Abstract: Based on mine safety environment anlalysis and danger source theory,coal mine danger source ontology by OWL is constructed.Exploring the intrinsic relation between mine danger source and gas explosion accident,we construct gas explosion fault tree.Moreover,by designing reasoning rules,we realize the reasoning process from the root danger source ontology to the status danger source ontology.Finally,from bottom to top,based on the basic events status,the probability of the top accident event is calculated according to the constructed gas explosion fault tree ontology.The proposed construction and reasoning methods of mine accident ontology can be used for assistant decision-making on gas explosion accident warning.The experimental results show that the constructed mine accident ontology can correctly implement the reasoning from the root danger source to the status danger source.Meanwhile,the probability of top accident event of the gas explosion fault tree ontology is exactly calculated.These results verify the effectiveness of OWL ontology technology in predicting coal mine gas explosion accidents.