Parametric inversion of viscoelastic media from VSP data using a genetic algorithm
Parametric inversion of viscoelastic media from VSP data using a genetic algorithm作者机构:Institute of Seismic Exploration Tsinghua University Beijing 100084 China
出 版 物:《Applied Geophysics》 (应用地球物理(英文版))
年 卷 期:2007年第4卷第3期
页 面:194-200页
核心收录:
学科分类:070801[理学-固体地球物理学] 07[理学] 0708[理学-地球物理学]
主 题:viscoelastic parameter inversion genetic algorithm VSP data
摘 要:Viscoelastic parameters are becoming more important and their inversion algorithms are studied by many researchers. Genetic algorithms are random, self-adaptive, robust, and heuristic with global search and convergence abilities. Based on the direct VSP wave equation, a genetic algorithm (GA) is introduced to determine the viscoelastic parameters. First, the direct wave equation in frequency is expressed as a function of complex velocity and then the complex velocities estimated by GA inversion. Since the phase velocity and Q-factor both are functions of complex velocity, their values can be computed easily. However, there are so many complex velocities that it is difficult to invert them directly. They can be rewritten as a function of Co and C∞ to reduce the number of parameters during the inversion process. Finally, a theoretical model experiment proves that our algorithm is exact and effective.