The Application Of Improved Extreme Learning Machine In The Prediction Of Coal And Gas Outburst

Coal and gas outburst is a serious dynamic disaster for the coal mine, which seriously threatens the safety of people's life and property and the safety of the production. Under the analysis of coal and gas outburst mechanism and various factors which influence the coal and gas outburst, this paper proposes an improved extreme learning machine model to predict the coal and gas outburst, and then explains for details bellow.Firstly, describe the current research situation and prediction method of the coal and gas outburst, as well as the research situation of extreme learning machine. On the basis of the discussion, it put forward the research scheme of this thesis.Secondly, based on discussing the mechanism and the influence factors of the coal and gas outburst, this part use the Analytic Hierarchy Process (AHP) to analysis the weights of the factors which influence the outburst. And take it as a basis for selecting the input variables of the prediction model. The AHP improves the accuracy and the computational efficiency by means of removing the redundant variables of the input variables.Thirdly, to optimize the input weights and the hidden layer thresholds by immune genetic algorithm on the basis of the structure and the algorithm of extreme learning machine. Simulation via wine data sets illustrates that the improved algorithm increase the stability and accuracy of the model, so it is of efficiency for predicting the coal and gas outburst.Finally, establish the predicting model of coal and gas outburst combining with the actual situation of the underground mine. At the same time, compared with other machine learning algorithm, it shows that the improved extreme learning machine has a better effect.