Study Of Key Technologies For Highway Bridge Weigh-in-motion

In the past few decades,the number of road vehicles,the vehicle weights and the volume of freight traffic in China have experienced rapid growth,resulting huge changes in the characteristics of the actual load on the highway.Therefore,accurately and quickly identifying the weights of passing vehicles is of great significance for obtaining accurate vehicle load model for bridge safety assessment and infrastructure design,and will play an important role in controlling overloaded vehicles to reduce the potential bridge damages or even failures.Bridge weigh-in-motion(BWIM)is a technology that utilizes the responses of existing bridges to calculate weights of the passing vehicles.It is a promising technique to solve the problems discussed previously due to the following advantages: high identification accuracy,no need to stop or slow down the vehicles to be weighed,easy installation and maintainence,no traffic interruption or damage to the road surface.However,the fact that additional axle detection device is required in addition to the weighing sensors has limited the practicability of existing BWIM systems.Besides,further studies are still needed to remove the restrictions posed on the bridge type and span length by the current BWIM systems.In this dissertation,in-depth studies are conducted on the important issues in BWIM researches: axle detection and weighing algorithm.A series of theories and methods for identifying vehicle speed,axle locations,axle spacings,axle weights and gross weight of vehicles are proposed.The proposed theories include the virtual simply supported beam theory,equivalent shear force theory,virtual axle concept,etc.The methods include virtual simple beam method and equivalent shear method for identifying the speed and axle of vehicles traveling at high speed without installing sensors on the road,virtual axle method,improved k-means clustering method and gradient-based VCG(Virtual axle & Clustering & Gradient method)method for synchronously identifying axle locations and weights of vehicles without any axle detector.To evaluate the accuracy and reliability of the proposed methods,three-dimensional finite element models for the vehicle and bridge were established.A numerical program was established and a scaled model test platform was built with high similarity in the laboratory to simulate the bridge-vehicle interaction.Then,numerical simulations and model experiments were conducted to verify the proposed methods and to perform parametric studies.The identification accuracy,efficiency and reliability under certain conditions with different vehicle models,lateral loading positions,driving speeds,road surface conditions,noise levels,single or multiple vehicle presence,are examined.The results show that the proposed method can accurately and efficiently obtain the key information of traffic loads,such as vehicle speed,axle spacing,axle weight and gross weight,and can achieve good performance under complex conditions.Ensuring the safety of highway bridges and facilitating the growing logistics and trading are the important basis for the stability of our country and society and sustainable economic development.The research results from this dissertation have broadened the applications of BWIM technologies regarding the type and span length of bridges and have improved the practicability of BWIM technologies.It will play a positive role in bridge safety related fields,including traffic load monitoring and monitoring and control of overloaded vehicles.