Research On Prediction And Elimination For Surface-related Multiple Of Marine Seismic Data

The suppression of multiple waves has always been a particularly important task in seismic data processing.Due to the existence of multiple waves,the quality of seismic data processing is severely affected.In marine seismic exploration,the influence of free surface multiples is particularly obvious.Therefore,how to suppress the free surface multiples while retaining the effective wave to the greatest extent has always been a key topic in marine seismic exploration.The thesis systematically summarizes and analyzes the current status of the multiple suppression method.The multiple suppression methods are mainly the following two types: one is based on the filtering method to suppress the multiple wave,the other is based on the wave equation to predict and suppress Multiple waves.The main research and implementation of this paper is surface-related multiple elimination method(SRME),which belongs to the wave equation method.The basic principles of the two steps of free surface multiple prediction and adaptive subtraction are analyzed in detail in this paper 3.The realization process,and finally the method is realized by writing a program independently.According to the principle of the SRME method,when the seismic data is incomplete,especially when the near-offset data is missing,it will have a great influence on the prediction of multiple waves by the SRME method,which in turn will affect the next step of adaptive subtraction.Therefore,this paper studies the problem of missing seismic data reconstruction,systematically summarizes and analyzes the current seismic data reconstruction methods,and focuses on the research and implementation of the hyperbolic Radon transform method to reconstruct seismic traces.In order to effectively improve the reconstruction accuracy and calculation speed,Introducing the idea of track equilibrium,the concrete steps of reconstructing the missing track based on the track equilibrium hyperbolic Radon transform are given.Finally,combined with the model data for the experiment,this paper first discusses the effect of the lack of near-offset data on the prediction of multiples by the SRME method in a simple seafloor model,and then uses the channel equalization hyperbolic Radon transform method to reconstruct the seismic trace,and the least square Compared with the hyperbolic Radon transform,the reconstruction accuracy is higher than that of the reconstruction method,which shows that the method can be used in the reconstruction of seismic data,and its accuracy can meet the needs of the SRME method when predicting multiple waves.On this basis,in this paper,the SRME method is used in two models to predict the free surface multiples,and then the following four adaptive subtraction algorithms are used to match and suppress the multiples: single channel,multi channel,pseudo multi channel and expansion Pseudo multi-channel method,and comparative analysis of the research results.The study found that the effect of using extended pseudo-multi-channel and pseudo-multi-channel matching to suppress multiple waves is more ideal than single-channel and multi-channel,especially when the multiple waves and the primary wave overlap.At the same time,in order to obtain a better matching suppression effect,the selection and analysis of the two important parameters that affect the matching suppression effect are also analyzed and compared.The four matching suppression methods studied in the paper all remove the effective wave at the overlap of the effective wave and the multiple wave,because these four matched filtering methods are alladaptive matched filtering methods based on the residual minimum energy criterion.How to retain the effective wave where the multiple and effective waves overlap and subtract the multiple has always been a problem of multiple suppression,so the study of the free surface multiple suppression method(SRME)based on the wave equation needs to be further studied In-depth.