T-S模糊神经网络应用于变参数系统的控制研究

Research on T-S Fuzzy Neural Network Control of Variable Parameters System

针对基于模型的传统控制策略在线性时变系统中的应用受到系统的时变性和不确定性限制,通常难以获得理想的控制性能这一问题,提出了线性时变系统的一种变参数系统模型。该模型具有有界性和不确定性特点,利用模糊神经网络具有的自学习能力强、模型依赖性小以及鲁棒性强的优点,提出一种基于遗传算法的T-S模糊神经网络控制器对其进行控制研究,并通过仿真实验证明了该模糊神经网络控制器对变参数系统控制的可行性与有效性,为线性时变系统的控制问题提供了一种新思路。

According to the problem that the application of traditional model-based control methods is limited by linear time varing and uncertain property of the system, and often cannot obtain good performance, a model of variable parameter system is proposed. Fuzzy neural network has the advantage of good learning ability, little dependence on model and strong robustness.Aiming at the uncertain and time varying nature of the model, a T-S fuzzy neural network controller based on genetic algorithm is proposed. Simulation result shows that this fuzzy neural network controller used to control of the above variable parameter system has feasibility and validity.So it proposes a new methos for control of linear timely system.