Prediction of Mine Gas Emission Rate using Support Vector Regression and Chaotic Particle Swarm Optimization Algorithm

Forecasting of gas emission rate in mine is a complicated problem due to its nonlinearity and the small quantity of training data. Support vector regression (SVR) can solve the problem with small samples, nonlinear and high dimensions. However, the precision of SVR is significantly affected by its parameter. In order to improve the mine gas emission rate accurately, an optimal selection approach of support vector regression parameters is proposed based on the chaotic particle swarm optimization algorithm (CPSO). A model based on the CPSO-SVR to predict the mine gas emission rate is established and the optimal parameters of SVR is searched by CPSO. The experimental data from a coal mine in China is used to illustrate the performance of proposed CPSO–SVR model. The results show that the proposed prediction model has better results than the artificial neural network (ANN) and traditional SVR algorithm under the circumstances of small sample. This indicates that the precision can meet the requirement of practical production and demonstrates that the CPSO is an effective approach for parameter optimization of SVR.