Application Of Basis Pursuit To Surface Waves Separation And Suppression

With the development of seismic exploration technology, exploration targets required more and more sophisticated,surface wave is a coherent interference, which is a signal to noise ratio greater impact on the seismic records.The traditional method of surface wave suppression only focus on one aspect to consider a single feature, or assumptions based only on certain conditions, despite the effective suppression of surface wave, however, while the effective suppression of surface wave, effective wave will also be different degrees of loss. Developed in recent years, the Basis Pursuit algorithm is that the area of sparse signal a new approach,which sought the most sparse representation of signal from the complete (overcomplete) function (base) set, that is expressing the original signal exactly with base as little as possible, thereby obtain the intrinsic nature of the signal characteristics. According to the characteristics of seismic signals,this paper applied this method to the seismic data processing, which provides a new research direction in order to btain high-resolution seismic data.This paper first analyzes the development and current research situation of separation and suppression method of surface wave, then introduced the normal method of removal of surface wave and the basis knowledge of signal sparse decomposition, analysis of the frequency estimation method which is based on BP, experimental results show the using of BP decomposition the signal can accurately estimate the frequency of the signal,which provides the basis of a certain method for BP separating seismic signals. In the third chapter, the separation of time-frequency of the synthesized one-dimensional seismic signals based on BP was achieved, experimental proved BP can effectively isolate the various sub-wave which composed seismic signal, separation in BP based on the seismic signal, a separation and suppression method of surface wave based on BP is bring forward on the basis of BP separating seismic signals, this method is based on the sparse decomposition principle, according as the frequency characteristics and energy characteristics of surface wave, according to the actual frequency range of two-dimensional seismic signals and structural characteristics of the seismic signal establishes adaptive atom dictionary, and selected the importance atoms which can approximate reconstruct the seismic signal form the atom dictionary, then according to decomposition coefficients and parameters of the selected importance atoms set the energy threshold and the range of main_frequency threshold adaptively, identify the atoms which compose surface wave form the decomposition results of every seismic signal and reconstruct the surface wave component of this seismic signal, using of this seismic signal subtract the surface wave component and get the seismic signal which its surface wave component is suppressed, to achieve and the effect of separating and suppressing surface wave. In the fourth chapter, using of the proposed method of BP to separation and suppression of surface wave separate and suppress the surface wave of synthetic two-dimensional seismic record, comparing with the traditional one-dimensional Fourier filtering and one-dimensional wavelet processing method, respectively subjective and objective evaluate the treatment effect of the method, results show that the method's superiority in the treatment effect. In the fifth chapter, using of the proposed method of BP to separation and suppression of surface wave separate and suppress the surface wave of the actual seismic profile, comparing with the traditional one-dimensional Fourier filtering and one-dimensional wavelet processing method, experimental results show this method still reflects its advantages in actual seismic data processing than the traditional methods.