An elastic wave full-waveform inversion method and apparatus

Abstract

The invention discloses an elastic wave full-waveform inversion method and apparatus. The method comprise the steps of establishing a convolution-type objective functional according to analog data and observation data; through adoption of an adjoint method based on the Lagrange theory, determining an adjoint equation of a time domain first order rate-stress elastic wave equation, adjoint epicenter items, a gradient expression of a convolution-type objective functional with regard to model parameters; determining a forward-propagating source wavefield according to the time domain first order rate-stress elastic wave equation; determining reverse-time extrapolation accompanying wavefield according to the adjoint equation and the adjoint focus items; determining the gradient of the convolution-type objective functional with regard to the model parameters according to the forward-propagating epicenter wavefield, the reverse-time extrapolation accompanying wavefield and the gradient expression; and carrying out multi-scale full-waveform inversion through utilization of low pass filtering characteristics of the convolution-type objective functional. According to the invention, influences by imprecise estimation of wavelets on inversion results can be eliminated, and problems that wavelets are difficult to accurately extract in practical data are effectively solved.

Description

technical field [0001] The invention relates to the technical field of petroleum seismic exploration velocity modeling, in particular to an elastic wave full waveform inversion method and device. Background technique [0002] The full waveform inversion method has high resolution and has great potential to improve the imaging quality. Since a single longitudinal wave cannot accurately simulate the propagation law of seismic waves in elastic media, and cannot explain and deal with issues such as shear wave conversion, in order to better describe the parameters of the underground medium and reconstruct the information of the underground structure, full elastic wave inversion is very important. important. The traditional elastic wave full waveform inversion method is based on the assumption that the wavelet is known. However, it is difficult to obtain accurate source wavelet in actual exploration. Inaccurate wavelet estimation will lead to phase mismatch between simulated data...

AI Technical Summary

Problems solved by technology

[0004] In summary, the current research on the wavelet problem of full waveform inversion mainly focuses on the inversion of acoustic wave waveforms in the frequency domain, and there is still a lack of research on the wavelet problem of full waveform inversion of elastic waves in the time domain

Images