参数在线跟踪的交流传动系统双神经网络自适应规划控制

Dual-Hopfield Neural Network based online parameters' tracking and adaptive dynamics programming control of AC drive system

本文将Hopfield神经网络引入交流传动系统的自适应控制 ,通过神经网络来规划交流调速系统的速度控制器动态输出 ,使速度控制具有对某些参数变化的一定程度的鲁棒性 ,对于不可控的负载转矩分量 ,加入作者先前所提出的参数自动跟踪神经网络[3 ] ,构成具有参数在线跟踪功能的交流传动双神经网络自适应规划控制模式 ,进一步提高了系统的控制性能。

In this paper, Hopfield Neural Network (HNN) is introduced into adaptive control of AC drive system, in which the dynamics of its speed controller is set by HNN programming result, thus making the speed control result robust to some parameters' changing. As to the uncontrollable load torque part, the adaptive parameters' tracking HNN is added into the system to make out the dual-HNN based adaptive AC drive system programming tracking control mode, thus improving the control performance of the system. These results are proved by simulation.