摘要
目前,电力系统连锁故障自动识别主要通过检测故障元件实现,由于缺乏对故障特征的聚类分析,识别效果不佳。为此,本文提出基于微弱信号检测的电力系统连锁故障自动识别方法,通过结合随机共振理论,调节多稳态结构参数,捕捉电力系统连锁故障的微弱信号,再提取信号特征,构建故障特征向量,最后对故障特征向量进行聚类分析,实现对故障的自动识别。在实验中,对提出的方法进行了识别精度的检验,实验结果表明,采用提出的方法识别故障特征时,容错率较高,具备较为理想的识别效果。
Currently,the automatic detection of cascading faults in power systems relies on fault component identification,which leads to poor recognition results due to the lack of fault feature clustering analysis.To address this issue,we propose an automatic recognition method for cascading faults in power systems based on weak signal detection.By leveraging stochastic resonance theory,adjust the multistable structural parameters to capture weak signals from power system cascading faults.Then extract signal features and construct fault feature vectors that are subsequently clustered to achieve automatic fault recognition.To evaluate our method,conducted an experiment to test its identification accuracy.The results demonstrate a high tolerance rate and an ideal recognition effect.
作者
蓝天宇
韩伟
孟令民
LAN Tianyu;HAN Wei;MENG Lingmin(NARI Technology Co.,Ltd.,Nanjing,Jiangsu 211106,China)
出处
《自动化应用》
2023年第10期59-61,共3页
Automation Application
关键词
微弱信号
电力系统
特征向量
故障识别
weak signal
power system
Eigenvector
fault identification
作者简介
蓝天宇,男,1993年生,硕士研究生,中级工程师,研究方向为电力系统及其自动化。
