基于SVM的潘三矿B8组与C13组煤开采中突水水源判别模型  

The discrimination model based on SVM of inrushed water sources in the coal mining on the level of B8 & C13

作　　者：钱家忠[1] 潘婧[1] 赵卫东[1] 陈陆望[1] 

机构地区：[1]合肥工业大学资源与环境工程学院,合肥230009

出　　处：《系统工程理论与实践》2011年第12期2425-2430,共6页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(40872166);教育部新世纪优秀人才计划(NCET-06-0541);合肥工大创新计划(2009HGCX0233)

摘　　要：矿井突(涌)水水源的快速识别是矿井水害有效防治的前提条件.为了更有效地区分潘三煤矿B8、C13组煤系突水水源,利用支持向量机(SVM)建立水源判别模型,并将其与模式识别领域发展比较成熟的BP神经网络判别模型对比,发现SVM法能够将煤系B8、C13组混和水源快速、有效地分开.研究结果表明:SVM法的分类函数结构简单,运算速度快,解决了在BP神经网络方法中无法避免的局部极值问题,对于B8、C13组煤系突水水源的区分有更好的适用性和优越性,为矿井水害防治提供了一种辅助决策手段.Rapid discrimination of the water bursting sources is the precondition of preventing waterinrush from coal floor effectively. To distinguish inrushed water sources in the coal mining on the level of B8 ＆ C13 more effectively, SVM was introduced to establish the water inrush discrimination model. By comparing with BP neural network model for discrimination in the development of the relatively mature field of pattern recognition, we found that SVM algorithm can distinguish the water on the level of B8 ＆ C13 quickly and effectively. The results show that, the classification function of SVM algorithm is simple and has fast operation. It has solved the local optimal problem which can not avoid in BP neural network algorithm. It has better applicability and superiority in the discrimination of the water bursting sources on the level of B8 ＆ C13. It provides an assistant means for decision-making to prevent water-inrush from coal floor.

关 键 词：突水水源判别 支持向量机 BP神经网络 煤矿 模型 

分 类 号：TD745.21[矿业工程—矿井通风与安全]