南京大学学报(自然科学版) ›› 2020, Vol. 56 ›› Issue (2): 284–289.doi: 10.13232/j.cnki.jnju.2020.02.015

• • 上一篇    下一篇

基于小波包能量熵和随机森林的级联H桥多电平逆变器故障诊断

陈石,张兴敢()   

  1. 南京大学电子科学与工程学院，南京，210023
  • 收稿日期:2019-09-17 出版日期:2020-03-30 发布日期:2020-04-02
  • 通讯作者: 张兴敢 E-mail:zhxg@nju.edu.cn

Fault diagnosis for cascaded H⁃bridge multilevel inverter based on wavelet packet energy entropy and random forest

Shi Chen,Xinggan Zhang()   

  1. School of Electronic Science and Engineering, Nanjing University, Nanjing, 210023, China
  • Received:2019-09-17 Online:2020-03-30 Published:2020-04-02
  • Contact: Xinggan Zhang E-mail:zhxg@nju.edu.cn

RichHTML

6

PDF (PC)

945

摘要：

为了提升级联H桥多电平逆变器故障诊断的准确性和高效性，提出一种基于小波包能量熵和随机森林的故障诊断方法.首先对级联H桥多电平逆变器的输出电压进行小波包分解，提取小波包能量熵构建故障特征；然后采用主成分分析法对故障特征进行维数约简，以降低诊断模型的训练时间；最后采用经参数调优后的随机森林模型对逆变器故障进行分类诊断.基于Matlab平台，将该诊断策略与传统的基于快速傅里叶变换的SVM(Support Vector Machine)方法以及基于小波变换的BP(Back Propagation)神经网络方法进行对比.仿真结果表明，针对级联H桥多电平逆变器中功率开关晶体管开路故障，基于小波包能量熵和随机森林诊断策略的故障识别率更高，可有效提升故障诊断率至97%左右.

关键词: 故障诊断, 级联H桥多电平逆变器, 小波包能量熵, 随机森林

Abstract:

Due to the difficulty in extracting fault features,the accuracy of open circuit fault diagnosis for power switching transistors in CHMLI (Cascaded H?bridge Multilevel Inverter) is not high. A new approach based on WPEE (Wavelet Packet Energy Entropy) and RF (Random Forest) is proposed in this paper. Firstly,the WPT (Wavelet Packet Transform) is applied for the decomposition of the output voltage signals of CHMLI,and the fault features are extracted from the energy entropy of decomposed signals. Then,the PCA (Principal component analysis) is utilized to reduce the original sample's dimension,a lower dimension of features can help accelerating the training time of diagnostic model. Finally,a RF classification model optimized by relevant parameters is used for fault recognition after trained by reduced features samples. The experimental result baesd on Matlab shows that,compared with traditional methods such as SVM (Support Vector Machine) based on FFT (Fast Fourier Transform) and BP (Back Propagation) neural network based on WT (Wavelet Transform),the strategy based on WPEE and RF has higher diagnosis accuracy as high as 97% with short operation time,which is suitable for open circuit of the power switch transistors.

Key words: fault diagnosis, cascaded H?bridge multilevel inverter, wavelet packet energy entropy, random forest

中图分类号: 

  • TP391

图1

级联H桥七电平逆变器Simulink仿真模型"

图2

H?bridge2上开关晶体管开路时电压波形"

图3

基于小波包能量熵和随机森林的诊断流程图"

图4

H?bridge2上开关晶体管开路故障电压的小波包能量熵"

图5

随机森林参数调优结果"

图6

随机森林模型诊断结果"

表1

三种故障诊断方法的性能对比"

诊断方法训练时间(s)测试时间(s)故障诊断率(%)
①0.3220.09497.08
②0.2670.10993.80
③1.1860.58387.85
1 Liu J F,Cheng K W E,Ye Y M. A cascaded multilevel inverter based on switched?capacitor for high?frequency AC power distribution system. IEEE Transactions on Power Electronics,2014,29(8):4219-4230.
2 Jung S M,Park J S,Kim H W,et al. An MRAS?based diagnosis of open?circuit fault in PWM voltage?source inverters for PM synchronous motor drive systems. IEEE Transactions on Power Electronics,2013,28(5):2514-2526.
3 Xu H,Wang T Z,Qi J,et al. RPCA?SVM fault diagnosis strategy of cascaded H?bridge multilevel inverters∥2014 First International Conference on Green Energy. Sfax,Tunisia:IEEE,2014:164-169.
4 Fu L J,Yang Q,Wang G X,et al. Fault diagnosis of power electronic device based on wavelet and neural network∥2016 Chinese Control and Decision Conference. Yinchuan,China:IEEE,2016:2946-2950.
5 Miao B B,Shen Y X,Wu D H,et al. Three level inverter fault diagnosis using EMD and support vector machine approach∥2017 12th IEEE Conference on Industrial Electronics and Applications (ICIEA). Siem Reap,Cambodia:IEEE,2017:1595-1598.
6 姜媛媛，程浩，崔江等. VMD小波能量的光伏逆变器软故障诊断. 电力系统及其自动化学报,2018,30(11):19-25.
Jiang Y Y,Chen H,Cui J, et al. Soft fault diagnosis for PV inverter based on variational mode decomposition and wavelet energy. Proceedings of the CSU?EPSA,2018,30(11):19-25.)
7 Huang Z J,Wang Z S,Zhang H G. A diagnosis algorithm for multiple open?circuited faults of microgrid inverters based on main fault component analysis. IEEE Transactions on Energy Conversion,2018,33(3):925-937.
8 Choi U M,Jeong H G,Lee K B,et al. Method for detecting an open?switch fault in a grid?connected NPC inverter system. IEEE Transactions on Power Electronics,2012,27(6):2726-2739.
9 Wang ZW,Zhang QH,Xiong JB,et al. Fault diagnosis of a rolling bearing using wavlet packet denosing and random forests. IEEE Sensors Journal,2017,17(17):5581-5588.
10 Li S L,Liu Z L. Application of improved wavelet packet energy entropy and GA?SVM in rolling bearing fault diagnosis∥IEEE International Conference on Signal Processing,Communications and Computing. Qingdao,China:IEEE,2018:1-5.
11 Liu Z G,Han Z W,Zhang Y,et al. Multiwavelet packet entropy and its application in transmission line fault recognition and classification. IEEE Transactions on Neural Networks and Learning Systems,2014,25(11):2043-2052.
12 Huang Z J,Wang Z S,Yao X S,et al. Multi?switches fault diagnosis based on small low?frequency data for voltage?source inverters of PMSM drives. IEEE Transactions on Power Electronics,2019,34(7):6845-6857.
13 Yang S F,Tang Y,Wang P. Seamless fault?tolerant operation of a modular multilevel converter with switch open?circuit fault diagnosis in a distributed control architecture. IEEE Transactions on Power Electronics,2018,33(8):7058-7070.
14 Dai C X,Liu Z G,Hu K T,et al. Fault diagnosis approach of traction transformers in high?speed railway combining kernel principal component analysis with random forest. IET Electrical Systems in Transportation,2016,6(3):202-206.
15 De Mello Oliveira A B,Moreno R L,Ribeiro E R. Short?circuit fault diagnosis based on rough sets theory for a single?phase inverter. IEEE Transactions on Power Electronics,2019,34(5):4747-4764.
16 Yan H,Xu Y X,Cai F Y,et al. PWM?VSI fault diagnosis for a PMSM drive based on the fuzzy logic approach. IEEE Transactions on Power Electronics,2019,34(1):759-768.
[1] 安　晶, 艾　萍, 徐　森, 刘　聪, 夏建生, 刘大琨. 一种基于一维卷积神经网络的旋转机械智能故障诊断方法[J]. 南京大学学报(自然科学版), 2019, 55(1): 133-142.
[2] 胡　淼，王开军，李海超，陈黎飞. 模糊树节点的随机森林与异常点检测[J]. 南京大学学报(自然科学版), 2018, 54(6): 1141-1151.
[3]  曹冬寅，王　琼*，张兴敢. 基于稀疏重构残差和随机森林的集成分类算法
[J]. 南京大学学报(自然科学版), 2016, 52(6): 1127-.
[4]  赵荣泳 1 , 李翠玲 2 ** , 高晓康 3 , 王昭云 4
.  电子镇流器故障诊断的变精度粗糙集模型*
[J]. 南京大学学报(自然科学版), 2010, 46(5): 494-500.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 龙柄翰, 徐伟华. 模糊三支概念分析与模糊三支概念格[J]. 南京大学学报(自然科学版), 2019, 55(4): 537 -545 .
[2] 柴变芳,魏春丽,曹欣雨,王建岭. 面向网络结构发现的批量主动学习算法[J]. 南京大学学报(自然科学版), 2019, 55(6): 1020 -1029 .
[3] 党政,代群威,安超,彭启轩,卓曼他,杨丽君. 静态水蚀条件下自然钙华预制块的溶出特性研究[J]. 南京大学学报(自然科学版), 2019, 55(6): 916 -923 .
[4] 王大洋. 渤南洼陷沙三下亚段烃源岩地球化学特征及差异性研究[J]. 南京大学学报(自然科学版), 2019, 55(6): 924 -933 .
[5] 周昊,沈庆宏. 基于改进音形码的中文敏感词检测算法[J]. 南京大学学报(自然科学版), 2020, 56(2): 270 -277 .
[6] 王俞策,曹剑,陶柯宇,李二庭,向宝力,施春华. 准噶尔盆地芦草沟组致密油系统油源对比与成藏非均质性研究[J]. 南京大学学报(自然科学版), 2020, 56(3): 322 -337 .

版权所有 © 2021 《南京大学学报(自然科学)》

地址：江苏省南京市鼓楼区汉口路22号，南京大学学报编辑部，210093

电话：025-83592704 , 传真：025-83592704        技术支持：北京玛格泰克科技发展有限公司