摘要
笔者在前人工作基础上提出的等厚薄层权重自适应迭代阻尼最小二乘法 ,能有效地解除对初始模型的依赖。理论模型与实际资料试算结果表明 ,用该算法对瑞雷波频散曲线进行反演 ,具有稳定性好、精度高、分辨能力强的特点 ,并能自动分层和反演地层参数。
The self-adaptively iterative damping least square method is proposed in this paper by combining equal thinner layers and weighting matrix. The new efficient algorithm can overcome such serious shortcomings existent in the classic least square method as the strict dependence on initial model, the remarkable man-made error, low accuracy, and the bad convergence velocity, which have a great adverse effect on the inversion of dispersion curves of Rayleigh wave. The inversion results of the new theoretical model and field dispersion data indicate that this inversion is characterized by good stability, high accuracy and strong resolution, and can automatically inverse shear velocity and thickness.
出处
《物探与化探》
CAS
CSCD
2003年第3期212-216,共5页
Geophysical and Geochemical Exploration
基金
"九.五"国家重点科技攻关项目 (NO .96 - 2 2 1- 0 1 -0 2 0 5 )
长江水利委员会综合勘测局科技发展基金项目 (NO .KY2 0 0 0- 4 )联合资助
关键词
频散曲线
自适应迭代阻尼因子
最小二乘法
权重矩阵
反演算法
dispersion curve
self-adaptively iterative damping factor
least square method
weighting matrix
inversion algorithm
