基于主元分析的瓦斯抽放系统故障检测方法

Fault Detection for Gas Drainage System Based on PCA

瓦斯抽放系统是煤矿安全生产的重要环节。为提高系统故障检测能力,迅速处理故障,本文提出一种基于主元分析的多元统计分析方法,利用SPE统计量、T2统计量以及得分向量判断瓦斯抽放监控系统运行过程,是否发生异常。通过实验获取煤矿瓦斯抽放系统测量参数并进行分析,验证了基于主元分析的过程检测方法在瓦斯抽放系统故障检测中效果良好。

The gas drainage system is an important part of coal mine production safety. In order to improve the system fault detection ability and deal with failure quickly, this paper presents a multivariate statistical analysis method based on principal component analysis. Through the measurement parameter correlation analysis, the paper eliminated the variable association and established the statistical model of state detec- tion. Then it adopted SPE statistic, T2 statistic and score vector to determine whether an exception occurs in the gas drainage monitoring sys- tem. The experimental analysis demonstrated that the method can be used for fault detection in a gas drainage system.