配电网暂态电能质量扰动的定位与识别研究  

Localization and Identification Research of Transient Power Quality Disturbance in Distribution Network

作　　者：何巨龙[1] 王根平[1,2] 刘丹[1] 唐友明[1] HE Ju-long WANG Gen-ping LIU Dan TANG You-ming(Key Laboratory of Intelligent Computing ＆ Information Processing of Ministry of Education, Xiangtan University, Xiangtan 411105, China Shenzhen Polytechnic, Shenzhen 518055, China)

机构地区：[1]湘潭大学智能计算与信息处理教育部重点实验室,湖南湘潭411105 [2]深圳职业技术学院,广东深圳518055

出　　处：《电气开关》2017年第4期15-18,共4页Electric Switchgear

基　　金：研发资金(JCYJ20140508155916430)资助

摘　　要：针对配电网暂态电能质量扰动的非平稳性、突变性和短时持续性的特点,提出一种基于小波变换和BP神经网络的扰动定位与识别方法。首先对暂态电能质量扰动信号进行小波分解,提取其高频系数得到扰动时刻信息,然后利用模极大值对扰动突变点峰值进行定位检测,再用扰动信号的能量分布序列来构造特征向量并用BP神经网络设计扰动识别器。MATLAB仿真结果表明,该方法对暂态电能质量扰动信号起止时刻定位快速,精度较高,能有效地识别五种电能质量扰动,识别率高。According to the non - stationary, short duration and fast change rate of transient power quality disturbance in distribution network, a method to localize and identify disturbance is p based on lifting wavelet and BP neural net- work. At first, high frequency coefficient are extracted, and fault time information is got through wavelet decomposing, then mutation peak of transient disturbances localized using wavelet modulus maxim. At last, feature vector is constructed through extracting energy distribution sequence of transient disturbance signal, and using BP neural network designs dis- turbance recognition device. MATLAB simulation results show that the proposed method can quickly localize the disturb- ances＇ start -stop time with relatively high accuracy, and can effectively identify five common power quality disturbances with high discrimination ratio.

关 键 词：暂态扰动 小波变换 BP神经网络 定位 识别 

分 类 号：TB172[生物学—生物工程]