IGCT三电平变流器故障识别及状态监测方法  

A Method of IGCT Three-level Converter Fault Identification and Condition Monitoring

作　　者：田凯 李风生 楚子林 俞智斌 李腾 王明玥 TIAN Kai;LI Fengsheng;CHU Zilin;YU Zhibin;LI Teng;WANG Mingyue(Tianjin Research Institute of Electric Science Co.,Ltd.,Tianjin 300180,China;National Engineering Research Center for Electric Drive,Tianjin 300180,China)

机构地区：[1]天津电气科学研究院有限公司,天津300180 [2]电气传动国家工程研究中心,天津300180

出　　处：《电气传动》2023年第4期45-50,共6页Electric Drive

基　　金：天津电气科学研究院有限公司科研基金(YF2022ZL002);国机研究院青年科研基金(TD2021ZK003)。

摘　　要：针对以IGCT为核心的中压三电平变流器短路故障保护以及运行过程中主回路器件参数变化监测的问题,提出一种同时实现IGCT三电平变流器故障识别及主回路运行状态监测方法。通过硬件电路检测其电感后端的电压波形并结合器件开关状态来预判是否发生短路故障,筛选电压波形数据中峰值电压与峰值电压时间作为特征值,推导主回路电气参数与峰值电压、峰值电压时间关系,形成对应数据训练神经网络权重构建数值拟合观测器,将实时采样波形数据输入数值拟合观测器在线辨识变流器主回路参数,实现当器件因虚焊、老化、发热等异常原因导致参数变化时予以报警提示。Aiming at the short circuit fault protection of medium voltage three-level converter with IGCT as the core and the monitoring of changes in main circuit device parameters during operation,a method was proposed to simultaneously realize fault identification of IGCT three-level converter and monitoring of main circuit operation status.The voltage waveform at the back end of the inductance was detected by the hardware circuit and combined with the switching state of the device to predict whether there is a short circuit fault.The peak voltage and peak voltage time in the voltage waveform data were selected as the characteristic values,and the relationship between the electrical parameters of the main circuit and the peak voltage and peak voltage time was derived to form the corresponding data training neural network weight to build a numerical fitting observer,real-time sampled waveform data was fed into the numerical fitting observer to identify the main circuit parameters of the converter online,so as to give an alarm when the parameters change due to abnormal reasons such as faulty soldering,aging,heating,etc.

关 键 词：三电平 集成门极换流晶闸管 故障识别 状态监测 

分 类 号：TM28[一般工业技术—材料科学与工程]