Research on SVM for Analyzing Coal-Gas Outburst

摘要：

This paper analyzes the patterns and amount of coalgas emission with support vector machine (SVM) via studying the correspondence between amount of coalgas emission in coal face and geological structure index. Then, the SVM recognition model for two patterns of coal-gas outburst, the H -SVMs method for multi-pattern coal-gas outburst and SVM for predicting amount of gas emission is built. The research results show: SVM is good method to recognize pattern of coal-gas outburst; and the SVM model is better to the BP model in predicting the amount of coal-gas emission because SVM is based on strict mathematical theory with simple structure and good generalization performance, and can reflect the weight of geological index in outburst pattern recognition through the parameter W in the decision function.

关键词： coal and gas outburst support vector machine (SVM) H-SVMs model recognition model

作者: SUN Yufeng LI Zhongcai CHEN Zhangliang

作者单位: School of Management Science & Engineering, Shandong Institute of Business & Technology, Yantai 264005, Shandong, China

会议类型: 国际会议

会议名称: The 2010 International Symposium on Safety Science and Technology(2010 安全科学与技术国际会议)

会议地点: 杭州

会议语种:英文

页码: 1644-1650

在线出版日期: 2010-10-26（万方平台首次上网日期，不代表论文的发表时间）