| Seismic exploration includes three steps: data collection,processing and interpretation.The data collection is to help us obtain the underground wave impedance information through seismic instruments.The processing is to obtain the results of the profile form by the special method through the collected wave impedance information,and then the interpreter makes a geological interpretation of the profile to obtain the underground area of the target area.Geological information.It can be seen that seismic data processing is a very important step in seismic exploration and a link to communicate the entire seismic exploration operation.Among them,denoising is the key in seismic data processing,and the traditional single denoising method can no longer meet the actual production needs because of different The types of noise have different characteristics such as generating conditions and frequency amplitude.Therefore,different denoising methods should be used for different noises,and various noises should be removed from a multi-domain perspective.For this purpose,I start with research on the principle of seismic wave propagation,analyzes and summarizes the propagation modes and characteristics of different types of seismic waves,introduces various parameters characterizing seismic waves and their meanings and internal relationships,so as to analyze seismic waves in depth.At the same time,according to the principle of "tracing back to the source",the noise is divided into regular noise and irregular noise according to its characteristics in the seismic records.A detailed description gives a fuller understanding of these noises.Secondly,based on different denoising theories,the paper introduces the principles and specific implementation steps of median filtering and Wiener filtering in detail,and analyzes their advantages and disadvantages.The non-linear characteristics of median filtering can reduce the interference of external input interference information on the filtering.The idea of minimum square difference provided by Wiener filtering has opened up ideas for the study of denoising technology.Starting from the one-dimensional Fourier transform theory,it is extended to the two-dimensional Fourier transform,fully introduces the FK filtering method and principle,evaluates its advantages and disadvantages,introduces the Heisenberg uncertainty principle,and explains the inherent shortcomings of Fourier transform analysis.In-depth study of the FX domain filtering method extended on the basis of the Fourier transform theory,and combining the least square difference idea in Wiener filtering,the mathematical reasoning of the FX domain prediction denoising technology and the FX domain fitting denoising technology is introduced in detail And the specific operation steps,discussed the applicable conditions of different methods.Through the time-frequency analysis method,time and frequency can be closely linked.According to the different performances of noise and effective signals in the time spectrum,a filter is designed to filter the signal in the time-frequency domain,and the accuracy of the time-frequency analysis is higher in the filtering process.Windowed Fourier Transform,Wavelet Transform and S Transform have always been important means of time-frequency analysis.Through theoretical analysis and model testing,this paper verifies that the effect of S-transform time-frequency analysis performs better among the three and has higher accuracy.Further suppressing the surface wave interference in the data by means of S-transform time-frequency filtering has obtained good results in the verification of theoretical models and actual data.Finally,the feasibility of these denoising methods is verified through models and actual data,the denoising effects of these methods are discussed,and the advantages and disadvantages of various methods are summarized in conjunction with the analysis of the applicable conditions of these denoising techniques,and seismic data processing Multi-domain denoising technology. |
PDF Full Text Request