Page 162 - 《振动工程学报》2026年第3期
P. 162

762                                振   动   工   程   学   报                               第 39 卷

                  [J]. arxiv preprint arxiv:1412.3474, 2014.         multi-source  domain  heterogeneous  model  transfer[J].
             [4] LONG M, CAO Y, WANG J, et al. Learning transfer⁃    Journal  of  Vibration  and  Shock,  2023,  42(24) :
                   able features with deep adaptation networks[C]// Inter⁃  257-266.
                   national  Conference  on  Machine  Learning.  2015:   [11] DE  SANTANA  CORREIA  A,  COLOMBINI  E  L.
                   97-105.                                           Attention, please! A survey of neural attention models
             [5] JIANG Y M, XIA T B, WANG D, et al. Joint adap⁃      in  deep  learning[J].  Artificial  Intelligence  Review,
                   tive transfer learning network for cross-domain fault di⁃  2022, 55(8): 6037-6124.
                   agnosis  based  on  multi-layer  feature  fusion[J].  Neuro⁃  [12] WANG Y P, WANG J B, ZHANG S, et al. Remain⁃
                   computing, 2022, 487: 228-242.                    ing  useful  life  prediction  model  for  rolling  bearings
             [6] HAN T, LIU C, YANG W G, et al. A novel adversar⁃    based  on  MFPE-MACNN[J].  Entropy,  2022,  24
                   ial learning framework in deep convolutional neural net⁃  (7): 905.
                   work  for  intelligent  diagnosis  of  mechanical  faults[J].   [13] VASWANI  A,  SHAZEER  N,  PARMAR  N,  et  al.
                   Knowledge-Based Systems, 2019, 165: 474-487.      Attention is all you need[C]// Advances in Neural In⁃
             [7] 苏浩,杨鑫,向玲,等 . 基于深度对比迁移学习的变工况                         formation Processing Systems  30 (NIPS 2017). 2017,
                   下 机 械 故 障 诊 断[J]. 振 动 工 程 学 报 , 2023, 36(3):      30: 5998-6008.
                   845-853.                                     [14] GRETTON  A,  BORGWARDT  K,  RASCH  M,  et
                   SU Hao, YANG Xin, XIANG Ling, et al. Mechanical   al.  A  kernel  method  for  the  two-sample-problem[C]//
                   fault  diagnosis  using  deep  contrastive  transfer  learning   Advances  in  Neural  Information  Processing  Systems
                   under variable working conditions[J]. Journal of Vibra⁃  19:  Proceedings  of  the  2006  Conference.  2006,  19:
                   tion Engineering, 2023, 36(3): 845-853.           513-520.
             [8] XIA Y, SHEN C Q, WANG D, et al. Moment match⁃  [15] GRETTON A, BORGWARDT K M, RASCH M J,
                   ing-based  intraclass  multisource  domain  adaptation  net⁃  et al. A kernel two-sample test[J]. The Journal of Ma⁃
                   work  for  bearing  fault  diagnosis[J].  Mechanical  Sys⁃  chine Learning Research, 2012, 13: 723-773.
                   tems and Signal Processing, 2022, 168: 108697.  [16] LESSMEIER C, KIMOTHO J K, ZIMMER D, et al.
             [9] YE  Z,  YU  J  B.  Deep  negative  correlation  multisource   Condition  monitoring  of  bearing  damage  in  electrome⁃
                   domains adaptation network for machinery fault diagno⁃  chanical drive systems by using motor current signals of
                   sis  under  different  working  conditions[J].  IEEE/  electric  motors:  a  benchmark  data  set  for  data-driven
                   ASME  Transactions  on  Mechatronics,  2022,  27(6):
                                                                     classification[C]// PHM  Society  European  Confer⁃
                   5914-5925.                                        ence. 2016.
             [10] 王玉静,夏林,康守强,等 . 基于多源域异构模型迁移
                   的滚动轴承故障诊断方法[J]. 振动与冲击, 2023, 42              第一作者: 王亚萍(1972―),女,博士,教授。
                  (24): 257-266.                                         E-mail: wypbl@163.com
                   WANG Yujing, XIA Lin, KANG Shouqiang, et al. A   通信作者: 江旭东(1977―),男,博士,讲师。
                   fault  diagnosis  method  of  rolling  bearings  based  on   E-mail: jxd_2023@163.com
   157   158   159   160   161   162   163   164   165   166   167