Study On Risk Assessment Of High-rise Building Fire Accident Based On Fuzzy Bayesian Network

In order to improve the utilization rate of land area,under the guidance of the development of modern science and technology level,the building is bound to develop towards a higher direction.At the same time,frequent fire accidents of high-rise buildings pose a serious threat to people's life and property.Therefore,it is necessary to evaluate the risk level of high-rise buildings.Based on high-rise building fire risk assessment as the guidance,high-rise building as the research object,on the risk theory,fuzzy set theory and bayesian network theory as theoretical basis,established a risk evaluation model based on fuzzy bayesian network,the purpose is to improve the management level and safety measures,for the high-rise building fire risk assessment provides the scientific method,so as to avoid fire accidents.First of all,this paper makes clear the content and technical route of this paper as well as the concept and characteristics of high-rise building fire accident risk through the study of relevant literature.A fuzzy Bayesian network risk assessment model for high-rise building fire accident risk assessment is proposed by analyzing the applicable scope,advantages and disadvantages of the existing risk assessment methods.Secondly,the paper studies the formation mechanism of high-rise building fire accident with the help of the theory of "three kinds of hazard sources",and makes clear the composition of risk system.Through the analysis and screening of risk indexes by means of typical accident analysis method and questionnaire survey,the risk evaluation index system of high-rise building fire accident with 18 basic evaluation indexes is constructed,and the evaluation indexes are explained.Finally,on the basis of the establishment of the fire accident risk evaluation index system of high-rise buildings,a basic fuzzy Bayesian network fire accident risk evaluation model of high-rise buildings is constructed.Calculation of conditional probability;Fuzzy Bayesian network inference analysis.The degree of risk is divided into four levels: very high,high,medium and low.At the same time,an empirical study is carried out with actual cases.The risk assessment results are basically consistent with the actual fire risk situation,which verifies the rationality and scientificity of the model.Through the combination of fuzzy theory and Bayesian network,the problem of uncertain risk factors and incomplete data in the fire accident risk assessment of high-rise buildings is well solved,and the bayesian network is also well extended,and a new idea is provided for the fire accident risk assessment of high-rise buildings in the future.