电工技术学报  2020, Vol. 35 Issue (10): 2216-2225    DOI: 10.19595/j.cnki.1000-6753.tces.190429
电力电子 |
基于能量谱熵及小波神经网络的有源中性点钳位三电平逆变器故障诊断
李兵1,2, 崔介兵1,2, 何怡刚1,2, 史露强1, 刘晓晖1
1. 合肥工业大学电气与自动化工程学院 合肥 230009;
2. 可再生能源接入电网技术国家地方联合工程实验室 合肥 230009
Fault Diagnosis of Active Neutral Point Clamped Three-Level Inverter Based on Energy Spectrum Entropy and Wavelet Neural Network
Li Bing1,2, Cui Jiebing1,2, He Yigang1,2, Shi Luqiang1, Liu Xiaohui1
1. College of Electrical Engineering and Automation Hefei University of TechnologyHefei 230009 China;
2. National and Local Joint Engineering Laboratory for Renewable Energy Access to Grid Technology Hefei 230009 China
全文: PDF (38957 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 功率开关器件是逆变器核心组成部分,其故障特征参数的提取及分类方法是故障诊断和预测的重要技术基础,对于提高逆变器可靠性具有重要意义。以有源中性点钳位(ANPC)三电平逆变器的IGBT开路故障诊断为例,首先提出一种基于能量谱熵及小波神经网络的故障诊断方法,采用ANPC三电平逆变器上、中、下桥臂电压作为测量信号,通过小波包能量谱熵提取桥臂电压信号特征,并利用核主成分分析对特征向量进行优化;其次利用自适应矩估计小波神经网络(A-WNN)建立故障字典;最后通过搭建实验平台验证了算法的可行性。实验结果表明,A-WNN具有故障特征辨识速度快、精度高等优点,适用于ANPC三电平逆变器实时故障诊断。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
李兵
崔介兵
何怡刚
史露强
刘晓晖
关键词 三电平能量谱熵小波神经网络故障字典    
Abstract:The power switching device is the core component of the inverter. The method of fault characteristic parameter extraction and classification is the technical basis for fault diagnosis and prediction, which is important for improving the reliability of the inverter. Taking the IGBT open circuit fault diagnosis of active neutral-point clamped (ANPC) three-level inverter as an example, a fault diagnosis method based on energy spectrum entropy and wavelet neural network is proposed. The upper, middle and lower arm voltages of ANPC three-level inverter are used as the measurement signals. The characteristics of bridge arm voltage signals are extracted by wavelet packet energy spectrum entropy, and the eigenvectors are optimized by kernel principal component analysis. Secondly, the adaptive moment wavelet neural network (A-WNN) is used to build a fault dictionary, Finally, the feasibility of the algorithm is verified by setting up an experimental platform. The experimental results show that A-WNN has the advantages of fast fault identification and high precision, which is suitable for real-time fault diagnosis of ANPC three-level inverter.
Key wordsThree-level    entropy of energy spectrum    wavelet neural network    fault dictionary   
收稿日期: 2019-04-16     
PACS: TM464  
通讯作者: 李 兵 男,1973年生,副教授,硕士生导师,主要研究方向为故障诊断、智能电网技术。E-mail: libinghnu@163.com   
作者简介: 崔介兵 男,1994年生,硕士研究生,主要研究方向为功率器件故障诊断。E-mail: cuijiebing_hg@163.com
引用本文:   
李兵, 崔介兵, 何怡刚, 史露强, 刘晓晖. 基于能量谱熵及小波神经网络的有源中性点钳位三电平逆变器故障诊断[J]. 电工技术学报, 2020, 35(10): 2216-2225. Li Bing, Cui Jiebing, He Yigang, Shi Luqiang, Liu Xiaohui. Fault Diagnosis of Active Neutral Point Clamped Three-Level Inverter Based on Energy Spectrum Entropy and Wavelet Neural Network. Transactions of China Electrotechnical Society, 2020, 35(10): 2216-2225.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.190429          https://dgjsxb.ces-transaction.com/CN/Y2020/V35/I10/2216