基于参数不确定性的预应力混凝土梁模型修正  

Model updating of prestressed concrete beams using parameters uncertainty

作　　者：房长宇[1,2] 张耀庭[1] 

机构地区：[1]华中科技大学土木工程与力学学院,湖北武汉430074 [2]广西工学院土建系,广西柳州545006

出　　处：《华中科技大学学报（自然科学版）》2011年第11期87-91,共5页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：贵州省交通厅基金资助项目(0231240305)

摘　　要：提出了基于参数不确定性的全预应力混凝土梁模型修正方法.灵敏度分析发现混凝土弹性模量和密度是影响梁的自振频率较大的参数.在贝叶斯理论中融入马尔可夫链-蒙特卡罗算法,实现了有限元分析程序和复合链抽样技术的协同工作,得到了模型参数后验分布的统计特征,并预测了梁的前三阶自振频率分布范围.数值分析表明:模型参数后验分布的标准差相对于先验分布有较大的改变,不同的观测数据源的计算参数后验估计存在差异;由本方法得到的前两阶自振频率中间位置与试验均值比较接近.Model updating method considering the uncertainty of structural parameters was proposed for full prestressed concrete beam.By sensitivity analysis,it was seen that the elastic modulus and density of concrete were important parameters to the beam′s natural frequencies.Markov chain-Monte Carlo（MCMC） algorithm was incorporated in Bayes method,and cooperation of finite element program and multiple chain sampling technology was realized,and statistical characteristics of structural parameters were obtained.Distribution ranges of the first three frequencies were predicted based on parameter estimation.Numerical simulation indicates that there are some changes in standard deviations of posterior distribution compared of those of prior distributions,and difference of each natural frequency used as witness resource exists when calculating the posterior estimation of structural parameters.In addition,as for the first two frequencies,the median values are in good agreement with the corresponding measured average values.

关 键 词：预应力混凝土 自振频率 模型修正 贝叶斯方法 马尔科夫链-蒙特卡罗法 不确定性估计 

分 类 号：TU378.2[建筑科学—结构工程]