基于GA参数整定的时变时滞系统灰色预测控制

针对一类数学模型未知且存在时变时滞的复杂系统,提出一种基于遗传算法参数整定的灰色预测控制方法。该方法采用BP神经网络对系统的时变时滞进行辨识,利用灰色预测算法对系统的输出进行预测,进而使用基于遗传算法整定PID控制器对系统进行输出反馈控制。该方法将灰色预测算法与遗传算法相结合,有效提高了控制器的自适应性。通过仿真实例,结果表明该方法能够对具有大时滞、大惯性、模型不确定等特点的复杂系统进行有效地控制。该方法是可行的、有效的。

Research on Grey-Model Prediction Control Based on GA

For a class of nonlinear systems with unknown input delay, a grey predictive control method based on Genetic Algorithm is proposed. By using BP neural network, the unknown input delay is identified, The system output is estimated by the grey predictive algorithm, and then use PID controller based on genetic algorithm to conduct output feedback control for the system.The method combines grey predictive algorithm with genetic algorithm, the adaptability of the controller is improved effectively. Through simulation, the results show that the method can effectively control complex system with a large delay, large inertia, model uncertainty and other characteristics. The method is feasible and effective.