基于广义生长-剪枝RBF神经网络的谐波源建模.pdfVIP

25 卷 16 期 中 国 电 机 工 程 学 报 Vol.25 No.16 Aug. 2005 2005 年8 月 Proceedings of the CSEE 2005 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2005) 16-0042-05 中图分类号：TM711 文献标识码：A 学科分类号：47040 基于广义生长-剪枝径向基函数 神经网络的谐波源建模 1 1 1 2 勇 ，程浩忠 ，葛乃成 ，黄广兵 ( 1. 上海交通大学电气工程系，200240 上海市 闵行区; 2. 南洋理工大学, 639798 新加坡) GENERALIZED GROWING AND PRUNING RBF NEURAL NETWORK BASED HARMONIC SOURCE MODELING 1 1 1 2 ZHAN Yong , CHENG Hao-zhong , GE Nai-cheng , HUANG Guang-bin ( 1. Shanghai Jiaotong University, Minhang District, Shanghai 200240, China; 2. School of Electrical and Electronic Engineering, Nanyang Technological University, Nanyang Avenue, 639798, Singapore) ABSTRACT: A novel generalized growing and pruning RBF 于电力系统的安全、经济运行具有重要的意义。为 neural network (GGAP-RBF) is proposed for modeling 了对系统的谐波进行有效的分析和治理，需要有适 nonlinear electric loads in steady-state frequency-domain. In 当的模型来表征谐波源的特性。电力系统中的谐波 the model, the nonlinearity mapping between harmonic 源大体上可分成两种类型，一类为电力电子设备， voltages and harmonics currents is established by GGAP-RBF. GGAP-RBF is a sequential learning algorithm so that model 它们在运行中按一定规律控制电路的通断，注入谐 can be established dynamically. Numerical results show that the 波电流到系统中。这类谐波源所产生的谐波电流可 proposed method, having the characteristics of short training 根据供电电压波形、设备的电路结构及参数和控制 time, high precision and dynamic modeling, is an effective [1]