基于CMAC的双馈水轮发电机系统控制策略研究

STUDY OF CONTROL STRATEGIES FOR HYDROGENERATOR SYSTEM WITH DOUBLY FED GENERATORS BASED ON CMAC NEURAL NETWORK

针对可调速双馈水轮发电机系统的不确定性、非线性和参数时变的特点,提出了一种采用小脑模型(CMAC)神经网络的自适应控制策略。该控制策略以系统动态误差和给定信号量作为CMAC的激励信号,并与自适应神经网络控制器相结合构成系统的复合控制。该文对双馈水轮发电机系统的稳态调节和暂态特性进行了数字仿真研究,并与常规的PID控制进行比较。结果表明,基于CMAC的自适应控制策略对系统模型结构和参数变化、负荷扰动都具有很好的适应性和鲁棒性,控制品质优良,是一种适于在线学习控制的双馈水轮发电机系统控制方法。

According to the features of uncertain and nonlinear as well as parameters time-variation for adjustable speed hydrogenerator system with doubly fed generators (DFG), a self-adaptive control strategy based on the cerebellar model articulation controller (CMAC) neural network is presented in this paper. Combining it with the adaptive neural network controllers, the multiplex control strategy uses the dynamic errors and given signals of the system as input parameters to the CMAC neural network. The digital simulation of the steady state regulation and transient operation characteristics for hydrogenerator system with DFG is studied by using the proposed control scheme. By comparison the conventional PID controller, the simulation results show that the proposed control scheme is of good adaptability and robustness against the changes of model, parameters and the external load disturbance, it has an excellent dynamic performance and thus is an on-line learning control method for the control of hydrogenerator system with DFG