| Building information model(BIM)is increasingly being applied throughout a bridge's management stage for various applications,such as structure monitoring,defect detection,building renovation and energy simulation.The bridge management system is complex involving various stages of building information and its basis is the bridge models database.There are many methods to obtain bridge information models,including the total station mapping,image photography,laser scanning and so on.And this thesis chooses the modeling method based on point cloud data.These as-is bridge BIM models are often constructed using dense 3D point measurements obtained from laser scanners using its rapid capture of original conditions.In order to create valid and complete model,the modeling process include: defining the shape and geometry,object recognition and describing the topological relationships between the constituent components.Laser scanning technology can obtain detailed geometric surface information,but extraction of topology and semantics is insufficient.BIM can make up for it as an integrator of information data.There are two aspects in the thesis:(1)Common BIM modeling methods are presented and evaluated;(2)A particular existing bridge BIM modeling method based on point clouds is proposed,the algorithm of boundary fitting and the information attribution are given.The collected raw data undergoes data downsizing,boundary detection,and bridge component categorization,resulting in the building components being recognized as different elements.Finally,the attribute information of the components is given,and the BIM model of the complete bridge is formed.The results of a test conducted on a bridge in HUST collected initial data to validate the technical feasibility and evaluate the performance of the proposed method,indicating that it can accelerate as-is bridge information modeling from point clouds,and also integrate the data needed for bridge management. |
PDF Full Text Request