| In seismic inversion,how to make full use of the kinematics and dynamics information of seismic data to obtain accurate results is a research hot spot in recent years.Conventional prestack AVA(amplitude variation with angle)inversion method obtains elastic parameters by using amplitude variation with offset,therefore it has higher accuracy than poststack inversion.However,the CRP gather after normal moveout(NMO)correction,as input data of prestack AVA inversion,which suffers from NMO stretching and distortion,results in the reduction of accuracy especially the density.The traveltime in seismic data only depends on the velocity structure of underground medium and has the characteristics of high signal-to-noise ratio.In order to solve the problem of stretching introduced by NMO correction and make full use of the kinematics information simultaneously,we focus on the role of traveltime in inversion and joint AVA inversion method with traveltime.Firstly,we compare and analyze the results of different input data through the particle swarm optimization(PSO),which proves that the traveltime information can effectively improve the stability and accuracy of inversion and reduce the dependence on the initial model to some extent.Secondly,we respectively test the performance of the AVA inversion when the initial model is acquired form the traveltime information and the root mean square velocity penalty term is added to the object function.Then we put forward the AVA inversion method before NMO correction.The proposed method avoids the impact of stretching and distortion introduced by NMO correction and makes full use of the traveltime information as well.Finally,inspired by the research idea of traveltime AVA inversion,we proposed a target-oriented NMO correction method based on sparse inversion for more effective subsequent processing and interpretation.Some examples are tested to prove the validity and stability of the proposed method. |
PDF Full Text Request