摘要
详细地介绍了基于遗传算法改进的BP神经网络,根据围岩变形量的时序变化特性,对上马基隧道围岩变形进行了进化神经网络预测,得出了用GA-BP神经网络建立的模型分析方法是隧道变形预测中的可行、简易、有效的方法的结论。
The article introduces BP neural network improved based on genetic algorithm, carries out evolutionary neural network forecast on wall reck deformation in Shangmaji tunnel according to time-interval changing characteristics of wall reck deformation, and draws a conclusion that the model analysis method build by GA-BP neural network is a feasible, simple and effective method in tunnel deformation forecast.
出处
《山西建筑》
2008年第4期364-365,共2页
Shanxi Architecture
关键词
隧道
遗传算法
围岩变形
神经网络
tunnel, genetic algorithm, wall reck deformation, neural network
