基于扩散模型和Informer的卫星转台故障数据生成方法&#x0002A;

兰青; 林宝军; 朱野

doi:10.7523/j.ucas.2025.025

中国科学院大学学报 >

0 2025003 - 2025003

DOI: https://doi.org/10.7523/j.ucas.2025.025

基于扩散模型和Informer的卫星转台故障数据生成方法^*

兰青 ,
林宝军 ,
朱野

展开

中国科学院微小卫星创新研究院,上海 201304;
中国科学院大学,北京 100049;
上海微小卫星工程中心,上海 201304

收稿日期: 2025-01-10

修回日期: 2025-04-17

网络出版日期: 2025-07-17

基金资助

^*中国科学院青年创新促进会优秀会员（Y2022078）资助

收起

Fault data generation method for satellite turntable based on diffusion model and Informer

LAN Qing ,
LIN Baojun ,
ZHU Ye

Expand

Innovation Academy for Microsatellite, Chinese Academy of Sciences, Shanghai 201304, China;
University of Chinese Academy of Sciences, Beijing 100049, China;
Shanghai Engineering Centre for Microsatellites, Shanghai 201304, China

Received date: 2025-01-10

Revised date: 2025-04-17

Online published: 2025-07-17

Fold

摘要

针对卫星遥测数据中故障数据样本少,导致故障检测和诊断模型表现受限的问题,本文提出了一种基于扩散模型和Informer架构的卫星转台故障数据生成方法。该方法结合时间序列分解技术与深度学习模型,旨在生成高质量且具备可解释性的时间序列数据。在扩散正向过程,通过数据分解实现了成分解耦,从而提升了生成数据的多样性和准确性。在逆向生成阶段,利用Informer解码器的稀疏自注意力机制,有效捕捉长时依赖关系和复杂的时序模式。该方法适用于无条件和有条件的故障数据生成,包括数据缺失的场景。实验结果表明,该方法在均方误差（MSE）、相关评分（CS）和相关系数（R）等指标上显著优于对抗生成网络和变分自编码器等方法。

关键词： 卫星转台; 遥测数据; 扩散模型; 时间序列分解; 故障数据生成

本文引用格式

兰青 , 林宝军 , 朱野 . 基于扩散模型和Informer的卫星转台故障数据生成方法^*[J]. 中国科学院大学学报, 0 : 2025003 -2025003 . DOI: 10.7523/j.ucas.2025.025

Abstract

This paper proposes a fault data generation method for satellite turntables based on a diffusion model and the Informer architecture, to address the issue of limited fault data samples in satellite telemetry, which restricts the performance of fault detection and diagnosis models. By integrating time series decomposition with deep learning models, the proposed method aims to generate high quality and interpretable time series data. Through data decomposition, the forward diffusion process achieves component decoupling, improving the diversity and accuracy of the generated data. During the reverse generation phase, the Informer decoder leverages a sparse self-attention mechanism to effectively capture long-term dependencies and complex temporal patterns. The method is suitable for both unconditional and conditional fault data generation, including scenarios with missing data. Experimental results demonstrate that the proposed method significantly outperforms Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) in terms of metrics such as Mean Squared Error (MSE), Correlation Score (CS), and Correlation Coefficient (R).

Key words： satellite turntable; telemetry data; diffusion model; time series decomposition; fault data generation

参考文献

[1] Xu J M, Li Z, Tang H H, et al.Significant influence of nonlinear friction torque on motion performance of tracking turntables[J]. Tribology International, 2019, 136:148-154. DOI: 10.1016/j.triboint.2019.03.059
[2] 王蜀泉, 赵光恒. 基于模糊控制的卫星大角度姿态机动控制方法研究[J]. 中国科学院研究生院学报, 2006, 23(1): 111-117. DOI: 10.3969/j.issn.1002-1175.2006.01.016.
[3] Wang J H, Ruan P, Xie Y J, et al.Design of U-shaped frame of spaceborne turntable based on multi-constraint topology optimization method[J]. Applied Sciences, 2024, 14(13): 5842. DOI: 10.3390/app14135842.
[4] 王泽, 姜斌, 程月华, 等. 一种基于生成对抗网络的卫星异常检测方法[J]. 空间控制技术与应用, 2023, 49(1): 113-120. DOI: 10.3969/j.issn.1674-1579.2023.01.013.
[5] 孙宇豪, 李国通, 张鸽. 距离相关系数融合GPR模型的卫星异常检测方法[J]. 北京航空航天大学学报, 2021, 47(4): 844-852. DOI: 10.13700/j.bh.1001-5965.2020.0041.
[6] Lei Y G, Yang B, Jiang X W, et al.Applications of machine learning to machine fault diagnosis: A review and roadmap[J]. Mechanical Systems and Signal Processing, 2020, 138: 106587. DOI: 10.1016/j.ymssp.2019.106587.
[7] Luo W, Yang W, He J L, et al.Fault diagnosis method based on two-stage GAN for data imbalance[J]. IEEE Sensors Journal, 2022, 22(22): 21961-21973. DOI: 10.1109/JSEN.2022.3211021.
[8] Ren Z J, Zhu Y S, Liu Z, et al.Few-shot GAN: Improving the performance of intelligent fault diagnosis in severe data imbalance[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 3516814. DOI: 10.1109/TIM.2023.3271746.
[9] Yan K, Su J Y, Huang J, et al.Chiller fault diagnosis based on VAE-enabled generative adversarial networks[J]. IEEE Transactions on Automation Science and Engineering, 2022, 19(1): 387-395. DOI: 10.1109/TASE.2020.3035620.
[10] Lu S F, Tang X L, Zhu Y J, et al.A cloud-edge collaborative intelligent fault diagnosis method based on LSTM-VAE hybrid model[C]//2021 8th IEEE International Conference on Cyber Security and Cloud Computing (CSCloud)/2021 7th IEEE International Conference on Edge Computing and Scalable Cloud (EdgeCom). June 26-28, 2021. Washington, DC, USA. IEEE, 2021: 207-212. DOI: 10.1109/cscloud-edgecom52276.2021.00045.
[11] Baptista M, Sankararaman S, de Medeiros I P, et al. Forecasting fault events for predictive maintenance using data-driven techniques and ARMA modeling[J]. Computers & Industrial Engineering, 2018, 115: 41-53. DOI: 10.1016/j.cie.2017.10.033.
[12] de Oliveira B N, Valk M, Marcondes Filho D. Fault detection and diagnosis of batch process dynamics using ARMA-based control charts[J]. Journal of Process Control, 2022, 111: 46-58. DOI: 10.1016/j.jprocont.2022.01.005.
[13] 陈予恕. 机械故障诊断的非线性动力学原理[J]. 机械工程学报, 2007, 43(1): 25-34. DOI: 10.3321/j.issn:0577-6686.2007.01.005.
[14] Mogren O.C-RNN-GAN: Continuous recurrent neural networks with adversarial training[EB/OL]. 2016: 1611.09904. DOI: 10.48550/arXiv.1611.09904.
[15] Yoon J, Jarrett D, Schaar M V D.Time-series Generative Adversarial Networks[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems. 2019: 494. DOI: 10.5555/3454287.3454781.
[16] Borji A.Pros and cons of GAN evaluation measures: New developments[J]. Computer Vision and Image Understanding, 2022, 215: 103329. DOI: 10.1016/j.cviu.2021.103329.
[17] Xia X, Pan X Z, Li N, et al.GAN-based anomaly detection: A review[J]. Neurocomputing, 2022, 493: 497-535. DOI: 10.1016/j.neucom.2021.12.093.
[18] Desai A, Freeman C, Wang Z H, et al.TimeVAE: A variational auto-encoder for multivariate time series generation[EB/OL]. 2021: 2111.08095. DOI: 10.48550/arXiv.2111.08095
[19] Rybkin O, Daniilidis K, Levine S.Simple and effective VAE training with calibrated decoders[C]//International conference on machine learning. PMLR, 2021: 9179-9189. DOI: 10.48550/arXiv.2006.13202
[20] 张思慧, 吴云龙, 张毅, 等. 一种基于联合变分自编码器的卫星重力数据粗差探测方法研究[J]. 武汉大学学报(信息科学版), 2024, 49(6): 986-995. DOI: 10.13203/j.whugis20230226.
[21] Yang L, Zhang Z L, Song Y, et al.Diffusion models: A comprehensive survey of methods and applications[J]. ACM Computing Surveys, 2024, 56(4): 1-39. DOI: 10.1145/3626235.
[22] Zhu Q B, Han J L, Chai K, et al.Time series analysis based on informer algorithms: A survey[J]. Symmetry, 2023, 15(4): 951. DOI: 10.3390/sym15040951.
[23] 戴前伟, 丁浩, 张华, 等. 基于变分模态分解和奇异谱分析的GPR信号去噪[J]. 吉林大学学报 (地球科学版), 2022, 52(3): 701-712. DOI: 10.13278/j.cnki.jjuese.20210243.
[24] 陶景桥, 孙小松, 李明. 自旋卫星测试转台精度分析[J]. 空间控制技术与应用, 2010, 36(2): 20-24, 30. DOI: 10.3969/j.issn.1674-1579.2010.02.004
[25] Zhang X R, Rane K P, Kakaravada I, et al.Research on vibration monitoring and fault diagnosis of rotating machinery based on Internet of Things technology[J]. Nonlinear Engineering, 2021, 10(1): 245-254. DOI: 10.1515/nleng-2021-0019.
[26] Anowar F, Sadaoui S, Selim B.Conceptual and empirical comparison of dimensionality reduction algorithms (pca, kpca, lda, mds, svd, lle, isomap, le, ica, t-sne)[J]. Computer Science Review, 2021, 40: 100378.DOI: 10.1016/j.cosrev.2021.100378.
[27] Zhang C, Jiang D F, Jiang K H, et al.A hierarchical multivariate denoising diffusion model[J]. Information Sciences, 2023, 648: 119623. DOI: 10.1016/j.ins.2023.119623.
[28] Chicco D, Warrens M J, Jurman G.The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation[J]. Peerj computer science, 2021, 7: e623. DOI: 10.7717/peerj-cs.623.
[29] Zhou H M, Deng Z H, Xia Y Q, et al.A new sampling method in particle filter based on Pearson correlation coefficient[J]. Neurocomputing, 2016, 216: 208-215. DOI: 10.1016/j.neucom.2016.07.036.
[30] Liao S J, Ni H, Sabate-Vidales M, et al.Sig-Wasserstein GANs for conditional time series generation[J]. Mathematical Finance, 2024, 34(2): 622-670. DOI: 10.1111/mafi.12423.
[31] Qahtan A, Wang S J, Zhang X L.KDE-track: An efficient dynamic density estimator for data streams[J]. IEEE Transactions on Knowledge and Data Engineering, 2017, 29(3): 642-655. DOI: 10.1109/TKDE.2016.2626441.
[32] 朱杰, 陈黎飞. 核密度估计的聚类算法[J]. 模式识别与人工智能, 2017, 30(5): 439-447. DOI: 10.16451/j.cnki.issn1003-6059.201705006.

Options

文章导航

摘要

本文引用格式

Abstract

参考文献

访问统计

联系我们

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献

访问统计

联系我们