RELIABILITY ANALYSIS OF EXISTING BRIDGE BASED ON NEURAL NETWORK-RESPONSE SURFACE

摘要：

Usually it is very difficult to get an analytical equation of limit state for reliability analysis of the existing bridge. Therefore, the paper puts forward a geometric analysis method of reliability analysis based on Neural Network-Response Surface. The method calculates the index β of reliability by using finite element numerical simulation, and has important implications for the reliability analysis of large and complex structures which can not to get an analytical equation of limit state. The numerical experiment shows that the neural network model can well fit the true limit state function and the method based on Neural Network-Response Surface has broad application prospects in the reliability analysis of existing bridges.

关键词： Neural network reliability response surface ezisting bridge

作者: Jian-Ying Chen Guo-Jing He

作者单位: School of Civil Engineering and Mechanics,Central South University of Forestry andTechnology,Changsh School of Civil Engineering and Mechanics,Central South University of Forestry and Technology,Changs

会议类型: 国际会议

会议名称: The 4th Internatiuonal Symposium on Lifetime Engineering of Civil Infrastructure(第四届土木工程结构生命周期国际学术研讨会)

会议地点: 长沙

会议语种:英文

页码: 444-447

在线出版日期: 2009-10-26（万方平台首次上网日期，不代表论文的发表时间）