Study On Wavelet Analysis Method For Blasting Seismic Signals

At present, there is growing concern in the construction of blasting vibration effects' testing, but for the collection of blasting vibration signals, many deficiencies in the data analysis still exist. Blasting vibration essentially is a non-stationary random process, and the theoretical basis of traditional Fourier transformation is not consistent with the process. After complex geotechnical media dissemination, blasting seismic waves contain various frequency components of the interferential wave, because of stacking of the interferential wave, blasting vibration analysis to the frequency characteristics of signals have became very difficult. The application of the traditional Fourier transformation is unable to get the accurate time-frequency characteristics of blasting vibration signals, for this reason, the application of new analytical methods become necessary, the appearance of wavelet analysis were just fill the inadequacy of signals' time-frequency analysis methods in the current. In this paper, on the basis of introduction of the theory for Fourier analysis and wavelet analysis, Fourier transformation method and wavelet analysis method are used to deal with signals. The article can be concluded as follows:(1) Theoretically expounded the basic principles and mathematical basis of wavelet analysis, Summarized the current analysis methods of blasting vibration signal. Through comparison wavelet analysis with Fourier analysis and short-time Fourier analysis, theoretically expounded wavelet analysis in the signal processing were better than Fourier transformation.(2) To conduct signals Fourier transformation, through spectrum analysis we can transform blasting vibration signals from the time domain to the frequency domain, and we can receive signal's frequency, distribution and the main vibration frequency and so on. But in the process of transformation has discarded the concept of time, we can not confirm the blasting vibration signals' frequency identity.(3) For true signal that contained interference and noise, the author bring forward threshold de-noise methods on the basis of the theory of wavelet analysis. And based on multi-criteria analysis principles, expounded how to carry out dispersewavelet analysis. Through wavelet decomposition and reconstruction method we can attain the true signal from original signal and we can see reconstructed signal have not contained interferential component. So we can conduct further analysis of blasting vibration signals with wavelet analysis. Explained wavelet analysis method has sub-frequency characteristics and de-noise functions.(4) Based on the wavelet time-frequency analysis method for blasting vibration signals, presented blasting vibration signal's energy analysis methods, and used wavelet packet decomposition method to signals, compiled the calculation procedures to attain the energy distribution of different frequency bands. Through the energy analysis of different frequency bands, we can attain the blasting vibration signals' frequency identity and original signals' part character.In the paper, wavelet threshold de-noise methods are used to deal with blasting vibration signals, and wavelet packet decomposition methods are used to conduct signals' time-frequency analysis and energy analysis of different frequency bands, which have achieved good results. Research shows the superiority of wavelet analysis methods in the analysis of the blasting vibration signal. We believe wavelet analysis methods have very good application prospects in the field of analysis of the blasting vibration signals.