Study On BIM-based Visualization,Prediction And Evaluation For Safety Monitoring Of Water Diversion Project

The long-distance water diversion projects have many characteristics,i.e.long operation period,large coverage,and numerous crossing buildings.To ensure the operational safety of such linear projects,it is critically important to conduct health monitoring,and realize timely evaluation and early warning.At present,the safety monitoring visualization of water diversion projects is biased towards the apparent appearance of buildings and has poor capacity of data carrying,which is not integrate with information such as structural design and electromechanical equipment.In the meanwhile,objective and accurate prediction as well as comprehensive evaluation of the operational status,which based on current monitoring information,are the key issues to be solved in the operation and management of the water diversion projects.Usually safety monitoring involves multiple evaluation criteria and multiple monitoring indicators.To achieve a more comprehensive evaluation of operation safety,various evaluation indicators as well as multiple evaluation results need to be considered in the comprehensive evaluation of the operational status.Based on the above questions,the main research contents and achievements of this thesis are as follows:(1)Combined with the advantages of building information modeling(BIM)technology,the BIM modeling method and visual management of water diversion projects was explored.Using Autodesk Revit as modeling software,a component family library of water diversion projects was constructed to realize the refined modeling of the 3D scene.In the meanwhile,the BIM model was coupled with safety monitoring information.A technical route for Web visualization of the BIM model integrated with safety monitoring information is developed.The network visualization management of the BIM model is realized through the Forge Viewer platform.With the proposed method,a case study was applied in the Tianjin South-to-North Water Diversion Project.3D virtual scene of the engineering area was established to assist safety monitoring analysis,which indicates that the proposed method can realized the dynamic,intuitive and efficient management of safety monitoring,and improved the level of the safety management of water diversion projects.(2)A method based on Deep Belief Network(DBN)for predicting and analyzing safety monitoring indicators is proposed.Based on the analysis of DBN structure and algorithm,the model structure is discussed by analyzing the influencing factors of the monitoring indicators of the water diversion projects.Then a DBN-based monitoring indicators prediction model was established.The effect of time window length to model accuracy was discussed.Combined with engineering examples,the prediction accuracy of the model was analyzed.The results verified the reliability of the constructed model and provided a new idea for the prediction of safety monitoring indicators of water diversion projects.(3)A comprehensive safety evaluation method for water diversion projects was proposed based on multiple criteria and multiple indicators.A multi-indicators comprehensive evaluation system for safety operation was constructed.Combined with various measurement methods of safety monitoring indicators,the multiple evaluation criteria of evaluation indicators were discussed and then the membership function was established.A workflow for multi-indicators fuzzy comprehensive evaluation was summarized.Furthermore,combined with the D-S evidence theory,a fusion method for the operational safety states of water diversion projects based on multiple evaluation criteria was proposed.A case study was carried out in the Tianjin South-to-North Water Diversion Project.The proposed method provided a scientific basis and decision support for the comprehensive evaluation of the operation safety monitoring for the water diversion projects.