活性染料耐洗色牢度支持向量机分类模型  

Support Vector Classification Model for Colour Fastness to Washing of Reactive Dyes

作　　者：黄磊[1,2] 禹新良 杨辉琼[1] HUANG Lei;YU Xin-liang;YANG Hui-qiong(College of Chemistry and Chemical Engineering,H unan Institute of Engineering,Xiangtan 411104,China;Dongguan Texwinca Textile &Garment Limited,Dongguan 523138,China;Key Laboratory of Ecological Textile Materials &Novel Dyeing and Finishing Technology,Hunan Institute of Engineering,Xiangtan 411104,China)

机构地区：[1]湖南工程学院化学化工学院,湘潭411104 [2]东莞德永佳纺织制衣有限公司,东莞523138 [3]湖南工程学院生态纺织材料及染整新技术湖南省高校重点实验室,湘潭411104

出　　处：《湖南工程学院学报（自然科学版）》2018年第4期62-66,71,共6页Journal of Hunan Institute of Engineering(Natural Science Edition)

基　　金：湖南省自然科学基金资助项目(12JJ6011);湖南省教育厅科研资助项目(16A047);化学生物传感与计量学国家重点实验室(湖南大学)开放课题(2016013)

摘　　要：当今人们对染料染色织物的耐洗色牢度越来越关注.报道了基于二元分类问题的支持向量机分类(support vector classification,SVC))模型预测活性染料分子耐洗色牢度等级.在测试123支活性染料的耐洗色牢度、溶解过程中的抽滤时间、pH值之后,采用粒子群优化(particle swarm optimization,PSO)算法搜寻SVC模型参数C与γ值.模型采用3个变量(活性染料的pH值、抽滤时间、及结构特征定性变量)作为SVC分类模型的输入.所建分类模型对训练集活性染料分子耐洗色牢度等级预测整体准确度为84.1%,对测试集染料分子整体预测准确度为77.1%.结果表明,所建模型精确、可靠,可以用于活性染料耐洗色牢度等级预测.Nowadays people are more concerned about the colour fastness to washing of textiles. This paper describes the pattern recognition used for color fastness to washing of reactive dyes by applying a support vector classification (SVC) technique for a two-class problem. After measuring the color fastness to wash-ing, filtering time and pH values of 123 reactive dyes, the optimal SVC model is obtained by applying the particle swarm optimization (PSO) algorithm. The model is based on three variables, which reflect the types of chromophoric groups, filtering time and pH values of reactive dyes. The classification model pos-sesses overall accuracies of 84.1% for the training set and 77.1% for the test set. The results show that the SVC is accurate and acceptable in predicting color fastness to washing of reactive dyes.

关 键 词：耐洗色牢度 粒子群优化算法 活性染料 支持向量机分类 

分 类 号：TS193.63[轻工技术与工程—纺织化学与染整工程] O642.1[轻工技术与工程—纺织科学与工程]