摘要
矿井突水水源的准确判别是目前矿井防治水害工作的关键之一。选取8种常规水化学指标作为样本标准库的基础上,考虑各水化学指标之间信息叠加的影响,引入核主成分分析(KPCA)和Fisher判别分析理论(FDA)并建立多因素耦合判别模型。研究结果表明:应用基于KPCA-FDA方法的判别法较单一FDA判别法具有更高准确性,消除各指标间信息叠加的影响,实现了矿井突水水源的快速、准确判别。
Accurate identification of mine water inrush source is one of the key points in mine prevention and control work. On the basis of 8 conventional hydro chemical indexes, the KPCA and Fisher discriminant analysis theory （FDA） were introduced and the multi factor coupling discriminant model was established. The results show that the discrimination method based on the KPCA-FDA method is more accurate than the single FDA discriminant method, eliminating the influence of information superposition among the indexes, and water inrush water source in the mine. realizing the rapid and accurate identification of the
作者
成荣秋
吴燕清
CHENG Rong-qiu;WU Yan-qing(China Coal Research Institute,Beijing 100013,China;Chongqing Research Institute,China Coal Technology and Engineering Group,Chongqing 400039,China)
出处
《煤炭技术》
CAS
2018年第11期194-195,共2页
Coal Technology
关键词
矿井突水
水源判别
核主成分分析
FISHER判别
mine water inrush
discrimination of water source
kernel principal component analysis
Fisher discriminant analysis
