摘要
电缆缓冲层烧蚀故障会引发绝缘介质的渐进性劣化,未及时干预的劣化进程可能导致电缆被击穿,最终威胁电网的安全稳定。针对传统的高压电缆缓冲层烧蚀故障的检测技术存在的不足,融合频域阻抗谱和机器学习,构建一种新型故障诊断模型。通过对电力电缆的分布参数模型进行仿真,得到正常、缓冲层烧蚀、局部老化和感性故障的首端频域阻抗谱;通过分析频域阻抗谱,可以识别有缓冲层烧蚀、局部老化的容性故障和感性故障,但无法准确识别电缆局部老化和缓冲层烧蚀;在频域阻抗谱的基础上,通过多层感知器对故障类型分类,实现了对缓冲层烧蚀故障的有效识别。
The ablation failure of cable buffer layers can induce progressive deterioration of insulation media.Untreated deterioration processes may lead to insulation failure events,ultimately threatening the safety and stability of power grids.Traditional detection technologies for high-voltage cable buffer layer ablation failures exhibit deficiencies.To this end,by integrating frequency-domain impedance spectroscopy with machine learning,a novel fault diagnosis model is constructed.Through simulations of the distributed parameter model of power cables,the frequency-domain impedance spectra at the cable head under normal conditions,buffer layer ablation,localized aging,and inductive faults are obtained.Analysis of these spectra can identify capacitive and inductive faults involving buffer layer ablation and localized aging,but fails to accurately distinguish between localized cable aging and buffer layer ablation.On the basis of frequency-domain impedance spectroscopy,a multilayer perceptron is employed to classify fault types,achieving effective identification of buffer layer ablation failures.
作者
汪洪亮
刘昕
徐浩焜
马明超
WANG Hongliang;LIU Xin;XU Haokun;MA Mingchao(School of Automation and Information Engineering,Xi’an University of Technology,Xi’an 710048,Shaanxi,China;School of Electrical Engineering,Xi’an University of Technology,Xi’an 710054,Shaanxi,China)
出处
《电网与清洁能源》
北大核心
2025年第6期99-105,共7页
Power System and Clean Energy
基金
国家自然科学基金(52077176)
国网陕西省电力有限公司科技项目(SGTYHT/23-JS-004)。
关键词
频域阻抗谱
多层感知器
缓冲层烧蚀
故障识别
frequency domain impedance spectrum
multilayer perceptron
ablation of the buffer layer
fault identification
