受变电站复杂环境的影响，无人机在变电站巡检时的飞行安全问题变得日益突出 。 目前，无人机的飞行状 态监测主要采用飞行控制系统，缺乏对飞行数据，特别是飞行轨迹数据的深入分析 。为了更深入地挖掘飞行控制 系统的数据，文中提出了异常检测和轨迹预警方法 。首先，基于子空间学习方法，学习保存在子空间中的原始数据 的局部时序信息，不仅降低了数据的计算复杂度，而且促进了异常数据的发现；其次，通过测量数据子空间向量的 变化来实现飞行数据的异常检测；最后，利用基于时间序列的 LSTM 神经网络实现无人机的飞行轨迹预测，只对会 导致无人机飞行安全的异常数据进行预警，提高了预警效率。
[1] Schumann J，Rozier K Y，Reinbacher T，et al. Towards real- time ，on-board，hardware-supported sensor and software health management for unmanned aerial systems[J]. Interna- tional Journal of Prognostics and Health Management，2015， 6（1）∶1-27.
[2] MELNYK I，YADAV P，STEINBACH M，et al.Detection of precursors to aviation safety incidents due to human factors [C]//Workshops of the 2013 IEEE International Conference on Data Mining. Piscataway：IEEE，2014∶407-412.
[3] KhalastchiE，Kaminka G A，Kalech M，et al.Online anomaly detection in unmanned vehicles[C]//Adaptive Agents and Multi-Agents Systems. International Foundation for Autono-mous Agents and Multiagent Systems，2011. DOI：10.5555/ 2030470.2030487.
[4] Freeman，P，Pandita，etal. Model-Based and Data-Driven Fault Detection Performance for a Small UAV[J]. IEEE/ ASME transactions on mechatronics，2013（4）∶18.
[5] HATICE V D，MURAT T，NIHAN K.Unmanned aerial vehi cles（UAVs）battery power anomaly detection using temporal convolutional network with simple moving average algorithm [C]//Proceedings of the 2022 International Conference on In novations in Intelligent Systems and Applications（INISTA）， Piscataway：IEEE，2022 ∶1-5.
[6] You J T，LIANG J，LIU D T. An adaptable UAV sensor data anomaly detection method based on TCN model transferring
[C]// Proceedings of 2022 Prognostics and Health Manage- ment Conference.Piscataway：IEEE，2022∶73-76.
[7] Y. He，Y. Peng，S. Wang，et al. "ADMOST：UAV flight data anomaly detection and mitigation via online subspace track - ing，"in IEEE transactions on instrumentation and measure- ment，vol.68，no.4，pp. 1035-1044，April 2019，doi：10. 1109/ TIM.2018.2863499.
[8]李苗 .基于GWO-LSTM算法的船舶轨迹预测研究[J].珠江水运， 2024，（12）∶49-51. DOI：10. 14125/j. cnki. zjsy. 2024.12.043.
[9] 张邦楚，杨朋坤，梁毅雍，等 .基于LSTM的空战无人机实时航迹 预测[J]. 战术导弹技术，2024，（03）∶72-83. DOI：10. 16358/j. issn.1009-1300.20240001.
[10] Shi Z，Xu M，Pan Q，et al.LSTM-based flight trajectory pre- diction[C]//2018 International Joint Conference on Neural Networks（IJCNN）.IEEE，2018 ∶1-8.
