首页|基于自适应阈值分割的动车组辅助供电系统单芯动力电缆故障定位方法研究

基于自适应阈值分割的动车组辅助供电系统单芯动力电缆故障定位方法研究

扫码查看
单芯动力电缆是动车组辅助供电系统能量传输通道之一,其故障后准确定位对检修具有重要意义.当前,电缆故障定位使用较为广泛的方法为时频域反射法,但在对反射信号进行时频分析过程中,产生的交叉项是制约定位准确度的重要原因.为此提出一种基于自适应阈值分割的Wigner-Ville变换方法:在70 m长的电缆样品中设置低阻、高阻、开路和短路四种缺陷故障样本,对时频域法检测到的信号使用Wigner-Ville变换进行时频分析,说明原始Wigner-Ville变换产生交叉项的原因,并据此提出基于自适应阈值分割的Wigner-Ville变换方法,解决交叉项干扰问题;将本文方法与当前交叉项抑制方法进行测试效果对比.结果表明:提出的基于自适应阈值分割的Wigner-Ville变换方法可以有效消除交叉项,且对于单个缺陷故障和多个缺陷的电缆均可适用;对于所设置的单个缺陷故障位置的定位误差均小于0.80%,对于双缺陷故障检测结果的定位误差小于0.40%,验证了本文方法的有效性.
Research on Fault Location for Single-core Power Cable of Auxiliary Power Supply System of EMU Based on Adaptive Threshold Segmentation
Accurate positioning in the case of failure of single core power cables,as one of the channels for energy trans-mission in the auxiliary power supply system of electric multiple units(EMU),is of paramount importance in mainte-nance procedures.Currently,the widely adopted method for cable fault location is the time-frequency domain reflectome-try.However,the cross terms generated by time-frequency analysis limit the accuracy of fault location.Thus,Wigner-Ville distribution based on adaptive threshold segmentation was proposed in this paper.To begin,four types of fault sam-ples,namely low resistance,high resistance,open circuit and short circuit,were embedded into a 70-meter cable sam-ple.Subsequently,Wigner-Ville distribution transformation was used to process these detected signals to explain the rea-son for the cross terms generated,leading to the proposal of the Wigner-Ville distribution based on adaptive threshold segmentation to mitigate cross-term.Finally,a comparative evaluation was performed between the new method and exist-ing cross-term suppression methods.The results show that the proposed method that effectively eliminates cross-terms is applicable to both single-fault and multi-fault cable.The positioning error for single-fault location is below 0.80%and that for dual-fault is less than 0.40%,validating the effectiveness of the method presented.

cable insulationdefect and fault locationtime-frequency reflectometrycross term elimination

王东阳、胡宇鹏、蒋伟辉、马含青、刘博凯、周利军

展开 >

西南交通大学电气工程学院,四川成都 611756

电缆绝缘 缺陷与故障定位 时频反射法 交叉项消除

国家自然科学基金

U1834203

2024

铁道学报
中国铁道学会

铁道学报

CSTPCD北大核心
影响因子:0.9
ISSN:1001-8360
年,卷(期):2024.46(4)
  • 24