摘要
采用旅行时线性/二次联合插值射线追踪算法(QLTI算法)得到各向异性地层的P波井间层析成像旅行时,在地层垂向速度和各向异性参数反演中引入了小生境遗传算法,采用共享函数完成种群进化过程中个体的选择操作,更好地维护了种群中个体的多样性,在一定程度上能避免陷入局部极值。数值算例表明,该方法能细致地描述地层各向异性信息,反演精度高、稳定性好。多尺度联合反演使得小炮检距下也能得到精度较高的各向异性参数。
The Quadratic/Linear Traveltime Interpolation （QLTI） ray tracing method is adopted to get the P-wave traveltime of crosswell seismic tomography in this paper. The Niche Genetic Algorithm is applied to the inversion of vertical velocity and anisotropic parameters. The sharing function is used in the population evolutionary process to select the individuals, which can well maintain the individual variety and keep off trapping in local minima to some extent. The numerical simulation shows that the precise anisotropic parameters and high stabilization can be obtained by using this inversion method. The multi-scale inversion can improve the precision of inverted results even in short offset.
出处
《工程力学》
EI
CSCD
北大核心
2006年第A01期58-61,共4页
Engineering Mechanics
基金
国家自然科学基金资助项目(10272064)
高等学校博士学科点专项科研基金资助课题(20030003061)
关键词
井间地震
反演
小生境遗传算法
射线追踪
各向异性
crosswell seismic tomography
inversion
Niche Genetic Algorithm
ray tracing method
anisotropic
