基于BDNM的桥梁结构可靠度预测

发布时间：2018-06-04 20:40

本文选题：桥梁健康监测 + 应力　；参考：《哈尔滨工业大学学报》2014年02期

【摘要】：为了结合监测极值应力和应力参数(平均值)的先验模型来对桥梁可靠度进行预测,认为极值应力随时间变化的动态测量为一个时间序列,并考虑到贝叶斯动态线性模型(BDLM)的局限性,引入贝叶斯动态非线性模型(BDNM)对时变极值应力进行预测.运用BDNM建立了极值应力的状态方程和监测方程,通过泰勒级数展开技术,将其近似转化为贝叶斯动态线性模型(BDLM),并通过贝叶斯因子来对应力信息进行监控,然后结合应力参数的先验信息,对极值应力的状态参数进行贝叶斯后验概率推断,建立动态模型对极值应力变化趋势进行预测.基于监测信息,考虑到变量估计主观认识的不确定性,引入折扣因子来确定状态误差方差.最后利用建立的BDNM和一次二阶矩(FOSM)可靠度方法,对结构可靠度进行预测,并通过实例验证了所建模型的合理性和适用性.
[Abstract]:In order to predict the bridge reliability with a priori model for monitoring extreme stress and stress parameters (mean value), it is considered that the dynamic measurement of extreme stress with time is a time series. Considering the limitation of Bayesian dynamic linear model (BDLM), the Bayesian dynamic nonlinear model (BDNM) is introduced to predict the time-varying extreme stress. The state equation and monitoring equation of extreme stress are established by using BDNM. By Taylor series expansion technique, it is approximately transformed into Bayesian dynamic linear model, and the stress information is monitored by Bayesian factor. Based on the prior information of stress parameters, Bayesian posteriori probability inference is carried out on the state parameters of extreme stress, and a dynamic model is established to predict the trend of extreme stress change. Based on monitoring information and considering the uncertainty of subjective cognition of variable estimation, a discount factor is introduced to determine the variance of state error. Finally, the reliability of the structure is predicted by using the established reliability method of BDNM and the first order second order moment FOSM, and the rationality and applicability of the model are verified by an example.
【作者单位】：哈尔滨工业大学土木工程学院;
【基金】：国家自然科学基金资助项目(50678057)
【分类号】：U441

【参考文献】