Risk Assessment of Water Conservancy Engineering Quality Based on Multi-source Data
Obtaining comprehensive,accurate and timely multi-source data is the basis of water conservancy project quality risk assessment.Considering that the correlation and wholeness of the causes of risky accidents will affect the reli-ability and authenticity of the assessment results,relying on the 32 key water conservancy projects in Zhejiang Province as the basis of multisource data,this paper initially determines the risk factors based on the literature survey and analysis of engineering examples,and selects the causal variables that lead to quality risk from three aspects fo quality of raw materi-als,quality of engineering entities,and quality of management of human supervision and behaviour.By analyzing the cor-relation between variables,the Bayesian network model(BN)based on water conservancy project quality risk assessment is constructed by using the Bayesian network software GeNle.Through the analysis of variable sensitivity and the most extensive causal chain of accidents,it is found that the high risk path of the key factors is listed as follows:the participant qualification examination→qualification qualification→human supervision module behavioural management quality→water conservancy project quality;The sensitivity of general unit key unit project self-inspection excellence rate,qualification examination of construction personnel passed,the quality of the incoming materials qualified,the construction data re-tained complete and the phase project acceptance passed is the highest,and its changes are more likely to lead to accidents and are important factors affecting the quality of water conservancy projects.Its research methods and results can provide a scientific basis for water conservancy project quality risk control and safety management.
risk assessmentBayesian networksconditional probabilitysensitivity analysisquality of hydraulic engineering
万方数据
维普
