|
个人信息Personal Information
教授
博士生导师
硕士生导师
任职 : 工业装备结构分析优化与CAE软件全国重点实验室主任、结构优化理论与应用国际联合研究中心主任
性别：男
毕业院校：大连理工大学
学位：博士
所在单位：力学与航空航天学院
学科：工程力学. 计算力学. 结构工程. 车辆工程
办公地点：力学楼302
联系方式：0411-84707267 ligang@dlut.edu.cn
电子邮箱：ligang@dlut.edu.cn
A hybrid sequential approximate programming method for second-order reliability-based design optimization approach
点击次数：
论文类型：期刊论文
发表时间：2017-05-01
发表刊物：ACTA MECHANICA
收录刊物：SCIE、EI、Scopus
卷号：228
期号：5
页面范围：1965-1978
ISSN号：0001-5970
摘要：The second-order reliability method (SORM) is an effective tool to evaluate the reliability, but its application for reliability-based design optimization faces unbearable computational cost. In this study, a new hybrid sequential approximate programming (HSAP) method is developed to calculate the optimum efficiently by developing a distance-checking criterion and a convex approximate method. Since the distance-checking criterion identifies the feasibility of the probabilistic constraint effectively, the proposed method combines the efficiency of the sequential approximate programming method and the accuracy of SORM. The convex approximate method is also constructed using the sensitivity and function value of the probabilistic constraint. So no additional computational cost is required in the optimization process. Five illustrative examples, including two mathematical examples and thee practical engineering examples, demonstrate the efficiency and accuracy of the proposed HSAP.