基于退化数据和离散动态贝叶斯网络的桥梁构件可靠性分析  

Reliability analysis of bridge components based on degradation data and discrete dynamic Bayesian networks

作　　者：陶明珠 TAO Mingzhu(Department of Basic Course Teaching,Economics of Hefei University,Hefei 2300ll,China)

机构地区：[1]合肥经济学院基础课教学部,安徽合肥230011

出　　处：《延边大学学报(自然科学版)》2025年第1期94-99,共6页Journal of Yanbian University(Natural Science Edition)

基　　金：安徽省科研计划编制项目(2024AH051547)。

摘　　要：为提高预测桥梁构件性能的准确性,构建了一个基于退化数据和离散动态贝叶斯网络的桥梁构件性能退化过程预测模型及安全性分析模型(BDS模型).BDS模型综合考虑了桥梁构件性能退化的动态演变过程和多因素的交互作用,并通过引入时变参数和时不变参数,以及利用条件概率表和“正向-反向”算法对桥梁性能的退化状态进行精确计算.研究表明:在监测的第10d,模型对2个监测点的预测精度分别达到了4.5和4.3MPa^(-2),显示出较高的预测精度.在第30d的监测中,2个监测点的极值抗弯承载力分别为32.36和32.34 kN·m,与实际值接近.结合实际监测数据对先验模型进行修正后,BDS模型预测得到的混凝土梁抗弯承载力在60a内优化了7.37%~37.93%.表明BDS模型能够有效地提高预测桥梁构件可靠性的精度,可为桥梁的性能退化预测提供良好参考.In order to improve the accuracy of predicting the performance of bridge components,a model for predicting the degradation process and analysing the safety of bridge components based on degradation data and discrete dynamic Bayesian network is constructed(bridge degradation and safety,BDS model).The BDS model takes into account the dynamic evolution of bridge component performance degradation and the interaction of multiple factors,and accurately calculates the degradation state of bridge performance by introducing timevarying and time-invariant parameters,as well as by using conditional probability tables and"forwardbackward"algorithms.Research shows the prediction accuracies of the model for the two monitoring points reach 4.5 MPa^(-2) and 4.3 MPa^(-2) at the 1oth day of monitoring,which show high prediction accuracies.In the 30th day of monitoring,the extreme bending capacity of the 2 monitoring points were 32.36 kN·m and 32.34 kN:m,which were close to the actual values.After correcting the a priori model with the actual monitoring data,the bending capacity of concrete girders predicted by the BDS model was optimised by 7.37% to 37.93% within 60 years.Thus the BDS model can effectively improve the accuracy of predicting the reliability of bridge members,and can provide a good reference for the prediction of the performance degradation of bridges.

关 键 词：桥梁构件 性能退化 动态贝叶斯网络 安全性分析 钢筋锈蚀 

分 类 号：U445.73[建筑科学—桥梁与隧道工程]