基于主成分聚类分析的煤层底板突水危险性预测  

Risk Prediction of Water Inrush from Coal Floor Based on Principal Component Clustering Analysis

作　　者：王静宇[1] 李翠平[1] 李仲学[1] 

机构地区：[1]北京科技大学金属矿山高效开采与安全教育部重点实验室,北京100083

出　　处：《中国安全科学学报》2013年第8期120-125,共6页China Safety Science Journal

基　　金：国家自然科学基金资助(51174260;51174032);教育部新世纪优秀人才支持计划资助项目(NCET-10-0225);中央高校基本科研业务费专项资金资助项目(FRF-TP-09-001A)

摘　　要：为准确预测煤层底板突水危险性,实现矿山安全生产,针对突水影响因素选取的主观性与因素之间的相关性,首先通过研究多元统计分析理论,建立基于主成分聚类分析的煤层底板突水危险性预测模型。然后用SPSS统计软件,对选取的样本数据进行主成分分析(PCA),定量分析各成分贡献率以确定主成分。将主成分作为系统聚类分析中的变量,结合主成分对聚类谱系图进行分析,预测样本煤层底板突水危险性。最后将本方法与其他2种方法的预测结果对比,分析存在差异的原因。In order to predict the risk of water inrush from coal floor accurately, and realize the safety of mine production, bearing the subjectivity of mine water inrush influencing factors selection and correlativity of influencing factors in mind, a model was built for predicting the risk of water inrush from coal floor based on principal component clustering analysis and multivariate statistical analysis theory. A PCA of the data on sample was made by using software SPSS. Principal components were chosen after quatitative anal- ysis of contribution rate of each component. The principal components were used as variables in hierarchi- cal clustering analysis. The risk of water inrush from coal floor of each sample was predicted by analyzing the clustering spectral pattern combined with principal components. Finally, a comparative analysis of prediction results obtained by using this method and two other prediction methods＇ was made and why some differences occur was explained.

关 键 词：煤层底板突水 影响因素 预测模型 主成分分析(PCA) 聚类分析 

分 类 号：X936[环境科学与工程—安全科学]