Prediction of Coal and Gas Outburst Based on PCA-BP Neural Network

In order to effectively predict coal and gas outburst,PCA was combined with neural network.Taking Pingdingshan 8th coal mine as the research object,data on the factors affecting coal and gas outburst were collected based on geological dynamic division method.Through PCA method,three principal components with a total contribution rate of more than 80 percent were extracted and used to take the place of the nine original factors.The principal components were regarded as the input parameters of BP neural network.The magnitude of coal and gas outburst was divided into four levels.A PCA-BP neural network prediction model was built.Typical samples were selected to train PCA-BP neural network and to verify the trained network.The verification result of the trained network shows that the prediction by the network conforms to reality.