【作者】 张新良； 刘坤；

【Author】 ZHANG Xin-liang,LIU Kun (School of Electronic Information and Electrical Engineering, Shanghai Jiaotong University, Shanghai 200030, China;Department of Automation, Nanjing Institute of Technology, Nanjing 210000, China.

【机构】 上海交通大学电子信息与电气工程学院； 南京工程学院自动化系；

【摘要】 提出一种基于非线性伺服电动机神经网络模型的自适应控制方法．使用外部自回归神经网络对伺服电动机的正向模型进行建模,使所建模型具有更好的泛化性．基于此模型设计了神经网络PID控制器,对控制器的系数进行在线自适应调整,对控制器权值训练常用的Specialized算法进行改进．仿真结果证明了所提出的建模和控制方法的有效性．更多还原

【Abstract】 A novel adaptive PID controller using neural networks has been proposed. The external recurrent neural network (ERNN) has been implemented to establish the forward model of the nonlinear servo-motor to be controlled and the experimental result shows the neural network model(NNI) established by ERNN has better generalization. An adaptive PID controller has been designed based on the N’NI, with the weights of it being updated online. Further, the commonly used specialized learning has been reformed. The numerical simulation results show that the proposed control method is simple and effectively.更多还原