王勇;付志红;张淮清;王好娜;侯兴哲;

• 1:输配电装备及系统安全与新技术国家重点实验室(重庆大学)
• 2:国家电网公司电能计量器具性能评估实验室

摘要(Abstract)：

电网基波是电能计量和电能质量评估的重要指标,提出了基于前馈神经网络的电网基波频率和幅值的高精度检测方法。根据数学推导得出:正弦信号过零点与其两侧对称两点的连线与时间轴交点的时间差,同频率满足单调关系,但并非严格的线性关系,而且与幅值无关,据此用前馈神经网络建立该时间差与频率的映射关系。Matlab仿真表明,提出的算法对频率的检测精度达到10?4级,幅值的检测精度高达10?5级,远远高于快速傅里叶变换和Hanning窗的插值算法;随机噪声和谐波对前馈神经网络测量精度的影响很小,该算法具有较强的抗干扰能力。

关键词(KeyWords)： 电网基波;前馈神经网络;基波频率;基波幅值

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(40874094)~~

作者(Author): 王勇;付志红;张淮清;王好娜;侯兴哲;

Email:

DOI: 10.13335/j.1000-3673.pst.2011.08.007

参考文献(References)：

• [1]谢小荣,韩英铎.电力系统频率测量综述[J].电力系统自动化,1999,23(3):54-58.Xie Xiaorong,Han Yingduo.An overview on power system frequencymeasurement[J].Automation of Electric Powe Systems,1999,23(3):54-58(in Chinese).
• [2]邱海锋,周浩.非同步采样下电网谐波分析方法的探讨[J].继电器,2008,36(1):57-62.Qiu Haifeng,Zhou Hao.Study on the approachs of electricalharmonic analysis for asynchronous sampling[J].Relay,2008,36(1):57-62(in Chinese).
• [3]胡铭,陈珩.电能质量及其分析方法综述[J].电网技术,2000,24(2):36-38.Hu Ming,Chen Heng.Survey of power quality and its analysismethods[J].Poewr System Technology,2000,24(2):36-38(inChinese).
• [4]薛蕙,杨仁刚.基于FFT的高精度谐波检测算法[J].中国电机工程学报,2002,22(12):106-110.Xue Hui,Yang Rengang.Precise algiorithms for harmonic analysisbased on FFT algorithm[J].Proceedings of the CSEE,2002,22(12):106-110(in Chinese).
• [5]林海雪.现代电能质量的基本问题[J].电网技术,2001,25(10):5-12.Lin Haixue.Main problems of modern power quality[J].PowerSystem Technology,2001,25(10):5-12(in Chinese).
• [6]Aghazadehr L.New technique for frequency and amplitude estimationof powersystem signals[J].Proceedings of the IEEE,2005,152(3):435-440.
• [7]Thomasdwp W.Evaluation of frequency tracking methods[J].IEEETrans on Power Delivery,2001,16(3):367-371.
• [8]Pradhanak R B.Power system frequency estimation using least meansquare technique[J].IEEE Trans on Power Delivery,2005,20(3):1812-1816.
• [9]刘涤尘,夏利民,商志会.基于人工神经网络的电网频率测量方法[J].电网技术,2000,24(8):42-45.Liu Dichen,Xia Limin,Shang Zhihui.An artificial neural networkapproach for measuring power system frequency[J].Power SystemTechnology,2000,24(8):42-45(in Chinese).
• [10]高大启.有教师的线性基本函数前向三层神经网络结构研究[J].计算机学报,1998,21(1):80-86.Gao Daqi.On structures of supervised linear basis functionfeedforward three-layered neural networks[J].Chinese Journal ofComputers,1998,21(1):80-86(in Chinese).
• [11]王群,吴宁,王兆安.一种基于人工神经网络的谐波测量方法[J].电网技术,1999,23(1):29-32.Wang Qun,Wu Ning,Wang Zhaoan.An artificial neural networkapproach for measure harmonic[J].Power System Technology,1999,23(1):29-32(in Chinese).
• [12]王小华,何怡刚.一种新的基于神经网络的高精度电力系统谐波分析算法[J].电网技术,2005,29(3):72-75.Wang Xiaohua,He Yigang.A new neural network based powersystem harmonics analy algorithm with high accuracy[J].PowerSystem Technology,2005,29(3):72-75(in Chinese).
• [13]江学军,唐焕文.前馈神经网络泛化能力的系统分析[J].系统工程理论与实践,2000,20(8):36-40.Jiang Xuejun,Tang Huanwen.System analysis for generalization ofmultilayer feedforward neural network[J].Systems EngineeringTheory&Practice,2000,20(8):36-40(in Chinese).
• [14]王小华,何怡刚.一种新的基于神经网络的高精度电力系统谐波分析算法[J].电网技术,2005,29(3):72-75.Wang Xiaohua,He Yigang.A new neural network based powersystem harmonic analysis algorithm with high accuracy[J].PowerSystem Technology,2005,29(3):72-75(in Chinese).
• [15]Laxmidhar B,Swagat K,Awhan P.On adaptive learning rate thatguarantees convergence in feedforward networks[J].IEEE Trans onNeural Networks,2006,17(5):1116-1125.
• [16]Ng S C,Cheung C C,Leung S H.Magnified gradient function withdeterministic weight modification in adaptive learning[J].IEEE Transon Neural Networks,2004,15(6):1411-1423.