摘要
文章研究结构可靠度敏感性,提出参数的相对敏感性分析方法,并基于该方法提出2种智能响应面法分析边坡稳定性。具体思路:由可靠指标对随机变量分布参数的相对敏感性分析,确定边坡可靠度主要影响参数;用2种智能响应面的网络模型近似替代响应量与基本变量间的隐式极限状态函数,根据蒙特卡罗模拟法,对网络模型进行可靠度分析,求解结构可靠度指标。基于可靠度敏感性的智能响应面法对均值和成层的边坡算例进行稳定性分析,并与传统响应面法相比较,结果表明:该方法分析边坡稳定性是准确的且具有较高的计算效率,并可以有效解决传统响应面法可靠度计算不收敛问题;且2种智能响应面法计算结果几乎相同,可以互为验证,确保计算结果的准确性。
An analysis of slope stability using two kinds of intelligent response surface methods is presented in this paper based on the relative sensitivity analysis of the structural reliability parameters.Specific ideas are as follows： to access the main influencing parameters of the slope reliability based on the relative sensitivity analysis of the random variable distribution parameters with reliability index;to replace the implicit limit state function of response and basic variables by the network models of two kinds of intelligent response surfaces;to analyze the reliability of the network models and solve the structural reliability index based on the Monte Carlo simulation method.The stability analysis of mean and stratified slope numerical examples is carried out by using the intelligent response surface method based on the sensitivity of reliability.Compared with the traditional response surface method,the results show that the presented method is accurate and has higher efficiency,and it can effectively resolve the non-convergence problem of the traditional response surface method for reliability analysis.The results of two kinds of intelligent response surface methods are almost the same and can verify each other,so the accuracy of the results can be ensured.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
北大核心
2011年第8期1196-1201,共6页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(40972194)
安徽高校省级自然科学研究计划资助项目(KJ2010B267)
关键词
可靠度
敏感性分析
智能响应面法
网络模型
reliability
sensitivity analysis
intelligent response surface method
network model
