【作者】 向微； 陈宗海； 张海涛； 秦廷； 王雷； 李明；

【Author】 XIANG Wei CHEN Zong-hai ZHANG Hai-tao QIN Ting WANG Lei LI Ming (Department of Automation,University of Science and Technology of China,Hefei,230027)

【机构】 中国科学技术大学自动化系；

【摘要】 传统的基于机理或局部线性化的控制策略不足以解决越来越复杂的控制问题,而神经网络用于控制也存在泛化能力差等缺陷,因此本文提出了一种将系统机理和RBF神经网络结合起来实现逆模控制的方法,一方面能发挥神经网络非线性逼近的强大功能,一方面利用系统机理信息指导NN的收敛方向,改进了神经网络的泛化能力。由此方法设计的逆模控制器,在保证控制精度的前提下,速度远快于标准RBFNN逆模控制器,且对扰动、时延、非线性及对象参数的摄动有较强的适应能力,具有良好的控制品质。更多还原

【Abstract】 Traditional control methods are not satisfied in more and more complex process control,and neural networks used in control may in short of generalization ability.In this paper,a novel structure combining process characteristic knowledge and RBFNN is presented to direct neural networks convergence and exert the excellent capability on nonlinear approach of neural networks.Simulation results show that the compute velocity of the backward model Controller using this hybrid RBFNN,while the control precision index is ensured,is far more faster than backward model controllers using common RBFNN.And also,this hybrid RBFNN backward model controller has excellent control quality,it shows good adaptation to disturbance,time delay,nonlinear and the drift of plant parameters.更多还原