Research On Application Of Lightweight Visualization Operation And Maintenance Management System Of Bridge Information Based On BIM

Bridge construction has the characteristics of large number and large span,which brings great pressure to the operation and maintenance.Building Information Modeling technology is the key to the digital transformation of the construction industry.BIM technology integrates the data in the building life cycle into the same digital 3D model in a visual way,which can realize the collaborative management of building information.This paper researched the solution of the Lightweight Visualization Operation and Maintenance Management System of Bridge based on Web-BIM.The main tasks are as follows:To solve the problems of large volume and high computing resource consumption of BIM 3D model,developed Revit API plug-in to parse IFC file,extracted the key data and packaged it;Researched and improved the approximate curvature edge folding and grid traversal algorithms,reduced the number of faces in the 3D model,and simplify the grid coding method.The combination of the two methods is applied to the model compression algorithm.The experimental results showed that the model’s volume can be reduced to less than 20%.Aiming at the problems of closed system and poor distributed capability of the C/S architecture 3D model rendering software system,a B/S architecture 3D model visualization rendering scheme was proposed.Researched Web GL graphics rendering engine;Designed the model rendering framework of the browser side,built the 3D model visualization platform.The experimental results showed that the 3D model can be loaded completely on browser,and the rendering time is reduced to less than 15%.To overcome the current problems of poor cross platform performance and high calculation service pressure of server based deep learning mode,combined with the Tensor Flow.js,framework,designed and implemented a browser based operation and maintenance parameter prediction platform.Researching deep learning algorithms and establishing a browser-based Attention-GRU fusion prediction model.The experimental results showed that the platform can operate stably and the average percentage error is about 5%.Researched the overall plan of the “BIM+IOT” system,designed information collection nodes,analyzed system requirements,designed the overall framework of the system,integrated the functions of each module,and realized the B/S architecture software system,which has the functions of visualization,interaction,and operation and maintenance parameter training and prediction.The system is highly scalable and supports cross-platform operation,which can provide managers with a safe and reliable bridge operation and maintenance information management method.