摘要
为了实现混沌运动的控制,应用神经网络模型预测控制的策略。首先通过系统辨识建立要控制的混沌系统的神经网络模型,并设计出神经网络模型预测控制器,然后通过该控制器对混沌运动进行预测控制,使混沌吸引子中的不稳定周期轨道被镇定到不动点。最后借助MATLAB平台进行数字仿真,结果表明该控制方法对混沌控制是有效的。
The Van der pol forced vibrator is a typical chaotic dynamic system. It is chaotic motion status under certain initial conditions. The neural network model is used for the chaotic prediction control of the system. First, it establishes the neural network model of the Van der pol forced vibrator with system identification. Then the prediction control is applied to the chaotic motion using the system model to control the unsteady periodic track in the chaotic attractor to the fixed point. Simulation results show that the method is not only effective, but also robust in the chaos control.
