基于最小二乘支持向量机的棉针织物活性染料湿蒸染色预测模型
|
摘要
针对棉针织物在用活性染料连续湿蒸染色过程中出现的染色条件对织物色光难以控制和预测,易导致染色织物不符合预期产品要求的问题,选用雷马素金黄RGB对棉针织物进行湿蒸染色,研究了元明粉和纯碱浓度、汽蒸时间对织物表观染色深度(K/S值)的影响,同时基于最小二乘支持向量机(LS-SVM),将这些影响因素作为预测模型的输入变量,织物K/S值作为输出变量,建立了多因素模型并进行预测。结果表明,织物K/S实验值和模型预测值的相关系数高达0. 999 6,平均相对误差小于1%,说明该模型具有较高的精度,该建模方法可用于预测织物K/S值,为棉针织物活性染料湿蒸染色工艺的优化提供参考。
Aiming at the problem of hard control and prediction of dyeing conditions on the color of dyed fabrics in the continuous wet-steaming dyeing of cotton knitted fabrics with reactive dye,the influences of sodium sulfate concentration,soda concentration,and steaming time on the color depth( K/S value) of the dyed fabrics were studied in the wet-steaming dyeing process of cotton knitted fabrics with Remazol golden yellow RGB. At the same time,based on least squares support vector machine( LS-SVM),using these factors as the input variables of the prediction model and the K/S value of fabric color depth as the output variable,a multi-factor model of K/S value was established to predict K/S value. The experiment results show that the correlation coefficient between the experimental value and the predicted value of the model is0. 999 6,and the mean relative error is lower than 1%,which indicates that the model has high accuracy.The modeling method can be applied to predict the K/S value of fabric,providing a basis reference for the optimization of the wet-steaming reactive dyeing process conditions for cotton knitted fabric.
Aiming at the problem of hard control and prediction of dyeing conditions on the color of dyed fabrics in the continuous wet-steaming dyeing of cotton knitted fabrics with reactive dye,the influences of sodium sulfate concentration,soda concentration,and steaming time on the color depth( K/S value) of the dyed fabrics were studied in the wet-steaming dyeing process of cotton knitted fabrics with Remazol golden yellow RGB. At the same time,based on least squares support vector machine( LS-SVM),using these factors as the input variables of the prediction model and the K/S value of fabric color depth as the output variable,a multi-factor model of K/S value was established to predict K/S value. The experiment results show that the correlation coefficient between the experimental value and the predicted value of the model is0. 999 6,and the mean relative error is lower than 1%,which indicates that the model has high accuracy.The modeling method can be applied to predict the K/S value of fabric,providing a basis reference for the optimization of the wet-steaming reactive dyeing process conditions for cotton knitted fabric.
引文
[1]李连颖,臧少玉,毛志平,等.平绒织物湿蒸短流程轧染工艺条件的优化[J].印染助剂,2010,27(1):45-48.LI Lianying,ZANG Shaoyu,MAO Zhiping,et al.The optimization of short wet-steam process in pad-dyeing panne fabric[J].Textile Auxiliaries,2010,27(1):45-48.
[2]李连颖,王天靖,陈志华,等.ECO活性染料湿蒸短流程和轧烘轧蒸染色工艺[J].印染助剂,2010,27(2):39-42.LI Lianying,WANG Tianjing,CHEN Zhihua,et al.Short wet-steam process and pad-dry-pad-steam dyeing process of ECO reactive dye[J].Textile Auxiliaries,2010,27(2):39-42.
[3]汪岚,金福江.活性染料上染率的多元预测模型分析[J].纺织学报,2008,29(8):78-80.WANG Lan,JIN Fujiang.Analysis of reactive dye uptake-rate of multivariate regression[J].Journal of Textile Research,2008,29(8):78-80.
[4]姜会钰,杨锋,王振东,等.BP神经网络预测活性染色织物K/S值[J].印染,2007(9):23-24.JIANG Huiyu,YANG Feng,WANG Zhendong,et al.K/S value prediction of the reactive dyeing with BP nerve network[J].China Dyeing&Finishing,2007(9):23-24.
[5]GESTEL T V,SUYKENS J A K,BAESENS B,et al.Benchmarking least squares support vector machine classifiers[J].Machine Learning,2004,54(1):5-32.
[6]WU Yonghong,SHEN Hui.Grey-related least squares support vector machine optimization model and its application in predicting natural gas consumption demand[J].Journal of Computational and Applied Mathematics,2018,338:211-220.
[7]MAO Xuefei,WANG Yijun,LIU Xiangdong,et al.Ahybrid feedforward-feedback hysteresis compensator in piezoelectric actuators based on least squares support vector machine[J].IEEE Transactions on Industrial Electronics,2018,65(7):5704-5711.
[8]FU Li,LUO Jun,CHEN Weimin,et al.LS-SVM-based surface roughness prediction model for a reflective fiber optic sensor[J].Chinese Optics Letters,2017,15(9):61-65.
[9]王建国,张文兴.支持向量机建模及其智能优化[M].北京:清华大学出版社,2015:32-33.WANG Jianguo,ZHANG Wenxing.Support Vector Machine Modeling and Intelligent Optimization[M].Beijing:Tsinghua University Press,2015:32-33.
[10]王克奇,杨少春,戴天虹,等.采用遗传算法优化最小二乘支持向量机参数的方法[J].计算机应用与软件,2009,26(7):109-111.WANG Keqi,YANG Shaochun,DAI Tianhong,et al.Method of optimizing parameter of least squares support vector machines by genetic algorithm[J].Computer Applications and Software,2009,26(7):109-111.
[11]齐亮.基于蚁群算法的支持向量机参数选择方法研究[J].系统仿真技术,2008,4(1):14-18.QI Liang.Parameters selection of support vector machine based on ant colony algorithm.system simulation technology[J].System Simulation Technology,2008,4(1):14-18.
[12]曾勍炜,徐知海,吴键.基于粒子群优化和支持向量机的电力负荷预测[J].微电子学与计算机,2011,28(1):147-149,153.ZENG Qingwei,XU Zhihai,WU Jian.Forecasting of electricity load based on particle swarm optimization and support vector machine[J].Microelectronics&Computer,2011,28(1):147-149,153.
[2]李连颖,王天靖,陈志华,等.ECO活性染料湿蒸短流程和轧烘轧蒸染色工艺[J].印染助剂,2010,27(2):39-42.LI Lianying,WANG Tianjing,CHEN Zhihua,et al.Short wet-steam process and pad-dry-pad-steam dyeing process of ECO reactive dye[J].Textile Auxiliaries,2010,27(2):39-42.
[3]汪岚,金福江.活性染料上染率的多元预测模型分析[J].纺织学报,2008,29(8):78-80.WANG Lan,JIN Fujiang.Analysis of reactive dye uptake-rate of multivariate regression[J].Journal of Textile Research,2008,29(8):78-80.
[4]姜会钰,杨锋,王振东,等.BP神经网络预测活性染色织物K/S值[J].印染,2007(9):23-24.JIANG Huiyu,YANG Feng,WANG Zhendong,et al.K/S value prediction of the reactive dyeing with BP nerve network[J].China Dyeing&Finishing,2007(9):23-24.
[5]GESTEL T V,SUYKENS J A K,BAESENS B,et al.Benchmarking least squares support vector machine classifiers[J].Machine Learning,2004,54(1):5-32.
[6]WU Yonghong,SHEN Hui.Grey-related least squares support vector machine optimization model and its application in predicting natural gas consumption demand[J].Journal of Computational and Applied Mathematics,2018,338:211-220.
[7]MAO Xuefei,WANG Yijun,LIU Xiangdong,et al.Ahybrid feedforward-feedback hysteresis compensator in piezoelectric actuators based on least squares support vector machine[J].IEEE Transactions on Industrial Electronics,2018,65(7):5704-5711.
[8]FU Li,LUO Jun,CHEN Weimin,et al.LS-SVM-based surface roughness prediction model for a reflective fiber optic sensor[J].Chinese Optics Letters,2017,15(9):61-65.
[9]王建国,张文兴.支持向量机建模及其智能优化[M].北京:清华大学出版社,2015:32-33.WANG Jianguo,ZHANG Wenxing.Support Vector Machine Modeling and Intelligent Optimization[M].Beijing:Tsinghua University Press,2015:32-33.
[10]王克奇,杨少春,戴天虹,等.采用遗传算法优化最小二乘支持向量机参数的方法[J].计算机应用与软件,2009,26(7):109-111.WANG Keqi,YANG Shaochun,DAI Tianhong,et al.Method of optimizing parameter of least squares support vector machines by genetic algorithm[J].Computer Applications and Software,2009,26(7):109-111.
[11]齐亮.基于蚁群算法的支持向量机参数选择方法研究[J].系统仿真技术,2008,4(1):14-18.QI Liang.Parameters selection of support vector machine based on ant colony algorithm.system simulation technology[J].System Simulation Technology,2008,4(1):14-18.
[12]曾勍炜,徐知海,吴键.基于粒子群优化和支持向量机的电力负荷预测[J].微电子学与计算机,2011,28(1):147-149,153.ZENG Qingwei,XU Zhihai,WU Jian.Forecasting of electricity load based on particle swarm optimization and support vector machine[J].Microelectronics&Computer,2011,28(1):147-149,153.