基于模糊模式识别的瓦斯含量指标异常预警技术  

Gas Content Pre-warning Technology Based on Fuzzy Pattern Recognition

作　　者：杨禹华[1] 钟震宇 蔡康旭[3] 

机构地区：[1]湖南科技大学建筑与城乡规划学院,湘潭411201 [2]广东科学院自动化工程研制中心,广州510070 [3]湖南科技大学能源与安全工程学院,湘潭411201

出　　处：《中国安全科学学报》2007年第9期172-176,共5页China Safety Science Journal

摘　　要：提出利用模糊模式识别技术对瓦斯涌出异常进行动态预警的新思路;通过对已知巷道的瓦斯含量监测数据和素描图进行分析,提取引起瓦斯涌出异常的前兆特征作为模式识别的特征向量,来对实时监测的巷道的工况进行判断,确定其是否存在瓦斯涌出异常的可能性。该方法将传统煤矿安全监控系统事后报警转变成事前预警,提高了系统的可用性。The paper puts forward the idea of using fuzzy pattern recognition technique to give dynamic warning on abnormal gas-effusion. First, the lane way＇s sketch map and monitoring data of gas content index are analyzed, and then the precursory features in abnormal gas-effusing area are extracted as the eigenvectors of pattern recognition so as to analyze the dynamic working situation in lane way and judge whether there is the possibility of abnormal gas-effusion. This method changes coalmine monitoring system＇s traditional way of after-alarming into early-warning, and thus enhances its feasibility.

关 键 词：模糊模式识别 瓦斯涌出异常 前兆特征 特征向量 动态预警 

分 类 号：X936[环境科学与工程—安全科学]