复杂地形条件下静校正的综合寻优  

HYBRID OPTIMIZATION OF STATIC ESTIMATION IN COMPLEX TOPOGRAPHY

作　　者：林依华[1] 张中杰[1] 尹成[2] 周熙襄[3] 

机构地区：[1]中国科学院地质与地球物理研究所,北京100101 [2]西南石油学院物探系,四川南充637001 [3]成都理工学院物探系,成都620061

出　　处：《地球物理学报》2003年第1期101-106,共6页Chinese Journal of Geophysics

基　　金：国家自然科学基金杰出青年基金项目 (4982 51 0 8) .

摘　　要：复杂地形 ,特别是山地条件下地震勘探中的大静校正问题是一个非线性的、具有多参数多极值的全局优化难题 .本文综合利用最大能量法、模拟退火与遗传算法的各自优势 ,提出一种具有局部收敛速度快、全局搜索能力强的综合寻优反演方法 .其中对产生伪静校正量的零空间现象进行空间平滑滤波剔除 .理论数值计算及实际资料的处理说明本方法适应能力强、静校正效果较好 .For the areas of complex topography, especially mountain area, static is a non-linear hybrid optimization problem with the presence of multi-parameters and multi-extremes. A hybrid optimization inversion method is developed to calculate large static by integrating stack power maximization, stimulated annealing and genetic algorithms. This method has a series of advantages, such as high-speed regional convergence and efficient global searching ability. A special smooth filtering operator is used to eliminate the pseudo-static null space phenomenon. Checking with synthetic and real seismic data shows that this method is adaptable and effective to process seismic data acquired in the areas with complex tomography.

关 键 词：山地勘探 静校正 零空间现象 地形 地震勘探 遗传算法 

分 类 号：P631.4[天文地球—地质矿产勘探]