Study On Methods Of Multiple Separation And Multiple Imaging

Compared to primaries,multiples with small reflection angle,longer travel time,wider covering range,can illuminate the shadow zone where primaries can't reach,thus they contain more extensive information about the subsurface,so study on methods of multiple imaging are especially important to the quality improvement of the structure image under the ground.However,multiple separation is the key to guarantee the precision of multiple imaging.Therefore,the study on methods of multiple separation and multiple imaging discussed in this paper has important research significance.Firstly,detailed knowledge of the related theory about multiples are given,then the research on multiple modeling are done,which can provide data foundation for the later research on the multiple separation and multiple imaging in this paper.During the study of multiple separation,deep research has been done based on the widely used Radon transform and surface-related multiple elimination(SRME),and several practical methods of multiple separation are developed.(1)a method of?-f domain high-resolution Radon transform is developed for multiple separation,it can significantly improve the computational efficiency at the same time ensure the accuracy of separation of multiples and primaries.(2)Based on the theory of SRME and focal transform,a method for separating primaries and different-order multiples are proposed in this paper,multiple image with high precision can be obtained by using the separated primaries and different-order multiples.(3)By the derivation of focal transform,an important conclusion can be found that the focal transform is essentially a process of multidimensional wavelet deconvolution,thus the new data recovered from multiples in the focal domain will have higher resolution than the original data,by using the non-stationary regression adaptive match filtering with sharping regularization to separate the multiples,high resolution data can be obtained.Form the point of improving resolution by multiples,this method has important research value in improving the quality of image.While during the study of multiple imaging,in view of the complexity of structure under the ground,intensive study on the multiple imaging methods respectively based on one-way wave-equation migration and reverse time migration has been done,and a series of feasible technical scheme has been proposed.(1)In the case of simple structure under the surface,so as to reduce the cost,we can use one-way wave-equation migration to realize the multiple imaging,by modifying the boundary condition combining with the frequency-space domain one-way wave-equation migration,multiple imaging can be completed;in order to remove the crosstalk artifacts generated during the migration,in this paper,the stereographic imaging condition is introduced to suppress the cross-correlation energy from unrelated events,which is completed by using local slope together with time to constraint the imaging,the feasibility of this method is verified by the model test,however,the practicability is still an open question.(2)Considering that the multiple separation may be difficult to implement or will cost huge computation cost under some certain conditions,to solve this problem,by theoretical derivation,an improved joint migration of primaries and multiples based on one-way wave-equation migration is presented in this paper.Extract wavelet from the seismic data first,then use the extracted wavelet and the recorded data as the source wavefield,use the recorded data as the receiver wavefield,through stereographic imaging condition,the joint migration of primaries and multiples can be realized,in this method,there is no need to separate multiples,and the multiples can supplement the primary image directly,but the extraction of wavelet also put forward new requirements for us.(3)In the case of complex structure under the surface,the multiple one-way wave-equation migration can't meet the demand of imaging accuracy for complex structure,therefore,the study on the reverse time migration of multiples is particularly important.Compared to the multiple one-way wave-equation migration,the multiple reverse time migration can better realize the image of complex structures,get more information about the subsurface,and improve the image precision;in order to suppress the crosstalk artifacts generated by the cross-correlation of unrelated events during the multiple reverse time migration,on the basis of analysis about the mechanism of the crosstalk noise,we propose a new method of multiple reverse time migration,firstly imaging the different-order multiples which are separated based on focal transform and SRME,then stack the images of different-order multiples to obtain the final image result,compared to the conventional multiple reverse time migration,this method can effectively remove the artifacts in the image result,improve the image accuracy of underground structures.(4)For the energy of internal multiples are extremely weak,so in the study of multiple imaging,only the migration of surface-related multiples are considered,however,the internal multiples are also real reflections from the subsurface like the surface-related multiples.Considering that the reverse time migration can image special waves,the reverse time migration for internal multiples is discussed in this paper,and it is concluded that reverse time migration can imaging the internal multiples,after further analysis,an important conclusion that the surface-related multiples containing internal multiples can also be correctly imaged by multiple reverse time migration are finally obtained.Combing the study on the multiple separation and multiple imaging,the feasibility,validity and practicability of these studied methods are finally reasonably verified by model examples and real data processing.