摘要
UPFC是最具有代表性的柔性交流输电系统(FACTS)装置,它集串、并联补偿为一体。通过能量缓冲装置的引入,使得UPFC中有功分布控制成为可能。该文采用模糊神经网络(FNN)来控制统一潮流控制器(UPFC)及能量缓冲装置;提出了一种改进的FNN控制器学习算法:在构成隶属函数的神经结构中采用遗传算法,在解模糊过程中采用最小二乘方法。模糊神经网络控制方法结合模糊理论与神经网络各自的优点,使其对于UPFC的控制具有更加灵活稳定和快速的特性和很强的鲁棒性,并使UPFC串并联侧协调控制更加可靠。通过大量的样本学习,验证了该控制系统能够确保各种运行模式下的UPFC正确工作,最后通过仿真证实了该方法的可靠性。
Unified Power Flow Controller (UPFC) is the most representative FACTS device, which has the functions of series and shunt compensation. The energy snubber can be used to control the real powers distributing, too. It is presented to control UPFC and its energy snubber by using the Fuzzy Neural Network (FNN). At the same time, because of the genetic algorithm抯 robust and self-adaptability, it is adopted as the learning algorithm of fuzzy neural network (Genetic Algorithm and Least-Square technique are introduced in making the network structure for the membership function and doing the defuzzification, respectively). The FNN control method is differ from the traditional control theory because it has the virtues of fuzzy theory and neural network. Therefore, this control method is more flexible, steady and robust though UPFC is a strongly nonlinear device. The coordination control of UPFCs series and shunt converter is more reliable. This FNN control system is trained using a great deal of sample learning, so it can make UPFC run in all kinds of mode rightly. At last, the reliability of this method is demonstrated by the system simulation.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2003年第10期83-88,共6页
Proceedings of the CSEE
