一种优化支持向量机回归算法的印刷工序损耗值预测方法  

A Prediction Method of Printing Process Loss Value based on Optimized Support Vector Regression

作　　者：彭来湖[1,2] 孙海涛 李建强 胡旭东[1] PENG Laihu;SUN Haitao;LI Jianqiang;HU Xudong(Zhejiang Key Laboratory of Modern Textile Equipment Technology,Zhejiang Sci-Tech University,Hangzhou 310000,China;Zhejiang Sci-Tech University Longgang Research Institute,Wenzhou 325000,China;College of Biomedical Engineering and Instrument Science,Zhejiang University,Hangzhou 310000,China)

机构地区：[1]浙江理工大学浙江省现代纺织装备技术重点实验室,浙江杭州310000 [2]浙江理工大学龙港研究院,浙江温州325000 [3]浙江大学生物医学工程与仪器科学学院,浙江杭州310000

出　　处：《软件工程》2023年第3期36-40,5,共6页Software Engineering

基　　金：浙江省博士后科研项目(ZJ2020004)。

摘　　要：针对印刷生产中物料需求计划的损耗值采用经验值的问题,提出一种优化支持向量机回归算法的印刷工序损耗值预测方法。通过皮尔逊相关系数量化特征值选取;采用布谷鸟搜索算法优化支持向量机回归算法的超参数选取,建立损耗预测模型;为验证模型的优越性,分别与不同的特征值选取方案、优化算法、回归算法的模型进行对比。实验结果表明该损耗预测方法具有更高泛化性和预测精度,决定系数、平均绝对百分误差、均方根误差分别为0.995、0.005、1.969,为解决后续相关问题提供了技术支持。Aiming at the empirical value problem of material demand planning loss value in printing production, this paper proposes a method to predict the loss value of printing process by optimizing Support Vector Regression. Firstly,the Pearson Correlation Coefficient is used to quantify the eigenvalue selection. Then, the Cuckoo Search algorithm is used to optimize the selection of super parameters of Support Vector Regression algorithm, and the loss prediction model is established. Finally, to verify the superiority of the model, it is compared with models of different eigenvalue selection schemes, optimization algorithms and regression algorithms. The experimental results show that the proposed loss prediction method has higher generalization and prediction accuracy, and the determination coefficient, average absolute percentage error, and root mean square error are 0.995, 0.005, and 1.969 respectively. It provides theoretical support for the follow-up related problems.

关 键 词：印刷工序 损耗预测 皮尔逊相关系数 支持向量机回归算法 布谷鸟搜索算法 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]