Prognosticating Remaining Useful Life of Electro-Mechanical Actuators Using a Multi-mode Transformer Model

doi:10.12382/bgxb.2022.0581

Abstract

Abstract:

Electro-mechanical actuators play an important role in next-generation spacecraft. To address the challenge of predicting the remaining useful life of electro-mechanical actuators, a failure prognostic algorithm based on a multi-mode Transformer model is proposed. Multi-channel sensor data are directly used as inputs to the Transformer model without feature extraction as a pre-processing step. The multi-mode Transformer uses multi-head attention to adaptively learn global features from various representation subspaces. The Encoder of the Transformer is utilized to extract features from different sensors across the time series in parallel and predict the remaining useful life directly. Simultaneously, the full Transformer, composed of an Encoder and a Decoder, is used to prognosticate key performance parameters of electro-mechanical actuators. A benchmark dataset is used to validate the effectiveness of the proposed model for electro-mechanical actuator failure prognostication. Experimental results reveal its advantage in accurate prediction and failure prognosis.

Key words: electro-mechanical actuator, failure prognostic, multimode data, attention mechanism

CLC Number:

V474

CHEN Zihan. Prognosticating Remaining Useful Life of Electro-Mechanical Actuators Using a Multi-mode Transformer Model[J]. Acta Armamentarii, 2023, 44(10): 2920-2931.

Figures/Tables 11

Fig.1 Architecture of the Transformer model

Fig.2 Architecture of the multi-mode Transformer model

Fig.3 An improved approach based on Granularity Consistent Attention

Fig.4 FLEA test bed

Fig.5 Temperature of motor housing X and Y with -40 lbs load

Table 1 Experiment conditions

负载类型	幅值/mm	频率/Hz	负载/lbs	最大速度/(m·s^-1)
正弦	80	0.25	-40	0.04

Fig.6 Overview of EMA degradation test data

Fig.7 Prediction of RUL

Table 2 Performance comparison of direct RUL methods

序号	模型	S分数
1	1DCNN	1503.86
2	LSTM	1387.51
3	多模态Transformer	656.69

Fig.8 Prediction of motor temperature

Table 3 Performance comparison of indirect RUL methods

序号	模型名称	MAPE
1	GPR	6.62
2	LSTM	5.01
3	原始Transformer	2.33
4	多模态Transformer	1.45

References 29

[1]	MAZZOLENI M, DI RITO G, PREVIDI F. Electro-mechanical actuators for the more electric aircraft[M]. Cham, Switzerland: Springer, 2021.
[2]	郭忠义, 李永华, 李关辉, 等. 装备系统剩余使用寿命预测技术研究进展[J]. 南京航空航天大学学报, 2022, 54(3): 341-364.
	GUO Z Y, LI Y H, LI G H, et al. Research progress on remaining useful life prediction technology of equipment systems[J]. Journal of Nanjing University of Aeronautics & Astronautics, 2022, 54(3): 341-364. (in Chinese)
[3]	DALLA VEDOVA M D L, GERMANÀ A, BERRI P C, et al. Model-based fault detection and identification for prognostics of electromechanical actuators using genetic algorithms[J]. Aerospace, 2019, 6(9): 94. doi: 10.3390/aerospace6090094 URL
[4]	ARRIOLA D, THIELECKE F. Model-based design and experimental verification of a monitoring concept for an active-active electromechanical aileron actuation system[J]. Mechanical Systems and Signal Processing, 2017, 94: 322-345. doi: 10.1016/j.ymssp.2017.02.039 URL
[5]	OSSMANN D, VAN DER LINDEN F L J. Advanced sensor fault detection and isolation for electro-mechanical flight actuators[C]// Proceedings of 2015 NASA/ESA Conference on Adaptive Hardware and Systems. Montreal, QC, Canada:IEEE, 2015.
[6]	DI RITO G, LUCIANO B, BORGARELLI N, et al. Model-based condition-monitoring and jamming-tolerant control of an electro-mechanical flight actuator with differential ball screws[J]. Actuators, 2021, 10(9): 230. doi: 10.3390/act10090230 URL
[7]	BALABAN E, SAXENA A, NARASIMHAN S, et al. Prognostic health-management system development for electromechanical actuators[J]. Journal of Aerospace Information Systems, 2015, 12(3): 329-344. doi: 10.2514/1.I010171 URL
[8]	BERRI P C, DALLA VEDOVA M D L, MAGGIORE P. A simplified monitor model for EMA prognostics[J]. MATEC Web of Conferences, 2018, 233: 00016. doi: 10.1051/matecconf/201823300016 URL
[9]	ZHANG Y J, LIU L S, PENG Y, et al. An electro-mechanical actuator motor voltage estimation method with a feature-aided Kalman filter[J]. Sensors, 2018, 18(12): 4190. doi: 10.3390/s18124190 URL
[10]	CHIRICO A J III, KOLODZIEJ J R. A data-driven methodology for fault detection in electromechanical actuators[J]. Journal of Dynamic Systems, Measurement, and Control, 2014, 136(4): 041025. doi: 10.1115/1.4026835 URL
[11]	刘俊, 王占林, 付永领, 等. 基于EEMD分解的直驱式机电作动器故障诊断[J]. 北京航空航天大学学报, 2012, 38(12): 1567-1571.
	LIU J, WANG Z L, FU Y L, et al. Fault diagnosis of direct-driven electromechanical actuator based on ensemble empirical mode decomposition[J]. Journal of Beijing University of Aeronautics and Astronautics, 2012, 38(12): 1567-1571. (in Chinese)
[12]	LIU H M, JING J Y, MA J. Fault diagnosis of electromechanical actuator based on VMD multifractal detrended fluctuation analysis and PNN[J]. Complexity, 2018, 2018: 9154682.
[13]	王剑, 王新民, 谢蓉, 等. 基于IMM-UKF方法的机电作动器突发性故障诊断研究[J]. 北京理工大学学报, 2019, 39(2): 198-202, 208.
	WANG J, WANG X M, XIE R, et al. Abrupt fault diagnosis for electro-mechanical actuator based on IMM-UKF[J]. Transactions of Beijing Institute of Technology, 2019, 39(2): 198-202, 208. (in Chinese)
[14]	王剑, 王新民, 谢蓉, 等. 基于DWNN的机电作动器渐变性故障诊断[J]. 北京航空航天大学学报, 2019, 45(9): 1831-1837.
	WANG J, WANG X M, XIE R, et al. Gradual fault diagnosis for electromechanical actuator based on DWNN[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019, 45(9): 1831-1837. (in Chinese)
[15]	李世晓, 杜锦华, 龙云. 基于一维卷积神经网络的机电作动器故障诊断[J]. 电工技术学报, 2022, 37(增刊): 62-73.
	LI S X, DU J H, LONG Y. Fault diagnosis of electromechanical actuators based on one-dimensional convolutional neural network[J]. Transactions of China Electrotechnical Society, 2022, 37(S): 62-73. (in Chinese)
[16]	REDDY K K, SARKAR S, VENUGOPALAN V, et al. Anomaly detection and fault disambiguation in large flight data: a multi-modal deep auto-encoder approach[C]// Proceedings of 2016 Annual Conference of the Prognostics and Health Management Society. Denver, CO, US: Prognostics and Health Management Society, 2016: 207-214.
[17]	YANG J, GUO Y Q, ZHAO W L. Long short-term memory neural network based fault detection and isolation for electro-mechanical actuators[J]. Neurocomputing, 2019, 360: 85-96. doi: 10.1016/j.neucom.2019.06.029 URL
[18]	SIAHPOUR S, LI X, LEE J. Deep learning-based cross-sensor domain adaptation for fault diagnosis of electro-mechanical actuators[J]. International Journal of Dynamics and Control, 2020, 8(4): 1054-1062. doi: 10.1007/s40435-020-00669-0
[19]	BALABAN E, SAXENA A, NARASIMHAN S, et al. Experimental validation of a prognostic health management system for electro-mechanical actuators[C]// Proceedings of AIAA Infotech@Aerospace. St. Louis, MI, US: American Institute of Aeronautics and Astronautics, 2011:AIAA2011-1518.
[20]	ZHANG Y J, LIU D T, YU J X, et al. EMA remaining useful life prediction with weighted bagging GPR algorithm[J]. Microelectronics Reliability, 2017, 75: 253-263. doi: 10.1016/j.microrel.2017.03.021 URL
[21]	ZHANG X Y, TANG L W, CHEN J S. Fault diagnosis for electro-mechanical actuators based on STL-HSTA-GRU and SM[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 3527716.
[22]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, CA, US: Neural Information Processing Systems Foundation, Inc., 2017: 6000-6010.
[23]	张玉杰, 彭宇, 刘大同. 飞机机电系统部件数据驱动健康状态在线估计方法综述[J]. 仪器仪表学报, 2022, 43(6): 118-130.
	ZHANG Y J, PENG Y, LIU D T. Review on data-driven health state on-line estimation methods for aircraft electromechanical system components[J]. Chinese Journal of Scientific Instrument, 2022, 43(6): 118-130. (in Chinese)
[24]	XIONG R, YANG Y, HE D, et al. On layer normalization in the transformer architecture[C]// Proceedings of the 37th International Conference on Machine Learning. Vienna, Austria:PMLR, 2020: 10524-10533.
[25]	HENDRYCKS D, GIMPEL K. Gaussian error linear units (gelus): arXiv:1606.08415[R/OL]. Ithaca, NY, US: Cornell University, 2020(2020-07-08). https://doi.org/10.48550/arXiv.1606.08415.
[26]	LIU F L, REN X C, ZHAO G X, et al. Rethinking and improving natural language generation with layer-wise multi-view decoding: arXiv:2005.08081[R]. Ithaca, NY, US: Cornell University, 2020(2020-08-29). https://arxiv.org/abs/2005.08081.
[27]	邵怡韦, 陈嘉宇, 林翠颖, 等. 小训练样本下齿轮箱故障诊断:一种基于改进深度森林的方法[J]. 航空学报, 2022, 43(8): 118-132.
	SHAO Y W, CHEN J Y, LIN C Y, et al. Gearbox fault diagnosis with small training samples: an improved deep forest based method[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 118-132. (in Chinese)
[28]	李国发, 王彦博, 何佳龙, 等. 机电装备健康状态评估研究进展及发展趋势[J]. 吉林大学学报(工学版), 2022, 52(2): 267-279.
	LI G F, WANG Y B, HE J L, et al. Research progress and development trend of health assessment of electromechanical equipment[J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(2): 267-279. (in Chinese)
[29]	LIU L S, WANG S J, LIU D T, et al. Entropy-based sensor selection for condition monitoring and prognostics of aircraft engine[J]. Microelectronics Reliability, 2015, 55(9/10): 2092-2096. doi: 10.1016/j.microrel.2015.06.076 URL