摘要
局部放电是导致电力设备绝缘劣化的主要成因,放电阶段的准确研判对于故障诊断和运维检修具有重要意义。本文提出一种基于局部放电日盲紫外信号相位特征的跟踪监测方法。首先,搭建典型绝缘缺陷的局部放电实验平台,使用光谱响应位于日盲紫外波段的光学传感器采集局部放电光信号;然后,通过分析正负极性脉冲次数随电压等级的增长趋势,将击穿前的局部放电过程划分为不同阶段,并对每个阶段的局部放电相位分布特征进行分析;最后,根据相基统计图谱构建数据集,采用反向传播神经网络和支持向量机对不同类型放电下的发展阶段进行跟踪。结果表明,不同放电阶段的光脉冲相位分布差异明显,反向传播神经网络模型的阶段跟踪效果较好,对悬浮放电的跟踪准确率可达98.8%。
Partial discharge(PD)is the primary cause of insulation degradation in power equipment,and it is significant for fault diagnosis and maintenance to identify the stage of PD accurately.In this paper,a PD trace monitoring method based on solar-blind ultraviolet phase characteristics is proposed.Firstly,the experimental platform for PD of typical insulation defects is constructed,and optical sensor with spectral response located in the solar-blind ultraviolet band is used to collect the PD optical signal.Then,the PD process before breakdown is divided into different stages by analysing the growth trend of the frequency of positive and negative polarity pulses with the voltage class,and the characteristics of the PD phase distribution in each stage are analysed.Finally,the dataset is constructed based on phase resolved partial discharge pattern,and the developmental stages under different discharge types are traced with the back propagation neural network and support vector machine.The results show that the phase distribution of light pulses in different stages varies significantly,and the stage tracing of the backpropagation neural network model is more effective,with an accuracy of up to 98.8%for floating discharges.
作者
杨燚虎
刘胤康
项恩新
段生江
王科
陈文良
杨金培
任明
YANG Yihu;LIU Yinkang;XIANG Enxin;DUAN Shengjiang;WANG Ke;CHEN Wenliang;YANG Jinpei;REN Ming(Dehong Electric Power Supply Company,State Grid Yunnan Electric Power Co.,Ltd.,Dehong 678400,China;State Key Laboratory of Electrical Insulation and Power Equipment,Xi’an Jiaotong University,Xi’an 710049,China;Electric Power Research Institute,State Grid Yunnan Electric Power Co.,Ltd.,Kunming 650217,China)
出处
《电工电能新技术》
北大核心
2025年第6期120-128,共9页
Advanced Technology of Electrical Engineering and Energy
基金
南方电网有限责任公司科技项目(YNKJXM20220132)。
关键词
局部放电
日盲紫外检测
相位分析
跟踪监测
反向传播神经网络
partial discharge
solar-blind ultraviolet detection
phase-based analysis
trace monitoring
back propagation neural network
