摘要
针对气流诱导分离技术的纸张分离率预测问题,提出了一种粒子群优化支持向量回归预测模型。首先,采用BBD(Box-Behnken Design)方法设计多因素正交试验,探究撕裂度、吹嘴流量、吹嘴口径、吹嘴与堆叠纸张间距共4个因素对纸张分离率的影响规律;然后,根据正交试验样本数据建立支持向量回归与粒子群优化-支持向量回归2类预测模型;最后,确定评价指标并对2类预测模型进行对比分析。结果表明,粒子群优化-支持向量回归预测模型有更低的均方根误差、平均绝对误差和平均偏差误差,且决定系数高于支持向量回归预测模型,为气流诱导分离技术的纸张分离率预测提供了一种有效的方法。
To address the prediction of paper separation rate in air-induced separation technology,a prediction model based on Particle Swarm Optimization and Support Vector Regression(PSO-SVR)is proposed.Firstly,Box-Behnken Design method was used to design a multi-factor orthogonal experiment to investigate the effects of four factors,namely tearing degree,mouthpiece flow rate,mouthpiece caliber,and spacing between mouthpiece and stacked paper on paper separation rate;then,two kinds of prediction models of support vector regression and particle swarm optimization-support vector regression are established according to the data of orthogonal experiment;finally,the evaluation index is determined and the two types of prediction models are compared and analyzed.The results show that the PSO model has lower root-mean-square error,mean absolute error and mean deviation error,and its coefficient of determination is higher than that of SVM model,which provides an effective method for paper separation rate prediction by air-induced separation technology.
作者
任龙
赵丽琴
郝仰彪
李彦璋
马清艳
冯亮
REN Long;ZHAO Liqin;HAO Yangbiao;LI Yanzhang;MA Qingyan;FENG Liang(School of Mechanical Engineering,North University of China,Taiyuan 030051,China;Shanxi Limin Industrial Co.,Ltd.,Taiyuan 030051,China)
出处
《机械与电子》
2025年第8期16-22,共7页
Machinery & Electronics
基金
山西省专利转化专项计划项目(202302007)。
