JPS63208913A - Vibration control device - Google Patents

Abstract

PURPOSE:To reduce both soft and hard loads by detecting the movement of a sample to convert it into a frequency area signal and comparing this area signal with a reference spectrum for correction of the frequency characteristics. CONSTITUTION:A sample 3 is attached to a movable table 2 of a vibration generator 1 and receives the necessary vibrations and impacts. An acceleration sensor 4 is attached to the table 2 and the acceleration of the sensor 4 is supplied to a vibration control device 16 via charge amplifier 5. The device 16 includes an input signal processing part 7, a PSD (power spectral density) converting part 8, a reference PSD pattern data storing part 9, a PSD comparing part 10, an output frequency characteristic memory part 11, and an adverse FFT processing part 12 and then synthesizes the random signals equivalent to a single frame to output them to a power amplifying part 15. Then, a random number generating part 17 which produces the white noises is provided together with a digital filter 18. The frequency area signal of the movement of the detected sample 3 is compared with a reference spectrum for correction of the frequency characteristics.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a control device for a vibration/shock testing device.

[Conventional technology]

Generally, vibration/shock test equipment is provided to test the reliability and durability against vibrations/shocks of various parts and devices subject to vibration, such as aircraft parts and automobile parts.

The control device of the conventional vibration/shock test equipment is
There is one described in Japanese Patent No.-18873.

This will be explained based on FIG.

A sample 3 is mounted on a table 2 of a vibration generator 1, and the required vibration and shock are applied thereto. An acceleration sensor 4 is attached to the table 2, and an electric signal proportional to the acceleration from the sensor 4 is amplified by a charge amplifier f5 and input to the vibration control device 6. In the vibration control device 6, first, the input signal is filtered in the input signal processing section 2, and after analog-to-digital conversion is performed, the data is stored as a block. Then, a power spectral density (PSD) comparator 8 performs fast 7-lier transform (FFT) processing on the data to obtain PSD.

Reference PSD 14 The reference PSD pattern stored in the data storage section 9 and the above PSD are compared by the 'i PSD comparator 10. This comparison result is temporarily stored in the output frequency characteristic storage section 11, and is stored in the inverse FFT processing section 12.
Inverse FFT processing is performed to synthesize random signals for one frame. Furthermore, a randomization processing section 13 performs randomization processing on this random signal to obtain a connected (non-periodic) random signal. Then, this random signal is sent to the power amplification unit 15 via the output processing unit 14.
This is what is output to.

[Problem that the invention seeks to solve]

All of the above conventional configurations are realized by digital computer functions, but the processes from FFT processing, PSD comparison, inverse FFT processing to randomization processing require extremely high speed and complex processing, and are completely pure random. Obtaining a signal is difficult in principle. Furthermore, this configuration is limited to control of random wave vibrations and cannot be applied as a control device for sine wave vibrations or the like.

The present invention aims to solve these problems.

[Means for solving problems]

The present invention includes a detection section that detects the movement of a sample attached to a vibrating section, a conversion section that converts a detection signal into a frequency domain signal, a data storage section that stores reference spectrum data, and a combination of the frequency domain signal and the reference spectrum data. a comparison section that compares the spectra, a signal generation section that generates an arbitrary continuous signal, an equalization section that corrects the frequency characteristics of this continuous signal, and an equalization section correction section that corrects the frequency characteristics of the equalization section. , and output means for outputting the output of the equalization section to the vibration drive section.

[Effect]

The detection section detects the movement of the sample, and the conversion section converts a detection signal representing the movement into a frequency domain signal. This frequency domain signal and the reference spectrum are compared by the full comparison section, and the frequency characteristic of the υ equalization section is corrected by the equalization section correction section based on the comparison result.

Any continuous signal generated by the signal generating section passes through this equalizing section, is corrected so that the movement of the sample approximates the movement based on the reference spectrum, and is outputted to the vibration driving section by the output means.

〔Example〕

A vibration control device according to an embodiment of the present invention will be explained based on FIG.

In the figure, 1 is an electrodynamic vibration generator, 2 is a movable table, 3 is a sample attached to table 2, 4 is an acceleration sensor that detects the acceleration of table 2 and outputs an acceleration signal, and 5 is an acceleration sensor. This is a charge amplifier that converts acceleration signals into voltage signals.

1b is a vibration control device. 7 is an input signal processing section. In this input signal processing section 7, the above voltage signal is
After passing through a signal filter, sampling is performed, and the analog signal is converted into a digital signal by an N-point converter, and the signal is stored in memory as one block of digital data (N points).As is well-known, low-pass filtering is This is to remove undesirable frequency components generated by sampling.The digital data for one block is processed by the well-known fast Fourier transform (Fast Fourier transform).
FFT)
The algorithm is used to calculate the now f worth vector density (Pow
er5pectralDensity: PSD
) is converted to

8 is a PSD conversion unit, which stores the above-mentioned one block worth of sampling data f(K) (K=0, 1, . . . , N
−1), the discrete Fourier transform F(J) (J = 0
, 1. ....

N-1), and further calculate PSD P(Jl, F(J
) is calculated as the square of the absolute value of Expressing these as a formula, P(J)=IF(J)12(J=0.1...,N/
2) (2) In equation (1), j2=-1
(imaginary unit).

As is well known, if the PSD of a random signal is measured only once, the value will vary each time it is measured, and in order to obtain stable data, the data measured several times must be smoothed by appropriate processing. The most common data processing method is to average the data and use it as an estimate of the true PSD.

Si (Jl = P(Jl
(3) Here, 5i (J) is the PSD found in the first control loop (this is called the control PSD), and - is the estimated value obtained by the smoothing operation. Reference PSD/# turn data storage unit, reference PSD
SD pattern P, (Jl (J = 0, 1...N/
2) is stored in the memory.

10 is a PSD comparison section, which compares the control PSD with the reference PSDP.
, [Jl is compared using equation (4), and the frequency characteristic required for the output random signal H1+ 1(') (
J = O* 1 m...N/2).

Reference PSD pattern Pr(J) (J=0, 1,
・-, N/2) are stored in memory in advance.

11 is an output frequency characteristic storage unit that temporarily stores comparison results;
12 is an inverse FF'T processing section that performs inverse FFT processing on the comparison results, and 17 is a random number generation section that generates a white line sound signal.

18 is a digital filter for realizing a transmission system having frequency characteristics H1+1(J). This will be explained below based on FIG. FIG. 3 is a model of a general signal transmission system. 5l(t) is the input signal, 5o(
tl is the output signal. ai(tl, 50(tl is a so-called time-series signal, with dimension
on) does not matter. 19 is a linear signal transmission system,
The transfer function is assumed to be H(ω) (ω=2πf, f is frequency). Under the above assumption, the relationship between input and output signals is determined by concalinion integral. Here, the function h(t) is an inverse Fourier transform of the transfer function H(-, and is called an impulse response function.

In equations (5) and (6), 51(t) and go(t) are both defined on the continuous time axis t, but in actual applications, they are defined on the discrete time axis n (n==Δt
-, where Δt is the sampling period, and Δt is the sampling number), and the results are obtained. Here, N is the total number of sampling data for one block.

As is clear from the above, by multiplying the input signal B'', (rn) by the impulse time series h(m) and performing a discrete condelusion operation according to equation (7),
It is possible to approximately realize a transfer system with a frequency characteristic of It can be easily determined by

The device that realizes equation (7) is an N-order FIR (Finite
This is well known as an ImpulseRssponse) type digital filter. The larger N is, the H(Jl
Although the degree of approximation between and H(ω) becomes better, the number of calculations and memory capacity increase, which limits the sampling time that can be realized with the same hardware. This digital filter can be realized using combiator software, but it can also be designed as an independent device by applying an inexpensive micro combiator. In this embodiment, the maximum value is N=1024, but it goes without saying that the degree of approximation can be improved if a larger N is applied as hardware progresses in the future.

Returning to FIG. 1 again, the operation of equation (8) above is performed in the inverse FFT processing section 12, and the digital filter 18 converts the self-sound signal from the random number generation section Z7 into t= i (m) of equation (7). As, the discrete con? of equation (7)? execute the revision. Output of digital filter 18, output signal processing section 20
After conditioning and clipping the output signal, it is converted into an analog voltage signal and sent to the input of the power amplifier 5. The power amplifier 15 amplifies the power of the input voltage signal and drives the vibration generator 1.
Random vibrations are generated on the movable table 2 to complete one round of the runo.

〔Effect of the invention〕

According to the present invention, there are the following effects.

(1) It is now possible to eliminate the conventional highly technical operation of damaging/damaging the output signal, reducing the burden on both the software and hardware of the host computer system, making it possible to realize a much cheaper system. It became.

(2) By introducing a method of equalizing the transmission system, it is possible to control sine wave vibration, shock wave vibration control, seismic waves, etc.
Expansion to a waveform reproduction system for real vibrations is now possible using exactly the same hardware.

(3) Furthermore, if the present invention is used for random vibration control, it becomes possible to excite random vibrations with a more natural pure random vibration waveform.

(4) The introduction of a transmission system equalizer can be easily extended to more complex vibration control systems, such as two-way or three-way simultaneous vibration control systems.

Multidirectional vibration control is basically the same as the method of the present invention except that the transfer function is not a matrix and the number of parameters to be controlled is increased.

(5) Latest US military regulations MIL-8TD-810D 5
"RANDOM ON R" specified in 14.3
ANDOM” and '5INEON RANDOM
It can be easily extended to even more complex types of vibration control called ”.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a vibration control device according to an embodiment of the present invention, FIG. 2 is a block diagram of a conventional vibration control device, and FIG. 3 is an explanatory diagram showing a model of a general signal transmission system. 1: Vibration generator, 2: Movable table, 3: Sample, 4: Acceleration sensor, 5: Charge amplifier, 7: Input signal processing section, 8: PSD conversion section, 9: Reference PSD I! Turn data storage section, io: PSD comparison section, 11: Output frequency characteristic storage section, 12: Inverse FFT processing section, 15: Power amplifier, 17: Random number generation section, 18: Digital filter,
20: Output signal processing section. Patent applicant: Shinken Co., Ltd. Figure 1 is a configuration diagram of the vibration control device according to the embodiment.

Claims (1)

[Claims] A detection section that detects the movement of the sample attached to the vibrating section, a conversion section that converts the detection signal into a frequency domain signal, a data storage section that stores reference spectrum data, and a comparison between the frequency domain signal and the reference spectrum. a signal generating section that generates an arbitrary continuous signal, an equalizing section that corrects the frequency characteristics of this continuous signal, an equalizing section that corrects the frequency characteristics of the equalizing section, and an equalizing section that corrects the frequency characteristics of the equalizing section. A vibrating device comprising: an output means for outputting an output of the above to a vibrating drive unit.