Page 147 - 《振动工程学报》2025年第8期
P. 147
第 8 期 刘 杰,等: 不均衡样本下轴承故障的 LSGAN-Swin Transformer 诊断方法 1787
140: 103689. main bearing fault with imbalanced small sample scenari‑
[12] WANG Z J, ZHAO J H, ZHANG R, et al. UATNet: os[J]. Proceedings of the CSEE, 2020, 40(2):
U-Shape attention-based transformer net for meteorolog‑ 563-574.
ical satellite cloud recognition[J]. Remote Sensing, [19] MAO X D, LI Q, XIE H R, et al. On the effectiveness
2021, 14(1):104. of least squares generative adversarial networks[J].
[13] 田应仲, 卜雪虎 . 基于注意力机制与 Swin Transformer IEEE Transactions on Pattern Analysis and Machine In‑
模型的腰椎图像分割方法[J]. 计量与测试技术,2021, telligence, 2019, 41(12): 2947-2960.
48(12):57-61. [20] YUAN W, XU W B. MSST-Net: a multi-scale adap‑
TIAN Yingzhong, BU Xuehu. Lumbar spine image tive network for building extraction from remote sensing
segmentation method based on attention mechanism and images based on swin transformer[J]. Remote Sensing,
swin transformer model[J]. Metrology & Measurement 2021, 13(23): 4743.
Technique, 2021, 48(12): 57-61. [21] GU Y, PIAO Z G, YOO S J. STHarDNet: swin trans‑
[14] 高涛,文渊博,陈婷,等 . 基于窗口自注意力网络的单图 former with HarDNet for MRI segmentation[J]. Ap‑
像 去 雨 算 法[J]. 上 海 交 通 大 学 学 报 , 2023,57(5): plied Sciences, 2022, 12(1): 468.
613-623. [22] XU Z Y, ZHANG W C, ZHANG T X, et al. Efficient
GAO Tao, WEN Yuanbo, CHEN Ting, et al. A sin‑ transformer for remote sensing image segmentation[J].
gle image deraining algorithm based on swin transformer Remote Sensing, 2021, 13(18): 3585.
[J]. Journal of Shanghai Jiao Tong University, 2023,57 [23] WANG J H, ZHANG Z Y, LUO L F, et al.
(5):613-623. SwinGD: a robust grape bunch detection model based
[15] 扶兰兰, 黄昊, 王恒, 等 . 基于 Swin Transformer 模型 on swin transformer in complex vineyard environment
的 玉 米 生 长 期 分 类[J]. 农 业 工 程 学 报 , 2022, 38 [J]. Horticulturae, 2021, 7(11): 492.
(14): 191-200. [24] SMITH W A, RANDALL R B. Rolling element bear‑
FU Lanlan, HUANG Hao, WANG Heng, et al. Clas‑ ing diagnostics using the Case Western Reserve Univer‑
sification of maize growth stages using the Swin Trans‑ sity data: a benchmark study[J]. Mechanical Systems
former model[J]. Transactions of the Chinese Society and Signal Processing, 2015, 64-65: 100-131.
of Agricultural Engineering, 2022, 38(14): 191-200. [25] MAO X D, LI Q, XIE H R, et al. Least squares gener‑
[16] 王璨, 武新慧, 张燕青, 等 . 基于移位窗口 Transform‑ ative adversarial networks[C]∥ Proceedings of the
er 网络的玉米田间场景下杂草识别[J]. 农业工程学 IEEE International Conference on Computer Vision(IC‑
报, 2022, 38(15): 133-142. CV). Venice, Italy: IEEE, 2017: 2813-2821.
WANG Can, WU Xinhui, ZHANG Yanqing, et al. [26] ZHOU Y Q, LIU Y, ZHOU H Y, et al. Wasserstein
Recognizing weeds in maize fields using shifted window distance feature alignment learning for 2D image-based
Transformer network[J]. Transactions of the Chinese 3D model retrieval[J]. Journal of Visual Communica‑
Society of Agricultural Engineering, 2022, 38(15): tion and Image Representation, 2021, 79: 103197.
133-142. [27] HUANG Y M, LEI H, LI X Y, et al. Quantum maxi‑
[17] GOODFELLOW I J, POUGET-ABADIE J, MIRZA mum mean discrepancy GAN[J]. Neurocomputing,
M, et al. Generative adversarial nets[J]. Advances in 2021, 454: 88-100.
Neural Information Processing Systems, 2014, 27: [28] SHAO S Y, MCALEER S, YAN R Q, et al. Highly
2672-2680. accurate machine fault diagnosis using deep transfer
[18] 黄南天, 杨学航, 蔡国伟, 等 . 采用非平衡小样本数据 learning[J]. IEEE Transactions on Industrial Informat‑
的风机主轴承故障深度对抗诊断[J]. 中国电机工程学 ics, 2019, 15(4): 2446-2455.
报, 2020, 40(2): 563-574.
HUANG Nantian, YANG Xuehang, CAI Guowei, et 通信作者: 刘 杰(1980—),女,博士,副教授。
al. A deep adversarial diagnosis method for wind turbine E-mail: starliujie@126.com

