JPH10197432A - Controller for material vibration testing machine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a controller for material vibration testing machine which can optimize the response time in real time by using an adaptive filter and adjust the gain of the control closed loop of a testing machine in real time so as to prevent the closed loop from becoming unstable. SOLUTION: The factor setting means 214 of a driving signal generating section 21 sets the factor of a first adaptive filter means 213 based on a response signal and a driving signal so that the filter means 213 may have the inverse function of the transfer function of a testing machine, corrects the driving signal so that the response signal may become equal to a target signal, and supplies the corrected driving signal in the next period. The factor setting means 233 of an adjusting signal generating section 23 sets the factor of a second adaptive filter means 231 based on the driving signals and a response signal containing an extrapolating signal component so that the filter means 231 can have the function corresponding to the transfer function of the testing machine. The gain of the testing machine is adjusted in accordance with the difference between the output signals of the second filter means 231 for extrapolating signals and a fixed filter means 232 having an ideal transfer function.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to, for example, a control device for an electro-hydraulic material vibration tester, and more particularly to a control device for controlling a response function between an input and an output of a material vibration tester by using an adaptive filter. The present invention relates to a control device for optimizing the material test and achieving accurate waveform reproducibility in a material test.

[0002]

2. Description of the Related Art In an electrohydraulic material vibration tester, a feedback loop, which is a basic control loop of the tester, forms an independent closed loop. In a conventional control device, a tentative test operation is performed at the start of a test by a test machine so that the loop response function has a unique optimum value including the characteristics of a test piece, and an optimum response value at that time is obtained to obtain a control closed loop. PID values were set (US Pat. No. 5,394,0).
No. 71).

If the frequency that governs the stability condition of the closed loop of the tester is, for example, several tens of hertz, control is impossible unless information on the response characteristics near the frequency is taken into the system. Therefore, even if the operating frequency is 1 Hertz, the information left in the adaptive filter includes not only information of 1 Hertz but also information of several tens Hertz necessary for stability determination in the filter coefficient. There must be. For this reason, there is a type in which a waveform necessary for determining the stability of the driving wave is inserted (Japanese Patent Publication No. Sho 5).
No. 7-3011).

[0004]

However, in a system in which the response characteristic correction system using an adaptive filter is always set to an optimum value (U.S. Pat. No. 5,394,071), each system constituting the system is required. Since the balance by the assignment of the response function between the elements, that is, the gain of the control closed loop determined by the PID value during the operation was not managed, the influence of the characteristic change due to fatigue and hardening of the specimen under the actual operation was not considered. As a result, there is a possibility that the optimum value changes with time, the closed loop becomes unstable, and a dangerous state such as oscillation occurs.

Further, if a waveform necessary for determining the stability of an operating wave is simply inserted, the tester is also operated by combining the inserted wave, so that there is a problem in application to a precise test. .

In view of the above-mentioned conventional problems, the present invention optimizes a response function between input and output in real time using an adaptive filter. An object of the present invention is to provide a control apparatus for a material vibration testing machine capable of adjusting a gain of a control closed loop of an optimum testing machine in real time so as to avoid making the closed loop unstable.

[0007]

According to a first aspect of the present invention, there is provided a control apparatus for a material vibration tester, which is operated by supplying a drive signal having a predetermined repetition cycle. A control device that outputs a response signal corresponding to a transfer function that changes with the progress, and controls a material vibration testing machine having a control closed loop whose gain is adjusted by input of an external adjustment signal, First adaptive filter means for correcting the drive signal so that a response signal is equal to a target signal which is a target of operation of the material vibration tester; the drive signal supplied to the material vibration tester; The first adaptive filter has an inverse transfer function of a transfer function of the material vibration testing machine based on a response signal obtained at an output of the material vibration testing machine in response to the drive signal. A first coefficient setting means for setting a coefficient of the data means, and a drive signal generating unit for supplying the drive signal corrected by the first adaptive filter to the material vibration testing machine in a next cycle; An extrapolation signal generator for generating an extrapolation signal to be supplied as a disturbance to the material vibration tester, a second adaptive filter means, and a transfer function of the material vibration tester based on the drive signal and the response signal. A second coefficient setting means for setting a coefficient of the second adaptive filter means so as to have a transfer function corresponding to the following, and transmission of an ideal material vibration tester exhibiting stable characteristics with respect to the extrapolated signal. Fixed filter means whose coefficients are set in advance so as to have a function, and a difference between signals obtained respectively at the outputs of the second adaptive filter means and the fixed filter means in response to the supply of the extrapolated signal. By It is characterized in that it comprises an adjustment signal generator for generating an adjustment signal for adjusting the gain of the material vibration testing machine.

In the above configuration, when a driving signal of a predetermined repetition cycle is supplied to the material vibration testing machine for operation, the first coefficient setting means of the driving signal generating section outputs a response signal obtained from the output of the material vibration testing machine. And the drive signal, the coefficient of the first adaptive filter means of the drive signal generation unit is set so as to have an inverse transfer function of the transfer function of the material vibration tester, and the response signal indicates the operation of the material vibration tester. The drive signal is corrected so as to be equal to the target signal to be a target, and the corrected drive signal is supplied to the material vibration testing machine in the next cycle. Therefore, even if the extrapolation signal generator supplies an extrapolation signal as a disturbance to the material vibration tester, the material vibration tester can perform a test according to the target signal without being disturbed by the extrapolation signal. it can.

In the adjustment signal generating section, the second coefficient setting means sets the coefficient of the second adaptive filter means based on the drive signal and the response signal including the extrapolated signal component.
The second adaptive filter means has a transfer function corresponding to the transfer function of the material vibration tester. By supplying an extrapolation signal, the signal of the signal obtained at the output of the second adaptive filter means and the fixed filter means is obtained. The difference is used to generate an adjustment signal for adjusting the gain of the material vibration tester.However, the fixed filter means determines the transfer function of the ideal material vibration tester that exhibits stable characteristics with respect to the extrapolated signal. The transfer function of the material vibration tester with respect to the extrapolated signal is also maintained at an ideal value by the adjustment signal based on the difference between the output signal of the second adaptive filter means and the output signal of the fixed filter means since the coefficient is set in advance. The test can be performed while performing.

According to a second aspect of the present invention, there is provided a control apparatus for a material vibration testing machine according to the present invention. An electrohydraulic having the servo controller for controlling the gain of a deviation signal corresponding to a difference between the drive signal and the response signal, and a servo valve operated by the gain-adjusted signal by the servo controller; And a vibration corresponding to the target signal is given to a material to be subjected to a vibration test by the actuator.

In the above configuration, the adjustment signal generating section is generated even when the characteristics of the extrapolated signal component change with the progress of the operation by applying the vibration according to the target signal to the material to be subjected to the vibration test by the actuator. The gain of the material vibration tester is adjusted by the adjustment signal, and the transfer function of the material vibration tester with respect to the extrapolated signal is always maintained in an ideal state.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. 1 and 2 show an embodiment of an apparatus for implementing a control device of a material vibration testing machine according to the present invention.

In FIG. 1, reference numeral 10 denotes a material vibration tester;
0 is a control device. The material vibration tester 10 operates according to a repetitive drive signal 100a having a low frequency of, for example, about 1 Hz supplied from the control device 20 and an extrapolation signal 100c having a repetition frequency of, for example, several tens of Hz. A response signal 200a is output. The response function or transfer function T of the material vibration tester 10 changes as the operation proceeds. The control device 20 controls the drive signal 10
In addition to generating and outputting 0a and the insertion signal 100c, an adjustment signal 100b for adjusting the gain of the control closed loop in the material vibration tester 10 in real time is generated and output. The generated response signal 200
Enter a. The drive signal 10 by the control device 20
0a, the adjustment signal 100b and the insertion signal 100c will be described later in detail.

The material vibration tester 10 has an addition point 11 to which the drive signal 100a from the control device 20 is supplied to one positive input. The addition point 11 adds a response signal 200a, which will be described later, supplied to the other negative input and the drive signal 100a to add a deviation signal 200 corresponding to the difference between the response signal and the drive signal.
b is output. The deviation signal 200b is a PID (proportion
al-integral-derivative) input to the servo controller 12, where the PID gain is adjusted, and the operation signal 200
Output as c. PID servo controller 12
(Proportional) I (integral) D (differential) parameters Kp, Ki,
It has gain adjusters 121 to 123 for adjusting Kd.
Among the parameters of the servo controller 12, a gain adjuster for adjusting the proportional gain Kp is automatically adjusted by an adjustment signal 100b from the control device 20, which will be described later.

An operation signal 200c output from the PID servo controller 12 is input to a servo amplifier 14 via a second addition point 13 and amplified by the servo amplifier 14, and then, for example, a servo signal of an electrohydraulic actuator 15 Applied to valve 16. In addition, the second addition point 13 outputs an extrapolation signal 100c from the control device 20 described later to the operation signal 200.
This is for superimposing on c.

The actuator 15 includes a cylinder 15a and a piston 1 movably accommodated in the cylinder 15a.
5b, and the piston 15b is operated by the pressure oil supplied into the cylinder 15a via the servo valve 16. The servo valve 16 adjusts the valve opening degree according to the magnitude of the operation signal 200c applied thereto, and controls the cylinder 15 that supplies pressure oil through the valve.
The cylinder chamber of a is determined. As a result, the piston 15b is moved at a predetermined speed in a predetermined direction in response to the application of the operation signal 200c.

A test piece T is provided at the lower end of the piston 15b.
P is connected to one end, and the other end of the test piece TP is
It is fixed via 7a. Therefore, the test piece TP
A tensile load and a compressive load are repeatedly applied by the reciprocating movement of the piston 15b. The load sensor 17a is a test piece T
The magnitude of the load applied to P is detected, and a load signal having a magnitude corresponding to the magnitude of the load is output. In addition, piston 5
A displacement sensor 17b that detects the amount of movement of the piston 15b is provided at the upper end of the position b. The displacement sensor 17b detects the amount of expansion and compression of the test piece TP and outputs a displacement signal having a magnitude corresponding to the amount of expansion and compression.

The load signal output from the load sensor 17a and the displacement signal output from the displacement sensor 17b are input to a signal switching section 18. The signal switching unit 18 includes a control device 20 described later.
One of the signals according to the current control mode is selected according to the selection signal from, and is output as a response signal 200a indicating the operation result of the material vibration tester 10 according to the drive signal 100a. It is applied to the input and supplied to the controller 20.

As described above, the control device 20 sends the drive signal 100a and the adjustment signal 10 to the material vibration tester 10.
0b, an extrapolation signal 100c and a selection signal 100d, and a response signal 200a from the material vibration tester 10. Specifically, as shown in FIG. 2, the control device 20 includes a drive signal generation unit 21 for generating a drive signal 100a, and the drive signal generation unit 21
a and the material vibration tester 1 in response to the drive signal 100a.
The drive signal of the next cycle is generated based on the response signal 200a input from 0. The control device 20 also adjusts an extrapolation signal generator 22 that generates an extrapolation signal 100c that is input as a disturbance to the material vibration tester 10, and a proportional gain Kp of the servo controller 12 in the material vibration tester 10. And a switching operation unit 24 that outputs a selection signal 100d for inputting a response signal according to the current control mode.

The drive signal generator 21 generates a target signal 100e for determining a repetitive load or displacement applied to the test piece TP in the material vibration tester 10, ie, a target value of the vibration, for the vibration test of the material. It has a target signal generator 211. The target signal 100e generated by the target signal generator 211 is applied to the + input of an addition point 212 to which the response signal 200a output from the signal switching unit 18 in the material vibration tester 10 is applied to the-input. Addition point 2
12, the target signal 100e and the response signal 200a
Is obtained, and based on the difference signal 100f, the target signal 100e and the response signal 200a are finally obtained.
Is generated, so that the driving signal 100a becomes equal to the driving signal 100a.

The difference signal 10 obtained at the output of the addition point 212
0f is supplied to an adaptive filter 213 as a first adaptive filter means comprising a finite impulse response (FIR) digital filter. The adaptive filter 213 has a coefficient for determining the frequency response function W, the coefficient of which is determined by a coefficient setting network 214 as a first coefficient setting means. Is set to have the inverse of. Applicable filter 21
The output of No. 3 is applied as a correction signal 100g to the other + input of the addition point 215 where the drive signal 100a is applied to one + input.

At the output of the addition point 215, a corrected drive signal 100h obtained by superimposing the correction signal 100g on the drive signal 100a is obtained. The corrected drive signal 100h is input to a delay circuit 216 as a delay unit that delays the time corresponding to one cycle of the drive signal 100a. Delay circuit 216
When the drive signal 100a for one cycle has been output, the corrected drive signal 100h is sequentially sent to the output as a new drive signal 100a and supplied to the material vibration testing machine 10.

The coefficient setting network 214 includes the adaptive filter 2
13, an adaptive filter 214a composed of the same finite impulse response (FIR) digital filter,
And LMS (Least-Mean-Square) algorithm network 214c. The input of the adaptive filter 214a includes
Response signal 200a obtained from output of material vibration tester 10
Is supplied. The signal 100i obtained at the output of the applicable filter 214a is applied to the minus input of the summing point 214b to which the drive signal 100a is applied to the plus input.

The difference signal 100j between the drive signal 100a and the signal 100i (the component of the drive signal 100a and the extrapolation signal 100c) is obtained at the output of the addition point 214b.
The difference signal 100j obtained at the output of the addition point 214b is L
It is provided to the MS algorithm network 214c. LMS
The algorithm network 214c generates a signal for setting the coefficient of the adaptive filter 214a based on the difference signal 100j such that the output of the summing point 214b becomes 0, and supplies the signal to the adaptive filters 214a and 213 to supply the signal to both of the adaptive filters 214a and 213. Adjust the coefficients in the same way.

The LMS algorithm network 21
The fact that the output of the addition point 214b becomes 0 by setting the coefficient by 4c means that the drive signal 100a before being applied to the material vibration tester 10 and the signal 100i after passing through the adaptive filter 214a are equal. Means
At this time, the frequency response function W of the adaptive filters 214a and 213 is equal to the inverse function of the transfer function T of the material vibration tester 10, and T = W ^-1 .

If there is a relationship of T = W ⁻¹ between the frequency response function W of the adaptive filter 213 and the transfer function T of the material vibration tester 10, the difference signal 1 obtained at the output of the summing point 212 is obtained.
00f through the adaptive filter 213, the correction signal 10
0g is added to the drive signal 100a at the addition point 215 to obtain a corrected drive signal 100h.
The difference signal 100f becomes 0 in the next cycle by supplying the drive signal 100a of the next cycle to the material vibration testing machine 10 via the drive signal 16. The fact that the difference signal 100f becomes 0 means that the response signal 200a becomes the target signal 100e.
And the material vibration tester 10 outputs the target signal 100e
Means that the vehicle is being driven according to the target value.

Since the response signal 200a includes the component of the extrapolated signal 100c inserted as a disturbance, the response signal 200a including the component of the extrapolated signal 100c and the target signal 100e is obtained, whereby the material vibration tester 10
Is the target signal 100 despite the insertion of the extrapolated signal 100c.
The operation according to e is performed.

The adjustment signal generator 23 includes an adaptive filter 231 and an observer filter 232 as second adaptive filter means including a finite impulse response (FIR) digital filter, and a second for setting coefficients of the adaptive filter 231. Coefficient setting network 23 as coefficient setting means
And 3. The adaptive filter 231 has a coefficient determining circuit 214 for determining a coefficient for determining the frequency response function H.
Is set to have a function that is equal to the transfer function T of the material vibration tester 10 that changes as the operation proceeds, whereas the observer filter 232 sets the extrapolated signal 100
advance coefficient to have a transfer function H ₀ of ideal material vibration test machine 10 shown stable characteristics is set for c.

The input of the adaptive filter 231 and the input of the observer filter 232 receive the same extrapolated signal 100 c as applied to the material vibration testing machine 10 as a disturbance.
2, response signals 100 k and 100 m to the extrapolated signal 100 c are generated at outputs of the adaptive filter 231 and the observer filter 232, respectively. The two response signals are applied to the + input and the − input of the addition point 234, respectively, and a difference signal 100n between the two response signals is output from the addition point 234 and input to the adjustment signal generation circuit 235. The adjustment signal generation circuit 235 generates an adjustment signal 100b corresponding to the difference signal 100n obtained at the output of the addition point 234.

The coefficient setting network 233 has an adaptive filter 233a composed of a finite impulse response (FIR) digital filter, an addition point 233b, and an LMS algorithm network 233c. The drive signal 100a to be applied to the material vibration tester 10 is input to the input of the adaptive filter 233a. The signal 100p obtained at the output of the applicable filter 233a is applied to the minus input of the addition point 233b to which the response signal 200a is applied to the plus input.

The difference signal 100q between the response signal 200a and the signal 100p is obtained at the output of the addition point 233b. The difference signal 100q obtained at the output of the addition point 233b
Is supplied to the LMS algorithm network 233c.
The LMS algorithm network 233c provides the difference signal 100q
, A signal for setting the coefficient of the adaptive filter 233a so that the output of the addition point 233b becomes 0 is supplied to the adaptive filters 233a and 231 to adjust both coefficients in the same manner.

The LMS algorithm network 23
The fact that the output of the addition point 233b becomes 0 by the setting of the coefficient by 3c means that the response signal 200a from the material vibration tester 10 and the signal 1 after passing through the adaptive filter 233a.
00p means that the frequency response function H of the adaptive filters 233a and 231 is equal to the transfer function T of the material vibration tester 10, and T = H
It becomes that.

If there is a relation of T = H between the frequency response function H of the adaptive filter 231 and the transfer function T of the material vibration tester 10, a response signal 100k based on the extrapolated signal 100c obtained at the output of the adaptive filter 231 Is an extrapolated signal 100c obtained by the transfer function T of the material vibration tester 10.
Can be regarded as a response signal. Therefore, the response signals 100k and 100m obtained at the output of the addition point 234
Is the transfer function T of the material vibration tester 10.
So that represents a what state against the transfer function H ₀ of ideal material vibration test machine 10 shown stable characteristics.

For example, when the difference signal 100n is 0,
The response of the adaptive filter 231 to the extrapolated signal, that is, the response of the material vibration tester 10 is maintained in the same stable state as the observer filter 232, and the proportional gain Kp holds the ideal value, and any adjustment is performed. The adjustment signal generation circuit 235 does not need to generate the adjustment signal 100b. On the other hand, when the difference signal 100n is not 0, an adjustment signal 100b corresponding to its polarity and magnitude is generated and supplied to the servo controller 12 in the material vibration tester 10 to adjust the proportional gain Kp.

When the proportional gain Kp in the material vibration tester 10 is adjusted as described above, the transfer function T of the material vibration tester 10 changes, and the response signal 200a also changes. Specifically, if H> H0, the difference signal 100
n indicates that the proportional gain Kp is larger than the standard value (see FIG. 3).
By generating the adjustment signal 100b that lowers p, the transfer function T of the material vibration tester 10, that is, the adaptive filter 2
31 can be operated so that the frequency response characteristic H of the extrapolation signal 100 c to the observer filter 232 becomes equal.

As described above, the drive signal generation unit 21
The principle of generating a drive signal corrected so that a test according to a target signal can be performed will be described based on the simplified diagram of FIG.
As an LMS algorithm for optimizing the transfer function of the material vibration tester 10, a steepest descent is used.
A coefficient updating algorithm based on the escent type) method is used.

Now, assuming that an arbitrary signal x passes through a circuit having a transfer function W and is supplied to the material vibration testing machine 10 as a signal having a function Wx, the output of the material vibration testing machine 10 having a transfer function T has a response TWx. A signal is obtained. This response signal T
Wx is applied to the addition point 214b via the adaptive filter 214a of the transfer function W, and a difference from the signal Wx is obtained. As a result, a signal of e = Wx (1-TW) is obtained at the output of the addition point. This is provided to the LMS algorithm network 214c.

The LMS algorithm network 214c has e
= 0, that is, 1−TW = 0.
The coefficient of 14a is set, so that the adaptive filter 214a has a transfer function W of T ⁻¹ , that is, an inverse function of the transfer function T of the material vibration testing machine 10.

The coefficient of the adaptive filter 213 is set by the LMS algorithm network 214c to be the same as the coefficient of the adaptive filter 214a, and has the same transfer function W as that of the adaptive filter 214a. Therefore, the target signal M generated by the target signal generator 211 is applied to the addition point 212, and a difference from the response signal TWx is obtained.
When this is passed through the adaptive filter 213, W
(M-TWx) is obtained. This is added to the drive signal Wx at the addition point 215, and the output of the addition point
(M-TWx) + Wx is obtained. This is equal to the drive signal Wx, and the formula W (M−TWx) + Wx = Wx holds. When rearranging this equation, W (M-TWx) = 0. In this equation, since W = T ⁻¹ , W (M−M−
x) = 0. Therefore, when M = x, the material vibration tester 10 can perform a vibration test according to the target signal M even when the transfer function T of the material itself changes during operation.

Now, assuming that the weight vector of the LMS algorithm at the time point K is W _K and the input vector is Y _K ,
The MS algorithm is represented by the following equation. During _{W K + 1 = W K +} 2μY K e K formula, e is the error signal, mu is a gain constant to adjust the adaptation speed and stability.

In general, even if the transfer function of the material vibration tester changes during an operation performed by supplying a drive signal having a predetermined repetition cycle, the transfer function is configured to maintain its characteristics in an ideal state. The response characteristic (closed loop characteristic) of the control loop in question may be maintained at an optimum value by PID control. In the present invention, this PID control can be performed in real time.

In the PID control, the response characteristic (frequency characteristic) of the closed loop is maintained at an optimum value.
Unless information in the frequency band is obtained, it is impossible to make a necessary judgment for control. For example, when the proportional element kp is used as a parameter, the change in the frequency characteristic can be maintained at the optimum response characteristic by manipulating Kp if the frequency f0 near the representative root of the input / output is known.

However, it is necessary to find a transfer function having a coefficient including the band of f0 in the coefficient of the transfer function of the material vibration tester. Therefore, an extrapolation signal of the frequency f2 (= f0) is inserted as a disturbance into the material vibration tester, and a response signal including the frequency f2 up to the material vibration tester is obtained. Since the extrapolated signal is originally a disturbance, the test cannot be performed as intended. However, since the drive signal generator 21 corrects the signal including the extrapolated signal component as the disturbance, the test is performed according to the target signal. What you do is not compromised.

In addition, since the coefficients of the adaptive filter obtained under such circumstances include those for the frequency f2 (= f0), the characteristics of the adaptive filter for the frequency f2 (= f0) have ideal characteristics. By comparing with the observer filter, you can know the deviation from the ideal characteristic,
By adjusting Kp, which is one element of the transfer function of the material vibration tester, to correct this deviation, the response characteristics of the material vibration tester to the frequency f2 can be kept optimal.

More specifically, when an adaptive filter whose coefficient is set so as to have a transfer function equal to the transfer function of the material vibration tester is equal to the observer filter, an output is obtained when an extrapolation signal is applied to these filters. Since the signals are equal and the difference is 0, it is possible to know whether the transfer function of the material vibration tester is larger or smaller than the observer filter based on the magnitude and polarity of the output difference. By increasing or decreasing Kp according to the difference, it becomes possible to keep Kp at an ideal value suitable for ideal characteristics.

[0046]

As described above, according to the first and second aspects of the present invention, a response function between input and output in real time is optimized using an adaptive filter. To prevent the optimum value from changing over time and making the closed loop unstable, adjust the gain of the optimum test machine in real time and always keep the transfer function of the material vibration test machine in an ideal state. And a controller for a material vibration tester capable of performing a stable test while holding the control device.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an entire embodiment of a control device of a material vibration testing machine according to the present invention.

FIG. 2 is a block diagram showing a specific form of a control unit in FIG. 1;

FIG. 3 is a graph showing disturbance characteristics of a material vibration tester.

FIG. 4 is a block diagram for explaining a principle of optimizing a response function between input and output of a material vibration tester in real time by using an adaptive filter.

[Explanation of symbols]

 Reference Signs List 10 material vibration tester 12 servo controller 15 electrohydraulic actuator 16 servo valve 20 controller 21 drive signal generator 213 first adaptive filter means (adaptive filter) 214 first coefficient setting means (coefficient setting network) 22 extrapolation signal generator 23 adjustment signal generator 231 second adaptive filter means (adaptive filter) 232 fixed filter means (observer filter) 233 second coefficient setting means (coefficient setting network)

Claims (2)

[Claims] An operation is performed by supplying a drive signal having a predetermined repetition period, a response signal is output in accordance with a transfer function that changes as the drive progresses, and a gain is adjusted by input of an external adjustment signal. A control device for controlling a material vibration testing machine having a closed loop control, wherein the first signal is corrected so that the response signal is equal to a target signal which is a target of operation of the material vibration testing machine. Filter means, based on the drive signal supplied to the material vibration tester, and a response signal obtained at the output of the material vibration tester according to the supplied drive signal,
Setting a coefficient of the first adaptive filter means so as to have a characteristic reflecting a transfer function of the material vibration testing machine;
A drive signal generating unit that supplies the drive signal corrected by the first adaptive filter to the material vibration tester in the next cycle, and a disturbance to the material vibration tester. An extrapolation signal generating unit for generating an extrapolation signal to be supplied; a second adaptive filter means; and a transfer function corresponding to a transfer function of the material vibration tester based on the drive signal and the response signal. A second coefficient setting means for setting a coefficient of the second adaptive filter means, and a coefficient previously set so as to have a transfer function of an ideal material vibration tester exhibiting stable characteristics with respect to the extrapolated signal. Fixed filter means, and the second filter means is provided in accordance with the supply of the extrapolation signal.
A material vibration test, comprising: an adjustment signal generating unit for generating an adjustment signal for adjusting a gain of the material vibration tester based on a difference between signals obtained respectively at the adaptive filter means and the output of the fixed filter means. Machine control device. 2. A servo controller for adjusting a gain of a deviation signal corresponding to a difference between the drive signal and the response signal, wherein the control closed loop of the material vibration testing machine has a gain adjusted by the servo controller. 2. The material according to claim 1, further comprising: an electrohydraulic actuator having a servo valve operated by a signal, wherein the actuator applies vibration according to the target signal to the material to be subjected to vibration test. Control device for vibration testing machine.