(1 云南大学建筑与规划学院， 昆明 650500； 2 浙江大学建筑工程学院， 杭州 310058)

［摘要］随着服役龄期的增加，既有建筑结构的力学性能在自然条件下会呈现出衰退趋势，形成耐久性损伤。结构耐久性研究存在时间跨度长、影响因素复杂、不确定性大等问题，目前尚未有成熟的随机理论成果对其准确预测，基于数据监测更新的随机模型在一定程度上为该问题的解决提供了新的思路。基于材料实测数据驱动的贝叶斯信息更新，发展了概率密度演化理论基础上的结构易损性分析方法，从而实现对结构实时动态性能的准确预判。考虑材料性能参数的随机性，以某7层RC框架结构为例，在5个地震动条带中均匀选取满足场地条件的20条地震动记录，综合考虑结构自身参数与外界激励的不确定性，形成6个目标龄期、4种性能水平下的地震易损性曲线，进而生成考虑龄期连续变化的时变易损性曲面，证明所提方法的工程实用性。 

［关键词］材料性能退化；RC框架结构；实测数据驱动信息更新；概率密度演化理论；时变易损性 

中图分类号：TU375 文献标识码：A文章编号：1002-848X(2021)15-0054-08

 Timedependent vulnerability analysis of RC frame structures based on material parameter updating

YANG  Sizhao1,  WANG  Xianjie1,2 

(1 School of Architecture and Urban Planning, Yunnan University, Kunming 650500， China；2 College of Civil Engineering and Architecture, Zhejiang University, Hangzhou 310058， China) 

Abstract: With the increase of curing age, the mechanical properties of the existing structure show a tendency of decay in the atmospheric environment that forming durability damage. There is no mature stochastic theory achievement to predict it accurately at present because the research on the structures durability has problems such as long time span, complex influencing factors and large uncertainly, etc. The random model based on data monitoring and updating provides a new idea for solving this problem to some extent. The Bayesian information updating method based on measured data of material was proposed, the structural vulnerability analysis method was established based on the probability density evolution method (PDEM), thus the real-time dynamic performance of structures was accurately predicted. A 7-layer RC frame structures was taken as an example, and the 20 ground motion records of satisfying the site conditions were uniformly selected in the 5 seismic belts to comprehensively considering the randomness of structural parameters and external excitation. Seismic vulnerability curves with 6 target ages and 4 performance levels were formed, and then the time-dependent vulnerability surfaces with the continuous change of curing age were established. The engineering practicability of the proposed method was verified. 

Keywords:material performance degradation； RC frame structure； information updating driven by measured data； probability density evolution method； time-dependent vulnerability

*广西防灾减灾与工程安全重点实验室系研究资助项目（2016ZDK009，2016JYB009），云南省教育厅科学研究基金资助项目（2018Y07）。

 作者简介：杨思昭，硕士，Email:yangsz0616@yeah.net；通信作者：王宪杰，博士，硕士生导师，Email: xianjiewang@ynu.edu.cn。