一种基于时频域重构误差的卫星遥测数据异常检测算法*

doi:10.7523/j.ucas.2025.047

中国科学院大学学报

一种基于时频域重构误差的卫星遥测数据异常检测算法^*

李超^1,2, 王林林^1†

1 中国科学院国家空间科学中心复杂航天系统电子信息技术重点实验室北京 101407;
2 中国科学院大学北京 101408

收稿日期:2025-01-13 修回日期:2025-07-14 发布日期:2025-07-16
通讯作者: ^†E-mail: wanglinlin@nssc.ac.cn
基金资助:
^*中国科学院重点部署项目（KGFZD-145-2023-15）、深空探测全国重点实验室项目（NKLDSE2023A003）和基础科研项目（JCKY2021130B016）资助

A satellite anomaly detection algorithm based on time-frequency domain reconstruction error

LI Chao^1,2, WANG Linlin¹

1 Key Laboratory of Electronic Information Technology for Complex Space Systems, National Space Science Center, Chinese Academy of Sciences, Beijing 101407, China;
2 University of Chinese Academy of Sciences, Beijing 101408, China

Received:2025-01-13 Revised:2025-07-14 Published:2025-07-16

摘要/Abstract

摘要： 卫星遥测数据是卫星在轨运行过程中产生的重要数据,全面反映了卫星的运行状态,检测其中异常对维护卫星的安全与稳定具有重要意义。在工程实践中,传统的人工预设阈值比对的检测方法应用广泛,但该方法对阈值范围内的异常模式难以有效识别。同时,一些复杂的异常检测算法在性能表现上仍存在局限性。为了解决上述问题,提出了一种基于时频域联合分析的异常检测方法,通过对遥测数据进行时频分解、特征提取与重构误差计算,实现了对异常数据的精准识别。实验结果表明,在多个公开的异常检测数据集上,该方法能高效捕获多种异常模式,并展现出较强的泛化能力,为卫星运行提供了一种可靠的异常检测手段。

关键词: 异常检测, 遥测数据, 时频变化, 重构误差

Abstract: Satellite telemetry data is crucial for reflecting the operational status of satellites during their in-orbit operation, and anomaly detection in such data is of significant importance for ensuring the safety and stability of satellites. In engineering practice, the traditional detection method of manual preset threshold comparison is widely used, but the method is difficult to effectively recognize the anomaly patterns within the threshold range. At the same time, some complex anomaly detection algorithms still exhibit limitations in performance. To address these issues, a novel anomaly detection method based on joint analysis in time-frequency domain is proposed. By performing time-frequency decomposition, feature extraction, and reconstruction error calculation on the telemetry data, this method enables accurate identification of abnormal data. Experimental results demonstrate that this approach can efficiently capture anomalous patterns across several publicly available anomaly detection datasets and exhibit strong generalization capability, providing a reliable anomaly detection solution for satellite operations.

Key words: anomaly detection, telemetry data, time-Frequency variation, reconstruction error

中图分类号:

TP399

李超, 王林林. 一种基于时频域重构误差的卫星遥测数据异常检测算法^*[J]. 中国科学院大学学报, DOI: 10.7523/j.ucas.2025.047.

LI Chao, WANG Linlin. A satellite anomaly detection algorithm based on time-frequency domain reconstruction error[J]. Journal of University of Chinese Academy of Sciences, DOI: 10.7523/j.ucas.2025.047.

参考文献

[1] Hundman K, Constantinou V, Laporte C, et al.Detecting spacecraft anomalies using LSTMs and nonparametric dynamic thresholding[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London United Kingdom. ACM, 2018: 387-395. DOI: 10.1145/3219819.3219845.
[2] Fu N, Zhang G H, Xia K Q, et al.Research on fault detection and principal component analysis for spacecraft feature extraction based on kernel methods[J]. Open Astronomy, 2022, 31(1): 333-339. DOI: 10.1515/astro-2022-0194.
[3] 王志远,孙鹏菊,王海波,等. 基于聚类分类算法的IGBT健康状态分类研究[J]. 电工电能新技术, 2021, 40(11): 1-8. DOI: 10.12067/ATEEE2103041.
[4] Park D, Hoshi Y, Kemp C C.A multimodal anomaly detector for robot-assisted feeding using an LSTM-based variational autoencoder[J]. IEEE Robotics and Automation Letters, 2018, 3(3): 1544-1551. DOI: 10.1109/LRA.2018.2801475.
[5] Zeng A L, Chen M X, Zhang L, et al.Are transformers effective for time series forecasting?[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(9): 11121-11128. DOI: 10.1609/aaai.v37i9.26317.
[6] Xu Z J, Zeng A,Xu Q. FITS: Modeling Time Series with 10k Parameters[EB/OL].2023: arXiv: 2307.03756.(2023-7-6)[2025-6-25]. https://arxiv.org/abs/2307.03756.
[7] Wu H X, Hu T G, Liu Y, et al. TimesNet: Temporal 2D-variation modeling for general time series analysis[EB/OL].2022: arXiv: 2210.02186.(2022-10-5)[2025-6-25]. https://arxiv.org/abs/2210.02186.
[8] 彭喜元, 庞景月, 彭宇, 等. 航天器遥测数据异常检测综述[J]. 仪器仪表学报, 2016, 37(9): 1929-1945. DOI: 10.19650/j.cnki.cjsi.2016.09.002.
[9] Nie Y Q, Nguyen N H, Sinthong P, et al. A time series is worth 64 words: long-term forecasting with transformers[EB/OL].2022. arXiv: 2211.14730.(2022-11-27)[2025-6-25]. https://arxiv.org/abs/2211.14730.
[10] Kim T, Kim J, Tae Y, et al. Reversible instance normalization for accurate time-series forecasting against distribution shift[C/OL]//International Conference on Learning Representations. Virtual, Online, 2022. (2022-1-29)[2025-6-25]. https://openreview.net/forum?id=cGDAkQo1C0p.
[11] Yu G Q, Zou J, Hu X W, et al. Revitalizing multivariate time series forecasting: Learnable decomposition with inter-series dependencies and intra-series variations modeling[EB/OL].2024: arXiv: 2402.12694.(2024-2-20)[2025-6-25]. https://arxiv.org/abs/2402.12694v5.
[12] 王爱玲. “太极一号”卫星2020年部分分系统工程参数[DS/OL]. V1. Science Data Bank, (2022)[2025-06-25]. https://doi.org/10.57760/sciencedb.o00009.00053. DOI:10.57760/sciencedb.o00009.00053.
[13] Tax D M J, Duin R P W. Support vector data description[J]. Machine Learning, 2004, 54(1): 45-66. DOI: 10.1023/B:MACH.0000008084.60811.49.
[14] Liu F T, Ting K M, Zhou Z H.Isolation forest[C]//2008 Eighth IEEE International Conference on Data Mining. December 15-19, 2008, Pisa, Italy. IEEE, 2008: 413-422. DOI: 10.1109/ICDM.2008.17.
[15] Ruff L, Vandermeulen R, Goernitz N, et al. Deep one-class classification[C]//Proceedings of the 35th International Conference on Machine Learning. July 10-15, 2018, Stockholm, Swede,PMLR 80, 2018: 4393-4402.(2018)[2025-6-25]. https://proceedings.mlr.press/v80/ruff18a.html.
[16] Zhou T, Ma Z Q, Wen Q S, et al. FEDformer: Frequency enhanced decomposed transformer for long-term series forecasting[EB/OL].2022: arXiv: 2201.12740.(2022-1-30)[2025-6-15]. https://arxiv.org/abs/2201.12740v3.
[17] Zhou B, Liu S H, Hooi B, et al.BeatGAN: Anomalous rhythm detection using adversarially generated time series[C]//Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. August 10-16, 2019. Macao, China. International Joint Conferences on Artificial Intelligence Organization, 2019: 4433-4439. DOI: 10.24963/ijcai.2019/616.
[18] Su Y, Zhao Y J, Niu C H, et al.Robust anomaly detection for multivariate time series through stochastic recurrent neural network[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage AK USA. ACM, 2019: 2828-2837. DOI: 10.1145/3292500.3330672.
[19] Xu J H, Wu H X, Wang J M, et al. Anomaly transformer: Time series anomaly detection with association discrepancy[EB/OL].2021: arXiv: 2110.02642.(2021-10-6)[2025-6-25]. https://arxiv.org/abs/2110.02642.
[20] Shen L F, Li Z C, Kwok J T. Timeseries anomaly detection using temporal hierarchical one-class network[C]// Proceedings of the 34th International Conference on Neural Information Processing Systems(NIPS '20). Vancouver, BC, Canada, 2020: 13016-13026. (2020-12-6)[2025-6-25]. https://dl.acm.org/doi/abs/10.5555/3495724.3496816.
[21] Lai K H, Zha D C, Xu J J, et al. Revisiting time series outlier detection: Definitions and benchmarks[DS/OL]//35th Conference on Neural Information Processing Systems(NeurIPS2021) Track on Datasets and Benchmarks. (2021-7-29)[2025-6-25]. https://openreview.net/forum?id=r8IvOsnHchr.

一种基于时频域重构误差的卫星遥测数据异常检测算法^*

A satellite anomaly detection algorithm based on time-frequency domain reconstruction error

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	许凯凯, 张锐. 基于SE-TCN的一维低采样卫星帆板温度遥测数据插补方法[J]. 中国科学院大学学报, 2023, 40(6): 810-820.
[2]	何利健, 张锐, 林晓冬. 基于DWT和双通道LSTM的卫星电池阵电流预测方法[J]. 中国科学院大学学报, 2023, 40(3): 415-421.
[3]	孙宇豪, 李国通, 张鸽. 一种基于相关概率模型的卫星异常检测方法[J]. 中国科学院大学学报, 2021, 38(3): 409-416.
[4]	桂树, 郭立, 陆海先, 谢锦生. 基于MUGG的轨迹建模与异常检测[J]. 中国科学院大学学报, 2013, 30(2): 244-250.
[5]	谢锦生, 郭立, 陈运必, 赵龙. 基于时空惊奇计算的视频异常检测方法[J]. 中国科学院大学学报, 2013, 30(1): 83-89.
[6]	潘剑锋, 刘守群, 奚宏生, 谭小彬. 一种利用动态控制流路径分析的隐藏恶意代码异常检测方法[J]. 中国科学院大学学报, 2010, 27(1): 138-143.