摘要
提出了一种新的基于三角基函数的人工神经网络算法,利用该算法可一次性获得电力系统基波及各次谐波的频率、幅值和相位,提出并证明了该神经网络算法的收敛定理,给出了利用该算法进行谐波分析的仿真实例。仿真结果表明,文中提出的谐波测量方法的计算精度极高,且计算量较小,因此在电力系统谐波测量中有较高的应用价值。
A triangle base function based new artificial neural network (ANN) algorithm was presented by which the frequencies, amplitudes and phases of fundamental and various orders of harmonics could be obtained once only. The convergence theorem of the presented ANN algorithm was proposed and proved, and the simulation cases of harmonic analysis by use of the presented algorithm were given. The simulation results showed that the presented harmonic measuring method was very accurate and the calculation amount was small, so it could be applied to power system harmonic measurement.
出处
《电网技术》
EI
CSCD
北大核心
2005年第3期72-75,共4页
Power System Technology
基金
国家自然科学基金资助项目(50277010)
湖南省教育厅科研项目(04C073)~~
关键词
电力系统
谐波分析
算法
神经网络
收敛定理
傅立叶变换
Algorithms
Computer simulation
Convergence of numerical methods
Electric power systems
Fast Fourier transforms
Neural networks