首页> 中文期刊> 《铁道科学与工程学报》 >基于动态贝叶斯模型的大悬臂展翅宽箱梁桥性能预测评估

基于动态贝叶斯模型的大悬臂展翅宽箱梁桥性能预测评估

         

摘要

Aiming at the wing-box-girder bridge with large-cantilever-wing,this paper presented the establish-ment of the dynamic Bayesian model for structure performance prediction by setting the maximum transverse stress of the roots of cantilever as the target layer and the live load ratio,the maximum negative temperature gra-dient,shrinkage strain as the sub-structure layer based on Dynamic Bayesian Network theory.The maximum transverse stresses of the roots of the cantilever at different times were obtained by use of ANSYS model.Through parametric studying,the model parameters were obtained.Then taking advantage of backward and forward rea-soning algorithm,transverse stress-time curve before updating was obtained.The accuracy of the model was veri-fied by comparing probability density distributions and the fitting curves of shrinkage,live load ratio,and nega-tive temperature gradient.Finally,how to combine the measured data and dynamic Bayesian model to update per-formance prediction of the structure was explained.%针对大悬臂展翅宽箱梁桥,基于动态贝叶斯理论,设定目标层为悬臂根部顶板最大横向应力,子结构层为活载比例大小、负温度梯度最大温差值、收缩应变,建立结构评估动态贝叶斯模型。利用ANSYS模型进行计算分析,得到横向悬臂根部顶板最大横向应力点各个时间段的横向应力值,进行参数学习,得到模型参数,采用前向后向算法进行推理,得到更新前的横向应力随时间变化曲线,通过对比分析收缩变形、活载比例、负温度梯度最大温差值概率密度分布与所设置的收缩变形、活载比例、负温度梯度最大温差值拟合曲线,可以验证模型的准确性。最后,说明了如何结合所测数据和动态贝叶斯模型来更新结构的性能预测。

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号