Researched on the method of gas outburst prediction based on D-S evidence theory

In this paper, a method of gas outburst prediction based on D-S evidence theory was presented. The hierarchical analysis method was used to order the weight of the indicators of gas outburst at first, then using the neural network and traditional forecasting methods to predict respectively, finally, these prediction results were fused by D-S evidence theory, and the final result will be obtained. In this paper, the data of gas outburst were validated the prediction model. The results showed that the prediction method proposed in this paper could make different methods compensate, so it can effectively avoid making wrong judgments by using a single prediction method. It also made the prediction result more reliable and accurate.