
第45卷第2期 2018年4月
矿业安全与环保
MINING SAFETY & ENVIRONMENTAL PROTECTION
温延新,于风频.基于KPCA-Fisher判别分析的煤炭自燃预测研究[J].矿业安全与环保,2018,45(2):49-53 文章编号:10084495(2018)02004905
Vol. 45 No. 2 Apr. 2018
判别分析的煤炭自燃预测研究
基于KPCA-Fisher关
温廷新,于风娥
(辽宁工程技术大学系统工程研究所,过宁销芦岛125105)
摘要:为了提高煤炭自燃危险性预测精度,提出了基于KPCA-Fisher判别分析的煤炭自燃预测模型。利用核主成分分析法(KPCA)对相关程度较高的特征指标进行非线性特征提取,将提取出的主成分作为Fisher判别模型的判别因子。选取宣东2号煤矿煤炭自燃的历史数据,以3:1的比例抽取训练集和测试集并代入该模型进行训练和测试,并将预测结果与传统的FDA、SVM和BPNN模型相比较。结果表明:KPCA能有效提取煤炭自燃特征指标,降低指标间信息允余,基于KPCA的Fisher判别模型用于煤炭自燃预测简单可行,准确率较高。
关键词:煤炭自燃;预测;核主成分分析;Fisher判别分析;回代估计法
中图分类号:TD75
文献标志码:A
Research onPrediction of Coal SpontaneousCombustionBased on
KPCA-Fisher Discriminant Analysis
WENTingxin,YUFenge
(System Engineering Institue, Liaoning Technical Unirersity, Huludao I25105, China)
Abstract: In order to improve the prediction accuracy of coal spontaneous combustion, a model based on KPCAFisher discriminant analysis was proposed to prediect coal spontaneous combustion, kemel principal component analysis ( KPCA) was used tononlinear feature extraction for characteristic indexes with higher comelation. The extracted principal components were used as the diseriminant factor of Fisher discriminant model. The historical data of coal spontaneous combustion in No. 2 Coal Mine of Xuandong was selected, and the model was trained and tested by extracting training set and test set with the ratio of 3 : 1 and the forecast results were compared with traditional FDA, SVM, BPNN method. The results showed that KPCA can extract the characteristic indexes of coal spontaneous combustion effectively, and reduce the information redundancy among the indexes. Using Fisher discriminant model based on KPCA to forecast coal spontaneous combustion is not only simple and feasible, but also with high accuracy.
Keywords : coal spontaneous combustion; prediction; kernel principal component analysis; Fisher discriminant analysis; backgeneration estimation
煤炭自燃火灾是煤矿安全开采过程中所面临的重要灾害之一。煤炭自燃预测与评价是防止煤炭自燃灾害的前提条件,及时、准确地预测煤炭自燃以便采取相应的措施来防止煤炭自燃事故发生已成为煤
收稿日期:20170605;201803-14修订
基金项目:国家自然科学基金项目(713711091);辽宁备社科基金项目(L14BTJ004)
作者简介:温廷新(1974一),男,山西太谷人,博士,教投,项士研究生导师,主要从事矿业系统工程、数据挖据等方面的研究工作。E-mail;wen_tx@163.com
通讯作者:于风姨(1992一),女,辽宁大连人,硕士研究
生,主要从事临息系统、数据挖掘等方面的研究工作。万方数据
矿安全生产中呕待解决的科学问题
目前,对煤炭自燃研究主要体现在两个方面:一方面是煤炭自然发火预测指标气体的优选,王从陆等[1]通过对新鲜煤样和氧化煤样的热解实验,测定实验煤样在不同热解温度下各气体的浓度,具体分析了指标气体浓度随温度的变化特性,为煤炭自燃预测指标气体的选取提供理论依据;吴兵等[2]通过自燃特性实验来研究煤炭自燃过程中的标志性气体,分析了各气体产生量随煤样温度变化的曲线,从而确定煤炭自燃预测指标气体。另一方面是煤炭自燃危险性预测的研究及优化,邵良杉(3)等将因子分析与支持向量机相结合并应用于煤炭自燃预测,较
· 49·