Research About Reconstruction Method Of Seismic Wave Field Data Compression

Before oil and gas resources exploitation, workers first need to ascertain the correct position of the distribution in the ground, that is, the need for oil and gas exploration.The seismic exploration as an important means of oil and gas exploration, plays a huge role in the oil and gas exploration.But with the expanding range of oil and gas exploration, ground and underground exploration work environment is growing increasingly complex, coupled with the impact of human and other factors, so that can not be collected to complete seismic exploration seismic data, which seriously affected the seismic data accurate imaging, seismic exploration giving correct proven oil and gas resources distribution position difficult.Therefore, to address the problem, if we want to be able to use artificial methods to incomplete seismic data reconstruction, to restore a complete seismic data?This article is from this issue of view, the development of new compressed sensing signal processing theory to seismic exploration, seismic waves of the collected field data were sparse representation, select the appropriate measurement matrix and the reconstruction algorithm for missing seismic data reconstruction, and ultimately the desired treatment effect.Compressed Sensing theory overcomes the shortcomings of traditional Nyquist sampling theorem, it is a signal compression process is also integrated into the process of sampling, that the sampling and compression at the same time, and only a low sampling rate of the signal sampling, it is possible to reconstruct the original signal meets certain precision.In view of this basic process,this paper focuses on converting sparse seismic field data reconstruction algorithm optimization and sparse.In the sparse conversion, in view of Curvelet transform locality, multi-scale and multi-directional Processing unsteady seismic wavefield data have greater advantages, so choose Curvelet transform sparse seismic field data conversion, and Curvelet transform can not only transform the seismic wave field data better sparse representation, but also to achieve the seismic data denoising and removing the phase axis interference effect by experiments described below.In terms of the sparse reconstruction algorithm optimization, optimization by solving a sparse l0 norm optimization problem to obtain an accurate reconstruction of the original signal, l0 norm is optimized for a underdetermined problem, not easily solved, is usually the case where l1 norm approximation l0 norm as projections a solvability of optimization problems,study convex sets projection theory(POCS) solve l1 norm optimization problems, lack of seismic wave field data reconstruction.Research designed based on compressed sensing seismic field data reconstruction method, and verified by experiment its practicality.The study found that the number of iterations and the threshold model are the two factors that influence the effects of reconstruction, so focused on their impact on the effectiveness of reconstruction.In the threshold model, the design of the linear threshold model and mathematical expression index threshold model, and the index threshold model is proposed to improve a new Exponential Decay threshold model, and use the threshold model iteration improves the computing speed but also to achieve a better reconstruction effect.