The Research Of Neighborhood Rough Set And Kernel Extreme Learning Machine In Coal And Gas Outburst Prediction

Coal and gas outburst are one of the most serious and destructive safety accidents in coal mining and production.It’s safety issue has become the top priority of our country energy security.Coal and gas outburst prediction is a complex,non-linear,high-dimensional problem that is affected by multiple factors.Traditional methods often fail to work.On the basis of analyzing the mechanism and influencing factors of coal and gas outburst,combining with the research status and research results of both domestic and international,the topic compares and improves the existing classification methods.The theory of rough sets and the theory of kernel extreme learning machine are applied to the prediction of coal and gas outburst.In this paper,the properties of coal and gas outburst decision information table are made by using the NRS theory.This means eliminating redundant information while maintaining data integrity.The data reduction dimension is realized.Then,the training samples and test samples are selected from the pre-treated samples.Train the prediction model of the kernel extreme learning machine and select the appropriate kernel function and parameters.Finally,the prediction model is applied to predict the risk of coal and gas outburst.Experimental results demonstrate that the coal and gas outburst prediction model based on NRS-KELM theory quicks convergence.The prediction accuracy is high and the performance is excellent.This suggests that this article uses the neighborhood rough set and kernel extreme learning machine theory used in the coal and gas outburst prediction method is feasible,in order to further explore practical and effective method of coal and gas outburst prediction provides a new,theoretical and engineering method.