unit 4, 5, 6 Function generator 7 Upper computer

Claims (4)

[Claims] 1. A deviation amount between a control symmetrical command input and both states is set as an applied input amount of the controlled object through a stabilizing compensating means including P (proportional) and I (integral). In the electric motor control device having the equivalent disturbance compensating means for calculating the equivalent disturbance of the controlled object through the first-order lag filter using the applied input amount and the state quantity of the controlled object, ) And I (integration) and the time constant of the first-order lag filter are multiplied by the deviation amount or its differential value by a weighting coefficient and added to make it variable according to the driving situation. Variable structure control method. 2. A variable structure control method according to the above control method, wherein a PI constant and a filter time constant are given by a function so as to be variable according to an operating condition. 3. The variable structure control method according to claim 3, wherein the PI constant and the filter time constant are selected with reference to a data table defined in advance so as to be variable according to the operating condition. . 4. A variable structure control method according to the above control method, wherein a PI constant and a filter time constant are provided from a host computer in accordance with an operating condition.