【摘要】 本文提出一种基于Hopfield的动态神经网络 ,以具有非线性时变特性的电液伺服速度系统为对象 ,采用趋化算法训练网络 ,进行辨识。辨识结果经过再输入正弦信号考察 ,结果表明 ,采用本文提出的方法是可行的和有效的 ,可以获得较好的效果更多还原

【Abstract】 In this paper, a dynamical neural network based on Hopfield neural network is presented. It is used to identify a electro-hydraulic velocity servo system having nonlinear time-vary characteristics. The DNN is trained by Chemotaxis algorithm and the finished network is tested by inputting again a sin signal. The results show that the proposed method is feasible and efficient.更多还原