逐步判别分析法在陶瓷原料分类中的应用  

THE APPLICATION OF STEPWISE DISCRIMINATION ANALYSIS IN THE CLASSIFICATION OF CERAMIC RAW MATERIALS

作　　者：胡祎[1] 周景纬[1] 胡真[1] 

机构地区：[1]景德镇陶瓷学院信息工程学院,景德镇333403

出　　处：《中国陶瓷》2010年第12期39-41,共3页China Ceramics

基　　金：景德镇市科技局软科学项目2009

摘　　要：在陶瓷原料分类中,陶瓷原料的9种化学成分(SiO2,Al2O3,Fe2O3,CaO,MgO,K2O,Na2O,TiO,和I.L)是最直接、容易精确测量和控制的常用的特征参数。本文通过逐步判别分析法优选SiO2,Fe2O3,CaO三个作用比较显著的参数建立21个训练样本的逐步判别分析模型,并对10个待判样本进行分类,得出分类结果。通过回判率检验结果表明,判别函数模型效果显著,误判率为0,可以在生产实际中推广应用。In the classification of ceramic raw materials, the nine chemical composition （SiO2, Al2O3, Pe203, CaO, MgO, K2O, Na2O, TiO, I.L） are the commonly used character parameters which are most direct and easy to measure accurately and controlled. This paper found a classification model of 21 training samples of ceramic raw materials based on stepwise discrimination analysis through choosing three effective parameters（SiO2, Fe2O3, CaO）.And then 10 new samples are classified through using the model. The results show that the classification model of stepwise discrimination analysis has excellent performance, and the accurate ratio is above 90% and the proposed model could be used widely in practice.

关 键 词：陶瓷原料 分类 逐步判别分析法 

分 类 号：TQ174.18[化学工程—陶瓷工业]