Research on quantitative risk assessment method of dust explosion based on Bayesian network

Abstract

In order to effectively prevent the occurrence of dust explosion accidents in industrial enterprises and reducing the risk of dust explosion accidents, the article builds dust explosion risk assessment index system for industrial enterprises, taking Fault Tree Analysis (FTA) and Event Tree Analysis (ETA) as theoretical basis and establishes dust explosion risk Bow-Tie Analysis (BT) diagram model. Mapping BT model into Bayesian Networks (BN) reveals the coupling mechanism of dust explosion risk nodes and constructs a generic dust explosion risk assessment BN model, which provides a quantitative assessment of the probability of occurrence of dust explosion accidents. The causative mechanism of dust explosion accidents is investigated by learning risk inference from this model. It is also demonstrated that appropriate safety barrier measures can effectively reduce the probability of accidents, thereby changing the accident level and improving the safety and reliability of the system. Finally, the validity and applicability of the model is verified through empirical analyses. The research results contribute to the scientific prediction of the probability of dust explosion risk and reasoning learning, which provides a scientific basis for enterprises to formulate effective dust explosion safety measures as well as to realize the hierarchical control of dust explosion risk.