不确定滞后系统的灰色预测神经元控制

Grey prediction neuron control of uncertain systems with time delay

针对不确定滞后系统建模和控制方面的困难,提出了利用神经网络对不确定滞后系统的滞后时间进行辨识,并在此基础上采用变步长灰色预测方法对不确定滞后系统的行为进行预测,同时结合单神经元控制器克服辨识误差、预测误差、系统干扰等不确定因素带来的影响,从而提高系统的自适应性。仿真结果表明,控制系统运行平稳,受不确定因素的影响较小,说明这种方法对于不确定的时变滞后系统具有良好的控制效果。

To overcome the difficult in modeling and controlling uncertain systems with time delay, a method of neural network is proposed to identify the delay time of such a system. Moreover, the step-varying grey prediction algorithm is used to predict future behaviors of uncertain systems with time delay. Combined with the neuron controller, this algorithm can get over the impacts of the identifying error, prediction error and system disturbance, and considerably improve the adaptability of the control system. Computer simulation results show that the control system using the proposed algorithm works stably and robustly, which shows that the algorithm has a good control performance over the uncertain systems with time delay.\;