The Construction And Application Of Urban Rail Risk Control Knowledge Graph Based On The Accident Report

Urban rail transit has become one of the main modes of transportation in the city.However,with the increasing complexity of urban rail transit system and the rapid growth of operating mileage,some safety problems have gradually emerged,bringing great challenges to the safe and reliable operation of the rail transit system.The current urban rail transit operation safety management mode is not perfect,there are such as risk identification,risk targeted prevention and control of some key problems,needs to break through for the key technology of active prevention and control risk,the research of urban rail accident accurate identification method of key risks,study the role of the relationship between risk and risk evaluation methods,The bayesian network model is established to realize the risk reasoning calculation,and the knowledge map is used to realize the efficient management and control of urban rail accidents and risk knowledge.Firstly,this paper makes full use of the characteristics of accident report reflecting the occurrence,development and evolution mechanism of urban rail safety accident risk,innovatively proposes an urban rail transit accident analysis paradigm based on "risk point→risk→event→accident",and establishes the construction mode layer of urban rail accident knowledge graph.Then,by means of hidden Markov model and Viterbi algorithm and deconstruction analysis of the components and functions of urban rail transit system,the urban rail transit risk domain thesaurus is established.Then,the extraction modes and rules of events,risk points,risks and their coupling effects are further established,and the risk knowledge is extracted and fused.Based on the fusion of entities and relationships,knowledge triples are constructed and stored in graph database to complete the construction of knowledge graph.Secondly,in order to fully explore the occurrence,development and evolution mechanism of urban rail fire accidents and realize the quantitative reasoning and analysis of risk,this paper proposes the establishment of knowledge network mapping Bayesian network model.Firstly,bayesian network modeling and structure learning are carried out according to knowledge graph network.Secondly,the maximum likelihood parameter(MLE)method is used to learn network parameters from actual data,and fuzzy expert evaluation and coupling conditional probability calculation method are introduced to correct the parameters.Finally,Ge Nle software is used to complete bayesian network modeling,and its inference engine can realize directional reasoning and causal analysis of risks,and analyze the posterior probability of accidents caused by important nodes,laying a solid theoretical foundation for accurate deduction of risk links before accidents and backtracking of key causative risk points after accidents.Finally,on the basis of above research,in-depth analysis of user needs,for urban rail management personnel,design the system architecture and function framework,to build urban rail risk control system,and based on the knowledge map of urban rail traffic accident and the risk control platform,realize the risk management,knowledge management,emergency management,and other functions,It solves the key problems of poor systematic risk management and insufficient knowledge support of current urban rail system operation security system,which is of great significance to improve the initiative and intelligence of urban rail transit system security.