The Research On Improvement Method Of Blended Seismic Data Imaging

Firstly, from the point of view of the frequency domain Helmholtz equation, a detaileddescription of the frequency domain modeling process and frequency domain blendedseismic acquisition forward process were given. Take "Mushroom" model as therepresentative, the frequency domain forward got frequency slice and conversion toseismic record in time domain. The use of phase encoding method in frequencydomain, the frequency domain hybrid model of excitation acquisition forward, thefrequency of different frequency slices. By solving the Helmholtz equation,reverse-time migration method in frequency domain is derived. Reverse timemigration method based on frequency domain, the frequency domain least-squaresreverse-time migration technology. Based on the Marmousi model, the typicalrepresentative, through the method of mixing the offset frequency domain andachieved good results. Comparative analysis of the factors of the frequency domainleast-squares reverse-time migration effect and advantages and disadvantages.The above methods are based on the theory of data, and achieved very good results.But there are many problems, practical seismic exploration for example: there aredifferent in the same area does not match the seismic data, seismic deficiencies,marine detector drift impact on data acquisition. These factors restrict the accuracy ofseismic exploration, seismic events caused by discontinuous, the signal-to-noise ratiois low, low resolution. Especially in the area of seismic exploration for the two time,and therefore must be high precision matching filter.Using the high precision matching processing to extract valuable information is moreand more concerned by geophysicists. Widely used in new and old data matching,different source data matching, important field of4D seismic monitoring, matchedfiltering. The traditional matched filtering method by a number of limitations, difficultto overcome the influence of noise. Match processing of seismic data in the same areain different years or source, the energy and phase features matching. Then for thesubsequent processing of seismic data. When the underground structure is verycomplex, how to carry out high precision data matching seismic data processing. Howto get high accuracy of underground structure from seismic data after matching,geophysicist very attention and concern of the problem.Based on the above problems, this paper puts forward the optimal L1norm in thewavelet domain matching processing, amplitude, waveform, phase consistencyproblems have been solved. The wavelet decomposition of seismic data, matching the L1norm optimal extraction of effective signal, improve the quality of data, highfidelity, clear wave group character, reflects the rich geological information. Themethod of matching of different ages of seismic data acquisition, processing effectson the frequency, phase, amplitude and other aspects of the conventional method aresignificantly improved, and provide basic information is very useful for geologicalresearch. Because the Curvelet transform sparsity better, than the wavelet transform ismore precise, the article further in the Curvelet domain using L1’s Curvelet domain atfine scale seismic data matching norm, seismic data waveform consistency is veryhigh after matching, noise can be suppressed.In seismic exploration, the affected factors, other factors such as surface geophones,seismic data are often missing word. The seismic data of different degrees of deletionof Curvelet domain missing trace interpolation, especially in the absence of channelnumber, the interpolation effect is obvious. Interpolation with high accuracy and highsignal-to-noise ratio and matching results show the superiority of the method. Noisedata on different levels of deletion in the Curvelet domain seismic loss traceinterpolation, not only the effect of data interpolation is obvious and interpolation datais consistent with the waveform, amplitude spectrum and phase spectrum of higher.This paper matching interpolation of real seismic data, achieved satisfactory results.Seismic data acquisition is often mixed excitation absence question, the seismic dataacquisition of the mixed excitation matching interpolation, seismic data with highprecision is obtained.In-depth study of the main process of repeatedly produced. And the multiple wavemigration imaging technique is studied. Provides a new idea of multiple wavemigration technology for high precision seismic exploration. Through the research ofhybrid excitation acquisition technique, is performed with mixed seismic multiples.Migration of multiple wave based on, the paper focuses on the hybrid multiple wavemigration imaging. The simultaneous excitation of acquisition multiples (notincluding a reflected wave) mixed multiple wave migration imaging is tectonicconcave model. This dissertation focuses on the study of a wave in single shot recordand multiple joint migration imaging, using this method can improve the imagingaccuracy. On the basis of this study also triggers a wave data acquisition and multiplejoint imaging, and gives also triggers a wave acquisition and multiple joint imaging. Amore accurate geological model was obtained on a wave mixing collection recordsand multiple joint imaging.At the multi-source simultaneous (blended) acquisition data, there are more problemsin migration imaging, derived from the theory of existence of crosstalk noise blendedacquisition migration (crosstalk). Through the numerical model data, also confirmedthe existence of the phenomenon of crosstalk noise mixed in migration imaging.According to the characteristics of crosstalk noise exists, puts forward a new method of adaptive weighted median filter to suppress the crosstalk noise of migrationimaging. The method is based on the basis of the former study, multileveltwo-dimensional improved median filtering window using the weighted. The designof a weight function is applied to multilevel two-dimensional improved median filterwindow. Based on the simulated data are tested and compared with other noisesuppression methods, show that this method can effectively.