ABSTRACT
Coal mine water disaster is a complicated phenomenon involving many unknown causes and complex relationships. It may lead to severe economic losses and casualties if occurring. In this study, the reports on coal mine water disaster accidents were manually collected and analyzed to acquire the keyword frequency. The accident impact factors (AIFs) of coal mine water disasters were identified under cause theory. The Bayes network model was established using a data-driven method to input 139 coal mine water disaster cases to study and analyze the impact relationship among AIFs under different accident levels. The study discovered some key factors impacting coal mine flood accidents, set the different states of the elements in scenario settings to analyze the evaluation relationship among AIFs quantitatively, and finally performed scenario simulation to understand the difference of AIF status in accidents at different levels. The water hazard problems in the accident reports were summarized. Serving as a reference for coal mining companies to investigate and rectify water hazards, this study can help coal companies develop preventive measures to reduce accident hazards.
