摘要
纸页干燥过程的排风温度、排风湿度和纸幅湿度是反映干燥部运行状态和工艺合理性的重要参数,其预测控制对干燥部的运行优化和节能降耗具有重要的意义,传统的机理预测模型面临一些过程变量难以准确确定的问题。随着工业大数据的发展,基于数据驱动的方法为建立纸页干燥过程关键参数的预测模型提供了一条有效的途径。通过相关性分析从生产数据中选取与关键参数相关的特征变量作为模型的输入,采用弹性网络算法建立前烘干燥部的排风温湿度和纸幅湿度预测模型。结果表明,采用弹性网络算法建模的预测结果更贴近实际变化趋势,预测模型的平均相对误差均小于7%,比SVR模型低2%以上。模型为验证干燥部操作变量的调控合理性提供了一种有效的验证方法,为工业用纸干燥过程的优化控制提供了新的依据。
The exhaust air temperature the exhaust air humidity and the paper humidity are important parameters that reflect the operating state and the rationality of the parameters of dryer section during paper drying,and its predictive control is of great significance to the energy saving and consumption reduction in dryer section of paper machine.The traditional mechanism prediction model faces the problem that some process variables are difficult to determine accurately.With the development of industrial big data,data-driven methods provide an effective way to establish the parameter prediction models for the process of paper drying.In this study,the characteristic variables related to the predictive parameters are selected from the production data as the inputs of the model through correlation analysis,and the prediction models of the exhaust air temperature、the exhaust air humidity and the paper humidity of the front dryer section are established by using the elastic network algorithm.The results show that the prediction results modeled by elastic network are closer to the actual change trend,and the average relative error of the parameter prediction models are less than 7%,which is more than 2%lower than that of the SVR model.The model provides an effective verification method for verifying the rationality of the control of operating variables in the drying section,and provides a new basis for the optimal control of the drying process of industrial paper.
作者
马亚运
洪蒙纳
李继庚
何正磊
满奕
MA Yayun;HONG Mengna;LI Jigeng;HE Zhenglei;MAN Yi(State Key Lab of Pulp and Paper Engineering,South China University of Technology,Guangzhou 510640,China;China-Singapore International Joint Research Institute,Guangzhou 510640,China)
出处
《造纸科学与技术》
2022年第2期1-6,共6页
Paper Science & Technology
基金
国家重点研发计划(2020YFE0201400)
国家自然科学基金(52000078)。
关键词
干燥部
弹性网络
预测模型
关键参数
dryer section
elastic net algorithm
prediction model
key parameters
