Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 往年索引

基于原子能量分布的配电网暂时过电压识别方法

来源:电工电气发布时间:2024-11-04 13:04 浏览次数:226

基于原子能量分布的配电网暂时过电压识别方法

廖宇飞
(国网福建省电力有限公司福州供电公司,福建 福州 350009)
 
    摘 要:针对配电网领域的过电压识别问题,提出一种基于原子能量分布的配电网暂时过电压识别方法。利用原子分解算法对母线零序电压进行分解,获得能反映过电压信号内部结构的原子及其参数;依据原子频率对原子进行划分和重构,构造原子能量作为特征向量,并将其输入支持向量机中实现暂时过电压识别。分别利用仿真模型、物理实验平台和现场实测数据对所提方法进行验证,结果表明:该方法所提取的特征量维数低、区分度高,且能表示暂时过电压的本质特征,结合分类器具有较高的识别准确率和较强的适应性。
    关键词: 配电网;暂时过电压;识别方法;原子分解;零序电压;原子能量;支持向量机
    中图分类号:TM711 ;TM863     文献标识码:A     文章编号:1007-3175(2024)10-0019-07
 
Identification Method of Temporary Overvoltage in Distribution Network
Based on Atomic Energy Distribution
 
LIAO Yu-fei
(Fuzhou Power Supply Company of State Grid Fujian Electric Power Co., Ltd, Fuzhou 350009, China)
 
    Abstract: Aiming at the problem of overvoltage identification in the field of current distribution networks, a method for temporary overvoltage identification method based on atomic energy distribution is proposed. Firstly, the atomic decomposition algorithm is used to decompose the zero sequence voltage of the bus to obtain the atoms and their parameters that reflect the internal structure of the overvoltage signal.Then, the atoms are divided and reconstructed according to the atomic frequency, and the atomic energy is constructed as a feature vector,which input into support vector machine to realize temporary overvoltage identification. Finally, the simulation model, physical experiment platform and field measurement data are used to verify the proposed method, and the results show that: the feature quantity extracted by the proposed method has low dimension, high discrimination and can represent the essential characteristics of temporary overvoltage. In addition,the combination classifier has higher recognition accuracy and stronger adaptability.
    Key words: distribution network; temporary overvoltage; identification method; atomic decomposition; zero sequence voltage; atomic energy;support vector machine
 
参考文献
[1] 廖宇飞,杨耿杰,高伟,等. 基于 AD-CNN 算法的配电网内部过电压识别技术[J] . 高电压技术,2019,45(10) :3182-3191.
[2] 杜林,郭良峰,司马文霞,等. 基于遗传算法的电网过电压分层模糊聚类识别[J] . 中国电机工程学报,2010, 30(10) :119-124.
[3] 杜林,李欣,王丽蓉,等. 电力系统暂时过电压多级支持向量机分层识别[J] . 电力系统保护与控制,2012, 40(4) :26-31.
[4] 杨勇,李立浧,杜林,等. 采用时频矩阵奇异值分解和多级支持向量机的雷电及操作过电压识别[J]. 电网技术,2012,36(8) :31-37.
[5] 杨庆,赵洪彬,司马文霞,等. 基于阈值判断和支持向量机的电网实测过电压识别[J] . 高电压技术,2016, 42(10) :3188-3198.
[6] WANG Jing, YANG Qing, SIMA Wenxia, et al.A Smart Online Over-Voltage Monitoring and Identification System[J].Energies,2011,4(4) :599-615.
[7] 黄艳玲,司马文霞,杨庆,等. 电力系统实测过电压信号的特征量提取与验证[J] . 高电压技术,2013,39(1) :60-66.
[8] GUO Moufa, YANG Nienche.Features-clusteringbased earth fault detection using singular-value decomposition and fuzzy c-means in resonant grounding distribution systems[J].International Journal of Electrical Power & Energy Systems,2017, 93 :97-108.
[9] 司马文霞,王荆,杨庆,等.Hilbert-Huang 变换在电力系统过电压识别中的应用[J] . 高电压技术,2010,36(6) :1480-1486.
[10] 金涛,许立彬,高伟,等. 一种基于 LCD-Hilbert 变换和奇异谱熵的配电网暂时过电压类型识别方法[J] .电机与控制学报,2018,22(11) :26-36.
[11] 杨冬锋,陈盛开,刘晓军,等. 基于自适应 VMD 和时-频分段能量熵特征的过电压信号识别[J] . 电网技术,2019,43(12) :4597-4604.
[12] 张英杰,龚庆武,李勋,等. 基于 PSO 的原子分解法在间谐波分析中的应用[J] . 电力系统保护与控制,2013, 41(15) :41-48.
[13] MALLAT S G, ZHANG Z F.Matching pursuits with time-frequency dictionaries[J].IEEE Transactions on Sigal Processing,1993,41(12) :3397-3415.
[14] 董飞飞,刘涤尘,廖清芬,等. 基于阻尼正弦原子分解的次同步振荡模态辨识[J] . 中国电机工程学报,2013,33(19) :119-125.
[15] AZAD H R L.An Application of Opposition Based Colonial Competitive Algorithm to Solve Network Count Location Problem[J].International Journal of Intelligent Systems and Applications,2013,6(1) :29-35.
[16] 黄艳玲,司马文霞,杨庆,等. 基于实测数据的电力系统过电压分类识别[J] . 电力系统自动化,2012,36(4) :1000-1026.
[17] 郭谋发,游林旭,洪翠,等. 基于 LCD-Hilbert 谱奇异值和多级支持向量机的配电网故障识别方法[J]. 高电压技术,2017,43(4) :1239-1247.
[18] 吴广宁,袁海满,宋臻杰,等. 基于粗糙集与多类支持向量机的电力变压器故障诊断[J] . 高电压技术,2017, 43(11) :3668-3674.
[19] 司马文霞, 冉锐, 袁涛, 等. 采用数学形态学的弧光接地过电压识别方法[J] . 高电压技术,2010,36(4) :835-841.
[20] 司马文霞,郭飞,杨庆,等. 变电站铁磁谐振仿真分析及抑制措施研究[J] . 电力自动化设备,2007,27(6) :22-26.