中山大学人工智能学院,广东 珠海 519082
詹明耀(2001年生),男;研究方向:故障诊断;E-mail: zhanmy7@mail2.sysu.edu.cn
孟云鹤(1978年生),男;研究方向:故障诊断;E-mail: mengyh7@mail.sysu.edu.cn
收稿:2026-01-04,
录用:2026-02-28,
网络首发:2026-04-02,
移动端阅览
詹明耀, 孟云鹤. 样本不平衡条件下航天器姿态控制的故障诊断[J/OL]. 中山大学学报(自然科学版)(中英文), 2026,1-11.
ZHAN Mingyao, MENG Yunhe. Fault diagnosis of spacecraft attitude control under sample imbalance conditions[J/OL]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2026, 1-11.
詹明耀, 孟云鹤. 样本不平衡条件下航天器姿态控制的故障诊断[J/OL]. 中山大学学报(自然科学版)(中英文), 2026,1-11. DOI: 10.11714/acta.snus.ZR20260004.
ZHAN Mingyao, MENG Yunhe. Fault diagnosis of spacecraft attitude control under sample imbalance conditions[J/OL]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2026, 1-11. DOI: 10.11714/acta.snus.ZR20260004.
针对航天器姿态控制系统在故障样本稀缺、样本不平衡条件下的诊断性能下降问题,提出一种融合时间序列图像编码与深度学习的故障诊断方法。首先,采用带梯度惩罚的Wasserstein生成对抗网络(WGAN-GP)对故障时序样本进行生成增广,以平衡训练集分布;其次,利用Gramian角和场、马尔可夫转移场及递归图编码方法将时序信号转换为二维图像;最后,构建二维卷积神经网络实现故障特征的提取与分类。实验结果表明,所提方法在所有样本不平衡比例下均能显著提升诊断准确率,验证了其对于样本不平衡场景的有效性与鲁棒性。
To address the performance degradation of fault diagnosis in spacecraft attitude control systems under conditions of scarce fault samples and imbalanced data distribution, this paper proposes a fault diagnosis method integrating time-series image encoding and deep learning. First, a Wasserstein Generative Adversarial Network with gradient penalty (WGAN-GP) is employed to augment fault time-series samples and balance the training set distribution. Subsequently, three encoding methods—Gramian Angular Summation Field, Markov Transition Field, and Recurrence Plot—are utilized to convert time-series signals into two-dimensional images. Then, a two-dimensional convolutional neural network is constructed to extract and classify fault features. Experimental results demonstrate that the proposed method significantly improves diagnostic accuracy under varying sample imbalance ratios, validating its effectiveness and robustness in imbalanced sample scenarios.
黄瑾 , 刘洋 , 钟麦英 , 等 , 2021 . 利用随机森林算法的卫星控制系统故障诊断 [J]. 宇航学报 , 42 ( 4 ): 513 - 521 .
梁寒玉 , 刘成瑞 , 刘文静 , 等 , 2024 . 基于序列-图像映射的航天器智能故障诊断方法 [J]. 飞控与探测 , 7 ( 1 ): 62 - 71 .
陆钦华 , 陈嘉宇 , 王旭航 , 等 , 2025 . 数据不平衡故障诊断:一种预训练数据增强方法 [J]. 测控技术 , 44 ( 1 ): 10 - 21 .
邱瑞 , 姚全营 , 刘鹏 , 等 , 2024 . 基于动态不确定因果图的航天器故障诊断方法 [J]. 航天器工程 , 33 ( 5 ): 9 - 14 .
沈毅 , 李利亮 , 王振华 , 2020 . 航天器故障诊断与容错控制技术研究综述 [J]. 宇航学报 , 41 ( 6 ): 647 - 656 .
孙权 , 彭飞 , 李宏胜 , 等 , 2024 . 样本不均衡下基于CGAN-CNN的逆变器故障诊断方法 [J]. 电源学报 , 22 ( 6 ): 318 - 326 .
张达 , 高君宇 , 丁腾欢 , 等 , 2023 . 基于时序二维化的航空传感器故障检测 [J]. 西北工业大学学报 , 41 ( 6 ): 1033 - 1043 .
ARJOVSKY M , CHINTALA S , BOTTOU L , 2017 . Wasserstein generative adversarial networks [C]// Proceedings of the 34th International Conference on Machine Learning . Sydney, NSW, Australia : 214 - 223 .
CHAPMAN P , ZENTGRAF P , JAFRY Y , 2003 . Drag-Free Control Design Including Attitude Transition for the STEP Mission [J]. Spacecraft Guidance, Navigation and Control Systems , 516 : 551 .
CHEN S , YANG R , ZHONG M , et al , 2023 . A random forest and model-based hybrid method of fault diagnosis for satellite attitude control systems [J]. IEEE Trans Instrum Meas , 72 : 3518413 .
GAO S , ZHANG W , HE X , 2020 . Observer‐based multiple faults diagnosis scheme for satellite attitude control system [J]. Asian J Control , 22 ( 1 ): 307 - 322 .
GALLON R , SCHIEMENZ F , MENICUCCI A , et al , 2025 . Convolutional neural network design and evaluation for real-time multivariate time series fault detection in spacecraft attitude sensors [J]. Adv Space Res , 76 ( 5 ): 2960 - 2976 .
GOODFELLOW I , POUGET-ABADIE J , MIRZA M , et al , 2014 . Generative adversarial nets [EB/OL]. MIT Press . DOI: 10.3156/JSOFT.29.5_177_2 http://dx.doi.org/10.3156/JSOFT.29.5_177_2 .
GREWAL M S , WEILL L R , ANDREWS A P , 2007 . Global Positioning Systems, Inertial Navigation, and Integration [M]. NewYork : John Wiley & Sons .
GULRAJANI I , AHMED F , ARJOVSKY M , et al , 2017 . Improved training of Wasserstein GANs [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems . Long Beach, California, USA : 5769 - 5779 .
GUO X G , TIAN M E , LI Q , et al , 2020 . Multiple-fault diagnosis for spacecraft attitude control systems using RBFNN-based observers [J]. Aerosp Sci Technol , 106 : 106195 .
HE W , LIN L , FU S , et al , 2025 . Differential contrast guidance for aeroengine fault diagnosis with limited data [J]. J Intell Manuf , 36 ( 2 ): 1409 - 1427 .
HEDAYATI M , RAHIMI A , 2025 . A hybrid framework for real-time satellite fault diagnosis using Markov jump-adjusted models and 1D sliding window Residual Networks [J]. Acta Astronaut , 228 : 1066 - 1087 .
HEDAYATI M S , BARZEGAR A , RAHIMI A , 2024 . Fault diagnosis and prognosis of satellites and unmanned aerial vehicles: A review [J]. Appl Sci-Basel , 14 ( 20 ): 9487 .
HEDAYATI M S , BARZEGAR A , RAHIMI A , 2024 . Mitigating data scarcity for satellite reaction wheel fault diagnosis with Wasserstein generative adversarial networks [C]// 2024 IEEE International Conference on Prognostics and Health Management . Spokane, WA, USA : 367 - 376 .
HENRY D , 2008 . Fault diagnosis of microscope satellite thrusters using H-infinity/H_ filters [J]. J Guid Control Dyn , 31 ( 3 ): 699 - 711 .
LeCUN Y , BENGIO Y , HINTON G , 2015 . Deep learning [J]. Nature , 521 ( 7553 ): 436 - 444 .
LEI C , MIAO C , WAN H , et al , 2023 . Rolling bearing fault diagnosis method based on MTF-MFACNN [J]. Meas Sci Technol , 35 ( 3 ): 035007 .
LI Y , HU Q , SHAO X , 2022 . Neural network-based fault diagnosis for spacecraft with single-gimbal control moment gyros [J]. Chin J Aeronaut , 35 ( 7 ): 261 - 273 .
MARKLEY F L , CRASSIDIS J L , 2014 . Fundamentals of spacecraft attitude determination and control [M]. NewYork : Springer .
MARWAN N , ROMANO M C , THIEL M , et al , 2007 . Recurrence plots for the analysis of complex systems [J]. Phys Rep , 438 ( 5/6 ): 237 - 329 .
MENG L , XIE J , CHENG L , et al , 2026 . Multi-condition bearing fault diagnosis method based on RP–MBRN [J]. Struct Health Monit , 25 ( 1 ): 456 - 467 .
POURTAKDOUST S H , FAKHARI MEHRJARDI M , HAJKARIM M H , et al , 2023 . Advanced fault detection and diagnosis in spacecraft attitude control systems: Current state and challenges [J]. J Aerosp Eng , 237 ( 12 ): 2679 - 2699 .
SUO M , ZHU B , AN R , et al , 2019 . Data-driven fault diagnosis of satellite power system using fuzzy Bayes risk and SVM [J]. Aerosp Sci Technol , 84 : 1092 - 1105 .
VENKATESWARAN N , SIVA M S , GOEL P S , 2002 . Analytical redundancy based fault detection of gyroscopes in spacecraft applications [J]. Acta Astronaut , 50 ( 9 ): 535 - 545 .
WANG Z , OATES T , 2015 . Imaging time-series to improve classification and imputation [EB/OL]. arXiv : 1506 . 00327 . https://arxiv.org/abs/1506.00327 https://arxiv.org/abs/1506.00327 .
XIAO B , YIN S , 2020 . A deep learning based data-driven thruster fault diagnosis approach for satellite attitude control system [J]. IEEE Trans Ind Electron , 68 ( 10 ): 10162 - 10170 .
YANG C L , CHEN Z X , YANG C Y , 2019 . Sensor classification using convolutional neural network by encoding multivariate time series as two-dimensional colored images [J]. Sensors , 20 ( 1 ): 168 .
YANG Y , 2012 . Spacecraft attitude determination and control: Quaternion based method [J]. Annu Rev Control , 36 : 198 - 219 .
ZHANG X , YANG K , 2025 . Transformer fault diagnosis method based on MTF and GhostNet [J]. Measurement , 249 : 117056 .
ZHAO H , LIU M , SUN Y , et al , 2024 . Automated design of fault diagnosis CNN network for satellite attitude control systems [J]. IEEE Trans Cybern , 54 ( 7 ): 4028 - 4038 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621
