Improving coal mine safety by preventing and controlling gas explosion accidents is crucial, and effective risk assessment plays a vital role in mitigating such risks. To address the limitations of existing dynamic control methods, uncertainty handling, and expression of risk, this study proposes a novel approach that combines fault tree analysis and fuzzy polymorphic Bayesian networks. This method categorizes risk factors into multiple states, introduces the concept of accuracy to refine fuzzy theory subjectivity, and utilizes Bayesian networks for real-time risk probability calculation and risk distribution assessment. It also suggests tailored prevention and control measures.
The study’s findings reveal that the current gas explosion risk probability at Wangzhuang coal mine stands at a concerning 35%. Among the risk factors, excessive ventilation resistance and spontaneous coal combustion are significant contributors to induced risk, with electric sparks being the most sensitive factor. Effectively addressing these key factors through prevention and control measures can substantially reduce the overall risk.
