ABSTRACT
In this study, using the weigh-in-motion (WIM) data collected from an operational long-span urban highway bridge structural heath monitoring (SHM) system located in Wuhan, China, the statistical characteristics of gross vehicle weight (GVW), axle weights and spacing of axles are obtained, and then a statistical fatigue traffic load model for urban highway bridges is also proposed based on the Miner linear accumulated damage hypothesis and the effective fatigue damage principle. The proposed statistical fatigue traffic model can be used for bridge fatigue design and assessment. On the basis of the types of vehicles, (1) Considering the double or multi-modal distribution characteristics of the traffic GVW, the statistical parameters of the double or multi-modal Gauss distribution are estimated using expectation maximization (EM) algorithm; (2) Considering the single or multi-modal distribution characteristics of the traffic axle weight, the statistical parameters of the single or multi-modal Gauss distribution are estimated using EM algorithm; (3) The expectation and standard deviation of the spacing of axles are directly calculated; (4) Considering the statistical characteristics of the weight and spacing of axle, a statistical fatigue traffic load model for urban highway bridges is proposed based on the Miner linear accumulated damage hypothesis and the effective fatigue damage principle.
