Research And Application Of Matching Pursuit And Basis Pursuit Analysis In Thin Interbedded Reservoirs

With the development of oil and gas exploration,the initial structural reservoirs that are easy to find.however,it is difficult to explore the reservoirs those highly concealed.the number of reservoirs stored based on structure is decreasing,and exploration targets are gradually shifting to find hidden reservoirs,especially thin reservoirs.Therefore,the fine characterization and recognition of thin reservoirs has become one of the important research directions.Based on the matching pursuit method,this thesis mainly studies the reflection inversion based on the basis pursuit algorithm and its application in thin interbedded reservoirs.This paper makes use of the matching pursuit algorithm and the basis pursuit algorithm in the theory of compressed sensing framework.Firstly,based on single-layer(monopole)matching pursuit algorithm realized the reflection inversion of seismic signals.The lateral continuity of the reflection obtained by the monopole matching pursuit algorithm is poor and the inversion ability of the tuned seismic signal is generated.This paper continues to study the double-layer(dipole)matching pursuit.This method is better than the monopole matching pursuit algorithm in signal decomposition accuracy,but the continuity of the obtained reflection needs to be improved.The reason is that the objective function of the matching pursuit algorithm is the norm optimization problem,the obtained solution is the local optimal solution,and the continuity is the requirement of the overall reflection coefficient.The idea of transforming the nonconvex optimization problem based on the objective function into a convex optimization problem.Secondly,this paper studies the application of basis pursuit in reflection inversion.The basis pursuit method is for the norm optimization problem,and the obtained solution is the global optimal solution.In the case where the wavelet is known,the effect of the basis pursuit is excellent,because the wavelet is unknown and time-varying in the actual seismic data inversion,the reflection obtained by the basis pursuit inversion is continuous.In the end,this paper proposes an improved basis pursuit algorithm.The improved basis pursuit algorithm is based on monopole matching pursuit to obtain time-varying wavelets to construct the atomic library of basis pursuit.This is more related to the propagation characteristics and laws of actual seismic waves.In accordance with this method,the method can effectively improve the inversion and recognition accuracy of thin reservoirs.Finally,the inversion accuracy and stability of the monopole matching pursuit algorithm,the dipole matching pursuit,the basis pursuit and the improved basis pursuit method are analyzed by theoretical model effects.At the same time,in order to verify the correctness and applicability of the algorithms,the inversion coefficient parameters obtained based on the actual seismic data are integrated,and the verification is consistent with the well data of the actual seismic data.The actual seismic data inversion results show that the inversion results of the improved basis pursuit algorithm are in good agreement with the well data,and the reflection is used to solve the relative wave impedance,thus providing an effective basis for thin interbed recognition and reservoir prediction.