JP7240340B2 - Vibration test equipment - Google Patents

Description

TECHNICAL FIELD The present invention relates to a vibration test apparatus for performing a vibration test on a test object.

2. Description of the Related Art In order to evaluate the seismic performance of a structure, there is known a test apparatus that performs a vibration test by vibrating a model simulating the characteristics of the structure (hereinafter also referred to as a "specimen"). Vibration test equipment often uses hydraulic actuators, which are excellent in responsiveness and driving force, because it is necessary to vibrate the test object by reproducing seismic waveforms.

The test pattern handled by the vibration test equipment is divided into low-frequency tests where the matching rate of waveforms in time series is important between the target value and response, and high-frequency tests where the matching rate of the energy given to the test object is important. can be broadly classified. In the low-frequency test, vibrations of several Hz or less, typically represented by seismic waveforms, are applied. On the other hand, in the high-frequency test, vibrations including vibrations of several tens of Hz or more are applied. Therefore, there is a demand for a vibration test apparatus that can cope with vibrations including various frequencies from low frequencies to high frequencies.

A vibration test apparatus performs a vibration test by driving a hydraulic actuator according to a predetermined test pattern (displacement command or acceleration command) defined by a user. Hydraulic actuators are variable elements in which the characteristics of hydraulic pressure, the friction of the driving part, and the like change. For this reason, if the vibration test apparatus is not designed appropriately for control, the response of the hydraulic actuator is delayed, and the test cannot be performed with a desired test pattern. In addition, since vibration test equipment is expected to be able to handle test patterns that include a variety of frequency bands, vibration tests can be performed over as wide a frequency band as possible while ensuring the stability of the control loop. There is a need.

The dynamics of the vibration tester can be described by differential equations, and the displacement response can be obtained by multiple integrations. Therefore, the Bode diagram (frequency characteristic diagram) of the closed-loop transfer function of the control system shows characteristics as shown in FIG.

FIG. 9 is a Bode diagram illustrating the response of a typical vibration tester. In FIG. 9, the upper diagram is a gain diagram and the lower diagram is a phase diagram.

FIG. 10A is an example of the target waveform and response waveform at frequency A shown in FIG. FIG. 10B is an example of the target waveform and response waveform at frequency B shown in FIG. FIG. 10C is an example of the target waveform and response waveform at frequency C shown in FIG. In FIGS. 10A to 10C, solid lines indicate target waveforms (target values), and dashed lines indicate response waveforms (actual responses).

In general, the response of the vibration tester shows a tendency that the higher the frequency, the lower the gain and the more the phase lags. For example, in the case of a sufficiently low frequency command, such as frequency A in FIG. 9, the actual response approximately matches the target value as shown in FIG. 10A. As the frequency increases to frequency B in FIG. 9, the actual response becomes smaller in amplitude and lags in phase than the target value, as shown in FIG. 10B. As the frequency is increased further to frequency C in FIG. 9, the actual response becomes even smaller in amplitude and more lagging in phase, as shown in FIG. 10C. Thus, unless the response of the control loop is properly tuned, the motion of the vibration tester cannot follow the command signal.

PID control, which is one of feedback controls, is widely known as a general control method. In PID control, the control response can be adjusted by adjusting gains corresponding to proportional, integral, and differential elements. However, in PID control, it is not easy to adjust the gain. Further, it is not easy to achieve both the stability of the control loop and the expansion of the excitation frequency band only by feedback control such as PID control.

In order to solve such problems, Patent Document 1 discloses a vibration test apparatus equipped with a stabilization compensator composed of a correction signal calculation unit that corrects a command signal input to a control loop and a phase lead compensator. It is The vibration test apparatus described in Patent Document 1 corrects the command signal based on the difference between the command signal and the actual response in the correction signal calculation unit, and further corrects the excitation frequency by improving the phase delay with the phase lead compensator. It has a configuration that widens the band.

JP 2008-102127 A

In the vibration test apparatus of Patent Literature 1, the correction signal calculation section compensates for the frequency characteristics of the controlled object using inverse Fourier transform, so it cannot be used for frequencies other than the frequency obtained during test excitation. Also, since the stabilizing compensator uses a phase lead compensator, the gain increases in a high frequency band above the phase lead frequency.

FIG. 11A is a Bode plot illustrating the response of a vibration tester utilizing a phase lead compensator. In FIG. 11A, the upper diagram is a gain diagram, the lower diagram is a phase diagram, and frequency D indicates the frequency with which the phase is advanced. Note that FIG. 11A shows the gain and phase shown in FIG. 9 with dashed lines. If the phase lead compensator is not properly designed, the vibration test apparatus equipped with the phase lead compensator exhibits frequency characteristics as indicated by the solid line in FIG. greater than 0 dB.

FIG. 11B is an example of the target waveform and response waveform at frequency D shown in FIG. 11A. In a vibration tester with an improperly designed phase lead compensator, the actual response will be approximately in phase with the target value but greater in amplitude than the target value, as shown in FIG. 11B. A vibration test apparatus with such a configuration may be driven at a high frequency with an amplitude (or velocity or acceleration) larger than the command value (target value). Since the amplitude (displacement), velocity, and acceleration are proportional to the vibration energy applied to the test object under test, it is not desirable to vibrate with an amplitude (or velocity or acceleration) greater than the command value.

From the above, the vibration test apparatus of Patent Document 1 is useful for test patterns mainly composed of low-frequency vibrations of several Hz such as seismic waves, but vibrates the test object with high-frequency vibrations. It can be judged that it is difficult to apply to the test pattern that

SUMMARY OF THE INVENTION An object of the present invention is to provide a vibration test apparatus capable of coping with a test pattern composed of vibrations including frequencies from low to high frequencies.

A vibration test apparatus according to the present invention includes a vibration target generation device that generates a target waveform, a sensor signal acquisition device, a controller that inputs the target waveform, and a test object that vibrates according to a command output from the controller. and a sensor provided in the vibrating section. The controller includes a feedback control calculation unit that calculates a control input based on the difference between the target waveform and a response waveform obtained by the sensor signal obtaining device from the sensor, and a control input that is calculated based only on the target waveform. and a feedforward control calculation unit. The feedforward control calculation unit includes a first compensating unit that operates for frequencies below the frequency ωp that can advance the phase without increasing the gain, and a frequency that provides a gain of 0 dB. and a second compensator that operates for frequencies equal to or higher than the lowest frequency ω2.

According to the present invention, it is possible to provide a vibration test apparatus capable of coping with a test pattern composed of vibrations including frequencies from low frequencies to high frequencies.

1 is a diagram schematically showing a vibration test apparatus according to an embodiment of the present invention; FIG. FIG. 3 is a functional block diagram of a controller in the vibration test apparatus according to Example 1 of the present invention; FIG. 9 is a block diagram showing another configuration example of the FF control calculation unit; FIG. 7 is a functional block diagram of a controller in the vibration test apparatus according to Example 2 of the present invention; FIG. 9 is a functional block diagram of another form of controller in the vibration test apparatus according to Example 2 of the present invention. 10 is a flow chart showing a process of determining parameters of the FF control calculation section in the vibration test apparatus according to Example 2. FIG. FIG. 7 is a diagram showing a flowchart of processing capable of changing the order of the transfer function representing the FF compensator F1(s) in the processing FC09 of the flowchart shown in FIG. 6; FIG. 8 is a functional block diagram of a controller in the vibration test apparatus according to Example 3 of the present invention; It is a Bode diagram explaining the response of a general vibration test apparatus. It is an example of a target waveform and a response waveform at frequency A shown in FIG. It is an example of a target waveform and a response waveform at frequency B shown in FIG. It is an example of a target waveform and a response waveform at frequency C shown in FIG. FIG. 4 is a Bode diagram illustrating the response of a vibration tester using a phase lead compensator; It is an example of a target waveform and a response waveform at frequency D shown in FIG. 11A. 4 is a Bode diagram showing the relationship between the frequency (test pattern) of the vibration test and frequency characteristics; FIG. FIG. 5 is a diagram showing the relationship between the amount of phase lead and the amount of gain increase; It is a figure which shows the example of the target waveform and response waveform in a vibration test. FIG. 10 is a diagram showing another example of target waveforms and response waveforms in a vibration test;

The vibration test apparatus according to the present invention can switch between processing for low-frequency vibration and processing for high-frequency vibration, and vibrates a test object with a test pattern composed of vibrations including frequencies from low to high frequencies. It can handle vibration tests that For example, for a test pattern composed of low frequencies such as seismic waveforms, it is possible to match the waveform in time series between the target waveform and the actual response, and add it to the test object (specimen). Vibration reproducibility can be improved. In addition, for test patterns composed of high frequencies such as strength tests for high-frequency vibration, the energy can be matched between the target waveform and the actual response, improving the reproducibility of the energy given to the test object. can be improved. Therefore, the vibration test apparatus according to the present invention can handle both low frequency tests and high frequency tests.

In the following description, feedforward is represented as "FF" and feedback is represented as "FB".

A vibration test apparatus according to an embodiment of the present invention will be described below with reference to the drawings. In the following embodiments, a single-axis hydraulically driven device will be exemplified for the sake of simplification of explanation, but the vibration testing device according to the present invention is limited to a single-axis hydraulically driven device. not something.

FIG. 1 is a schematic diagram of a vibration test apparatus according to an embodiment of the present invention. The vibration test apparatus is hydraulically driven, and includes a vibration target generator 1 , a controller 2 , and a vibrator 11 . The vibrating unit 11 is a control target of the controller 2, and includes a servo amplifier 3, a servo valve 4, a hydraulic source 5, a hydraulic cylinder 6, a hydraulic piston 7, a coupling 8, and a table 9. Vibrate the test object according to the command.

The vibration target generation device 1 is a device that generates a target vibration waveform (target waveform), and can be configured by a signal generator or the like. The excitation waveform generated by the excitation target generation device 1 may be a displacement command waveform, a velocity command waveform, or an acceleration command waveform. The excitation target generation device 1 can generate excitation waveforms of arbitrary patterns, such as waveforms that reproduce earthquakes that actually occurred in the past and waveforms that are designed to excite specific behaviors.

The controller 2 includes a sensor signal acquisition device 2b and a control command calculation device 2a, inputs the target waveform generated by the vibration target generation device 1, and performs various control calculations for moving the vibration excitation unit 11 according to the target waveform. Execute. The sensor signal acquisition device 2b acquires values (experimental data) detected by sensors S01 to S05b, which will be described later. Based on the excitation waveform (target waveform) generated by the excitation target generation device 1 and the experimental data acquired by the sensor signal acquisition device 2b, the control command calculation device 2a calculates the operation amount of the vibration excitation unit 11 and controls the servo amplifier. Output the command to 3. The controller 2 can be configured by, for example, a computer.

The servo amplifier 3 converts a command (voltage value) output from the controller 2 into a current value for driving the servo valve 4 .

The servo valve 4 adjusts the flow rate of pressure oil flowing from the hydraulic source 5 to the hydraulic cylinder 6 by opening and closing the valve according to the current value received from the servo amplifier 3 . As shown in FIG. 1, the servo valve 4 has two ports and supplies pressure oil to the hydraulic cylinder 6 from one of the ports.

A temperature sensor S01 that detects the temperature of pressure oil is provided between the hydraulic source 5 and the servo valve 4 .

Pressure oil is supplied to the hydraulic cylinder 6 from the hydraulic source 5 via the servo valve 4 . The hydraulic cylinder 6 is divided into two parts. One portion is supplied with pressure oil from one port of the servo valve 4 . The other portion is supplied with pressure oil from the other port of the servo valve 4 .

The hydraulic piston 7 is provided inside the hydraulic cylinder 6 and driven by pressure oil supplied to the hydraulic cylinder 6 . The driving direction of the hydraulic piston 7 is changed depending on which of the two portions of the hydraulic cylinder 6 is supplied with pressure oil.

Between the servo valve 4 and the hydraulic cylinder 6, flow rate sensors S02a and S02b for detecting the flow rate of pressure oil discharged from the servo valve 4 are provided. Since the flow path of the pressure oil changes depending on the driving direction of the hydraulic piston 7, the flow sensors S02a and S02b to be used may be selected according to the driving direction of the hydraulic piston 7. FIG.

The hydraulic piston 7 vibrates the table 9 by applying force to the table 9 via the coupling 8 . The hydraulic piston 7 has at least one of a displacement sensor S04 and a speed sensor S03. When only the displacement sensor S04 is provided, the differential value of the detection value of the displacement sensor S04 can be used as the velocity. When only the speed sensor S03 is provided, the integrated value of the detection value of the speed sensor S03 can be used as the displacement.

The hydraulic cylinder 6 includes pressure sensors S05a and S05b for detecting the hydraulic pressure of the hydraulic piston 7. As shown in FIG.

A structure (specimen) to be tested is placed on the table 9 .

The vibration test apparatus vibrates the test piece based on the excitation waveform (target waveform) generated by the vibration target generation device 1 and the experimental data detected by the sensors S01 to S05b, thereby performing the vibration test on the test piece. to implement.

The vibration test apparatus can also have an operation terminal 10 connected to the controller 2 . The operation terminal 10 can be operated by an operator (worker), can transmit commands input by the operator to the controller 2, and can display the calculation results of the controller 2 to the operator. By inputting commands via the operation terminal 10, the operator can change the settings of the controller 2 and sequentially monitor the operation status of the vibration test apparatus. For the operation terminal 10, for example, a normal personal computer (PC), a tablet computer, a mobile information terminal, or the like can be used. The operation terminal 10 may be connected via a network instead of being physically connected to the controller 2 . Also, the operation terminal 10 may be incorporated in the controller 2 and may have a configuration that makes it unportable.

Hereinafter, the features of the vibration test apparatus according to the present invention will be described in detail with reference to specific examples. In addition, in the following embodiment, in order to simplify the explanation, a configuration in which the controller 2 performs displacement control in which an excitation waveform (target waveform) is given by displacement will be exemplified. The sensor signal acquisition device 2b of the controller 2 acquires the displacement of the hydraulic piston 7 detected by the displacement sensor S04 as a response waveform (experimental data). The controller 2 can also perform velocity control in which the excitation waveform is given as velocity, and acceleration control in which the excitation waveform is given as acceleration. The present invention also includes a vibration test apparatus in which the controller 2 performs these controls.

FIG. 2 is a functional block diagram of the controller 2 in the vibration test apparatus according to Example 1 of the present invention.

The control command arithmetic unit 2a includes a feedback control arithmetic unit (FB control arithmetic unit) C01 and a feedforward control arithmetic unit (FF control arithmetic unit) C02. The output of the control command arithmetic unit 2a is calculated by the sum of the output of the FB control arithmetic unit C01 and the output of the FF control arithmetic unit C02. The controller 2 performs feedback control by the FB control calculation unit C01 and feedforward control by the FF control calculation unit C02, that is, two-degree-of-freedom control.

Based on the difference between the displacement target waveform r generated by the vibration target generation device 1 and the actual displacement y (response waveform) obtained by the sensor signal acquisition device 2b from the displacement sensor S04, the FB control calculation unit C01 Calculate control inputs. The FB control calculation unit C01 can be configured by commonly used PID control or the like.

The FF control calculation unit C02 calculates a control input based only on the displacement target waveform r generated by the vibration target generation device 1 . The FF control calculator C02 includes a first compensator C02a that operates in a low frequency region and a second compensator C02b that operates in a high frequency region. The FF control calculator C02 outputs the sum of the output of the first compensator C02a and the output of the second compensator C02b.

The first compensator C02a includes a low-pass filter (LPF) and an FF compensator F1(s). The second compensator C02b includes a high pass filter (HPF: High Pass Filter) and an FF compensator F2(s). The LPF causes the low frequency band signal to pass through the FF compensator F1(s). The HPF causes the high frequency band signal to pass through the FF compensator F2(s). It is assumed that the frequency band passed by the LPF is lower than the frequency band passed by the HPF. That is, the cutoff frequency of the LPF is assumed to be lower than the cutoff frequency of the HPF.

The FF control calculation unit C02 can switch between the operations of the first compensator C02a and the second compensator C02b according to the frequency of the target displacement waveform r by the LPF and HPF.

The configuration of the FF control calculation unit C02 is not limited to outputting the sum of the output of the first compensation unit C02a and the output of the second compensation unit C02b. For example, the FF control calculation unit C02 can also have a configuration as shown in FIG.

FIG. 3 is a block diagram showing another configuration example of the FF control calculation unit C02. The FF control calculator C02 includes a first compensator C02a, a second compensator C02b, a waveform determiner C02e, and a switch C02d.

The first compensator C02a does not have an LPF, but only an FF compensator F1(s). The second compensator C02b does not have an HPF and only has an FF compensator F2(s). The waveform determination unit C02e changes the output value of the FF control calculation unit C02 by switching the switch C02d based on the frequency component included in the displacement target waveform r. The frequency at which the switch C02d switches can be determined in the same manner as the cutoff frequencies of the LPF and HPF.

By switching the switch C02d, the signal in the low frequency band passes through the FF compensator F1(s) and the signal in the high frequency band passes through the FF compensator F2(s). The FF control calculation unit C02 shown in FIG. 3 uses the waveform determination unit C02e and the switch C02d to perform the first compensation unit C02a according to the frequency of the displacement target waveform r, similarly to the FF control calculation unit C02 shown in FIG. and the operation of the second compensator C02b.

The FF control calculation unit C02 shown in FIG. 3 can obtain the effects of the LPF and the HPF by the waveform determination unit C02e and the switch C02d. . However, the FF control calculation section C02 shown in FIG. 3 may have an LPF and an HPF, like the FF control calculation section C02 shown in FIG.

Returning to FIG. 2, specific design guidelines for the LPF and FF compensator F1(s) of the first compensator C02a and the HPF and FF compensator F2(s) of the second compensator C02b will be described.

As described above, in the low-frequency test, it is preferable that the waveforms in time series match between the target displacement waveform r and the actual displacement y, so that there is no phase delay and the gain neither increases nor decreases. is desirable. On the other hand, in high frequency testing, it is important that the energy match between the displacement target waveform r and the actual displacement y, so the phase may be delayed if the gain neither increases nor decreases.

In designing the first compensator C02a, it is important to improve the phase lag. As a method for improving the phase lag, first-order phase lead compensation of Equation (1) is widely known. In equation (1), s is the Laplace transform operator and α<1.

If the phase lead compensation of equation (1) is designed to advance the phase by φa at the angular frequency ωa, the gain Ka increases at the angular frequency ωa and the gain Kb increases at the angular frequency of 1/(αT) or higher. are given by equations (2) and (3), respectively. Note that T is a parameter that determines the frequency band in which the phase advances, and satisfies 1/T<ωa<1/(αT).

Note that the relationship ω=2πf is established between the frequency f and the angular frequency ω, and unit conversion is required. However, for convenience of explanation, unless a clear distinction is required, frequency and angular frequency will be referred to as "frequency" without distinction.

FIG. 13 shows the phase lead amount φa and the gain increments Ka and Kb when K=1 in equations (2) and (3) (that is, when K=1 in equation (1)). FIG. 4 is a diagram showing relationships; It can be seen from FIG. 13 that the gain increase amount Ka increases as the phase advance amount φa increases. If K is set to a value smaller than 1 in equation (1), the increase in gain can be suppressed, but the gain in the low frequency region is reduced. Therefore, when applying a phase lead compensator to a vibration test apparatus, setting K to a value smaller than 1 is not practical.

The characteristic that the gain increases as the phase advances as described above generally holds, not limited to the phase advance compensation shown in Equation (1).

Note that increasing the gain is unavoidable in order to improve the phase lag. The phase lag φ1 and the gain increase K1 at the highest angular frequency ω1 used in the low-frequency test are given when K=1. (2), it is necessary to satisfy the following relational expression (4).

FIG. 12 is a Bode diagram showing the relationship between the frequency (test pattern) of the vibration test and frequency characteristics. As described above, in the low-frequency test, the waveforms in time series must match between the target displacement waveform r and the actual displacement y (i.e., there must be no phase lag and neither increase nor decrease in gain). ), and for high frequency testing, it is preferable to have energy matching between the displacement target waveform r and the actual displacement y (ie, the gain neither increases nor decreases with phase lag).

In the frequency characteristics of FIG. 12, let ωp be the angular frequency at which the left side and the right side of Equation (4) are equal, that is, the angular frequency at which Equation (5) holds. A phase delay at the angular frequency ωp is represented by φp, and a gain increase amount is represented by Kp.

This angular frequency ωp is set to be equal to or higher than the highest angular frequency ω1 used in the low frequency test and equal to or lower than the lowest angular frequency ω2 used in the high frequency test (ω1≦ωp≦ω2). If the FB control calculation unit C01 is appropriately designed, the angular frequency ωp can be set so as to satisfy ω1≦ωp≦ω2. Hereinafter, the angular frequency ωp may be called "predetermined frequency".

As shown in FIG. 12, when the angular frequency is ωp (ω1≦ωp≦ω2), the phase delay is φp and the gain is −Kp. Therefore, if the phase lead compensation is performed at the predetermined frequency ωp at which equation (5) holds to eliminate the phase delay, the gain increases by Kp to 0 dB and does not exceed 0 dB. That is, if phase lead compensation is performed with an angular frequency of ωp and a gain increase of Kp, the phase can be advanced without increasing the gain to eliminate the phase lag.

When the angular frequency ωp cannot be set so as to satisfy ω1≦ωp≦ω2 (when ωp<ω1 or ωp>ω2), as described in the third embodiment, the feedback performed by the FB control calculation unit C01 is Control parameters (for example, each gain of PID control) are adjusted.

The following description is based on the premise that the FB control calculation unit C01 is appropriately designed and the feedback control is adjusted so that the predetermined frequency ωp satisfies ω1≦ωp≦ω2.

In this embodiment, the operator, through the operation terminal 10, controls the predetermined frequency ωp, the FF compensator F1(s) of the first compensator C02a, and the FF compensator F2(s) of the second compensator C02b. (for example, control parameters as described in the second embodiment) are set.

The first compensator C02a is designed to improve response for low frequency tests. For this reason, it is desirable that the LPF have a cutoff frequency equal to the highest angular frequency ω1 used in the low frequency test.

The second compensator C02b is designed to improve response for high frequency tests. For this reason, it is desirable that the HPF has a cutoff frequency that is the lowest angular frequency ω2 used in high frequency tests. ω2 is a frequency higher than the highest angular frequency ω1 of the low frequency test.

Since the cutoff frequency of the LPF is ω1 and the cutoff frequency of the HPF is ω2 (>ω1), the output of the HPF becomes 0 (zero) during the low frequency test, and the FF compensator F2(s) does not operate. , FF compensator F1(s) only operates. During the high frequency test, the output of the LPF becomes 0 (zero), the FF compensator F1(s) does not operate, and only the FF compensator F2(s) operates. By adopting such a configuration, an appropriate compensator can be operated according to the frequency band of the test pattern.

The operator sets the predetermined frequency ωp via the operation terminal 10 so as to satisfy the relationship ω1≦ωp≦ω2.

The FF compensator F1(s) of the first compensator C02a can be designed to eliminate the phase delay without increasing the gain because ω1 is a frequency equal to or lower than the predetermined frequency ωp. That is, the first compensator C02a is designed to advance the phase without increasing the gain at frequencies below ωp. For example, the FF compensator F1(s) may be designed to perform the same processing as phase lead compensation. To design the FF compensator F1(s), the operation of checking the frequency response after adjusting the parameters may be repeated, or the parameters may be designed using experimental data as described in Example 2. good too.

In the FF compensator F2(s), since ω2 is a frequency equal to or higher than the predetermined frequency ωp, an attempt to eliminate the phase delay entails an increase in gain. However, in the high-frequency test, between the displacement target waveform r and the actual displacement y, priority is given to energy matching over time-series waveform matching. For this reason, in the FF compensator F2(s), as long as the closed loop system is stable, it can be considered that priority is not given to improving the phase delay.

14A and 14B are diagrams showing examples of target waveforms and response waveforms in vibration tests. FIG. 14A shows an example in which the target waveform (displacement target waveform r) and the response waveform (actual displacement y) match in phase but do not match in amplitude (gain). . FIG. 14B shows an example in which the target waveform and the response waveform do not match in phase but match in amplitude. In the high-frequency test, the waveform shown in FIG. 14B is more consistent in energy between the target waveform and the response waveform than the waveform shown in FIG. 14A, and the reproducibility of the energy applied to the specimen is excellent. can be judged.

From the above point of view, the FF compensator F2(s) of the second compensator C02b is designed to have a gain of 0 dB at frequencies equal to or higher than ω2. In order to design the FF compensator F2(s), similar to the design of the FF compensator F1(s), the operation of checking the frequency response after adjusting the parameters may be repeated. Parameter design may be performed using experimental data.

From the above, the first compensator C02a operates when the frequency of the target waveform r is equal to or lower than the predetermined frequency ωp. In other words, the first compensator C02a operates on frequencies below the predetermined frequency ωp at which the phase can be advanced without increasing the gain. The second compensator C02b operates when the frequency of the target waveform r is equal to or higher than the lowest frequency ω2 used in the high frequency test. That is, the second compensator C02b operates with respect to the frequency at which the gain becomes 0 dB and is equal to or greater than ω2.

The vibration test apparatus according to the present embodiment has the above configuration, and can handle vibration tests in which a test object is vibrated with a test pattern composed of vibrations including frequencies from low frequencies to high frequencies. It is possible to cope with both a test pattern composed of very low frequencies and a test pattern composed of high frequencies such as a strength test against high-frequency vibration.

A vibration test apparatus according to a second embodiment of the present invention will be described. Below, the configuration different from that of the vibration test apparatus according to the first embodiment will be mainly described. In the vibration test apparatus according to the first embodiment, the operator sets the predetermined frequency ωp and the control parameters of the FF compensator F1(s) of the first compensator C02a and the FF compensator F2(s) of the second compensator C02b. . In the vibration test apparatus according to this embodiment, the controller 2 obtains the predetermined frequency ωp based on the obtained experimental data, and sets the control parameters of the FF compensator F1(s) and the FF compensator F2(s).

As shown in FIG. 1, the vibration test apparatus includes sensors S01-S05b, and signals detected by these sensors S01-S05b are stored in the controller 2. FIG.

FIG. 4 is a functional block diagram of the controller 2 in the vibration test apparatus according to this embodiment. The controller 2 of this embodiment has a configuration in which an operation switching unit C02c and a control response adjustment unit C03 are added to the controller 2 (FIG. 2) of the first embodiment.

The operation switching unit C02c is provided in the FF control calculation unit C02 and has an LPF and an HPF. In the controller 2 according to the first embodiment, the LPF and HPF are provided in the first compensator C02a and the second compensator C02b, respectively. In the controller 2 of this embodiment, the LPF and HPF are not provided in the first compensation section C02a and the second compensation section C02b, but are provided in the operation switching section C02c independently of them. The operation switching unit C02c can switch the operations of the first compensating unit C02a and the second compensating unit C02b according to the frequency of the target waveform r by the LPF and HPF.

The control response adjustment unit C03 is provided in the control command calculation device 2a, includes a frequency response analysis unit C03a, a first parameter design unit C03c, a second parameter design unit C03d, and a third parameter design unit C03b, and an FF control calculation unit Adjust the control parameters of C02.

The frequency response analysis unit C03a acquires the detection values (response waveforms) of the sensors S01 to S05b obtained in the experiment using the vibration test device from the sensor signal acquisition device 2b, and uses the detection values to determine the frequency response of the test piece. Then, the relationship between the gain and the phase with respect to the frequency (frequency characteristic as shown in FIG. 9) is obtained, and the predetermined frequency ωp (the angular frequency at which equation (5) holds) is calculated. As in this embodiment, when the vibration test apparatus is driven under displacement control, the frequency response analysis unit C03a analyzes the displacement data of the vibration test apparatus (actual displacement y acquired by the displacement sensor S04). to calculate the predetermined frequency ωp.

The third parameter design unit C03b uses the predetermined frequency ωp calculated by the frequency response analysis unit C03a to determine cutoff frequencies of the LPF and HPF of the operation switching unit C02c. Specifically, the cutoff frequency of the LPF is set to a frequency lower than ωp, and the cutoff frequency of the HPF is set to a frequency higher than ωp. The parameters (cutoff frequencies) of the LPF and HPF determined by the third parameter design section C03b are reflected in the LPF and HPF of the operation switching section C02c.

As described in Example 1, when the cutoff frequency of the LPF is set to the highest angular frequency ω1 used in the low frequency test, and the cutoff frequency of the HPF is set to the lowest angular frequency ω2 used in the high frequency test, The cutoff frequencies of the LPF and HPF are known in advance. In such a case, the control response adjustment section C03 does not have to include the frequency response analysis section C03a and the third parameter design section C03b as shown in FIG.

FIG. 5 is a functional block diagram of another form of controller 2 in the vibration test apparatus according to the present embodiment. The controller 2 shown in FIG. 5 has a configuration that does not include the frequency response analysis unit C03a and the third parameter design unit C03b in the controller 2 shown in FIG. In the configuration shown in FIG. 5, the control response adjuster C03 determines the predetermined frequency ωp so that ω1≦ωp≦ω2.

Returning to FIG. 4, the description of the control response adjustment unit C03 is continued.

The first parameter design unit C03c acquires the detection values (response waveforms) of the sensors S01 to S05b obtained in the experiment using the vibration test device from the sensor signal acquisition device 2b, and uses the detection values to obtain the first compensation unit C02a. adjust the control parameters for the FF compensator F1(s) of . Specifically, when the FF compensator F1(s) is represented by the transfer function of the second-order delay system of Equation (6), the first parameter design unit C03c calculates the control parameters ρ0 to ρ4 to adjust.

Existing data driven control can be used for adjusting these parameters ρ0 to ρ4. Data-driven control is a method of directly designing control parameters that minimize the value of an arbitrary evaluation function based on experimental data, and is described, for example, in the following documents.
Kaneko, Nakamura, Ikezaki: A New Approach to Updating Feedforward Controllers in Two Degrees of Freedom Control Systems, Transactions of the Society of Instrument and Control Engineers, Vol. 54, No. 12, pp. 857-874 (2018).

The FF compensator F1(s) of the first compensator C02a is designed so that the waveforms in time series match between the target displacement waveform r and the actual displacement y (response y). Therefore, it is desirable that the first parameter design unit C03c calculate the parameters ρ0 to ρ4 so as to minimize the value of the evaluation function J1(ρ) represented by Equation (7). Equation (7) is an evaluation function generally used in data driven control.

In equation (7), the subscript i is the index of the experimental data, i=1 means the initial time, and i=N means the end time. Also, y(ρ) is the response obtained when the control parameters ρ0 to ρ4 are changed. However, ρ means a vector, and ρ=[ρ0, ρ1, ρ2, ρ3, ρ4] in Equation (7). Also in the following description, ρ is defined as a vector.

Since the evaluation function J1(ρ) in Equation (7) is represented by the cumulative value of the square of the difference between the value of the target waveform r and the value of the response y(ρ) at each time, if there is a phase delay in the response y value increases. Therefore, when the value of the evaluation function J1(ρ) is minimized, the values of the control parameters ρ0 to ρ4 are obtained so as to improve the phase lag. Also, when the target waveform r and the response y are in phase and the gain (amplitude) is inconsistent, the value of the evaluation function J1(ρ) increases. Therefore, by minimizing the value of the evaluation function J1(ρ), not only the phase but also the gain can be matched. This is a characteristic that can be realized because the frequency is in the low frequency region below the predetermined frequency ωp.

The control parameter ρ obtained by the first parameter design section C03c is reflected in the FF compensator F1(s) of the first compensator C02a.

The second parameter design unit C03d acquires the detection values (response waveforms) of the sensors S01 to S05b obtained in the experiment using the vibration test device from the sensor signal acquisition device 2b, and uses the detection values to obtain the second compensation unit C02b. adjust the control parameters for the FF compensator F2(s) of . Specifically, when the FF compensator F2(s) is expressed by the transfer function of the second-order delay system of Equation (8), the second parameter design unit C03d calculates the control parameters θ0 to θ4 to adjust.

These parameters θ0 to θ4 can be adjusted by minimizing the value of the evaluation function using existing data-driven control, like the parameters ρ0 to ρ4. However, as described above, the FF compensator F2(s) is a compensator for high-frequency tests, and does not improve the phase delay, but rather the energy difference between the displacement target waveform r and the actual displacement y (response y). used to match Therefore, it is desirable to use an evaluation function different from the formula (7) for adjusting the parameters θ0 to θ4, for example, it is desirable to use the evaluation function J2(θ) expressed by the formula (9).

In equation (9), y(θ) is the response obtained when the control parameters θ0-θ4 are changed. However, θ means a vector, and θ=[θ0, θ1, θ2, θ3, θ4] in equation (9). Also in the following description, θ is defined as a vector.

The evaluation function J2(θ) of Equation (9) is expressed by the difference between the sum of squares of the values of the target waveform r and the sum of squares of the values of the response y(θ) at each time. Therefore, even if the phases do not match between the target waveform r of the displacement and the actual displacement y (response y), as long as the gain (amplitude) matches, the value of the evaluation function J2(θ) is become smaller. Therefore, the second parameter design unit C03d minimizes the value of the evaluation function J2(θ) so that the energy between the target displacement waveform r and the actual displacement y (response y) matches. Values of the parameters θ0 to θ4 can be calculated.

The control parameter θ obtained by the second parameter design section C03d is reflected in the FF compensator F2(s) of the second compensator C02b.

The experimental data used by the frequency response analysis unit C03a, the third parameter design unit C03b, the first parameter design unit C03c, and the second parameter design unit C03d may all be the same. However, considering the characteristics of the analysis unit and the design unit, it is better to prepare experimental data individually for each of the frequencies handled by each, so that desirable control characteristics can be obtained.

FIG. 6 is a flow chart showing the process of determining the parameters of the FF control calculation section C02 in the vibration test apparatus according to this embodiment. Processing of the frequency response analysis unit C03a, the third parameter design unit C03b, the first parameter design unit C03c, and the second parameter design unit C03d included in the control response adjustment unit C03 will be described with reference to FIG.

In FC01, it is selected whether or not the controller 2 has the frequency characteristic data of the test piece. If the vibration test apparatus has already performed a vibration test on the test piece and the controller 2 has performed frequency analysis of the test piece and has frequency characteristic data, the process transitions to FC06. On the other hand, when the controller 2 does not have the frequency characteristic data of the test piece because the vibration test apparatus is newly installed or the test piece is changed, the process transitions to FC02. Also, when the frequency characteristics of the test piece are to be acquired again for periodic maintenance or the like, the transition is made to FC02. This selection is preferably made by the operator using a GUI (Graphical User Interface) on the operation terminal 10 .

In FC02, in order to acquire the frequency characteristics of the test piece, a vibration test is performed on the test piece with a wide frequency test pattern including a desired frequency range. A chirp waveform or a random waveform is desirable for the excitation waveform in the test pattern. The operator operates the controller 2 using the GUI of the operation terminal 10 to vibrate the specimen and perform the vibration test. Vibration may be automatically stopped in accordance with the end time of the test pattern, or may be stopped by an operator's instruction through the operation terminal 10 .

In FC03, the control command calculation device 2a acquires experimental data obtained by the sensor signal acquisition device 2b from the start to the end of the vibration test in FC02. If noise is included in the acquired experimental data, the control command calculation device 2a may perform noise removal processing on the acquired data.

In FC04, the frequency response analysis unit C03a analyzes the data acquired by the control command calculation device 2a in FC03, analyzes the frequency response, and determines the relationship between the gain and phase with respect to frequency (frequency characteristics as shown in FIG. 9). Then, the predetermined frequency ωp is calculated. In this process, the operator does not operate.

In FC05, the third parameter designing unit C03b sets parameters relating to the LPF and HPF of the operation switching unit C02c according to the frequency analysis result in FC04. Specifically, the third parameter design unit C03b uses the predetermined frequency ωp obtained in FC04 to determine the cutoff frequencies of the LPF and HPF. In this process, the operator does not operate.

In FC06, it is selected whether or not it is necessary to design the FF compensator F1(s) that operates during the low frequency test. If the test apparatus executes only the existing test pattern and the FF compensator F1(s) has already been obtained, the FF compensator F1(s) does not need to be designed, so the process transitions to FC10. If the test apparatus is new and the FF compensator F1(s) has not been designed, the FF compensator F1(s) must be designed, so the process transitions to FC07. Further, even if the FF compensator F1(s) has already been designed, when driving the test apparatus with a new test pattern, redesigning the FF compensator F1(s) can be expected to improve the control performance. Transition to FC07. It is desirable for the operator to select whether or not the design of the FF compensator F1(s) is necessary using the GUI on the operation terminal 10. FIG.

In FC07, the specimen is subjected to a vibration test in a low frequency (ie, low frequency) test pattern. This vibration test is desirably performed with a test pattern (for example, a test pattern that reproduces seismic waveforms) in a low-frequency test that is actually performed. However, if there is a risk of damage to the test piece, the vibration test may be performed by lowering the gain of the test pattern to the extent that the test piece is not damaged. Using the GUI of the operation terminal 10, the operator selects the test pattern for executing the vibration test and the signal level (gain) of the test pattern, and executes the vibration test in the same manner as in the FC02.

In FC08, the control command arithmetic unit 2a acquires experimental data in the same manner as in FC03.

In FC09, the first parameter designing unit C03c uses the experimental data obtained in FC08 to adjust control parameters for the FF compensator F1(s) of the first compensating unit C02a. For example, the first parameter design unit C03c obtains the parameter ρ in Equation (6) so as to minimize the value of the evaluation function J1(ρ) represented by Equation (7). In this processing, the operator does not need to operate, but the operator may operate when changing the order of the transfer function representing the FF compensator F1(s) as described later.

In FC10, it is selected whether or not it is necessary to design the FF compensator F2(s) that operates during the high frequency test. If the test apparatus executes only the existing test pattern and the FF compensator F2(s) has already been obtained, the FF compensator F2(s) does not need to be designed, so the process transitions to FC14. If the test apparatus is new and the FF compensator F2(s) has not been designed yet, the FF compensator F2(s) must be designed, so the process transitions to FC11. Further, even if the FF compensator F2(s) has already been designed, when the test apparatus is driven with a new test pattern, redesigning the FF compensator F2(s) can be expected to improve the control performance. Transition to FC11. It is desirable for the operator to select whether or not the design of the FF compensator F2(s) is necessary using the GUI on the operation terminal 10. FIG.

In FC11, the specimen is subjected to a vibration test in a high frequency (ie, high frequency) test pattern. This vibration test is desirably performed with a test pattern for a high-frequency test that is actually performed (for example, a test pattern for a limit excitation test). However, if there is a risk of damage to the test piece, the vibration test may be performed by lowering the gain of the test pattern to the extent that the test piece is not damaged. Using the GUI of the operation terminal 10, the operator selects the test pattern for executing the vibration test and the signal level (gain) of the test pattern, and executes the vibration test in the same manner as in the FC02.

In FC12, the control command calculation unit 2a acquires experimental data in the same manner as in FC03.

In FC13, the second parameter design unit C03d uses the experimental data acquired in FC12 to adjust the control parameters related to the FF compensator F2(s) of the second compensator C02b. For example, the second parameter design unit C03d obtains the parameter θ in equation (8) so as to minimize the value of the evaluation function J2(θ) represented by equation (9). In this processing, the operator does not need to operate, but the operator may operate when changing the order of the transfer function representing the FF compensator F2(s) as described later.

In FC14, the controller 2 reflects the obtained parameters (LPF and HPF cutoff frequencies, ρ, and θ) in the FF control calculation unit C02.

In the flowchart shown in FIG. 6, the processing FC06 to FC09 for designing the FF compensator F1(s) that operates during the low frequency test and the processing FC10 to FC09 for designing the FF compensator F2(s) that operates during the high frequency test The FC13 may be performed in the reverse order.

In the above description, the FF compensator F1(s) of the first compensator C02a and the FF compensator F2(s) of the second compensator C02b have the second-order lags shown in equations (6) and (8), respectively. It was assumed to be expressed by the transfer function of the system. Naturally, the FF compensator F1(s) and the FF compensator F2(s) are not limited to those represented by second-order lag systems. As the order of the FF compensator increases, the possibility of reducing the value of the evaluation function represented by Equations (7) and (9) increases. However, if the order of the FF compensator is large, the time for calculating the control parameters ρ and θ increases, so it is not desirable to have the FF compensator with a larger order than necessary.

Therefore, the first parameter design unit C03c that designs the FF compensator F1(s) and the second parameter design unit C03d that designs the FF compensator F2(s) obtain the control parameter ρ and the control parameter θ, respectively. It is desirable to be able to change the order of the transfer function to

An example of a configuration in which the first parameter design unit C03c can change the order of the transfer function representing the FF compensator F1(s) of the first compensator C02a will be described below. The second parameter design section C03d can also change the order of the transfer function representing the FF compensator F2(s) of the second compensator C02b by adopting the same configuration as the first parameter design section C03c.

FIG. 7 shows the transfer function representing the FF compensator F1(s) in the process FC09 (the process in which the first parameter design unit C03c adjusts the control parameters of the FF compensator F1(s)) in the flowchart shown in FIG. FIG. 10 is a diagram showing a flowchart of processing that can change the order;

In FC09a, the first parameter design unit C03c performs optimization calculation to minimize the value of the evaluation function J1(ρ), and adjusts the control parameter ρ related to the FF compensator F1(s) with the default order. The order in the initial setting may be a second-order lag as shown in equation (6) or a first-order lag.

In FC09b, the controller 2 displays on the operation terminal 10 the result of the control parameter ρ obtained in FC09a. For example, the controller 2 uses the value of the evaluation function J1(ρ) obtained with the control parameter ρ obtained in FC09a and the value of the evaluation function J1(ρ) obtained in the predictive simulation of the vibration test executed using the control parameter ρ obtained in FC09a. The response received is displayed on the GUI of the operation terminal 10 .

In FC09c, the operator looks at the results displayed on the GUI of the operation terminal 10 and determines whether or not the FF compensator F1(s) has achieved the required performance. For example, when the value of the evaluation function J1(ρ) is sufficiently close to 0 (or the value of the evaluation function J1(ρ) is less than the threshold value), the operator selects the FF compensator F1(s) has achieved the required performance. If the operator determines that the required performance has been achieved, the processing of FC09 is terminated. If the operator determines that the required performance has not been achieved, the process transitions to FC09d.

In FC09d, the first parameter design unit C03c increases the order of the FF compensator F1(s). For example, the operator operates the operation terminal 10 and selects a transfer function having a required order from the transfer functions stored in advance in the controller 2, so that the first parameter design unit C03c can perform the FF compensator The order of F1(s) can be increased. For example, if the order of the FF compensator F1(s) is 2 when FC09c determines that the required performance is not achieved, the first parameter design unit C03c transfers the FF compensator F1(s) The function is expressed by a transfer function of a third-order delay system as in Equation (10), and the parameters ρ0 to ρ6 in Equation (10) are adjusted.

The number of control parameters for the transfer function of the FF compensator F1(s) is five, ρ0 to ρ4, if the FF compensator F1(s) is represented by the transfer function of the second-order delay system of Equation (6). However, if the FF compensator F1(s) is represented by the transfer function of the third-order delay system of Equation (10), there are seven ρ0 to ρ6. For this reason, the FF compensator F1(s) increases the optimization calculation time (calculation to minimize the value of the evaluation function J1(ρ)) due to an increase in the number of parameters, but can realize a response with a high degree of freedom. .

In FC09e, the first parameter design unit C03c performs optimization calculations to minimize the value of the evaluation function J1(ρ) in the same manner as in FC09a, Adjust the control parameter ρ. As described above, FC09e requires more computation time than FC09a. After the control parameter ρ is obtained in FC09e, the process returns to FC09b.

In FC09b, the controller 2 performs predictive simulation of the vibration test executed using the value of the evaluation function J1(ρ) obtained with the control parameter ρ and the control parameter ρ before and after the order is increased. is displayed on the GUI of the operation terminal 10 . The operator compares the results before and after increasing the order displayed on the GUI of the operation terminal 10 to determine whether the FF compensator F1(s) has achieved the required performance.

When the processing from FC09b to FC09e is repeated a plurality of times, even if the degree is increased, the results displayed on the GUI of the operation terminal 10 before and after increasing the degree almost match. If such a result is obtained, it is unlikely that the control performance of the FF compensator F1(s) will be improved even if the order is further increased. Therefore, it is desirable that the controller 2 has a function of notifying the operator that the results match when the results before and after increasing the order match.

In the vibration test apparatuses according to the first and second embodiments, the FB control calculation unit C01 is appropriately designed, and it is assumed that the predetermined frequency ωp satisfies ω1≦ωp≦ω2. If the operator is familiar with control adjustment, such FB control can be designed easily. However, if the operator is an unskilled person, it is expected that many man-hours and time will be required to adjust the FB control.

In order to solve this problem, the vibration test apparatus according to the third embodiment of the present invention has a configuration in which the FB control calculation section C01 is designed so that the predetermined frequency ωp satisfies ω1≦ωp≦ω2.

FIG. 8 is a functional block diagram of the controller 2 in the vibration test apparatus according to this embodiment. The controller 2 of the present embodiment has a configuration in which a fourth parameter design section C03e is added to the control response adjustment section C03 in the controller 2 (FIG. 4) of the second embodiment.

The fourth parameter design unit C03e acquires the detection values (response waveforms) of the sensors S01 to S05b obtained in the experiment using the vibration test device from the sensor signal acquisition device 2b, and uses the detection values to obtain the FB control calculation unit C01. control parameters (for example, each gain of PID control). The FB control calculation unit C01 performs feedback control using this control parameter.

The processing of the fourth parameter design unit C03e will be briefly described below.

The fourth parameter designing unit C03e executes optimal control design using existing data-driven control, like the first parameter designing unit C03c and the second parameter designing unit C03d. Specifically, the fourth parameter design unit C03e adjusts the parameters of the PID control of the FB control calculation unit C01 by minimizing the value of the evaluation function, and obtains the predetermined frequency ωp that satisfies ω1≦ωp≦ω2. .

When the FB control calculation unit C01 is configured by PID control as in Equation (11), the fourth parameter design unit C03e sets each of the proportional gain Kp, the integral gain Ki, and the differential gain Kd by data-driven control. Use and design. η in equation (11) is a parameter relating to the differential time.

The fourth parameter design unit C03e designs the transfer function Gdes(s) so that the ideal response yd used for data-driven control has desired frequency characteristics, and the evaluation function J3(K) expressed by Equation (12): Each gain is calculated so as to minimize

In equation (12), K means a vector of gains, and K=[Kp, Ki, Kd]. Practically, when the transfer function Gdes(s) of the ideal response is designed as in Equation (14), the predetermined frequency ωp satisfies ω1≦ωp≦ω2 in many cases.

The fourth parameter design unit C03e designs the FB control calculation unit C01 so that the predetermined frequency ωp satisfies ω1≦ωp≦ω2. It is desirable to use data obtained from vibration experiments at In this way, there is no need to acquire new experimental data necessary for designing the FB control calculation unit C01, so there is an advantage that the burden on the operator does not increase.

The fourth parameter design unit C03e adjusts the FB so that the Bode diagram of a closed loop system that performs only feedback control satisfies the characteristics shown in FIG. 12 (for example, the predetermined frequency ωp satisfies ω1≦ωp≦ω2). Adjust the control parameters of the control calculation unit C01. Therefore, the processing of the fourth parameter designing unit C03e needs to be executed prior to the processing of the first parameter designing unit C03c and the second parameter designing unit C03d that adjust the parameters of the feedforward control.

It should be noted that the present invention is not limited to the above embodiments, and various modifications are possible. For example, the above embodiments have been described in detail in order to facilitate understanding of the present invention, and the present invention is not necessarily limited to aspects having all the described configurations. Also, part of the configuration of one embodiment can be replaced with the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. Moreover, it is possible to delete a part of the configuration of each embodiment, or to add or replace another configuration.

DESCRIPTION OF SYMBOLS 1... Vibration target generation apparatus, 2... Controller, 2a... Control command calculation apparatus, 2b... Sensor signal acquisition apparatus, 3... Servo amplifier, 4... Servo valve, 5... Hydraulic source, 6... Hydraulic cylinder, 7... Hydraulic piston , 8... coupling, 9... table, 10... operation terminal, 11... vibration unit, C01... FB control calculation unit, C02... FF control calculation unit, C02a... first compensation unit, C02b... second compensation unit, C02c ... operation switching section, C02d ... switch, C02e ... waveform determination section, C03 ... control response adjustment section, C03a ... frequency response analysis section, C03b ... third parameter design section, C03c ... first parameter design section, C03d ... second parameter Design part, C03e... Fourth parameter design part, S01... Temperature sensor, S02a, S02b... Flow rate sensor, S03... Speed sensor, S04... Displacement sensor, S05a, S05b... Pressure sensor.

Claims (4)

a vibration target generation device that generates a target waveform;
a controller that includes a sensor signal acquisition device and inputs the target waveform;
a vibrating unit that vibrates the test object according to a command output from the controller;
a sensor provided in the vibrating unit;
with
The controller is
a feedback control calculation unit that calculates a control input based on a difference between the target waveform and a response waveform obtained from the sensor by the sensor signal obtaining device;
a feedforward control calculation unit that calculates a control input based only on the target waveform;
with
The feedforward control calculation unit is
a first compensating unit that operates for a frequency equal to or lower than the frequency ωp that can advance the phase without increasing the gain;
a second compensator that operates at a frequency at which the gain is 0 dB and that is equal to or higher than the lowest frequency ω2 used in the high frequency test;
comprising a
A vibration test apparatus characterized by: The controller includes a control response adjustment unit that adjusts control parameters of the feedforward control calculation unit,
The feedforward control calculation unit includes an operation switching unit that switches operations of the first compensation unit and the second compensation unit according to the frequency of the target waveform,
The control response adjustment unit is
a frequency response analysis unit that analyzes the frequency response of the test object using the response waveform and calculates the frequency ωp;
a first parameter design unit that adjusts a control parameter of the first compensator using the response waveform;
a second parameter design unit that adjusts a control parameter of the second compensator using the response waveform;
a third parameter design unit that determines, using the frequency ωp, a frequency at which the operation switching unit switches the operations of the first compensator and the second compensator;
comprising a
The vibration test apparatus according to claim 1. The control response adjustment unit includes a fourth parameter design unit that adjusts control parameters of the feedback control calculation unit using the response waveform,
The fourth parameter design unit obtains control parameters for the feedback control calculation unit such that the frequency ωp satisfies ω1≦ωp≦ω2, where ω1 is the highest frequency used in the low-frequency test,
The processing of the fourth parameter design unit is executed prior to the processing of the first parameter design unit and the second parameter design unit,
The vibration test apparatus according to claim 2. An operation terminal connected to the controller,
The operation terminal is capable of transmitting commands input by an operator to the controller, and displaying calculation results of the controller,
The controller can change the order of the transfer function representing the first compensator and the order of the transfer function representing the second compensator by the operation of the operation terminal by the operator.
The vibration test apparatus according to claim 1.

Images