利用RBF神经网络的电力网非线性滤波分析  

Nonlinear filtering analysis of power grid based on RBF neural network

作　　者：白云[1] 柴钰[1] 

机构地区：[1]西安科技大学电气与控制工程学院,陕西西安710054

出　　处：《西安科技大学学报》2008年第1期141-144,149,共5页Journal of Xi’an University of Science and Technology

摘　　要：针对电力系统现有的谐波分析方法及其局限性,提出了基于RBF神经网络的非线性滤波分析方法。重点阐述了基于RBF算法的谐波分析原理,并给出了利用该算法进行非线性负载谐波电流滤除的仿真实例。通过仿真实验,分析了RBF神经网络滤波的稳态精度和信号变化时的实时跟踪效果。仿真结果表明了该方法的有效性和易实现性,为电力系统谐波的治理提供了新思路。Aiming at the existing method of harmonic wave analysis and its limitations in power system, a new nonlinear filtering analysis method based on RBF neural network is proposed. A theory is discussed for analysis of harmonic wave based on RBF algorithm. In order to test the algorithm＇s validity, the simulation example for filtering nonlinear load harmonic current is presented. Through the simulation, steady-state precision and real-time tracking effect of filtering is analyzed. Simulation resuits show that the method is effective and easy to realize. It provides a new thought to control harmonic wave in power system.

关 键 词：RBF神经网络 傅立叶算法 谐波分析 非线性滤波 

分 类 号：TM711[电气工程—电力系统及自动化]