摘要
基于均匀设计、有限元法、人工神经网络和免疫遗传算法建立了新的岩质边坡结构面参数的反演方法.按照均匀设计要求,确定数值模拟方案;用有限元程序计算出相应的神经网络训练样本,建立边坡变形的神经网络预测模型,再利用免疫遗传算法进行反演分析,其中反演过程适应度的计算则采用已训练好的神经网络预测来替代有限元数值仿真,大大缩短了计算时间.通过实际工程的算例分析,反演结果比较理想.
A new backward analysis method for mechanical parameters of fault surface in slope was developed based on uniform testing design, finite element method, artificial neural network and immunization-genetic algorithm. According to uniform testing design, the value levels of the mechanical parameters were chosen, and simulation schemes were arranged; the related analytical samples for neural network were given by FEM calculations. Thus, a BP neural network which was used to forecast displacement of the slope＇ s character points was erected and trained. The physical and mechanical parameters could be analyzed backwards by immunization-genetic algorithm. In this algorithm the trained BP neural network was used to calculating the fitness value instead of the FEM method and the calculation time was much reduced. An actual engineering using immunization-genetic algorithm to solve the mechanical parameters of fault surface was analyzed. The result is reasonable and the backward analysis method is feasible and accurate.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2008年第9期977-982,共6页
Journal of China Coal Society
关键词
边坡
反分析
免疫遗传算法
人工神经网络
slope
back analysis
immunization-genetic algorithm
artificial neural network
