A model for bridge vehicular flow based on vehicular inflow and load effect measurements

摘要：

　　Vehicular loading is one of the key factors in bridge reliability and life-cycle assessment.It can be decomposed into two components: vehicular inflow and flow.The vehicular inflow component describes the information about each vehicle passing through the weigh-in-motion(WIM)system.The vehicular flow component describes the information about the real-time position and velocity of each vehicle in the bridge.In practice,the vehicular inflow component is measured by a bridge weigh-in-motion(WIM)sensing system,while the flow component cannot be directly measured.Based on the vehicular inflow and load effect measurements,this paper attempts to utilize the time-delay neural network(TDNN)to construct a surrogate model for the vehicular flow.It turns out that the TDNN achieves satisfactory performance in vehicular flow modelling.

关键词： Bridge vehicular load effect Machine learning Nagel-Schreckenberg model Time-delay neural network Weigh-In-Motion

作者: He-Qing Mu Stephen Wu Hou-Zuo Guo Tian-Yu Zhang

作者单位: Associate Professor,School of Civil Engineering and Transportation,State Key Laboratory of Subtropic Assistant Professor,The Institute of Statistical Mathematics,Tokyo 190-8562,Japan Student,School of Civil Engineering and Transportation,South China University of Technology,Guangzho

会议类型: 国际会议

会议名称: The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)

会议地点: 青岛

会议语种:英文

页码: 1870-1874

在线出版日期: 2018-07-22（万方平台首次上网日期，不代表论文的发表时间）