室内变压器有载分接开关吊芯装置故障自动化监测方法  

Automatic Monitoring Method for Fault of Hanging Core Device of Indoor Transformer On-load Tap-changer

作　　者：周利兴 林华斌 吴盈莹 楼翔 郑建吉 ZHOU Li-xing;LIN Hua-bin;WU Ying-ying;LOU Xiang;ZHENG Jian-ji(Ningbo Qixin Electrical Technology Co.,Ltd.,Ningbo 315500 China)

机构地区：[1]宁波启新电气科技有限公司,浙江宁波315500

出　　处：《自动化技术与应用》2025年第3期171-175,共5页Techniques of Automation and Applications

基　　金：国家电网公司科技项目(5804DY170912)。

摘　　要：为了提高室内变压器有载分接开关吊芯装置故障监测的准确性,结合故障时异常振动特性,提出室内变压器有载分接开关吊芯装置故障自动化监测方法。构建机械振动信号采集的空间解析式,提取振动信号的空间特征向量;计算有载分接开关时的过渡电路电压变化,获取吊芯装置振动信号的时序特征;构建装置故障自动监测模型,归一化处理信号的故障阈值分量;基于神经网络迭代训练学习,实现吊芯装置故障自动化监测。实验结果可知:通过振动信号能量分布可以较准确地区分故障特征,该方法的故障监测误差小于10^(-5),监测准确性较高。In order to improve the accuracy of fault monitoring for core suspension device of indoor transformer on load tap changer,an automatic fault monitoring method for the core lifting device of indoor transformer on load tap changer is proposed in combination with the abnormal vibration characteristics during the fault.The spatial analytic expression of mechanical vibration signal acquisition is constructed,and the spatial feature vector of vibration signal is extracted.It calculates the voltage change of the transition circuit when the on load tap changer is operating,and obtains the time sequence characteristics of the vibration signal of the core lifting device.The automatic fault monitor model of the device is constructed,and the fault threshold component of the signal is normalized.Based on the iterative training and learning of neural network,the automatic fault monitoring of core lifting device is realized.The experimental results show that the fault characteristics can be accurately distinguished by the energy distribution of vibration signal,the fault monitoring error of this method is less than 10^(-5),and the monitoring accuracy is high.

关 键 词：室内变压器 有载分接开关 吊芯装置 自动化监测 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置] TN06[自动化与计算机技术—控制科学与工程]