基于数据驱动的纸浆洗涤过程优化控制  

Optimal Control of Pulp Washing Process Based on Data Driven

作　　者：赵钢[1] ZHAO Gang(General Education School,Xi’an Aeronautical Polytechnic Institute,Xi’an 710089,China)

机构地区：[1]西安航空职业技术学院,西安710089

出　　处：《造纸科学与技术》2020年第5期35-39,共5页Paper Science & Technology

摘　　要：详细介绍了纸浆洗涤过程优化控制方案,模拟了纸浆洗涤的整个过程,并在此基础上通过神经网络动态模型对洗洗过程进行辨识,分别确定的动态模型和稳态模型下的模型结构参数和模型训练参数,提出了BP算法与L-M算法相结合的优化控制方案。经实验研究发现,研究所设计的两步神经网络法能够对纸浆洗涤过程进行充分地辨识。The optimal control scheme of pulp washing process was introduced in detail,and the whole process of pulp washing was simulated.On this basis,the neural network dynamic model was used to identify the washing process.The structural parameters and model training parameters of the dynamic model and the steady model were determined respectively,and the optimal control scheme combining BP algorithm and L-M algorithm was proposed.The experimental results showed that the two-step neural network method designed by this research institute could fully identify the pulp washing process.

关 键 词：纸浆洗涤 数据驱动 神经网络 优化控制 

分 类 号：TS75[轻工技术与工程—制浆造纸工程]