陈皇熹(1997),男,硕士在读,研究方向为电力设备故障检测(E-mail:
方春华(1980),男,博士,副教授,研究方向为电力设备故障检测
普子恒(1987),男,博士,副教授,研究方向为高压电气设备故障检测
快速准确定位电缆局放故障位置对电缆系统安全稳定运行具有重要意义。针对传统行波定位法在长电缆局放定位时反射信号难以识别以及存在时间同步误差的问题,提出一种基于变分模态分解-维格纳威尔分布(VMD-WVD)相位法的电缆局放双端定位方法,通过准确标记双端局放信号波头时刻实现双端局放信号的同步,利用相位定位实现长电缆局放入射信号的双端同步。文中建立了长线路PSCAD模型,分析故障位置、电缆长度和采样率对VMD-WVD相位法在长电缆定位的精度影响。结果表明,在上述3种因素影响下,VMD-WVD相位法的平均定位精度分别为0.54%,0.85%,0.69%,高于传统行波定位法。文中研究成果为长电缆局部放电精准定位提供一种全新思路。
Obtaining the cable partial discharge fault location quickly and accurately is of great significance to the safe and stable operation of the cable system. In order to resolve the problem of the traditional traveling wave positioning method on time synchronization, a double-ended positioning of partial discharge for cable based on variational mode decomposition-Wigner Ville distribution (VMD-WVD) phase method is proposed to reduce the influence of reflected waves identification in long-cable. It is marked the wave head of the double-ended partial discharge signal accurately to achieve the synchronization of the double-ended partial discharge signal by the proposed method. Through phase positioning, the advantages that synchronizing the incident signal of double-ended partial discharge in the long cable are realized. It is built a PSCAD model of a long line to analyze the positioning accuracy of the VMD-WVD phase method under the influence of fault location, cable length and sampling rate in long cables. The results show that the average positioning accuracy by the VMD-WVD phase method for the above three influencing factors is 0.54%, 0.85% and 0.69% respectively, which is higher than the value by the traditional traveling wave positioning method. The research results provide a new idea for the precise positioning of partial discharge in long cables.
局部放电(partial discharge, PD)监测是一种评估电缆绝缘状态的重要方法。该方法通过分析检测PD信号,对PD源位置进行快速准确定位,从而保障电力系统的稳定运行[
目前国内外针对电力电缆PD定位的研究较多,电力电缆故障检测手段以行波法为主。时域反射法(time domain reflectometry, TDR)是应用最为广泛的一种单端行波法,被应用于地质勘查、线路测距及PD定位等[
为此,文中提出一种基于变分模态分解(variational mode decomposition, VMD)-维格纳威尔分布(Wigner-Ville distribution, WVD)相位法的电缆局放双端定位。VMD是一种非递归分解模型,能反映信号奇异性特征,具有稳定性好、运算效率高和良好鲁棒性等优点[
VMD可将局放信号分解为一系列有限带宽固有模态函数(intrinsic mode function, IMF),分解后的IMF包含大量故障信息[
式中:
引入Lagrange乘数法和惩罚因子
利用式(4)对局放信号模态函数和中心频率进行更新。
式中:
式中:
由式(1)-式(5)可知,VMD使用迭代搜索变分模型最优解,确定局放信号模态函数及其中心频率,最终局放信号自适应分解为
WVD定义为信号瞬态自相关函数的傅里叶变换[
式中:
由式(7)、式(8)可知,局放信号在
使用VMD算法对局放信号分解,需预设VMD的分解层数
(1) 初始化分解层数
(2) 对局放信号进行
(3) 从
(4) 定义能量差绝对变化率
(5) 使用WVD求出该IMF的WVD瞬时能量分布,WVD瞬时能量分布幅值时刻即为局放信号波头时刻。
数字采样后的局放信号为离散信号,设局放信号为
式中:
重复式(10)可得各个点的相位谱。考虑局放信号传播过程的相位衰减特性,在电缆首、末端局放信号相位如式(11)和式(12)所示。
式中:
首、末端相位常数可利用式(13)分别计算得到。
式中:
式中:
综上所述,VMD-WVD相位法的具体步骤为:
(1) 在电缆首、末两端安装传感器,采集信号并对信号进行去噪提取并分割出首、末端局放信号。
(2) 根据式(9)将局放信号变换成频域信号。
(3) 将频域信号代入式(10)依次计算出首、末端局放脉冲相位谱。
(4) 对局放信号进行VMD-WVD处理得到信号波头时刻。
(5) 在相位谱中提取波头时刻和幅值时刻相位,代入式(14)以计算故障位置,局放故障位置可从
为验证VMD-WVD在实际工程中波头相位提取的精度,文中依托某供电公司电缆带电检测服务现场采集的局放信号进行验证。局放带电检测现场电缆型号YJV22-64/110 kV,电缆总长2 km。使用高频电流传感器环接在电缆屏蔽层接地线采集局放信号,其检测带宽能达30 MHz,采集系统采样率设置为200 MHz。局放信号经高频电流传感器采集后经50 Ω同轴电缆传输至采集系统,最后传输至电脑端进行处理。现场检测采集的2组信号如
现场采集的原始局放信号
Original partial discharge signals collected on site
由
局放信号波形
Waveforms of partial discharge signals
由于配网结构复杂,且受现场环境噪声等影响较大,局放信号波头检测较为困难[
能量差绝对变化率
Absolute change rate of energy difference
绝对变化率 |
绝对变化率 |
|
3 | 0.069 2 | 0.117 0 |
4 | 0.086 5 | 0.217 0 |
5 | 0.143 3 | 0.358 6 |
6 | 0.359 5 | 0.710 9 |
7 | 0.832 0 | 0.971 5 |
8 | 6.455 2 | 1.718 0 |
9 | 0.518 2 | 0.675 9 |
10 | 0.299 3 | 0.352 5 |
由
局放信号的VMD结果
VMD results of the partial discharge signals
局放信号的WVD瞬时能量分布
WVD instantaneous energy distribution of the partial discharge signals
局放故障定位受故障位置、电缆长度及采样率等因素影响。如
仿真模型
Simulation model
文献[
式中:
文中采用交联聚乙烯电缆作为PSCAD电缆模型,电缆模型的结构示意如
电缆模型
Cable model
首、末端局放信号和相位谱及WVD瞬时能量
Partial discharge signals of the first-end terminal-end, phase spectrums and instantaneous energy of WVD
为验证波速,以100 m为步长,设置电缆长度
不同局放位置的定位误差
Positioning error of different fault distances
保持电缆长度不变,局放位置的改变对3种定位方法误差影响较小,VMD-WVD相位法误差分别为0.62%,0.53%,0.48%,0.47%,0.58%,平均误差为0.54%,定位精度优于单端行波法和双端行波法。改变电缆长度,保持
不同电缆长度的定位误差
Positioning error of different cable lengths
随着电缆长度增加,3种方法的定位误差呈增大趋势。VMD-WVD相位法平均定位误差为0.85%,长距离传播中波形色散影响导致相位误差,但平均误差小于1%,能够满足实际需求。
为了研究采样率对定位精度影响,改变采样率,设置
不同采样率的定位误差
Positioning error of different sampling rates
VMD-WVD相位法平均定位误差0.69%,在1 MHz采样率时定位误差达1.08%,但在100 MHz及以上时,误差保持0.5%左右。采样率降低,采样间隔增大,造成两端信号波头实际时刻与采样时刻的误差。故使用VMD-WVD相位法时,须采用高采样率。
文中阐述了VMD-WVD的基本原理,在此基础上针对传统行波定位法在长电缆局放定位中时间同步和反射信号难识别的问题,提出一种基于VMD-WVD相位法的电缆局放故障双端定位,最终得出以下结论:
(1) 文中提出的VMD-WVD相位法不仅保留传统双端定位法只须采集入射局放信号的优点,且用相位代替波速,不必对波速进行验证。
(2) 经验证VMD-WVD相位法能准确提取局放入射信号波头时刻与相位,实现长电缆入射局放信号双端同步的优点,能够有效减小长电缆定位中时间难同步和反射局放信号难识别带来的影响。
(3) 文中所提的VMD-WVD相位法在故障位置、电缆长度、采样率3种影响因素下平均定位精度分别为0.54%,0.85%,0.69%,定位精度高于传统定位方法,采用100 MHz及以上采样率能获得更高的定位精度。
徐学伟, 李静鹏, 马驰, 等. 基于UHF信号特征参数指纹图谱的局部放电定位方法[J]. 电力系统保护与控制, 2019, 47(24): 84-91.
XU Xuewei, LI Jingpeng, MA Chi, et al. A partial discharge localization method based on fingerprint spectrum of UHF signal feature parameter[J]. Power System Protection and Control, 2019, 47(24): 84-91.
田妍, 王勇, 杨森, 等. 基于线性预测分析时差估计的特高频局部放电定位方法[J]. 高压电器, 2019, 55(12): 181-186.
TIAN Yan, WANG Yong, YANG Sen, et al. UHF partial discharge location method based on time delay estimation of linear prediction analysis[J]. High Voltage Apparatus, 2019, 55(12): 181-186.
刘哲, 苏菲, 王兴振, 等. 基于分布式无线TEV传感器的电缆终端局部放电监测系统[J]. 电测与仪表, 2019, 56(17): 102-108.
LIU Zhe, SU Fei, WANG Xingzhen, et al. Partial discharge monitoring system based on cable terminal with distributed wireless TEV sensors[J]. Electrical Measurement & Instrumentation, 2019, 56(17): 102-108.
WADI A L, AL-MASRI W, SIYAM W, et al. Accurate estimation of partial discharge location using maximum likelihood[J]. IEEE Sensors Letters, 2018, 2(4): 1-4.
SHAFIQ M, KⅡTAM I, TAKLAJA P, et al. Identification and location of PD defects in medium voltage underground power cables using high frequency current transformer[J]. IEEE Access, 2019, 7: 103608-103618.
高树国, 刘贺晨, 范辉, 等. 考虑波速特性的小波变换模极大值法的电力电缆局部放电定位研究[J]. 电网技术, 2016, 40(7): 2244-2250.
GAO Shuguo, LIU Hechen, FAN Hui, et al. PD location method of power cable based on wavelet transform modulus maxima considering wave characteristics[J]. Power System Technology, 2016, 40(7): 2244-2250.
YII C C, ROHANI M N K H, ISA M, et al. Multi-end PD location algorithm using segmented correlation and trimmed mean data filtering techniques for MV underground cable[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2017, 24(1): 92-98.
周凯, 汪先进, 赵世林, 等. 非接触式中继装置用于长电缆局部放电定位研究[J]. 电网技术, 2020, 44(1): 369-376.
ZHOU Kai, WANG Xianjin, ZHAO Shilin, et al. Research on partial discharge positioning of long cable based on non-contact relay device[J]. Power System Technology, 2020, 44(1): 369-376.
张若兵, 陈子豪, 杜钢. 基于行波法的振荡波电压下电缆局部放电自动定位算法[J]. 高电压技术, 2019, 45(4): 1289-1296.
ZHANG Ruobing, CHEN Zihao, DU Gang. Partial discharge automatic locating algorithm for cable under oscillatory wave voltage conditions based on traveling wave method[J]. High Voltage Engineering, 2019, 45(4): 1289-1296.
张若兵, 金森, 杜钢. 基于EMD与DTW算法的振荡波下电缆局部放电定位方法[J]. 高电压技术, 2020, 46(1): 273- 281.
ZHANG Ruobing, JIN Sen, DU Gang. Method of partial discharge localization for cable under oscillatory wave voltage conditions based on EMD and DTW algorithm[J]. High Voltage Engineering, 2020, 46(1): 273-281.
LAN S, HU Y Q, KUO C C. Partial discharge location of power cables based on an improved phase difference method[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2019, 26(5): 1612-1619.
张怿宁. 架空-电缆混合输电线路故障定位方法综述[J]. 电力工程技术, 2020, 39(6): 44-51.
ZHANG Yining. A review of fault location methods for hybrid overhead and cable transmission lines[J]. Electric Power Engineering Technology, 2020, 39(6): 44-51.
柴鹏, 周灏, 张煜, 等. 基于双端行波法的电缆线路短路故障定位改进[J]. 中国电力, 2020, 53(11): 168-174.
CHAI Peng, ZHOU Hao, ZHANG Yu, et al. Short-circuit fault location for power cables based on improved two-terminal traveling wave method[J]. Electric Power, 2020, 53(11): 168- 174.
高艳丰, 朱永利, 闫红艳, 等. 一种新型的输电线路双端行波故障定位方法[J]. 电力系统保护与控制, 2016, 44(8): 8-13.
GAO Yanfeng, ZHU Yongli, YAN Hongyan, et al. A new fault location method of transmission lines based on double-terminal traveling wave[J]. Power System Protection and Control, 2016, 44(8): 8-13.
DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.
高佳程, 朱永利, 郑艳艳, 等. 基于VMD-WVD分布与堆栈稀疏自编码网络的局放类型识别[J]. 中国电机工程学报, 2019, 39(14): 4118-4129.
GAO Jiacheng, ZHU Yongli, ZHENG Yanyan, et al. Pattern recognition of partial discharge based on VMD-WVD and SSAE[J]. Proceedings of the CSEE, 2019, 39(14): 4118-4129.
WANGX J, XUE Y J, ZHOU W, et al. Spectral decomposition of seismic data with variational mode decomposition-based Wigner-Ville distribution[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(11): 4672-4683.
胡开庚. 基于Wigner Ville分布的故障行波信号分析方法研究[D]. 长沙: 长沙理工大学, 2019.
HU Kaigeng. Research on fault traveling wave signal analysis method based on Wigner Ville distribution[D]. Changsha: Changsha University of Science & Technology, 2019.
王桥梅, 吴浩, 胡潇涛, 等. 基于VMD多尺度模糊熵的HVDC输电线路故障识别方法[J]. 电力系统及其自动化学报, 2021, 33(5): 134-144.
WANG Qiaomei, WU Hao, HU Xiaotao, et al. Fault recognition method for HVDC transmission line based on VMD multi-scale fuzzy entropy[J]. Proceedings of the CSU-EPSA, 2021, 33(5): 134-144.
刘洋, 曹云东, 侯春光. 基于经验模态分解及维格纳威尔分布的电缆双端故障定位算法[J]. 中国电机工程学报, 2015, 35(16): 4086-4093.
LIU Yang, CAO Yundong, HOU Chunguang. The cable two-terminal fault location algorithm based on EMD and WVD[J]. Proceedings of the CSEE, 2015, 35(16): 4086-4093.
卿宗胜, 高云鹏, 吴聪, 等. 基于自适应VMD和改进功率谱估计的球磨机负荷特征提取[J]. 仪器仪表学报, 2020, 41(5): 234-241.
QING Zongsheng, GAO Yunpeng, WU Cong, et al. Feature ex- traction method of ball mill load based on the adaptive variational mode decomposition and the improved power spectrum analysis[J]. Chinese Journal of Scientific Instrument, 2020, 41(5): 234-241.
余浩, 庄池杰, 曾嵘, 等. 提高暂态电压测量精度的自适应小波去噪方法[J]. 高电压技术, 2020, 46(6): 2139-2147.
YU Hao, ZHUANG Chijie, ZENG Rong, et al. Adaptive wavelet de-noising method for improving the accuracy of transient overvoltage measurement[J]. High Voltage Engineering, 2020, 46(6): 2139-2147.
李波. 基于变分模态分解及魏格纳威尔分布的电缆故障定位方法研究[D]. 广州: 华南理工大学, 2017.
LI Bo. Research of cable fault location based on variational mode decomposition and Wigner-Ville distribution[D]. Guangzhou: South China University of Technology, 2017.
谢敏, 周凯, 何珉, 等. 基于时间反演技术的电力电缆局部放电定位方法[J]. 中国电机工程学报, 2018, 38(11): 3402-3409.
XIE Min, ZHOU Kai, HE Min, et al. Partial discharge location for power cable based on the time reversal technique[J]. Proceedings of the CSEE, 2018, 38(11): 3402-3409.