高压断路器分合闸振动突变点自动化辨识技术  

Automatic Identification Technology for Sudden Vibration Points During the Opening and Closing of High-voltage Circuit Breakers

作　　者：李敬光 刘宏 罗松林 刘树安 LI Jingguang;LIU Hong;LUO Songlin;LIU Shu’an(Dongguan Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Dongguan 523008,China)

机构地区：[1]广东电网有限责任公司东莞供电局,东莞523008

出　　处：《自动化与仪表》2024年第12期110-113,118,共5页Automation & Instrumentation

基　　金：广东电网有限责任公司东莞供电局(031900KC22120001(GDKJXM20222630))。

摘　　要：为了及时发现潜在安全隐患,提出高压断路器分合闸振动突变点自动化辨识方法。将采集到的高压断路器分合闸振动信号输入到支持向量机中,以获得正常振动信号和疑似振动突变信号,提取出疑似振动突变信号;利用CEEMDAN分解算法获得各疑似振动突变信号的分量,并根据信号分量的方差贡献率保留重要分量;对重要分量的谱峭度展开计算,并设置谱峭度阈值,将分量谱峭度大于阈值的信号判定为振动突变信号,以此完成振动突变点的自动化辨识。实验结果表明,该方法的信号分量提取性能较高、突变点辨识效果较好。In order to timely detect potential safety hazards,an automated identification method for the sudden change point of high-voltage circuit breaker opening and closing vibration is proposed.Input the collected high-voltage circuit breaker opening and closing vibration signals into the support vector machine to obtain normal vibration signals and suspected vibration mutation signals,and extract suspected vibration mutation signals.Using the CEEMDAN decomposition algorithm to obtain the components of each suspected vibration mutation signal,and retaining important components based on the variance contribution rate of the signal components.Calculate the spectral kurtosis of important components and set a spectral kurtosis threshold.Signals with component spectral kurtosis greater than the threshold are identified as vibration mutation signals,thus completing the automated identification of vibration mutation points.The experimental results show that the method has high performance in signal component extraction and good identification effect for mutation points.

关 键 词：高压断路器分合闸 振动突变点辨识 支持向量机 CEEMDAN分解 谱峭度 

分 类 号：TM561[电气工程—电器] TP274[自动化与计算机技术—检测技术与自动化装置]