Coal And Gas Outburst IGSA-SVM Prediction Model And Its Application

With the increase of mining depth and mining strength increase,increasing gas disaster in coal mine, the mine gas disasters become the most important factors restricting our high-efficient exploitation of coal mine safety. The coal and gas outburst is one of the most serious disasters,serious threat to the life safety of the underground work personnel,it restricts the development of coal industry of fast and good;On the other hand, the coal mine gas constant emissions into the atmosphere, easily to atmospheric polution and waste a lot of high quality clean energy. Therefore, accurate prediction and effective defense to highlight the accident is a very important work.This paper generalizes the main gas outburst prediction method at home and abroad, in turn, on the basis of a large number of literature and accident report, based on these highlights the research and analysis of accident, the influence factors of coal and gas outburst is divided into 6 primary influence factors and 12 secondary factors, to further explore the various influence factors in the process of coal and gas outburst, the role of for the prediction of coal and gas outburst work provides the more theoretical basis. For prominent influence factors, and the complicated relationship between the traditional prediction methods have been unable to achieve the needs of the mine safety production at present.In order to effectively solve this problem, the in-depth study of gravity algorithm(GSA), support vector machine(SVM)and differential evolution algorithm(DE)on the basis of relevant theories, in order to improve the early stage of the GSA algorithm global searching ability and poor local search ability of faults, late will be coordinated groups of DE algorithm search and search capability of mutation strategy is introduced into the algorithm of gravity, puts forward the improved gravity algorithm(IGSA), and combined with support vector machine, establish IGSA-SVM prediction model of coal and gas outburst. Using the prediction model, multiple coal mine in Shanxi the typical experimental data as the research object, and using traditional single index prediction method to determine with outburst danger area to forecast, and the SVM model and GSA-SVM model predicted results compared with the results show that the IGSA-SVM prediction model of coal and gas outburst than the SVM model and the GSA-SVM model has high forecast accuracy, short time consuming, as well as good generalization performance. This article selects the chariot coal mine tunneling faces in shaqu coal mining working face of coal mine and coal and gas outburst experimental data as the research object, according to the coal mine geological conditions and the prominent influence factors analysis, finally determine the chariot coal mine adopts the consistence coefficient(f), drilling of coal gas emission velocity(q0), gas radiation initial velocity(p), drilling cuttings gas desorption index(K1) and the maximum amount of drilling cuttings(S) as an index of outburst prediction, and the largest selection in shaqu coal mine drilling cuttings volume(S) and the consistence coefficient(f) of coal, coal desorption index(K1), borehole gas emission velocity(q0) as a sensitive index of coal and gas outburst prediction, and use it as a prediction model of the input data. Through the numerical simulation model, if the output and the actual measurement result is the same as a result, then use the method to predict coal and gas outburst is feasible.Method described in this article, compared with the traditional prediction methods, Not only can realize the high precision of prediction of coal and gas outburst, and also for saving the cost of the mine safety production of coal mine, more improve coal mine safety production and laid a solid foundation.