Prediction Model Of Coal And Gas Outburst Based On Support Vector Machines

Coal and gas outburst is hidden danger that leads to major coal mine accidents. In coal and gas outburst prevention, the research is of practical significance for increasing the accuracy of prediction of coal and gas outburst risk, reducing the blindness and enhancing targeted in prevention measures.The prediction of coal and gas outburst is a complicated,nonlinear and high dimensional problem,and also synthetically affected by multi factors. It is often difficult to solve with traditional methods. It is an active research direction for the mine geologists to search an effective method for prediction of coal and gas outburst.A new way to solve prediction of coal and gas outburst is put forward in this thesis. Mechanism and influencing factors of coal and gas outburst were analyzed and the latest Statistics Learning Theory—Support Vector Machines(SVM) was used,a prediction method of coal and gas outburst based on the SVM is studied and proposed.The main research achievement can be inc1uded as fol1ows: first of all, the mechanism and influencing factors of coal and gas outburst were analyzed,Several multi- class -classification methods based on SVM were researched and compared about the performance, after that ,with one-against-one mechanism, improved the existing classical algorithm from the classification level. The multi-class Classification algorithm required this prediction model based on SVM was implemented with MATLAB Programming,Finally,According to the circumstances of physical geography in the selected area of coal mine, factors as minimum combination for prediction were extracted from coal and gas outburst affection factors set by feature selection,grid search and the cross validation were adopted to search the best parameters of SVM,and suitable kernel function and parameters in the forecast were selected.This paper establishes a prediction model of coal and gas outburst.This experiment obtains better effect in accordance with procedure in this forecast.And completed the experimental,the results show that the based on SVM has effectively increased the forecasting precision compared with a variety of prediction algorithms. It is feasible to predict outburst of coal and gas. The method provides reference value to further explore the practical and effective prediction of coal and gas outburst.