Application Of Wavelet Analysis In Experiments Of Ventilation Turbulence LDA And CTA

Wind speed distribution is the basis for calculating the distribution of mine gas,fire gas,dust,the distribution of temperature and humidity.Mine flow belongs to the category of turbulence.Research on turbulent flow velocity in mines has attracted more and more attention in recent years.The author's task group used Doppler laser speed meter LDA and hot wire anemometer CTA to test the characteristics of undisturbed single-point wind speed,cross-section wind speed characteristics of the mean-value roadway,and disturbance characteristics of the switch.The probability average method was used to solve the single-point average wind speed time series and the average wind speed distribution of the section.The characteristic of the disturbance waveform of the damper switch was extracted and the identification problem of down hole wind disturbance was studied.The experimental results show that there is a certain deviation between the test results and the theoretical calculation results.The reason for the deviation is not only the problem of the experiment itself,but also a more important reason is the white noise in the collected data.The LDA and CTA test signal transmission process is subject to Gaussian white noise pollution such as AC power supply and signal processing devices.These white noises cannot be eliminated by setting the system SNR.This paper uses wavelet transform to study the Gaussian white noise elimination of LDA and CTA experiments problem.The time-series values of the single-point three-dimensional wind speed fluctuations at the subtle level were obtained.In the seven kinds of wavelet basis functions,the DB and SYM wavelet basis functions were used to perform discrete decomposition experiments on the wavelet.Experimental eliminated noising experiments were performed using four thresholds: hard thresholds,soft thresholds,fixed thresholds,and extreme thresholds.Three types of wavelet decomposition layers,3,4,and 5,were used to reconstruct the eliminated wavelet.The experimental results show that:(1)Single point wind speed signal of straight laneway selects sym5 function for wavelet decomposition,and the number of decomposition layers is 5;when threshold factor value is used to select soft or hard threshold,the eliminated noising effect is best when using wavelet packet method,and the eliminated noised wind speed The time series waveform is closer to the theoretical waveform;(2)The full-section wind speed signal of the straight laneway also selects the sym5 function for wavelet decomposition,and the number of decomposition layers also takes 5,and the default value of the global threshold is used as the threshold and wavelet analysis as the eliminated noising method.Effectively remove the noise;(3)Wind speed signal with damper switch disturbance selects sym5 function for wavelet decomposition,and the number of decomposition layers is 5;when threshold factor value is used to select soft or hard threshold,the eliminated noising effect is best when using wavelet packet method,and the eliminated noised wind speed The time series waveform is closer to the theoretical waveform.