Research On Seismic Prediction Method Of Glutenite Reservoir Based On Complete Ensemble Empirical Mode Decomposition

With the expansion of exploration and development of oil and gas fields,subtle reservoirs such as glutenite reservoirs have become important targets for oil and gas exploration.The glutenite reservoir has the characteristics of fast phase change,strong heterogeneity and large difference in oil and gas.The special geological features determine that it has many difficulties compared with structural oil and gas reservoirs,both in describing ideas and in researching techniques and methods.Therefore,the formation of a set of methods suitable for the prediction of glutenite reservoirs will be of guiding significance for the exploration and development of future oil and gas reservoirs.To solve the above problems,this paper proposes a seismic prediction method based on the Complete Ensemble Empirical Mode Decomposition for the sedimentary subfacies and sedimentary microfacies of glutenite,which improves the seismic prediction accuracy of sedimentary subfacies and sedimentary microfacies and combines the impedance inversion results to clarify the distribution disciplinarian of the reservoir sand bodies.According to the sedimentary tectonic evolution history,regional geological data and a small amount of well data of the basin,the sedimentary model of the basin is inferred.Combined with the results of single well phase division and the analysis of lithology and physical parameters.The three-phase correspondence relationship of "well logging facies-rock physical facies-seismic reflection characteristics" of different sedimentary subfacies of glutenite body is summarized.We study the empirical mode decomposition(EMD)algorithm and its iterative decomposition process and analysis the main problems of the empirical mode decomposition(EMD)algorithm.and two improved algorithms are introduced for the problem of EMD.Two improved algorithms—the ensemble empirical mode decomposition(EEMD)and the complementary ensemble empirical mode decomposition(CEEMD),are introduced for the problem of EMD,and the improved algorithm is tested by synthetic signals.The results show that the improved algorithm CEEMD not only solves the modal aliasing problem existing in EMD,but also has good completeness and better computational efficiency than EEMD.The CEEMD method is used to divide the seismic data into several natural modal functions,and the characteristics of each sedimentary subfacies and sedimentary microfacies on different modal components are analyzed by high-precision time-frequency analysis.Then extract various seismic attributes from the original seismic data and various modal components,and find out the sensitive characteristic properties that can effectively reflect the sedimentary subfacies and sedimentary microfacies of glutenite through logging and geological understanding.Based on the fuzzy logic-based attribute fusion algorithm and attribute classification algorithm,the sedimentary subfacies and sedimentary microfacies are predicted.Based on the distribution disciplinarian of the sedimentary subfacies and sedimentary microfacies,combined with the inversion results,the favorable reservoirs in the target interval are predicted.The prediction results show that the sedimentary facies belt predicted by this method accords with the geological understanding of the sedimentary features of the area,and the sand distribution law of the inversion is also consistent with this,and it has achieved good application results in the prediction of favorable blocks in glutenite reservoirs.