基于递归特性分析和BP神经网络的电能质量扰动识别与定位(英文)Based on Recursive Feature Analysis and BP Neural Network Analyze Identification and Location of Power Quality Disturbances
王曦;毕贵红;祖哲;
摘要(Abstract):
提出将递归图和定量递归分析技术引入非平稳电能扰动信号处理研究中,通过分析递归图和递归特性,选择递归量化参数定量地表征电能扰动信号的递归特性,利用定量递归分析提取的不同参数,并运用BP神经网络进行扰动训练,最终实现区分不同的电能扰动模式。进一步选择信号递归度这一特征参数,初步检测了电能扰动信号的端点。仿真实验结果表明,该方法对电压暂降、暂升、中断、振荡暂态、电压尖峰和暂态谐波等多种暂态扰动信号有较好的识别与定位,识别率在98%以上。
关键词(KeyWords): 递归图;电能质量;定量递归分析;神经网络
基金项目(Foundation): 云南省自然科学基金资助项目(2009CD028)
作者(Author): 王曦;毕贵红;祖哲;
Email:
DOI: 10.13335/j.1000-3673.pst.2012.08.035
参考文献(References):
- [1]王丽霞,何正友,赵静.基于数学形态学的电能质量扰动检测和定位[J].电网技术,2008,32(10):63-68.Wang Lixia,He Zhengyou,Zhao Jing.Detection and location ofpower quality disturbance based on mathematical morphology[J].Power System Technology,2008,32(10):63-68(in Chinese).
- [2]方群会,刘强,周林,等.模式分类方法在电能质量扰动信号分类中的应用综述[J].电网技术,2009,33(1):1-36.Fang Qunhui,Liu Qiang,Zhou Lin,et al.Survey on application ofpattern classification in power quality disturbance signalsclassification[J].Power System Technology,2009,33(1):1-36(inChinese).
- [3]易吉良,彭建春,谭会生.基于广义S变换的短时电能质量扰动信号分类[J].电网技术,2009,33(5):22-27.Yi Jiliang,Peng Jianchun.Classification of short-time power qualitydisturbance signals based on generalized S-transform[J].PowerSystem Technology,2009,33(5):22-27(in Chinese).
- [4]易吉良,彭建春,谭会生.基于改进S变换估计电能质量信号的噪声水平[J].电网技术,2009,33(20):71-75.Yi Jiliang,Peng Jianchun,Tan Huisheng.Noise estimation of powerquality signal based on modified S-transform[J].Power SystemTechnology,2009,33(20):71-75(in Chinese).
- [5]赵静,何正友,贾勇,等.基于高阶累积量的暂态电能质量扰动分类研究[J].电网技术,2011,35(5):103-109.Zhao Jing,He ZhengYou,Qian Qingquan,et al.Classification oftransient power quality disturbances based on high-order cumulants[J].Power System Technology,2011,35(5):103-109(in Chinese).
- [6]耿云玲,何怡刚,王群,等.采用小波变换检测和定位电能质量扰动[J].继电器,2005,33(8):38-40.Geng Yunling,He Yigang,Wang Qun,et al.Detection and locationof power quality disturbances by wavelet transform[J].Relay,2005,33(8):38-40(in Chinese).
- [7]李天云,赵妍,李楠.基于EMD的Hilbert变换应用于暂态信号分析[J].电力系统自动化,2005,29(04):49-52.Li Tianyun,Zhao Yan,Li Nan,et al.Apply empirical modedecomposition based Hilbert Transform to power system transientsignal analysis[J].Automation Of Electric Power Systems,2005,29(4):49-52(in Chinese).
- [8]黄奂,吴杰康.基于独立分量分析的电力系统瞬时电压畸变信号判别方法[J].电网技术,2009,33(6):5-12.Huang Huan,Wu JieKang.A method to recognize power systeminstantaneous voltage distortion signal based on independentcomponent analysis[J].Power System Technology,2009,33(6):5-12(in Chinese).
- [9]Faure P,Kom H.A new method to estimate the koimogorov entropyfrom recurrence plots:its application to neuronal signals[J].PhysicaD-Nonlinear Phenomena,1998,122:265-279.
- [10]Riley M A,Balasubramaniam R,Turvey M T.Recurrencequantification analysis of postural fluctuations[J].Gait&Posture,1999,9:65-78.
- [11]Marwan N,Wessel N,Meyerfeldt U,et al.Recurrence-plot-basedmeasures of complexity and their application to heart-rate-variabilitydata[J].Physical Review,2002,6602:6702-6720.
- [12]Casdagli M C.Recurrence plots revisited[J].Physica NonlinearPhenomena,1997,108:12-44.
- [13]吕干云,冯华君,汪晓东,等.基于S变换和人工神经网络的电能质量扰动识别[J].仪器仪表学报,2006,2:2281-2283LüGanyun,Feng Huajun,Wang Xiaodong,et al.PQ disturbancesidentification based on S-transform and ANN[J].Chinese Journal ofScientific Instrument,2006,2:2281-2283
- [14]Liu Yong.Recurrent plot analysis of leakage current for evaluatingand monitoring outdoor insulator performance[D].TianJin:TianJinUniversity,2009.
- [15]Webber C L,Marwan N.Simpler methods do it better:success ofrecurrence quantification analysis as a general purpose data analysistool[J].Physics Letters,2009,373:753-3756.Received date:2011-11-07.