The Research On The Application Of Fuzzy Rough Set And Support Vector Machines In Coal And Gas Outburst Prediction

As one of the natural disasters, coal and gas outburst is a serious threat to mine safety and production. It has a sudden and devastating effect. It is the critical base of preventing and controlling gas disaster effectively, developing and utilizing gas resources efficiently to predict the coal and gas outburst risk accurately. At the same time, it is also the key technical issues and important research topic in the field of mine safety. This paper combined Fuzzy Clustering theory, Rough Set theory and Support Vector Machine learning algorithm, and established the coal and gas outburst prediction model of multiple classifiers based on Fuzzy Rough Set and Support Vector Machine. Firstly, Fuzzy C-means Clustering theory is used to discretize the original data. And Rough Set theory is used to acquire coal and gas outburst information from the discretized decision table. It means that using Rough Set theory to remove redundant information and reduce the dimensionality of data in the premise of maintaining data integrity. Then selecting the training samples set and test samples set from the pre-treated samples, and training the Support Vector Machine multi-classification prediction model, selecting the appropriate kernel function and parameters. In the end, prediction model of multiple classifiers is used to predict the coal and gas outburst hazard level. The experimental results show that the coal and gas outburst prediction model based on Fuzzy Rough Set and Support Vector Machine need short training time, but has high classification accuracy and good generalization performance. The application of combining Fuzzy Rough Sets and Support Vector Machine on coal and gas outburst prediction has great important meanings in ensuring the coal mine safety production and the lives of coal miners.