【作者】 沙道航； 杨华勇；

【Author】 SHA Daohang YANG Huayong State Key Laboratory of Fluid Power Transmission and Control Zhejiang University, Hangzhou, 310027

【机构】 浙江大学流体传动及控制国家重点实验室；

【摘要】 本文介绍一种采用两个BP神经网络的非线性系统控制的结构及原理、一个用来对被控系统进行在线辨识,另一个用做非线性自适应控制器。并对具有未知外部负载干扰的电液位置伺服系统进行了动态仿真,讨论了神经元激励函数形状因子、神经网络节点数以及神经网络训练次数对控制系统性能的影响。更多还原

【Abstract】 In this paper, a scheme of nonlinear system control using neural networks is proposed to the Electro-hydraulic servo systems subjected unknown external load disturbances and parameters variations. This control system consists of two parts, a controller using the neural networks and an identification using another neural networks. The simulation show that the performance of control system is independent of the plant parameters and the external disturbances when the structure of neural networks is selected appropriately. The affection of the factor of Sigmoid function shape, the unit numbers of neural networks and the learning times of neural networks are discussed.更多还原