摘要
建立了基于RBF(径向基函数)网络模型的响应面方法,用以解决实际工程中常用的功能函数无法明确表达的可靠度计算问题,介绍了该方法的建立过程,并将其应用到膨胀土大坝的稳定性分析中,表明用RBF网络模型的响应面方法预测边坡稳定性是可行的。
Usually it is very difficult to get an analytical expression of the limit state equation in the reliability analysis for stability of the slope. So the response surface method based on the RBF （radio basis function） neural network is established to simulate the reliability problems with the closed formed performance function in the practical engineering. And the course of the establishment is put out detailedly. Finally, the method is proved by the using of the slope reliability analysis of the expansive soil dam.
出处
《土工基础》
2007年第5期56-58,共3页
Soil Engineering and Foundation
基金
湖北省自然科学基金资助项目(2006ABA237)
关键词
响应面法
RBF神经网络
稳定性分析
response surface, radio basis function neural network, slope reliability analysis
作者简介
任青阳，男，1975年生，博士后，武汉大学土木建筑工程学院，主要从事岩土本构关系数值计算、仿真和结构动力分析等工作。
