论文部分内容阅读
随着并行计算技术的发展,非线性反演计算效率在不断提高,但对于基于单点搜索的非线性反演方法,其并行算法的实现则是一个难题。本文将群体搜索的思想引入到基于单点搜索的非线性反演方法,构建了并行算法,以量子蒙特卡罗方法为例进行了二维地震波速度反演及实际资料波阻抗反演,并测试了使用不同节点数进行计算的效率。计算结果表明:该并行算法在理论和实际资料反演中是可行的和有效的,具有很好的通用性;算法计算效率随着使用节点数的增加而提高,但算法计算效率的提高幅度随着使用节点数的增加逐渐减小。
With the development of parallel computing, the efficiency of non-linear inversion is increasing. But for the nonlinear inversion based on single point search, the parallel algorithm is a difficult problem. In this paper, the idea of group search is introduced into the non-linear inversion method based on single point search, and a parallel algorithm is constructed. The 2D Monte Carlo method is used as an example to perform 2D seismic velocity inversion and actual data wave impedance inversion. The efficiency of using different number of nodes for calculation. The calculation results show that the parallel algorithm is feasible and effective in the inversion of theoretical and practical data and has good generality. The computational efficiency of the algorithm increases with the increase of the number of nodes used, but the efficiency of the algorithm increases with The increase in the number of nodes used gradually decreases.