基于LS-SVM的织物烘干含水率模型的应用  

Research and Application of Fabric Moisture Content Model Based on LS-SVM

作　　者：赵迪[1] 金福江[1] 叶斌[1] 

机构地区：[1]华侨大学信息科学与工程学院,福建厦门361021

出　　处：《江南大学学报（自然科学版）》2015年第1期6-10,共5页Joural of Jiangnan University (Natural Science Edition)　

基　　金：国家自然科学基金项目(61143005);福建省产学研重大专项项目(2011H5019);福建省泉州市科技计划重点项目(2011G8)

摘　　要：针对织物烘干过程中工艺参数设定把握不准,导致织物过烘或未烘透造成能源浪费或织物质量下降的问题,分别采用多元非线性回归、扩展的GM(1,1)、最小二乘支持向量机建立织物干燥过程中的含水率预测模型,并通过实验验证方法的有效性。实验结果表明,相对于其他两种模型,基于最小二乘支持向量机的含水率预测模型可以准确学习织物干燥过程中的非线性关系,预测值平均误差低至1.8%。因此,该模型是准确的,可以为烘干环节生产工艺的选取提供依据。It is hard to grasp the parameter setting in the fabric drying process,which will cause over-drying or under-drying,and lead to energy waste or the decreasing of the quality. To solve this problem,this paper,uses the multivariate non-linear regression,extended GM（ 1,1） and the least squares support vector machines（ LS-SVM） to establish the prediction model of moisture content（ dry base） during the drying process,then draws the conclusion through experiments that compared with other two models,the least squares based support vector machine model could effectively learn the non-linear relationship in the drying process for predicting moisture content,and the average prediction accuracy is as low as 1. 8%. As a result,LS-SVM could effectively predict moisture content and provide theoretical basis for the selection of production parameters in the actual drying process.

关 键 词：织物烘干 含水率 预测 建模 最小二乘支持向量机 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TS115.1[自动化与计算机技术—控制科学与工程]