JPH05115189A - Torsional system control system employing state space method - Google Patents

Abstract

PURPOSE:To improve torsional vibration-proof performance and transient characteristics by employing disturbance torque estimated through an equivalent disturbance torque estimating mechanism in the configulation of an observer under lowest order state thereby estimating internal state variables of the torsional system. CONSTITUTION:In a torsional system where a driver 3 is coupled through a resilient shaft 4 with a load 5, transient load speed response lags behind motor speed in a disturbance torque estimating mechanism 6 because of the resilient shaft 4. But the lag is small as compared with the response of entire system, and thereby resiliency of the shaft 4 is neglected in the estimation of disturbance torque. Thus estimated equivalent disturbance torque is employed in the design of a lowest order state observer while furthermore a load speed and a torsional torque are estimated and normal state feedback control is adapted to the torsional system in order to select a desired pole arrangement thus determining a state feedback gain 12. According to the system, torsional vibration of shaft is suppressed and transient characteristics can be improved.

Description

Detailed Description of the Invention

[0001]

[Industrial application] In a motor drive system for industrial application, an elastic element is included in a torque transmission mechanism to form a resonance system in relation to the moment of inertia of the motor itself and a load. Normally, since the torque transmission mechanism is rigidly coupled, the influence of the resonance system on control can be ignored, but in the case of flexible coupling, it cannot be ignored.

When the proportional-plus-integral control (PI control) which is usually used for a soft-coupling system is applied to try to enhance the quick response, torsional vibration of the shaft occurs. Further, when the disturbance torque or the fluctuation of system parameters is large, the output is adversely affected.

The present invention solves such a problem by
The present invention relates to a control system that applies state feedback control to a torsion system to improve torsional vibration isolation and transient characteristics.

In order to realize the state feedback control, it is necessary to know all the state variables. The present invention relates to the idea of a new equivalent disturbance torque estimation mechanism in order to realize a minimum dimensional state observer in the case where state variables cannot be detected.

[0005]

2. Description of the Related Art An induction motor directly connected to a shaft having an elastic body and a load constitute a torsion system as shown in FIG. Each symbol in the figure has the following meaning.

[Outer 1]

In FIG. 2, since the damping of the electric motor 3, the load 5 and the shaft 4 is extremely small, it can be practically ignored.
Further, since the electric motor 3 is subjected to instantaneous space vector control by the PWM inverter 2, its response is much faster than that of the torsion system. Therefore, the torque constant Kt is assumed to have no time delay in FIG. 2 (see Takahashi et al. New High Speed Torque Control Method for Induction Motor Based on Instantaneous Slip Frequency Control ", Electron Theory B, 106-1, 1986). Therefore, when the conventional PI speed control is applied to this torsion system, its block diagram is shown in Fig. 3.
become that way.

When the block of the torsion system 9 surrounded by the one-dot chain line in FIG. 3 is rewritten into the state equation representation by the state space method, the torsion system 9 is represented by the one-dot chain line in FIG. From this FIG. 4, a state equation like the equation (1) is obtained.

[Equation 1] here,

[Equation 2]

The characteristic equation is the equation (2).

[Equation 3]

Obviously this system itself is a resonant system. Further, it has been confirmed that when the PI gain is increased in order to improve the responsiveness of the rotation speed, the torsional vibration of the shaft occurs near the resonance frequency ω ₀ .

When the torsional vibration occurs, there is a risk that the shaft will be threaded, the shaft will be broken, and the like. In the past, it was considered that PI control cannot realize active torsional vibration isolation control with high responsiveness. The control response frequency was set lower than the resonance frequency ω ₀ (the PI gain was lowered to slow the response), and control was performed so as not to give a sudden change. Further, the resonance frequency ω ₀ is increased by increasing the elastic coefficient of the one axis.

As described above, the conventional PI control cannot perform active torsional vibration isolation control.

[0012]

In order to suppress the torsional vibration of the shaft and improve the control response of the load speed, all poles of the system must be in the desired stable arrangement.

Therefore, it has been attempted to improve the characteristics by applying state feedback control to this torsion system. It should be noted that the principle of the state feedback control has been described in many documents, and will not be described here.

On the other hand, in order to realize the state feedback control as described above, it is necessary to know all the states.

[Outside 2] The minimum dimensional observer can be designed by the Gopinath method,
It is necessary to know the disturbance torque.

Therefore, generally, a method has been proposed in which the derivative value of the disturbance torque is set to zero, or the disturbance torque is simply assumed to be zero. However, in both cases, the output of the state observer is large due to the disturbance torque or parameter fluctuations. Since an error occurs, the transient characteristics may be deteriorated and the system may become unstable.

[0016]

FIG. 1 shows a principle diagram of a state feedback control for facilitating the understanding of the technical idea of the present invention for solving the above-mentioned problems. Determination of a state feedback gain, equivalent disturbance torque And the minimum dimension state observer. The same reference numerals as those in FIG. 2 indicate the same parts, 1 is a state feedback control unit, 6 is an equivalent disturbance torque estimation unit, and 7 is a minimum dimension state observer unit.

1. Determination of State Feedback Gain As described above, in order to suppress the torsional vibration of the shaft and enhance the responsiveness of the rotation speed, all poles of the system must be arranged at desired positions. .. On the other hand, in the speed control system, the rotation speed needs to follow the command value without a steady error. In this case, the configuration of the closed loop state feedback control system incorporating the integral element is as shown in FIG.

In FIG. 4, K _I is an integral gain,
F is a state feedback gain.

[Equation 4]

FIG. 5 is a block representation of the above equation of state. This is the same as that shown in FIG. 4 only after variable conversion.

Although the state feedback gain can be determined by the optimum regulator design method, it is necessary to give an appropriate evaluation function and solve the Riccati equation. Riccati in general
Solving equations is difficult and choosing an evaluation function is not so easy. Here, as an example, the feedback gain is obtained by giving the characteristic root (pole) of the system from the requirements of the suppression of torsional vibration and the transient response characteristics.

From equation (5), the characteristic equation is

[Equation 5] Can be asked for.

In selecting the poles, it is generally necessary to select a large absolute value of the poles in equation (7) in order to improve the transient characteristics of the system. At this time, a large input manipulated variable is required and the drive system The torque capacity of is increased. in this case,
The poles may be selected so that the input operation amount does not become excessive.

2. Equivalent Disturbance Torque Estimation As described above, in order to realize the speed control system shown in FIG. 3, all state variables must be fed back.

[Outside 3] The minimum dimensional state observer can be designed by the Gopinath method, but it is necessary to know the disturbance torque. Therefore, in general, a method has been proposed in which the derivative value of the disturbance torque is set to zero, or the disturbance torque value is simply set to zero (Oishi et al. ”, Electron Theory B, 106-2, pp.111-118, 1986, and SG Kulkar
"Robust Control for Vibration Sup" by ni & Y.Hori
compression in2-Mass System "IEE-Japan JAS, pp579 ~ 5
82, 1990). However, in both cases, a large error occurs in the output of the state observer due to disturbance torque or parameter fluctuation, which may deteriorate the transient characteristics and cause the system to become unstable.

As a solution to this problem, a disturbance torque estimating mechanism is proposed and described below. Since the shaft has elasticity, the response of the load speed lags behind the motor speed during a transition, but this delay is considered to be smaller than the response of the entire system. Therefore, the elasticity of the shaft may be ignored when estimating the disturbance torque. Then, the load speed matches the motor speed, and the load torque estimating mechanism is shown in FIG.

From FIG. 6, the estimated value of the disturbance torque is (10)
It can be obtained like the formula.

[Equation 6]

3. Minimum Dimensional State Observer When the estimated equivalent disturbance torque is used to design the minimum dimensional state observer by the equations (1) and (2), the following equation is obtained.

[Equation 7] Is obtained as. Therefore, the configuration of the state observer is as shown in FIG. 7, and the unknown state variable can be estimated.

The above problem can be solved by going through the above three steps.

[0028]

In order to know the internal state of the torsion system (load speed and torsion torque), the equivalent disturbance torque estimation mechanism is first used to estimate the equivalent disturbance torque. Then, a minimum-dimensional state observer is constructed using the estimated disturbance torque, and the internal state variable of the torsion system is estimated by it.

Since all the state variables of the torsion system can be estimated in this way, the usual state feedback control is adapted to the torsion system and the desired pole placement is chosen to determine the state feedback gain.

[0030]

FIG. 8 shows a block diagram of a torsional system state feedback control according to the present invention when speed control is performed.

Here, the parameters of the twisting system are as follows,

[Equation 8]

The method of selecting constants for this control system and the control results thereof are shown below. In consideration of the maximum torque of the induction motor, the characteristic root is selected as shown below.

[Equation 9]

FIGS. 9 to 12 show simulation results of the response of the speed command value and the rotational speed at the time of sudden change of the disturbance torque in the case of the state feedback control. These figures show the estimated disturbance torque value and the output value of the state observer. The actual output values are also shown.

FIG. 9 shows the case where all the system parameters match the nominal value. From this, it can be seen that there is no overshoot and good results are obtained even during sudden load changes.

[Outside 4] That is because the state observer of FIG. 6 is essentially an integrator.

Next, the influence of system parameter fluctuation will be considered.

[Outside 5]

FIG. 10 shows the response of the rotational speed when the elastic coefficient of the shaft is halved to the nominal value, and FIGS. 11 and 12 show the case where the load inertia is ⅓ and 3 times the nominal value, respectively. Is the speed response of. It can be seen that in both cases, there was no vibration, and that fairly good characteristics were obtained.

[0037]

In the past, when the state feedback control was applied to the torsion system, the disturbance torque could not be estimated. Therefore, the torsion system state detector was required. The control system becomes complicated and the cost also increases. Also, when the state variable cannot be detected, the state variable has to be estimated approximately. Therefore, a large error may occur in the output of the state observer due to disturbance torque or parameter fluctuation, which may deteriorate the transient characteristics and cause the system to become unstable. When the state feedback control is applied to the torsion system, it is necessary to know all the state variables.

According to the present invention, since the disturbance torque can be estimated, the load speed and the torsion torque can be estimated. Therefore,
A detector for these state variables is not required, the control system has a simple structure, and the cost can be reduced. Further, the state feedback control can be applied even when the state variable cannot be detected. Further, since fluctuations in system parameters are regarded as disturbances, robust and stable control can be realized not only with disturbance torques but also with parameter fluctuations.

[Brief description of drawings]

FIG. 1 is a state feedback control principle diagram shown to facilitate understanding of a technical idea of the present invention.

FIG. 2 is a diagram showing an electric motor drive torsion system model having a resonance system.

FIG. 3 is a PI control block diagram of a torsion system.

FIG. 4 is a configuration diagram of a state feedback control system.

FIG. 5 is a block diagram showing the state equation of equation (5).

FIG. 6 is a diagram showing a disturbance torque estimating mechanism.

FIG. 7 is a configuration diagram of a state observer.

FIG. 8 is a control block diagram of an embodiment of the present invention.

FIG. 9 is a diagram showing a simulation result of a speed response due to a speed command value and a 100% sudden load change when all system parameters of the state feedback control system match the nominal values.

FIG. 10 is a diagram showing a simulation result of a speed feedback value and a speed response due to a sudden change in 100% load when the elastic coefficient of the shaft is set to a half of the nominal value in the state feedback control system.

FIG. 11 shows the speed command value and 10 when the load inertia of the state feedback control system is set to one third of the nominal value.
It is a figure which shows the simulation result of the speed response by 0% load sudden change.

FIG. 12 is a diagram showing a simulation result of a speed response due to a speed command value and a 100% load sudden change when the load inertia of the state feedback control system is set to three times the nominal value. 1 State feedback control section 2 Inverter section (torque coefficient generation section) 2'Nominal torque coefficient 3 Motor section 3'Motor damping neglected motor 4 Torsional shaft section 4'Axial damping neglected axis 5 Load section 5 ' Load that ignores load damping 6 Equivalent disturbance torque estimation unit 7 Minimum dimension state observer unit 8 Proportional integration unit 9 Torsional system unit 10 Integrator 11 Integration constant 12 State feedback gain 13 Torsional system that ignores elastic coefficient of axis 14 Disturbance torque Estimator 15 Integration block 16 Minimal-dimensional state observer in Fig. 7

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical display location G05D 19/02 D 8811-3H H02P 5/00 X 7315-5H

Claims (2)

[Claims] 1. A state in which a state feedback gain is multiplied by an internal state variable of a controlled object such as a motor speed, a torsion torque, and a load speed of a controlled object of a torsion system in which an electric motor and a load are coupled by an elastic element. In a torsional control method that configures a feedback control system, and further includes an integral element in the difference between the speed command value and the load speed to configure a closed-loop feedback speed control system, ignore the elasticity of the shaft and input the motor command. And a speed of the electric motor to form an equivalent disturbance observer, and estimate the disturbance torque that is the input of the control target. 2. A load velocity and a torsion torque or a torsion angle of the torsion control system are estimated by a minimum-dimensional state disturbance observer using the disturbance torque estimated value obtained by the estimation mechanism, and the state feedback speed is estimated by the estimated value. The torsion system control method according to claim 1, wherein the control system is configured.