Construction Safety Risk Prediction Based On Fault Tree And Bayesian Network

The application of safety risk prediction in building construction has gradually attracted the attention of scholars,and an important indicator of safety risk probability component construction safety.The construction safety risk research is based on the design of the construction project and the safety risk factors in the construction process,and uses computer technology to analyze the law of risk factor data,thereby reducing the probability of occurrence of risk events.Based on the security risk factor database established by MySQL,this paper designs the function of fault tree analysis module based on B/S framework risk factor database system,deeply studies and designs the function of Bayesian network prediction module,and completes the security of risk factor database management system.Functional analysis of the risk prediction module.The construction construction safety risk is analyzed,the fault tree analysis is carried out based on the risk source analysis,and the fault tree model example and the modeling process are displayed.Using the designed functional modules to derive the risk factors,risk sources and risk levels of the FTA for the risk event,taking the“safety risk event occurrence” as the top-level event,according to the owner's risk,the design party risk,the construction party risk,and the supervision unit risk.Environmental risk 5 risk sources establish fault trees and complete the construction of top,middle and bottom events.Based on the fault tree,the Bayesian network mathematical model is established,and the "security risk event occurrence" risk event demonstration is also a network reasoning process.The prior probability of the underlying risk factor in the fault tree is calculated by the security risk factor database system.Through the established Bayesian network model,the relationship between security risk factors in different risk sources is found out,and the joint risk distribution is calculated by taking the risk of the designer and the risk source of the construction party as examples.,the probability interval used to guide the prediction of security risk events.Based on the security risk factor database system applicable security risk prediction field,the application database system assists security risk prediction.The method of fuzzy prediction of security risk events is adopted.The rationality of fuzzy prediction in the general confidence interval is verified by the design source risk source in the fault tree model.Theapplication of fuzzy prediction of security risk events in risk factor correlation is proposed.Taking the joint system consisting of two risk sources: design side risk and construction side risk as an example,the triangular fuzzy number suitable for the fuzzy prediction range of security risk events is selected,and the security risk prediction result and the Bayesian network model prediction result are carried out.The comparison verifies the feasibility of the security risk factor database system module.