| 引用本文: | 王晓卫,魏春燕,王毅钊,等.基于护层接地线电流和分段TKEO的电缆早期故障检测方法[J].电力系统保护与控制,2025,53(13):47-58.[点击复制] |
| WANG Xiaowei,WEI Chunyan,WANG Yizhao,et al.Early fault detection method for cables based on sheath grounding current and segmented TKEO[J].Power System Protection and Control,2025,53(13):47-58[点击复制] |
|
| 摘要: |
| 电缆早期故障放电持续时间短，故障特征微弱，难以触发常规的过电流保护，故障查找困难。若不及时检测和处理，故障会逐步恶化，并在一定时间内发展成永久性故障。针对中压配电网极易发生电缆早期故障且难以灵敏可靠地检测故障的问题，提出了一种基于护层接地线电流和分段Teager-Kaiser能量算子(Teager-Kaiser energy operator, TKEO)的电缆早期故障检测方法。首先，采用最优变分模态分解(optimal variational mode decomposition, OVMD)对护层接地线电流信号进行分解，利用中心频率法确定最优分解数，获得本征模态函数(intrinsic mode function, IMF)。然后，计算各模态的脉冲因子和余弦相似度，遴选出最具故障特征的模态分量。其次，结合分段TKEO的导数分析信号的能量变化特征。最后，计算每段TKEO的导数最大值与前一段TKEO导数最大值的比值，设置合理阈值进行故障检测。仿真结果表明，该方法能准确检测电缆状态，实现早期故障检测，保证了检测的灵敏性与可靠性。 |
| 关键词: 早期故障 分段Teager-Kaiser能量算子 脉冲因子 余弦相似度 |
| DOI：10.19783/j.cnki.pspc.241222 |
| 投稿时间：2024-09-09修订日期：2024-11-27 |
| 基金项目:国家自然科学基金项目资助(52177114，61403127)；国网陕西省电力有限公司科技项目资助(5226KY23001U) |
|
| Early fault detection method for cables based on sheath grounding current and segmented TKEO |
| WANG Xiaowei1,WEI Chunyan1,WANG Yizhao2,ZHANG Zhihua2 |
| (1. School of Electrical Engineering, Xi’an University of Technology, Xi’an 710054, China;
2. Institute of Electric Power Research of State Grid Shaanxi Electric Power Co., Ltd., Xi’an 710199, China) |
| Abstract: |
| Early faults in cables are characterized by short-duration discharges and weak fault signatures, making them difficult to trigger conventional overcurrent protection and complicating fault detection. If not detected and addressed in time, such faults can gradually deteriorate into permanent failures. To address the challenge of sensitively and reliably detecting early-stage cable faults that are prone to occurring in medium-voltage distribution networks, this paper proposes a cable early fault detection method based on the sheath grounding wire current and the segmented Teager-Kaiser energy operator (TKEO). First, optimal variational mode decomposition (OVMD) is applied to decompose the sheath grounding wire current signal. The optimal decomposition number is determined using the central frequency method, yielding a set of intrinsic mode function (IMF). Then, the pulse factor and cosine similarity of each mode are calculated, and the mode component with the most pronounced fault characteristics is selected. Next, the energy variation characteristics of the signal are analyzed using the derivative of the segmented TKEO. Finally, the ratio of the maximum derivative value of each TKEO segment to that of the previous segment is calculated, and a suitable threshold is set for fault detection. Simulation results demonstrate that this method can accurately detect cable conditions, enabling early fault detection and ensuring both sensitivity and reliability in detection. |
| Key words: early fault segmented Teager-Kaiser energy operator pulse factor cosine similarity |
