摘要
针对有限元蒙特卡罗法计算量大的弊端和边坡工程稳定功能函数不能显式表达的可靠度分析问题,文章提出了应用于边坡工程稳定可靠度分析的基于神经网络的随机有限元分析法的基本思路。神经网络具有高度非线性的映射能力,可用来逼近结构响应量与随机变量的映射关系。借助这种方法计算出边坡工程的可靠指标,并通过寻求失稳模式进行整体稳定可靠度分析。最后对一工程实例进行几种方法的计算分析,表明该方法用于边坡整体稳定可靠度分析是切实可行的,可供工程科学技术人员参考。
In reliability analysis of the slope stability, due to the disadvantage of excessive computation of the finite element Monte Carlo method and the problem of impossible explicit expression by slope function, the basic idea about application of the Monte-Carlo finite element method using artificial neural networks（ANN） to analyze the reliability of slope stability is presented in this paper. ANN is used to approximate the implicit mapping between structural response variables and basic random variables. Then, through calculating the corresponding reliability index and seeking the instability pattern, the whole reliability of the slope stability is analyzed. Finally, the feasibility of this method used to analyze the whole reliability of the slope stability is indicated by one practical example calculated and analyzed using several methods, which can serve as reference to the scientific and technical professionals.
出处
《地下空间与工程学报》
CSCD
2006年第4期578-581,共4页
Chinese Journal of Underground Space and Engineering
关键词
边坡稳定
蒙特卡洛模拟
有限元
神经网络
可靠度分析
slope stability
Monte-Carlo simulation
finite element
artificial neural networks
reliability analysis
