摘要
BOPP薄膜厚度控制系统是一个复杂的多变量耦合系统,通过对其工艺过程分析可得到三通道厚度控制系统传递函数模型,利用相对增益矩阵对其进行耦合性分析。由于现有的解耦算法无法完全消除系统间的耦合影响,使用径向基函数模型(RBF)对系统进行解耦设计。针对其存在训练过程长的问题,使用快速自学习算法、附加微分项与粒子群优化算法(PSO)对RBF模型进行优化,提升系统的抗干扰能力与响应速度。实验结果表明,提出的ASRBFD方法抗干扰能力强、RBF模型训练效率高,系统解耦性能优良。
The BOPP film thickness control system is a complex multivariable coupling system.The three-channel transfer function matrix model was obtained by analyzing the process,and the coupling analysis was carried out using the relative gain matrix.Since the existing decoupling algorithm could not completely eliminate the coupling effect,the radical basis function(RBF)model was used to decouple the system.Aiming at the long training process of RBF decoupling algorithm,fast self-learning,additional differential term and particle swarm optimization(PSO)were used to optimize the RBF model to improve the anti-interference and response speed of the system.Experiment results show that the ASRBFD method has strong anti-interference ability,high RBF model training efficiency and excellent decoupling performance.
作者
廖雪超
陈振寰
邓万雄
LIAO Xue-chao;CHEN Zhen-huan;DENG Wan-xiong(School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,China;Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System,Wuhan University of Science and Technology,Wuhan 430065,China)
出处
《计算机工程与设计》
北大核心
2021年第4期1079-1088,共10页
Computer Engineering and Design
基金
国家自然科学基金项目(61502359)。
关键词
解耦控制
RBF神经网络
快速自学习算法
附加微分项
粒子群优化算法
decoupling control
RBF neural network
fast self-learning
additional differential term
particle swarm optimization
