Study On Prediction Of Coal And Gas Outburst

The security situation of coal production is grim currently, and the biggest threat factor in coal production is coal and gas outburst. Coal mine workers at home and abroad faced a study topic is to find a way to predict coal and gas outburst effectively, and provide accurate and reliable basis for the mine workers in order to avoid or reduce the loss of life and property caused by coal and gas outburst. Therefore, study of coal and gas outburst prediction model has great practical significance.Experts and technical staff having been tried variety of methods to solve the problem, such as:the acoustic detection prediction, perforated prediction, electromagnetic wave detection prediction and so on. But the ideal solution is to forecast the coal and gas outburst making it possible to make the response. In recent years, SVM developed very well and made a lot of achievements in the field of data mining and forecasting. Coal and gas outburst is highly nonlinear process with multi-influencing factors, and SVM has advantages in solving the small sample, high dimension, nonlinear problems. So SVM was used to the prediction of coal and gas outburst in this paper.Various factors affecting the coal and gas outburst was analyzed and the theory and development of SVM was introduced in this paper. This paper put forward PSO-LS-SVM and GA-LS-SVM based on SVM to improve the training speed and classification accuracy of coal and gas outburst prediction, to meet the real-time needs of coal and gas outburst prediction.