基于粗糙集和聚类的采空区煤自燃火灾预报

采用标志气体分析法对煤自燃火灾预报时存在特征维数较高、特征之间存在冗余及人为划分温度段的不合理性等问题,文中提出基于粗糙集和聚类的采空区煤自燃火灾预报方法。即使用粗糙集对原始样本去除冗余和特征维数约简,再用聚类方法对约简后的特征进行聚类得到各温度段的特征中心,并使用模式识别的方法,确定出煤自燃标志气体特征其与温度段特征中心的相似性,从而实现采空区遗煤自燃状态的识别和早期预报。

Forecast of spontaneous combustion fire in goaf based on rough set and cluster

There exist some problems such as higher dimension of feature, redundancy existing in features and the irrationality of the artificial divided temperature pariods when using mark gas analysis method to forecast the spontaneous combustion fire, this paper proposed a approach to forecast spontaneous combustion fire in goaf based on rough set （RS） and k-means （ KMEANS ） , which gets the feature centers of different temperature pariods by using RS to eliminate redundancy and reduce feature dimension from original sample, using clustering methods to cluster the feature reduced samples, then use the method of pattern recognition to determine the similarity between the feature of coal spontaneous mark gas and temperature periods, which have realized identification and early warning of residual coal spon- taneous combustion state in goaf.