Theoretical Research On Methods Of High Efficient Seismic Data Acquisition

Nowadays, along with the high-density, high-fold and wide-azimuth seismic data acquisition methods being widely used to overcome the increasingly complex and sophisticated situation of oil and gas exploration targets, the acquisition period is becoming much longer than before and the acquisition cost is becoming higher and higher. Under the premise of quality assurance of the acquired seismic data, we have carried out the research of high efficient seismic data acquisition method based on the sparse representation theory of seismic data. A theoretical framework of high efficient seismic data acquisition has been preliminarily establishment, as well as some innovative results have been achieved.The sparsity of signals has been widely applied in the fields of signal compression and image processing, which is regarded as the foundation of developing high efficient seismic data acquisition methods. Firstly, the problems of conventional seismic data acquisition method based on the Nyquist sampling theorem are analyzed, and then the mathematical and physical foundation of seismic data’s sparsity is revealed according to the general formula of wave equation’s solution. Under the guidance of sparsity prior information, we carry out an exploratory research of high efficient sampling (or acquisition for real data) methods of seismic data and present an optimized piecewise-random sampling method with a spectral characteristics of "blue noises". Based on the above theory and method, a reconstruction strategy with the sparsity constraint is developed. A particular two-step recovery approach by combining sparsity-promoting method and hyperbolic Radon transform is also put forward for the seismic data with approximate hyperbolic traveltime characteristics. Sparse representation theory, high efficient acquisition methods based on sparsity constraint and reconstruction method based on sparsity constraint of seismic data have been improved in this thesis and constituted the enhanced theory of high efficient seismic data acquisition.Under the guidance of the high efficient acquisition theory, the thesis also carries out the research of specific implementation strategies of high efficient acquisition. First, we propose the high efficient acquisition network designing method by applying the above random sampling scheme with sparsity constraint to change the regular and dense layout of sources and receivers in the conventional acquisition network. We then propose the high efficient seismic data acquisition method based on single sources and the high efficient seismic data acquisition method based on blended sources according to the differences of the sources’shooting modes:single shooting of the former and blended shooting of the latter. Considering the irregularities of the efficient acquisition network, a reconstruction procedure is needed for the data acquired by high efficient acquisition method, which is used to obtain the seismic data on the regular and dense acquisition network to satisfy the requirements of the subsequent processing in accordance with the existing means. According to the high efficient seismic data reconstruction method based on sparsity constraint (a part of the high efficient seismic data acquisition theory) and the improved separation method of blended seismic data in our work, we develop the reconstruction approaches for the seismic data acquired by two types of high efficient acquisition methods (six submethods). What’s more, the direct migration and imaging method of seismic data acquired by high efficient acquisition methods is studied and the impacts of blended shooting in the high efficient acquisition based on blended sources on the imaging result is analyzed. In the end of the thesis, we carry out the numerical tests based on the international Marmousi model. The results show that the theoretical framework of high efficient seismic data acquisition theory established in this thesis is of important significance to improve the efficiency of the current seismic data acquisition and reduce the exploration costs, and the data reconstruction method, a component of the theoretical framework, can provide us with high-precise reconstruction results for the following data processing procedures, which indicates a very broad development and application prospects of the theoretical framework in the future.