基于LS-SVM的结构可靠度评估  

An Assessment of Structural Reliability by Using LS-SVM

作　　者：陈铁冰[1] 

机构地区：[1]深圳职业技术学院建筑与环境工程学院,广东深圳518055

出　　处：《中国农村水利水电》2009年第9期130-132,136,共4页China Rural Water and Hydropower

基　　金：广东省自然科学基金(06028131)

摘　　要：针对大型复杂结构极限状态方程一般难以显式表达的特点,提出了基于最小二乘支持向量机(the leastsquare support vector machine,LS-SVM)的结构可靠度评估方法。该方法采用均匀抽样法抽取随机变量样本,应用确定性有限元求解器进行数值计算。将样本数据进行训练,利用最小二乘支持向量机建立随机变量与结构响应之间的非线性映射关系,模拟结构极限状态方程。通过计算极限状态方程值和偏导数值,求解优化问题,计算结构可靠指标。结果表明,该方法能够评估隐式极限状态方程的结构可靠度,具有较高的计算精度和较好的计算效率。An approach assessing the structural reliability by using the least square support vector machine （LS-SVM） is proposed in this paper when implicit limit state functions are normally encountered in the complicated structures. Random variables such as material properties, physical dimensions and loads are sampled by uniform sampling. The LS-SVM can be trained by using a small set of numerical values obtained from a deterministic finite element analysis of structures and sample data mentioned above. The trained LS -SVM can map the structural responses and random variables and the limit state functions of structures can be approximated by using the LS-SVM. Then the values and partial derivatives of the implicit limit state functions can be calculated. So the reliability index of structures can be calculated by solving an optimization problem. Numerical examples are given. The results show that the proposed approach is applicable to evaluate the structural reliability involving implicit limit state functions. The proposed approach is accurate and efficient.

关 键 词：结构可靠度 最小二乘支持向量机 隐式极限状态方程 

分 类 号：TB114.3[理学—概率论与数理统计] TP181[理学—数学]