JP2612697B2 - Vibration control device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a control device for a vibration / shock test apparatus.

[Conventional technology]

2. Description of the Related Art In general, a vibration / impact test apparatus is provided for testing the reliability and durability of various parts and devices that receive vibration such as aircraft parts and automobile parts against vibration / impact.

An example of a conventional system and apparatus for controlling vibration is disclosed in Japanese Patent Publication No. 52-18873 (hereinafter referred to as a conventional example).

This prior art invention utilizes a precise analysis of vibration when subjecting a specimen to be tested to vibration in a vibration test environment or on a vibration device, as described in the detailed description of the invention in the publication. The purpose of the present invention is to provide a system for numerically controlling a vibration test environment or a vibration device.
5, the device is controlled and driven by a random signal such as a vibration test environment or a shaking table, whereby a sample (sample) on the vibration test environment or the vibration device is provided. To give a vibration with the expected output spectral density to the sample (sample)
Is detected by the acceleration sensor and converted into a digital signal. The output spectral density analysis is performed by first performing a Fourier transform on the digital signal, the output spectral density is compared with a desired or predetermined output spectral density, and a random signal is generated using the comparison result. When this signal is converted into a time-domain signal, a vibration test environment or a vibration device is driven to control the vibration. A device that generates a random digital signal representing a phase angle generates a random driving signal of a random parameter, and uses this phase angle together with a comparison result between the actual spectral density and a desired spectral density to generate a random digital signal. Create a signal. This signal is converted to a time-domain variable by performing an inverse Fourier transform on the random digital signal. The result of the inverse Fourier transform is converted into an analog signal, and the analog signal is used to drive and control a vibration test environment or apparatus. Although the Fourier transform and the inverse Fourier transform and the comparison are performed by a multipurpose digital computer, the calculation of the Fourier transform and the inverse Fourier transform can be performed by using a special digital computer. Therefore, the purpose of the prior art is to use a precise analysis of vibration when a sample (sample) to be tested is placed on a vibration test environment or a vibration device and subjected to a vibration test, so that the vibration test environment or the vibration device is used. A system and apparatus for controlling numerically are provided.

An example of the conventional device will be described with reference to a configuration diagram shown in FIG. The sample 3 is placed on the movable table 2 of the vibration generator 1.
Is installed and required vibration and impact are applied. An acceleration sensor 4 is attached to the possible table 2,
The electric signal proportional to the acceleration from the acceleration sensor 4 is amplified by the charge amplifier 5 and input into the vibration control device 6 indicated by a broken line. The vibration control device 6 includes an input signal processing unit 7, a PSD conversion unit 8, a reference PSD pattern data storage unit 9, a PSD comparison unit 10, an output frequency characteristic storage unit 11, an inverse FFT processing unit 12, a randomizing process 13, and an output signal processing unit. 14 are connected and configured as shown in the figure. The electric signal input into the vibration control device 6 is filtered in an input signal processing section 7 and converted into an analog-to-digital signal, and the data is stored as a block. The data is subjected to a fast Fourier transform (FFT) process by a PSD conversion unit 8 for converting the data into a power spectrum density (PSD) to obtain a PSD, and the reference PSD pattern stored in the reference PSD pattern data storage unit 9 and PSD and PS
The comparison is performed by the D comparison unit 10. The comparison result is temporarily stored in the output frequency characteristic storage unit 11, and is subjected to FFT processing in the inverse FFT processing unit 12 to synthesize a random signal for one frame. By randomly connecting a plurality of the random signals for each frame, a randomizing process 13 is performed to obtain a continuous (non-periodic) random signal. Then, the random signal is output to the power amplifying unit 15 via the output processing unit 14.

[Problems to be solved by the invention]

All of the configurations of the invention of the prior art described above are realized by the functions of a digital computer.
The processes from T processing, PSD comparison, inverse FFT processing to randomizing processing require extremely high speed and complicated processing, and it is difficult in principle to obtain a complete pure random signal. Further, in this configuration, only the control of the random wave vibration is limited and cannot be applied as a control device for the sine wave vibration or the like. The present invention aims to solve these problems.

[Means for solving the problem]

To solve the above problems, the present invention is configured as follows. That is, as described in the claims, 1. an electric signal from an acceleration sensor for detecting a vibration amount of a sample mounted on a movable table of an electrodynamic vibration generator is input to a vibration control device, In the vibration control device, an output obtained by performing an electrical process on the input in the vibration control device is applied to the electrokinetic vibration generator via a power amplifier, and a predetermined vibration is applied to the sample. A / D conversion of the input signal from the acceleration sensor, an input signal processing unit that processes the signal into a frequency domain signal, a PSD conversion unit that converts the signal into a PSD, and a reference PSD pattern data storage unit that is stored separately. A PSD comparing unit that compares the control PSD with the reference PSD, creates a spectrum having the frequency characteristics necessary for equalization, temporarily stores and holds the spectrum, an output frequency characteristic storing unit, and a random number generating unit that generates a white noise signal And multi-channel digital A filter, an inverse FFT processing unit, and an output signal processing unit are connected and formed, and a signal from the random number generation unit is passed through the multi-channel digital filter, and the digital filter is subjected to a signal necessary for the equalization. The vibration control device is characterized in that it has a means for equalizing.

To further clarify the means to solve the problem,
The present invention will be supplementarily described including effects and effects in comparison with the conventional example.

The configuration of the portion surrounded by the dashed line shown in the conventional example of FIG. 5 is composed of an input spectrum storage device 38 and a phase angle transmitter 31.
, A multiplier 41, and a vibrator input storage device 40. The operation can be summarized as follows. This hardware reads out one frame of output data in the memory at random and multiplies it by a window function. The operation is performed when each is added. The conventional example is a system that drives a vibration test device even with a random signal designed to give a predetermined spectral density to a sample (index) placed on a vibration device. The operation required to control this system is a digital type. However, even with a digital computer, it takes a certain amount of calculation time to perform an extremely large number of calculations, and therefore, there is a problem in the real-time property required for this device. In particular, the randomizing process for producing a continuous signal requires a huge amount of calculation, and there is a drawback that the real-time property is lost.

On the other hand, in the present invention, as shown in FIG.
Between the T processing unit 12 and the output, as shown, the coefficient memory 5
0, multiplier 51, delay element 52, digital random number generator 1
7 and the clock generator 53 were configured in the range shown by the dotted line 55 and output via the D / A converter 54, and this part was newly replaced. In summary, a digital filter having an extremely large number of channels has a frequency characteristic and is set instantaneously so as to have a characteristic opposite to the movement of the sample. In other words, the digital filter is provided with an equalizing function and outputs a signal obtained by equalizing the white noise signal into a spectrum having the required frequency characteristics, thereby eliminating the need for calculation time such as randomizing processing and real time characteristics. This has the effect of being improved.

[Action]

Since the present invention is configured as described above, the electric signal input to the input signal processing unit of the vibration control device is stored in a memory as digital data, and is converted into a power spectrum density (PSD) by a PSD conversion unit, In the PSD comparison unit, the data is compared with the reference PSD pattern data, the comparison result is inversely subjected to an FFT processing unit, and based on information from the inverse FFT processing unit, a transmission digital filter having a predetermined frequency characteristic is obtained. After passing, the sample vibrates so as to approximate the vibration based on the reference PSD pattern data.

〔Example〕

A vibration control device according to an embodiment of the present invention will be described with reference to FIG.

In FIG. 1, 1 is an electrodynamic vibration generator, 2 is a possible table, 3 is a sample mounted on the movable table 2, 4 is an acceleration sensor that detects the acceleration of the sample 3 and outputs an acceleration signal, and 5 is It is a charge amplifier that converts an acceleration signal into a voltage signal.

Reference numeral 16 in the configuration surrounded by broken lines denotes a vibration control device. In the vibration control device 16, 7 is an input signal processing unit, and 8 is a PSD.
Conversion unit 9, reference PSD pattern data storage unit 10, PSD comparison unit 11, output frequency characteristic storage unit 12, inverse FFT processing unit 1,
Reference numeral 7 denotes a random number generation unit, reference numeral 18 denotes a digital filter, reference numeral 20 denotes an output signal processing unit, reference numeral 15 denotes a power amplifier, and the above-described units and devices are connected in a straight line.

FIG. 3 is an explanatory view showing a model of a general signal transmission system, and 19 is an equalizer. FIG. 4 is a diagram for explaining the detailed principle of the portion related to the digital filter 18 of the embodiment (FIG. 1) of the present invention. In FIG. 4, 12 is an inverse FFT processing unit, 17 is a digital random number generator, and 50 is A coefficient memory, 51 is a multiplier, 52 is a delay element, 53 is a clock generator, 54 is a D / A converter, and the range indicated by a dotted line 55 is a newly replaced configuration of the present invention, which is like a straight line. Although they are connected, the description is omitted as described above.

The following description will be made with reference to FIGS. 1, 3, and 4.

The input signal processing unit 7 samples the voltage signal after passing it through a low-pass filter, and converts the analog signal into a digital signal by an A / D converter.
It is stored in the memory as digital data for one block (N points). Low-pass filtering, as is well known,
This is for removing an undesired frequency component generated by sampling. The one block of digital data is converted to a well-known Fast Fourier Transform (Fast Fourier Transform).
It is converted into a power spectral density (Power Spectral Donsity; PSD) using a Transform; FFT algorithm.

Numeral 8 denotes a PSD conversion unit, which obtains a discrete Fourier transform F (J) (J = 0,1, J) from the sampling data f (K) (K = 0,1,..., N−1) for one block. …, N-1)
Further, PSD P (J) is obtained as the square of the absolute value of F (J). When these are expressed by formulas, P (J) = | F (J) | ² (J = 0,1..., N / 2) (2) In the equation (1), j ² = −1 (imaginary unit).

As is well known, the value of the PSD of the random signal varies with each measurement using only one data, and some measurement data must be smoothed by appropriate processing to obtain stable data. The most common data processing method is to calculate the average value of the data and use it as an estimate of the true PSD.

S _i (J) = ▲ ▼ (3) where S _i (J) is the PSD obtained in the i-th control loop (this is called control PSD), and Δ is obtained by the smoothing operation. Represents the estimated value.

Reference numeral 9 denotes a reference PSD pattern data storage unit, which is a memory storing reference PSD patterns _Pr (J) (J = 0, 1,..., N / 2).

Numeral 10 denotes a PSD comparison unit, which compares the control PSD with the reference PSD P _r (J) according to the equation (4), and obtains a frequency characteristic H _{i + 1} (J) required for the output random signal (J = 0, 1, ... N / 2).

The reference PSD pattern _Pr (J) (J = 0, 1,..., N / 2) is stored in a memory in advance.

11 is an output frequency characteristic storage unit for temporarily holding a comparison result,
Reference numeral 12 denotes an inverse FFT processing unit that performs an inverse FFT process on the comparison result, and reference numeral 17 denotes a random number generation unit that generates a white noise signal.

Reference numeral 18 denotes a digital filter for realizing a transmission system having a frequency characteristic _{Hi + 1} (J). Hereafter the third
Description will be made based on the drawings. FIG. 3 is a model of a general signal transmission system. s _i (t) is an input signal, and s _o (t) is an output signal. s _i (t) and s _o (t) are so-called time-series signals, and may have any dimension. Reference numeral 19 denotes a linear signal transmission system whose transfer function is H (ω) (ω = 2
πf, f is a frequency). Under the above assumptions, the relationship between the input and output signals is the convolution integral Are related by Here, the function h (t) is H
This is an inverse Fourier transform of the transfer function (ω), which is called an impulse response function.

In the equations (5) and (6), both s _i (t) and s _o (t) are defined on the continuous time axis t. However, in actual applications, the discrete time n (n = Δt · K, where Δt is the sampling period and K is the sampling number) Becomes Here, N represents the total number of sampling data for one block.

As is apparent from the above, according to the equation (7), the input signal s _i (m) is multiplied by the impulse time series h (m), and a discrete composition operation is performed.
(J) It is possible to approximately realize a transmission system having a frequency characteristic as follows. The impulse time series h (m) is calculated by Expression (8), which is a well-known inverse FFT in the inverse FFT processing unit 12.
It is easily determined by processing. A device that realizes the expression (7) is known as an Nth-order FIR (Finite Impulse Response) type digital filter. The larger the N, the higher the H
Although the degree of approximation of (J) and H (ω) is good, since the number of operations increases and the memory capacity increases,
The sampling time that can be realized by the same hardware is limited. This digital filter can be realized by software of a computer, but can also be designed as an independent device by applying an inexpensive microcomputer. In the present embodiment, N = 1024 is set as the maximum value. However, it is needless to say that the approximation can be improved if a larger N is applied with the future progress of hardware.

Returning to FIG. 1, the operation of the above equation (8) is performed by the inverse FFT processing unit 12, and the digital filter 18 converts the white noise signal from the random number generation unit 17 into s _i (m) of the equation (7). , (7) is performed. The output of the digital filter 18 is subjected to output gain adjustment and clipping in the output signal processing unit 20, then converted into an analog voltage signal, and sent to the input unit of the power amplifier 15. The power amplifier 15 increases the power of the input voltage signal, drives the vibration generator 1, generates random vibration on the movable table 2, and completes a single loop.

〔The invention's effect〕

 According to the present invention, the following effects can be obtained.

(1) It is possible to eliminate the conventional operation of randomizing an extremely sophisticated output signal, thereby reducing the burden on both software and hardware such as a host computer system, and realizing a cheaper system. In addition, the real-time property has been significantly improved.

(2) According to the present invention, sine wave vibration control, shock wave vibration control, and extension to a waveform reproduction system of actual vibrations including seismic waves, etc., can be realized with exactly the same hardware.

(3) Further, if the present invention is used for random vibration control, more natural vibration of pure random vibration can be achieved.

(4) The digital filter of the present invention can be easily extended to more complex vibration control systems, for example, two-way and three-way simultaneous excitation control systems. The multi-directional vibration control is basically the same as the method of the present invention except that the transfer function becomes a matrix and the parameters to be controlled increase.

(5) It is easy to extend even more complicated types of vibration control called "RANDOM ON RANDOM" specified in the latest US standard MIL-STD-810D 514.3.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a vibration control device according to an embodiment of the present invention, FIG. 2 is a configuration diagram of a conventional vibration control device, FIG. 3 is an explanatory diagram showing a general signal transmission system model, FIG. FIG. 1 is an explanatory view of the detailed configuration principle according to FIG. 1 of the embodiment of the present invention, and FIG. 5 is a detailed explanatory view of a conventional example. 1: vibration generator, 2: movable table, 3: sample, 4: acceleration sensor, 5: charge amplifier, 7: input signal processing unit, 8: PSD conversion unit, 9: reference PSD pattern data storage unit, 10: PSD Comparison part, 1
1: output frequency characteristic storage unit, 12: inverse FFT processing unit, 14: output signal processing unit, 15: power amplifier, 17: random number generation unit, 18: digital filter, 20: output signal processing unit, 25: vibrator system ,
26: AD converter, 29: PSD comparison device, 30: Reference PSD, 31: Phase angle transmitter, 34: DA converter, 35: Vibrator output storage device, 36: Average output PSD storage device, 37: Output spectrum Density processor, 38: input spectrum storage device, 39: inverse Fourier transform processor, 40: vibrator input storage device, 41: multiplier, 50: coefficient memory, 51: multiplier, 52: delay element, 53: clock generator , 54: D / A converter

Claims (1)

(57) [Claims] 1. An electric signal from an acceleration sensor for detecting a vibration amount of a sample mounted on a movable table of an electrodynamic vibration generator is input to a vibration control device, and the input signal is input into the vibration control device. The vibration control device, which applies an output obtained by performing electrical processing to the electrodynamic vibration generator via a power amplifier to apply a predetermined vibration to the sample, comprising: An input signal processing unit that performs A / D conversion and processes the frequency domain signal, a PSD conversion unit that converts the signal into a PSD, a separately stored reference PSD pattern data storage unit, a control PSD and the reference PSD, To create a spectrum having the frequency characteristics required for equalization, and temporarily store and retain it
A PSD comparison unit, an output frequency characteristic storage unit, a random number generation unit that generates a white noise signal, a multi-channel digital filter, an inverse FFT processing unit, and an output signal processing unit are connected to each other. A means for passing said signal through said multi-channel digital filter and equalizing said signal with a signal necessary for said equalization by said digital filter.