DOI:10.3969/j.issn.1003-5060.2024.10.017
基于 Kriging 模型的大跨度拱桥时变地震易损性分析
郑攀,何沛祥
(合肥工业大学土木与水利工程学院,安徽合肥230009)
摘要
文章以桥墩钢筋锈蚀机理为基础, 探究了基于氯离子环境侵蚀引起的钢筋直径及力学性能的退化规律, 同时考虑地震动和结构参数的变异性, 因分析计算量很大, 引入 Kriging 模型和拉丁超立方抽样 (Latin hypercube sampling, LHS) 法建立易损性曲线, 并利用 Monte Carlo 抽样方法形成全寿命周期内的桥梁系统损伤概率曲面。运用 OpenSEES 程序引入时变效应, 对某大跨度拱桥建立有限元模型。分析结果表明: Kriging 模型能够较精确地代替有限元进行分析, 能够显著减小计算量; 对于大跨度拱桥而言, 由 Monte Carlo 抽样得到的全桥体系易损性来评价桥梁的抗震性能更加合理; 在全寿命设计基准期内, 随着服役时间增加, 桥墩损伤概率明显增大。该研究可为类似桥梁结构全寿命抗震性能设计提供参考。
关键词
大跨度拱桥;Kriging模型;时变易损性;Monte Carlo抽样;体系易损性
中图分类号:U441.3
文献标志码:A
文章编号:1003-5060(2024)10-1404-08
Time-varying seismic vulnerability analysis of long-span arch bridge based on Kriging model
ZHENG Pan, HE Peixiang
(School of Civil and Hydraulic Engineering, Hefei University of Technology, Hefei 230009, China)
Abstract
This paper explores the degradation law of reinforcement diameter and mechanical properties caused by chloride erosion based on the corrosion mechanism of pier reinforcement and considers the variability of ground motion and structural parameters. Due to the large amount of analysis and calculation, Kriging model and Latin hypercube sampling (LHS) method are introduced to establish the vulnerability curve, and the method of Monte Carlo sampling is used to form the damage probability surface of the bridge system in the whole life cycle. OpenSEES program is used to establish finite element model for a long-span arch bridge, in which the time-varying effect is introduced. The analysis results show that Kriging model can accurately replace the finite element analysis, and can significantly reduce the calculation amount. For long-span arch bridges, it is more reasonable to evaluate the seismic performance of bridges by the vulnerability of the whole bridge system obtained by Monte Carlo sampling. During the life-cycle design reference period, the pier damage probability increases significantly with the increase of service time. The research can provide reference for the life-cycle seismic performance design of similar bridge structures.
Keywords
long-span arch bridge; Kriging model; time-varying vulnerability; Monte Carlo sampling; system vulnerability
收稿日期:2023-02-23
修回日期:2023-03-14
基金项目:合肥工业大学产学研校企合作资助项目(W2020JSFW0263)