山西大学学报（自然科学版）2025，Vol.48Issue(6)：1142-1151,10.DOI:10.13451/j.sxu.ns.2024031
针对剩余静校正的慢速扩展遗传算法
Slow Expansion Genetic Algorithm for Residual Static Correction
摘要
Abstract
In the residual static correction phase of seismic exploration,the genetic algorithm is susceptible to noise interference when processing low signal-to-noise ratio data and quickly falls into a local optimum.This paper proposes a probabilistic statistics-based slow expansion genetic algorithm(SEGA).The common middle point gathers are used as input.The seismic traces and wave-mixing model traces are correlated with each other to obtain the cross-correlation matrices of the shot and receiver points,respective-ly.The probability matrices of the static correction are obtained by using the proposed transformation function.Based on the value of the probability matrix,the population is initialized,crossed over,and mutated.The stacked section is updated by selecting the opti-mal chromosome.The slow expansion of the phase space is used to redefine the upper and lower limits of the correction.The above steps are repeated until convergence achieved.The experimental results show the proposed method can significantly improve the im-aging quality of stacked section on seismic data collected in a mountainous area.The stacked energy is improved by 6.61%and 5.55%compared with the maximum stacked energy method and genetic algorithm,respectively.关键词
互相关/最大能量法/模型道/相空间Key words
cross-correlation/maximum stack-power method/reference trace/phase space分类
地质学引用本文复制引用
胥可,钱坤,于洪,闵帆..针对剩余静校正的慢速扩展遗传算法[J].山西大学学报（自然科学版）,2025,48(6):1142-1151,10.基金项目
国家自然科学基金(62136002 （）
61876027) （）
南充市科技局项目(23XNSYSX0062 （）
23XNSYSX0013) （）
