论文部分内容阅读
突水是影响煤矿安全生产的重要灾害之一,及时确定突水水源是进行水灾治理的关键,采用核主成分分析(Kernel Principle Component Analysis,KPCA)结合朴素贝叶斯分类方法(Naive Bayes Classifier,NBC)对煤矿突水水源进行判别.首先采用核主成分分析方法对淮南孔集煤矿78条水化学成分数据进行降维处理,选取其中5种成分数据进行分析,然后采用朴素贝叶斯方法对突水水源进行识别.通过与模糊综合评价方法进行比较,所得结果除了老空水外其它含水层的识别结果均优于或者等于模糊综合评价法所得结果,表明基于KPCA-NBC的煤矿突水水源判别方法具有较高的实用性和有效性.
Water inrush is one of the important disasters that affect coal mine safety production. The key to flood control in time is to determine the water inrush. By using Kernel Principle Component Analysis (KPCA) and Naive Bayes Classifier NBC) to discriminate coal mine water inrush sources.Firstly, the principal component analysis (PCA) was used to reduce the chemical composition of 78 water samples in Huai Nan Kong Kongji coal mine, five of them were selected for analysis, and then Naive Bayes method Water inrush sources.Compared with the fuzzy comprehensive evaluation method, the results of the recognition results of other aquifers other than the old empty water are better than or equal to those obtained by the fuzzy comprehensive evaluation method, which shows that the KPCA-NBC-based coal mine water inrush source Discriminant method has high practicability and effectiveness.