Identification of Compressed Air Leakage and Reflected Ultrasound Based on Spectrum Analysis for Vibration and Shock Liao Pingping, Cai Maolin, Xu Qiyue, Shi Yan (School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China) Problem; measuring the direct and reflected ultrasonic signals, the ultrasonic signal generated by the actual leakage source is reflected by the reflector, and is detected by the ultrasonic sensor, and the detector will mistakenly believe that there is a leak source at the reflector. If the direct ultrasonic signal and the reflected ultrasonic signal can be distinguished by digital signal processing, such a false detection can be avoided. In order to study the characteristics of the two ultrasonic signals, the direct and reflected ultrasonic signals were measured. The leaking holes were self-made, and the pressure reducing valve and the throttle valve were used to adjust the line pressure and flow, respectively, to adjust the amount of leakage.
2.3 Measurement principle Ultrasonic waves generated by compressed air leakage In the frequency range of about 40 kHz, the difference between the ultrasonic wave generated by the leakage and the ambient noise floor is large, as shown. Therefore, an ultrasonic sensor with a rated frequency of 40 kHz is generally used. this. When measuring the reflected ultrasonic signal, the direction of the sensor is directly opposite to the direction of propagation of the reflected ultrasonic wave, the leakage aperture and the ultrasonic leakage point ultrasonic sensor filtering and amplifying the circuit data ffif: the sensor is separated by a baffle to ensure that only the reflected ultrasonic signal Received by the ultrasonic sensor as shown. The ultrasonic propagation distance in both direct and reflected conditions is about 2.3 m. this. It can be seen from the figure that the ultrasonic waves generated by the leakage source are irregular fluctuations and irregular fluctuations, so it is difficult to find the difference between the two in the time domain.
3.2 Frequency domain feature analysis The collected multiple sets of direct and reflected ultrasonic time domain signals are subjected to Fast Fourier Transform (FFT), because the sampling frequency is higher (fs=240 kHz), in order to obtain higher frequency resolution, to ensure Analyze the required spectrum information. For FFT, the number of input samples is 219, and the frequency resolution is about 0. 46Hz. The FFT is obtained by MATLAB software and the spectrum is obtained. Only the spectrum of the frequency range of 35 ~ 50kHz is reserved. The amplitude of the other frequency segments is almost zero). To 2 are the spectrum spectra of typical direct ultrasonic signals and reflected ultrasonic signals of different reflecting surfaces, respectively.
It can be seen from ~2 that the direct ultrasonic signal shows a monotonously increasing (left) or monotonically decreasing (right) trend on both sides of the center frequency, and the spectral energy is mainly concentrated near the center frequency. However, the reflected ultrasonic signal has no obvious peak near the center frequency, showing a ''multimodal shape'), and the spectral energy is distributed in a relatively wide frequency range between 38 and 47 kHz, and the spectral trend has fluctuations. The reason is mainly because the reflection coefficient of the incident ultrasonic signals facing different frequencies is different. 3. There are obvious differences in the spectral shape between the direct and reflected ultrasonic signals, which are mainly reflected in two aspects: relative to the spectrum of the reflected ultrasonic signals, The spectrum of the direct ultrasonic signal is concentrated near the center frequency.
The spectrum low-frequency trend of the direct ultrasonic signal is “single-peakâ€, that is, the fluctuation is small; the spectrum low-frequency trend of the reflected ultrasonic signal is “multi-peakâ€, that is, the fluctuation is large.
The above analysis shows that there is a clear difference in the spectrum between the leaked direct and reflected ultrasonic signals. According to the difference between the above two aspects, the identification algorithm is designed to identify the direct and reflected ultrasonic signals.
4 Identification algorithm In order to effectively identify the direct and reflected ultrasonic signals, according to the two different points of the above-mentioned direct and reflected ultrasonic signal spectrum, the following identification algorithm is designed to obtain the horizontal center of gravity formed by the reflection axis of the identifiable reflected ultrasonic signal. Starting from the coordinates (center frequency), take N windows of appropriate length to the left, and calculate the average value of the square of the spectral amplitude F(n =1, 2, 3N) in the window. F represents the energy distribution of different frequency segments on the spectrogram. .
The adjacent Yn values ​​are compared in turn. When the value of Y +1 is greater than the value of F, the variance A of the two is calculated, and then multiplied by the weighting coefficient R to obtain REFn, and all REFn are added, and their sum REF is as follows Show: REF reflects the degree of fluctuation of Y; that is, the wave characterizing the spectrum. In the industrial field, the ultrasonic leak detector often detects both direct and reflected ultrasonic signals, and the superimposed signal of the direct and reflected signals is called a mixed signal. . Since the mixed signal has both the characteristics of the reflected and direct signals, the mixed signal and the reflected signal are similarly obtained when calculating by the equation (3), and only the reflected ultrasonic signal is desirably removed, so the right side is subtracted in the equation (3). To KD, where K is a constant coefficient; D is the ratio of the energy in the frequency range of ±1 kHz near the center frequency to the total energy in the frequency range of 35 to 50 kHz, reflecting the concentration of spectral energy near the center frequency.
The final calculation formula is that the larger the value of the reflection characteristic coefficient RCC, the more likely the detected ultrasonic signal is the reflected signal; the smaller the value of the RCC, the greater the possibility that the detected ultrasonic signal is a direct or mixed signal.
It can be seen that the recognition algorithm uses the volatility of the amplitude of different frequency segments in the spectrogram of the reflected ultrasonic signal to identify, the recognition result is not affected by the absolute amplitude of the ultrasonic signal, and the parameter design is the same for different reflective surfaces. That is to say, the recognition algorithm does not depend on the distance of the ultrasonic signal, the incident angle and the material of the reflective surface. Therefore, it is also suitable for the actual situation that the leakage point distance, the incident angle and the reflective surface material are unknown in the industrial field. Sex.
5 Calculation examples and verification The above algorithm is used to analyze and calculate the data of multiple sets of leakage ultrasonic signals. The results are shown in Table 1.
Type No. Direct Metal Anti-Wall Shot Sponge Cloth As can be seen from Table 1, all reflected ultrasonic signals have RCC values ​​greater than 200, and all direct ultrasonic signals RCC are below 100. Therefore, the RCC value of the ultrasonic signal can be accurately judged. Whether it is a reflected ultrasound signal.
6 Conclusion In order to avoid the problem of false detection of leakage source caused by reflected ultrasonic signal and improve the accuracy of ultrasonic detection technology of leakage source, this paper studies the direct and reflected ultrasonic signals and designs two signal recognition algorithms. The conclusions are as follows: direct and The reflected ultrasonic signal has a distinct difference in the spectral shape. The spectrum of the direct ultrasonic signal is concentrated near the center frequency, and the spectrum of the reflected ultrasonic signal is distributed near the center frequency, which is relatively scattered. In addition, the spectrum low-frequency trend of the direct ultrasonic signal is “single-peakâ€, that is, the fluctuation is small; the spectrum low-frequency trend of the reflected ultrasonic signal is “'multi-peakâ€, that is, the fluctuation is large.
By designing the recognition algorithm, the reflection characteristic coefficient RCC is obtained. The RCC value of the direct ultrasonic signal is below 100, and the RCC value of the reflected ultrasonic signal is above 200. Therefore, the identification algorithm designed in this paper can effectively distinguish between direct and reflected ultrasonic signals.
The research results provide an important way to improve the accuracy of the leak source ultrasonic testing technology and promote the development of this technology.
Since the weighting coefficients in the recognition algorithm proposed in this paper need to be set according to experience and subjective, the artificial intelligence-based recognition methods such as artificial neural network and support vector machine should be adopted in future research.
Speed Chain Conveyor Systems is one series of our Conveyor Systems , is also named pallets conveyors. The Home Appliance Assembly Line, like Tv Assembly Line, computer Assembly Line , oven assembly line are usually designed to Speed Chain Conveyor systems. In the speed chain conveyor systems, the semi-finished products on the pallets is driven by free flow chain, goes through every work station to fix all components into a completed products. The special structure of free flow chain make it continuously running but at the same time every work station can be stopped independently by manually controlling the stopper cylinders. This movement of the speed chain conveyor systems is convenient to operate and can make sure one assembling process are finished before next process. We can design the speed chain conveyor systems according to your products and your plant. Welcome to contact us for detail discussion.
Speed Chain Conveyor Systems
Speed Chain Conveyor Systems,Chain Conveyor System,Speed Chain Conveyor,Drive Pallet Chain Conveyor
Shenzhen Hongdali Machinery Co.,Ltd , https://www.conveyortop.com