Early Warning about Coal Mine Safety Based on Improved PNN-DS Evidence Theory

This paper establishes a two-level information fusion model based on improved probabilistic neural network and DS evidence theory to achieve the purpose of early warning of coal mine safety risk levels and apply the principle of information fusion to safety early warning problems. Firstly, this paper improves the traditional probabilistic neural network is improved in this paper. Then, input the sample data the sample data is input into the improved probabilistic neural network to obtain the basic probability assignment. Finally, the output of the neural network is fused by DS evidence theory. The historical data of a coal mine work face are trained and tested by MATLAB. The results show that the model established in this paper has a higher prediction accuracy, and realizes the purpose of early warning of coal mine safety risk level better, which has is meaningful and significant for the safety management of coal mine.