| Vertical Seismic Profiling(VSP)technology has the advantages of strong identification ability,high exploration accuracy and good imaging effect,and can clearly reflect the characteristics of subsurface geological structure and petrophysical properties.It has become a powerful technology to solve the problem of underground structure imaging and oil and gas reservoir description in oil and gas exploration.Therefore,it is necessary to study the data processing method of VSP.In this paper,the basic principle of the directionlet transform is studied in depth,and the multi-directional identification characteristics of the directionlet transform are used to apply it to the separation processing of the upstream and downstream waves of the VSP data for the first time,and the improvement scheme is proposed in view of its limitations in the process of processing.Namely,the separation method of VSP upstream and downstream waves based on radon domain directionlet transform.The following are the main research contents and achievements.The first part is the research on the numerical simulation of VSP forward modeling.Two VSP forward numerical simulation methods are studied,which are wave equation method based on Tesseral software and segment by segment iterative ray tracing method.By comparing the differences between the two forward modeling methods and combining with the requirements of the test model in this paper,the iterative ray tracing method was selected to simulate the VSP forward modeling,and three different geological models were established,including horizontal formation,inclined formation and inter-well VSP model.The second part is the research of traditional VSP upstream and downstream waves separation method.Two kinds of upstream and downstream waves separation techniques of VSP are studied,and the separation effects of the two methods are tested by using forward VSP model respectively.The results show that the linear radon transform method cannot completely separate the upstream and downstream waves,and the wave fields are influenced by the false frequency and truncation effect during the transform process,and the separated upstream and downstream waves in-phase axis is not clear and coherent.Compared with the original VSP record,the separated wavefield profile by curvelet transformation method has fuzzy in-phase axis,and the waveform distortion of single-channel record is larger,which will lose some effective signals.The third part is the research of VSP upstream and downstream waves separation method based on radon domain directionlet transform.The basic principle of directionlet transformation is mainly studied,including integer lattice theory,multi-coset decomposition and anisotropic wavelet transform.After analyzing the problems existing in the application of the transform to the separation of upstream and downstream waves in VSP,an improved scheme--the separation method of upstream and downstream waves based on radon domain directionlet transform is proposed,and the specific implementation process of the algorithm is given,which provides a relatively sufficient theoretical demonstration for the subsequent processing of VSP theoretical model and practical data.The fourth part is the processing of theoretical model and practical VSP data.The proposed directionlet transform separation method based on radon domain is applied to VSP theoretical models under different conditions,such as horizontal,inclined and inter-well,and compared with the results of linear radon transform method and curvelet transform method,the results show that the method in this paper is superior to the traditional method in separation effect.The main reason is that the characteristics of the upstream and downstream waves field after separation are more obvious,and the in-phase axis is clearer and the continuity is better.Finally,the method in this paper is applied to the processing of practical VSP data,and the separation results are consistent with those of the theoretical model,and better than those of the linear radon transform method.In conclusion,the superiority of the method proposed in this paper is verified by the separation of VSP theoretical model and practical data. |