Gas Emission Quantity Prediction Based on Particle Swarm Optimization of SVM and Deep Learning Network

In order to improve the accuracy and efficiency of coal and gas outburst prediction, based on the principle of deep learning network, SVM and the particle swarm optimization algorithm, the deep learning network was combined with particle swarm optimization of support vector machine neural network for the prediction of the situation of coal and gas outburst, more profound characteristics of variables can be learned from raw data through deep learning. Then the prediction models were established for a few characteristics by feature extraction which denotes the raw data by using PSO-SVM method to predict the emission concentration of gas instead of the raw data. Through mining workface of a certain coal mine in China to analyze and predict, the results showed that the approach met the requirement that reduced dimensionality of the raw data, and the predict accuracy was improved greatly. Meanwhile, the running time and efficiency of the algorithm was improved through this method.