基于FAVMD-时变峰度的电缆局部放电多传感器在线检测与定位
|
摘要
在利用多传感器测量法进行电缆局放故障在线定位时,由于强环境噪声的干扰,局放初至脉冲的拾取误差较大或无法拾取,造成精确定位失败。针对这一实际工程问题,提出一种基于FAVMD-时变峰度的电缆局放初至脉冲检测方法。首先利用相关系数法自适应确定最优基本模态个数,将其应用于变分模态分解,将局放信号和噪声信号快速分解到不同的基本模态中,实现噪声信号的有效分离。然后运用时窗能量比方法确定局放时窗,并采用一种新的时变峰度法在局放时窗内精确拾取局放初至脉冲。最后运用多传感器测量法实现电缆局放源的在线定位。实验结果表明,相对于现有方法,该算法适应噪声能力强,定位精度高,在信噪比为-16 dB的强噪声环境下,相对定位误差仅为0.23%,在实际工程中有良好的应用前景。
The multi-sensor measurement method is used for online positioning of the cable partial discharge faults,due to the interference of strong environmental noise,the pick-up error of the local discharge first arrival pulse is large or cannot be picked up,resulting in accurate positioning failure. Aiming at the practical engineering problem,a method for initial detection of cable partial discharge based on FAVMD-time varying kurtosis is proposed in this paper. Firstly,the correlation coefficient method is used to determine the optimal basic mode number adaptively,and that is applied to the variational mode decomposition. Then,the PD signal and the noise signal are quickly decomposed into different basic modes to achieve effective separation of noise signals. Afterwards,the time window energy ratio method is used to determine the PD time window,and a new time varying kurtosis method is used to accurately pick up the PD pulse in the PD window. Finally,the multi-sensor measurement method is used to realize the online positioning of the cable local discharge source. The experimental results show that compared with the existing methods,the noise adaptability of the algorithm proposed is stronger. Furthermore,the positioning accuracy of the algorithm is higher. Under the strong noise environment with signal-to-noise ratio of-16 dB,the relative positioning error is only 0.23%,and shows that the algorithm proposed in this paper is more suitable for engineering applications.
The multi-sensor measurement method is used for online positioning of the cable partial discharge faults,due to the interference of strong environmental noise,the pick-up error of the local discharge first arrival pulse is large or cannot be picked up,resulting in accurate positioning failure. Aiming at the practical engineering problem,a method for initial detection of cable partial discharge based on FAVMD-time varying kurtosis is proposed in this paper. Firstly,the correlation coefficient method is used to determine the optimal basic mode number adaptively,and that is applied to the variational mode decomposition. Then,the PD signal and the noise signal are quickly decomposed into different basic modes to achieve effective separation of noise signals. Afterwards,the time window energy ratio method is used to determine the PD time window,and a new time varying kurtosis method is used to accurately pick up the PD pulse in the PD window. Finally,the multi-sensor measurement method is used to realize the online positioning of the cable local discharge source. The experimental results show that compared with the existing methods,the noise adaptability of the algorithm proposed is stronger. Furthermore,the positioning accuracy of the algorithm is higher. Under the strong noise environment with signal-to-noise ratio of-16 dB,the relative positioning error is only 0.23%,and shows that the algorithm proposed in this paper is more suitable for engineering applications.
引文
[1] 刘洋,曹云东,候春光. 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.
[2] 张姝,林圣,唐进等.基于双层阻抗模型的三相单芯电缆自恢复故障定位[J]. 电工技术学报,2016,31(17):1-10.
[3] 张磊祺,盛博杰,姜伟,等. 交叉互联高压电缆系统的局部放电在线监测与定位[J]. 高电压技术,2015,41(8):2702-2715.
[4] 张磊. 基于XLPE电缆局部放电检测的传感器研究[D]. 天津:天津大学,2009.
[5] 刘晓鸥. 用于局部放电在线监测的高频电流传感器研究[D]. 上海:上海交通大学,2007.
[6] 孙抗,郭景蝶,杨延举,等. 基于AIC准则和时窗能量比的电缆局部放电在线检测与定位[J]. 传感技术学报,2017,30(8):1209-1214.
[7] 张俊民,周小猛,魏娟等. 基于小波变换的导线绝缘故障定位方法[J]. 电工技术学报,2012,27(5):99-104.
[8] 孙抗,郭景蝶,马星河. 高压电缆局部放电小波包-峰度法在线检测与定位[J]. 电子测量与仪器学报,2017,31(7):1099-1106.
[9] 徐俊明,王芳宗,尹星等. 基于Hilbert-Huang变换的行波法高压输电线路故障定位[J]. 电力系统保护与控制,2012,40(2):88-93.
[10] 张玲玲,廖红云,曹亚娟,等. 基于EEMD和模糊C均值聚类算法诊断发动机曲轴轴承故障[J]. 内燃机学报,2011,29(4):332-336.
[11] 康守强,王玉静,杨广学,等. 基于经验模态分解和超球多类支持向量机的滚动轴承故障诊断方法[J]. 中国电机工程学报,2011,31(14):96-102.
[12] Konstantin Dragomiretskiy,Dominique Zosso. Variational Mode Decomposition[J]. IEEE Transactions on Signal Processing,2014,62(3):531-544.
[13] 高艳丰,朱永利,闫红艳,等. 基于VMD和TEO的高压输电线路雷击故障测距研究[J]. 电工技术学报,2016,31(1):24-33.
[14] 武英杰. 基于变分模态分解的风电机组传动系统故障诊断研究[D]. 北京:华北电力大学,2016.
[15] 吕中亮. 基于变分模态分解与优化多核支持向量机的旋转机械早期故障诊断方法研究[D]. 重庆:重庆大学,2016.
[16] 胡永泉,尹成,潘树林,等. 改进的时变斜度峰度法微地震信号识别技术[J]. 石油物探,2012,51(6):625-632.
[17] 秦晅,宋维琪. 基于时窗能量比与互信息量的微地震初至拾取方法[J]. 物探与化探,2016,40(2):374-379.孙抗(1982-),江苏徐州人,副教授,主要研究方向为设备在线监测、智能电器等。
[2] 张姝,林圣,唐进等.基于双层阻抗模型的三相单芯电缆自恢复故障定位[J]. 电工技术学报,2016,31(17):1-10.
[3] 张磊祺,盛博杰,姜伟,等. 交叉互联高压电缆系统的局部放电在线监测与定位[J]. 高电压技术,2015,41(8):2702-2715.
[4] 张磊. 基于XLPE电缆局部放电检测的传感器研究[D]. 天津:天津大学,2009.
[5] 刘晓鸥. 用于局部放电在线监测的高频电流传感器研究[D]. 上海:上海交通大学,2007.
[6] 孙抗,郭景蝶,杨延举,等. 基于AIC准则和时窗能量比的电缆局部放电在线检测与定位[J]. 传感技术学报,2017,30(8):1209-1214.
[7] 张俊民,周小猛,魏娟等. 基于小波变换的导线绝缘故障定位方法[J]. 电工技术学报,2012,27(5):99-104.
[8] 孙抗,郭景蝶,马星河. 高压电缆局部放电小波包-峰度法在线检测与定位[J]. 电子测量与仪器学报,2017,31(7):1099-1106.
[9] 徐俊明,王芳宗,尹星等. 基于Hilbert-Huang变换的行波法高压输电线路故障定位[J]. 电力系统保护与控制,2012,40(2):88-93.
[10] 张玲玲,廖红云,曹亚娟,等. 基于EEMD和模糊C均值聚类算法诊断发动机曲轴轴承故障[J]. 内燃机学报,2011,29(4):332-336.
[11] 康守强,王玉静,杨广学,等. 基于经验模态分解和超球多类支持向量机的滚动轴承故障诊断方法[J]. 中国电机工程学报,2011,31(14):96-102.
[12] Konstantin Dragomiretskiy,Dominique Zosso. Variational Mode Decomposition[J]. IEEE Transactions on Signal Processing,2014,62(3):531-544.
[13] 高艳丰,朱永利,闫红艳,等. 基于VMD和TEO的高压输电线路雷击故障测距研究[J]. 电工技术学报,2016,31(1):24-33.
[14] 武英杰. 基于变分模态分解的风电机组传动系统故障诊断研究[D]. 北京:华北电力大学,2016.
[15] 吕中亮. 基于变分模态分解与优化多核支持向量机的旋转机械早期故障诊断方法研究[D]. 重庆:重庆大学,2016.
[16] 胡永泉,尹成,潘树林,等. 改进的时变斜度峰度法微地震信号识别技术[J]. 石油物探,2012,51(6):625-632.
[17] 秦晅,宋维琪. 基于时窗能量比与互信息量的微地震初至拾取方法[J]. 物探与化探,2016,40(2):374-379.孙抗(1982-),江苏徐州人,副教授,主要研究方向为设备在线监测、智能电器等。