基于分布式并行遗传算法的相对波阻抗反演方法  

Relative wave impedances inversion based on distributed parallel genetic algorithm

作　　者：黄进腊[1] 李庆春[2] 樊婷婷[1] 

机构地区：[1]西北大学大陆动力学国家重点实验室/地质学系,陕西西安710069 [2]长安大学应用地球物理研究所,陕西西安710054

出　　处：《西北大学学报（自然科学版）》2010年第5期855-859,共5页Journal of Northwest University（Natural Science Edition）

基　　金："863"国家高科技计划基金资助项目(2005AA615010)

摘　　要：目的研究地震相对波阻抗反演的优化方法。方法比较遗传算法二进制编码、实数编码在最优个体保存策略、随机均匀分布选择、分散交叉、高斯变异和分布式并行方案下的全局寻优能力。结果实数编码分布式并行遗传算法优于基本遗传算法,用该方法对数值模型和实际地震剖面进行波阻抗参数反演,验证了分布式并行遗传算法的有效性、优越性。结论基于分布式并行遗传算法的相对波阻抗反演,为遗传算法在地震属性的优化研究中提供了一种改进思路。Aim To improve the relative wave impedance inversion precision of seismic attribute optimization.Methods Compare the ability of global optimized between binary code and real code of genetic algorithm based on the genetic operation,which was the best optional optimization preserving strategy,stochastic uniform selection,scattered crossover,Gaussian mutation and distributed parallel genetic algorithm.Results Real-coded distributed parallel genetic algorithm was better than simple genetic algorithm.It brought forward global optimized wave impedances inversion founded on parallel genetic algorithm and checked out the effectiveness of parallel genetic algorithm by using numerical model and the practical seismic section.Conclusion The method of wave impedances inversion can provide an improved train of thought on seismic attribute optimization based on genetic algorithm.

关 键 词：分布式并行遗传算法 群迁移 相对波阻抗反演 实数编码 

分 类 号：P631[天文地球—地质矿产勘探]