Study On Noise Suppression And Reverse Time Migration For Virtual Shot Gathers In Passive Seismic Exploration

In recent years,with the rapid development of seismic interference technology,passive source seismic exploration methods have received increasing attention from researchers.Although passive seismic data usually have disadvantages such as low frequency,weak energy,and low signal-to-noise ratio,the the virtual shot gathers reconstructed reconstructed by seismic interference technology still contains much effective information reflecting underground structures.Therefore,passive seismic exploration can be a low-cost supplementary method to active seismic exploration in target areas where active seismic exploration conditions are difficult.Compared with passive seismic surface wave exploration,passive seismic reflection wave exploration has more potential in velocity information extraction,exploration depth,and highresolution imaging.Therefore,it is of good theoretical research significance and wide application prospects to carry out relevant research on passive seismic reflection wave exploration.The virtual shot gathers are similar to active shot gathers,which can be processed by conventional reflection seismic data processing methods to enable passive seismic reflection wave exploration.The migration is an effective method to obtain imaging of subsurface structures,and among many migration algorithms,reverse time migration has the advantages of being unrestricted by dip angle,adapting to lateral strong velocity variations,and imaging accurately for complex velocity structures.However,it is often difficult to obtain satisfactory imaging results by directly performing conventional reverse time migration on the virtual shot gathers.In order to improve the imaging quality of reverse time migration of passive seismic data,we first discuss the basic principle of the passive seismic interferometry technology and reverse time migration.Then,during the experimental process of passive seismic data interferometry,we discusse the influencing factors of virtual shot gathers reconstruction and further analyzes the problems of conventional passive seismic reverse time migration.The results show that,the quality of virtual shot gathers is the key to achieving passive source reflected wave exploration,and reverse time migration imaging of passive seismic data mainly has some problems,such as low signal-to-noise ratio,weak deep energy,low resolution and serious low-frequency migration artifact.So,we propose targeted solutions from the perspectives of passive source noise suppression,deep imaging amplitude energy recovery,high-resolution imaging conditions and migration low-frequency artifact suppression.Finally,the proposed filtering method,imaging condition and low-frequency suppression algorithm are further improved in the process of passive seismic data application to obtain a more complete set of reverse time migration method for passive seismic data.The details of the research can be summarized as follows:Passive seismic noise is a kind of complex noise with strong coherence,which is generated by the same mechanism as the effective signal,both of which are generated by seismic interference reconstruction,and it is difficult to achieve a better suppression effect by conventional filtering methods.In this paper,we proposed a passive seismic noise attenuation method based on local similarity coefficients.The characteristics of the effective signal and noise in the virtual shot gather are analyzed by a multi-channel comparison method.It is found that there is a significant similarity difference between the effective signal and the noise in the common shot gathers and the common receiver gathers.Based on the local similarity coefficient spectra obtained from the calculation,a weighted filtering operator with local similarity coefficients as the core is constructed.In the filtering operator,the weighting coefficients of the effective signal are larger,while the weighting coefficients of the noise are smaller.Through the threshold processing of the weighting coefficients,the noise can be attenuated and the effective signal can be extracted.Using the focal transformation can further improve the filter operator’s ability to suppress the coherent noise,and the effective signal has better separability from the noise in the focal domain.In this paper,a specific iterative denoising process is designed by continuously updating the filter operator and denoising results alternately,and satisfactory denoising results can be obtained with only a few iterations of calculation.Through the numerical experiment of theoretical data and field data,it is proved that the passive seismic noise attenuation method based on local similarity coefficients proposed in this paper can effectively suppress the passive seismic noise,so as to obtain high-quality virtual shot gathers.In addition,the filtering method in this paper uses the similarity difference between the effective signal and the noise for noise suppression,which is more consistent with the signal characteristics of the passive seismic data.Therefore,it is more suitable for application in passive seismic data compared with conventional filtering methods.There are the problems of weak deep energy and low resolution in reverse time migration imaging results of passive seismic data.Therefore,we use the excitation amplitude imaging condition with the advantage of high resolution for imaging in reverse time migration,and further solves the problem that the conventional excitation amplitude imaging condition has poor imaging capability for complex structures,and finally proposes a global normalized multi-value excitation amplitude imaging condition suitable for passive seismic data imaging.The excitation amplitudes obtained by the constraints of amplitude threshold,propagation travel time and propagation direction are more consistent with the definition of excitation criterion and can effectively guarantee the accuracy of imaging results.The introduction of multi-valued excitation amplitudes into the imaging condition can overcome the shortcomings of conventional excitation amplitude imaging condition that is weak in handling multiple paths of wavefield propagation.The global source normalization process after rewriting the ratio form of the imaging condition into the correlation form can not only obtain the reflection coefficient imaging results,but also overcome the problem of low stability when the imaging condition is applied in the passive seismic reverse time migration,and enhance the deep imaging energy.By using the new imaging condition,the passive seismic reverse time migration result in complex models such as Marmousi and Sigsbee2 a can obtain higher resolution imaging results than cross-correlation imaging conditions and better describe the fine structure of complex models.The effective suppression of low-frequency artifacts in migration results can further improve the quality of passive seismic reverse time migration imaging.The full wavefield decomposition algorithm based on Hilbert transform is improved by using the directional characteristics of excitation amplitude and low memory consumption,and a wavefield decomposition algorithm based on the receiver wavefield at excitation time is proposed,which avoids the problem of extra wavefield extrapolation due to the construction of the the analytic time wavefield.We verify that the excitation amplitude is a noise-free downgoing source wavefield imaging information through theoretical analysis and numerical simulation,and the low-frequency artifacts in the imaging are caused by the backward reflection of the receiver wavefield.Combining the time-bin of excitation amplitude,the sparse storage of wavefield,and the windowed Hilbert transform,we implement a wavefield decomposition algorithm based on the receiver wavefield at excitation time to obtain the upgoing wavefield imaging information.Numerical examples show that the proposed wavefield decomposition algorithm used in excitation amplitude imaging can effectively suppress the low-frequency noise in the migration results,and further improve the reverse time migration imaging results of passive seismic data.