电子电气工程与控制

一种基于GAMF-CNN的航空发电机整流器故障诊断技术

  • 崔江 ,
  • 周凡 ,
  • 陈永凡 ,
  • 于立 ,
  • 张卓然
展开
  • 南京航空航天大学 自动化学院,南京 210016
.E-mail: cuijiang@nuaa.edu.cn

收稿日期: 2024-03-18

  修回日期: 2024-03-31

  录用日期: 2024-04-22

  网络出版日期: 2024-05-08

基金资助

国家自然科学基金(51977108)

A technique for aerospace generator rectifier fault diagnosis based on GAMF-CNN

  • Jiang CUI ,
  • Fan ZHOU ,
  • Yongfan CHEN ,
  • Li YU ,
  • Zhuoran ZHANG
Expand
  • College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China

Received date: 2024-03-18

  Revised date: 2024-03-31

  Accepted date: 2024-04-22

  Online published: 2024-05-08

Supported by

National Natural Science Foundation of China(51977108)

摘要

针对航空发电机整流器实际故障样本数据较少等问题,提出一种基于格拉姆角积场-卷积神经网络(GAMF-CNN)的航空发电机整流器故障诊断技术。首先,采集整流器故障原始信号并进行预处理,将一维时序信号转化为格拉姆角积场图像,从而可以使故障诊断问题转变为图像识别问题;其次,借助深度迁移学习理念,采用卷积神经网络将仿真获取的故障特征知识迁移至缺少故障数据的发电机整流器中;最终,解决小样本数据下的航空发电机整流器故障诊断问题。经实验验证并与现有一些方法对比发现,所设计方法能以较高准确率实现故障二极管的诊断和定位。

本文引用格式

崔江 , 周凡 , 陈永凡 , 于立 , 张卓然 . 一种基于GAMF-CNN的航空发电机整流器故障诊断技术[J]. 航空学报, 2024 , 45(24) : 330398 -330398 . DOI: 10.7527/S1000-6893.2024.30398

Abstract

To address the problems such as less actual sample data of aerospace generator rectifier faults, a fault diagnosis technique based on Gramian Angular Multiply Field-Convolutional Neural Network (GAMF-CNN) is presented. First, original rectifier fault signals are collected and preprocessed, and the one-dimensional time series signals are transformed into GAMF images, so that the fault diagnosis problem can be transformed into an image recognition problem. Second, with the help of deep transfer learning concept, a convolutional neural network is used to transfer the fault feature knowledge obtained from the simulation to the real generator rectifier that lacks fault data. Finally, the aerospace generator rectifier fault diagnosis problem with small sample data is solved. Experimental verification and comparison with some existing methods find that the propsoed method can realize diagnosis and localization of faulty diodes with high accuracy.

参考文献

1 张卓然, 于立, 李进才, 等. 飞机电气化背景下的先进航空电机系统[J]. 南京航空航天大学学报201749(5): 622-634.
  ZHANG Z R, YU L, LI J C, et al. Advanced aircraft electric motor system in the context of aircraft electrification[J]. Journal of Nanjing University of Aeronautics and Astronautics201749(5): 622-634 (in Chinese).
2 于立. 多电发动机高速双凸极起动发电机系统关键技术研究[D]. 南京: 南京航空航天大学, 2019: 22-23.
  YU L. Research on key technology of high-speed double convex pole starter generator system for multi-electric engine[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2019: 22-23 (in Chinese).
3 冯小宝. 电励磁双凸极电机电动运行故障诊断与容错控制[D]. 南京: 南京航空航天大学, 2021: 1-3.
  FENG X B. Electrically excited double convex pole motor electric operation fault diagnosis and fault tolerant control[D].Nanjing: Nanjing University of Aeronautics and Astronautics, 2021: 1-3 (in Chinese).
4 SHARIDA A, BAYHAN S, ABU-RUB H, et al. Voltage-sensorless open-switch fault-tolerant control of three-phase T-type rectifier[J]. IEEE Transactions on Power Electronics202338(12): 15365-15376.
5 KOU L, LIU C, CAI G W, et al. Fault diagnosis for three-phase PWM rectifier based on deep feedforward network with transient synthetic features[J]. ISA Transactions2020101: 399-407.
6 CHEN J L, HAO L L, LI H Z, et al. Time–frequency characteristics analysis and diagnosis of rotating rectifier faults in multiphase annular brushless system[J]. IEEE Transactions on Industrial Electronics202370(4): 3233-3244.
7 JIAO N F, HAN X, WEI Z H, et al. Online fault diagnosis for rotating rectifier in wound-rotor synchronous starter–generator based on geometric features of current trajectory[J]. IEEE Transactions on Industrial Electronics202168(4): 2952-2963.
8 唐军祥. 航空三级式发电机旋转整流器故障诊断技术研究[D]. 南京: 南京航空航天大学, 2017: 5-7.
  TANG J X. Research on fault diagnosis technology of rotating rectifier of aviation three-stage type generator[D].Nanjing: Nanjing University of Aeronautics and Astronautics, 2017: 5-7 (in Chinese).
9 TANTAWY A, KOUTSOUKOS X, BISWAS G. Aircraft power generators: Hybrid modeling and simulation for fault detection[J]. IEEE Transactions on Aerospace and Electronic Systems201248(1): 552-571.
10 HUANG C, YUAN H W, MA Z, et al. The fault diagnosis of aircraft power system based on inverse problem of fuzzy optimization[J]. Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering2016230(6): 1059-1074.
11 夏一文, 张卓然, 张健, 等. 基于反电势电流的电励磁双凸极电机驱动电路单管开路故障诊断研究[J]. 电工技术学报202035(23): 4888-4897.
  XIA Y W, ZHANG Z R, ZHANG J, et al. A study on single-tube open-circuit fault diagnosis of electrically excited biconvex pole motor drive circuit based on counterpotential current[J]. Journal of Electrotechnology202035(23): 4888-4897 (in Chinese).
12 赵刚, 周波. 基于小波包分析的双凸极无刷直流发电机系统故障诊断[J]. 中国电机工程学报200828(8): 121-126.
  ZHAO G, ZHOU B. Fault diagnosis of doubly salient brushless DC generator system based on wavelet packet analysis[J]. Proceedings of the CSEE200828(8): 121-126 (in Chinese).
13 JORDAN S, APSLEY J. Open-circuit fault analysis of diode rectified multiphase synchronous generators for DC aircraft power systems[C]?∥ 2013 International Electric Machines & Drives Conference. Piscataway: IEEE Press, 2013: 926-932.
14 SHI X D, LI P J, ZHU J J, et al. Modeling and feather extraction of rotating rectifier faults of Aircraft integrated drive generator[J]. Journal of Applied Sciences201313(16): 3128-3136.
15 XU L, CAO M Y, SONG B Y, et al. Open-circuit fault diagnosis of power rectifier using sparse autoencoder based deep neural network[J]. Neurocomputing2018311: 1-10.
16 崔江, 冯赛, 张卓然, 等. 基于BLS的无刷发电机旋转整流器特征提取技术研究[J]. 中国电机工程学报202040(12): 4004-4013.
  CUI J, FENG S, ZHANG Z R, et al. Research on feature extraction technique of rotating rectifier for brushless generator based on BLS[J]. Proceedings of the CSEE202040(12): 4004-4013 (in Chinese).
17 陈文杰, 崔江. 基于LSTM的航空发电机整流电路诊断技术[J]. 电机与控制应用202350(4): 85-90.
  CHEN W J, CUI J. Diagnosis technology of aero-generator rectifier circuit based on LSTM[J]. Electric Machines & Control Application202350(4): 85-90 (in Chinese).
18 PAN S J, TSANG I W, KWOK J T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks201122(2): 199-210.
19 LONG M S, WANG J M, DING G G, et al. Transfer feature learning with joint distribution adaptation[C]?∥ 2013 IEEE International Conference on Computer Vision. Piscataway: IEEE Press, 2013: 2200-2207.
20 WANG J D, CHEN Y Q, HAO S J, et al. Balanced distribution adaptation for transfer learning[C]∥2017 IEEE International Conference on Data Mining (ICDM). Piscataway: IEEE Press, 2017: 1129-1134.
21 WANG J D, FENG W J, CHEN Y Q, et al. Visual domain adaptation with manifold embedded distribution alignment[C]?∥Proceedings of the 26th ACM international conference on Multimedia. New York: ACM, 2018: 402-410.
22 KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM201760(6): 84-90.
23 韩淞宇, 邵海东, 姜洪开, 等. 基于提升卷积神经网络的航空发动机高速轴承智能故障诊断[J]. 航空学报202243(9): 625479.
  HAN S Y, SHAO H D, JIANG H K, et al. Intelligent fault diagnosis of aero-engine high-speed bearings using enhanced CNN[J]. Acta Aeronautica et Astronautica Sinica202243(9): 625479 (in Chinese).
24 WANG Z, OATES T. Imaging time-series to improve classification and imputation[C]∥International Joint Conference on Artificial Intelligence. Buenos Aires: IJCAI, 2015: 3939-3945.
25 KEOGH E, CHAKRABARTI K, PAZZANI M, et al. Dimensionality reduction for fast similarity search in large time series databases[J]. Knowledge and Information Systems20013(3): 263-286.
文章导航

/