基于支持向量机算法的造纸过程磨后纤维形态软测量模型  

Support Vector Machine Algorithm Based Soft Measurement Model for Post-refining Fiber Morphology of Papermaking Process

作　　者：江伦 满奕[1,2] 李继庚 洪蒙纳[1] 孟子薇[1] 朱小林 JIANG Lun;MAN Yi;LI Jigeng;HONG Mengna;MENG Ziwei;ZHU Xiaolin(State Key Laboratory of Pulp and Paper Engineering,South China University of Technology,Guangzhou,Guangdong Province,510640;Shenzhen Xinyichang Technology Co.,Ltd.,Shenzhen,Guangdong Province,518000)

机构地区：[1]华南理工大学制浆造纸工程国家重点实验室,广东广州510640 [2]深圳新益昌科技股份有限公司,广东深圳518000

出　　处：《中国造纸学报》2020年第2期52-58,共7页Transactions of China Pulp and Paper

基　　金：制浆造纸工程国家重点实验室开放基金(201830)。

摘　　要：本研究提出了一种基于支持向量机(SVM)算法的造纸过程磨后浆料纤维形态软测量模型,该模型利用原始浆板纤维形态参数和磨浆参数作为输入,用于在线软测量生产过程中的磨后浆料纤维形态。结果表明,采用SVM算法进行建模时,7种磨后浆料纤维形态软测量模型的平均相对误差在2.87%~5.61%之间,均优于采用偏最小二乘回归(PLS)算法的建模效果(平均相对误差3.09%~6.60%),模型精度良好,满足生产中纤维形态实时检测对误差的要求。In this study,a soft measurement model of post-refining fiber morphology in a papermaking process based on support vector machine algorithm(SVM)was proposed.The model used the parameters of original pulp sheet and refining as input for online soft measurement of post-refining fiber morphology.The results showed that when SVM was used for modeling,the average relative error of the seven kinds of soft measurement models of post-refining fiber morphology was between 2.87%and 5.61%,which was better than the modeling based on PLS algorithm(the average relative error was between 3.09%and 6.60%),and the model precision was good,which met the error requirements of real-time fiber morphology measurement in production.

关 键 词：磨浆 纤维形态 软测量 SVM算法 

分 类 号：TS752[轻工技术与工程—制浆造纸工程]