基于多尺度自适应注意力网络的剩余寿命预测

doi:10.7527/S1000-6893.2022.26918

Abstract

Abstract:

Remaining useful life prediction is the key to and premise of intelligent maintenance decision-making of complex equipment systems. The potential relationship among different index data brings challenges to prediction accuracy， while parameter selection also increases model prediction difficulty. We use the multi-scale adaptive attention network method to fuse the feature relationship among data from vertical and horizontal dimensions， respectively， and give the piecewise nonlinear target function to improve the prediction accuracy and reduce the root mean square error. The adaptive mechanism is employed to automatically select the size of convolution kernel， improving the efficiency of network training. Empirical analysis of Company-Modular aero-propulsion system simulation data sets prove the effectiveness of this method in remaining useful life prediction of complex systems.

Key words: remaining useful life, multi-scale learning, adaptive attention, nonlinear target function, complex systems

CLC Number:

V235.13

Bin LIU, Jing XU, Meiling HUO, Xueying CUI, Xiufeng XIE, Donghui YANG, Jia WANG. Remaining useful life prediction based on multi-scale adaptive attention network[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(5): 226918.

Figures/Tables 19

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

Table 2

Table 3

Table 4

Fig.5

Fig.6

Fig.7

Table 5

Table 6

Fig.8

Fig.9

Fig.10

Fig.11

Table 7

Table 8

References 24

1	沈保明，陈保家，赵春华，等. 深度学习在机械设备故障预测与健康管理中的研究综述［J］. 机床与液压， 2021， 49（19）： 162-171.
	SHEN B M， CHEN B J， ZHAO C H， et al. Review on the research of deep learning in mechanical equipment fault prognostics and health management［J］. Machine Tool & Hydraulics， 2021， 49（19）： 162-171 （in Chinese）.
2	许艳雷，邱明，李军星，等. 基于SKF-KF-Bayes的滚动轴承剩余使用寿命预测方法［J］. 振动与冲击， 2021， 40（19）： 26-31， 40.
	XU Y L， QIU M， LI J X， et al. Remaining useful life prediction method of rolling bearing based on SKF-KF-Bayes［J］. Journal of Vibration and Shock， 2021， 40（19）： 26-31， 40 （in Chinese）.
3	DAI Y， CHENG S， GAN Q J， et al. Life prediction of Ni-Cd battery based on linear Wiener process［J］. Journal of Central South University， 2021， 28（9）： 2919-2930.
4	WANG Y D， ZHAO Y F， ADDEPALLI S. Remaining useful life prediction using deep learning approaches： A review［J］. Procedia Manufacturing， 2020， 49： 81-88.
5	SATEESH BABU G， ZHAO P L， LI X L. Deep convolutional neural network based regression approach for estimation of remaining useful life［J］. International Conference on Database Systems for Advanced Applications， 2016， 9642： 214-228.
6	LI X， DING Q， SUN J Q. Remaining useful life estimation in prognostics using deep convolution neural networks［J］. Reliability Engineering & System Safety， 2018， 172： 1-11.
7	SAXENA A， GOEBEL K， SIMON D， et al. Damage propagation modeling for aircraft engine run-to-failure simulation［C］∥ 2008 International Conference on Prognostics and Health Management. Piscataway： IEEE Press， 2008： 10423504.
8	SZEGEDY C， LIU W， JIA Y Q， et al. Going deeper with convolutions［C］∥ 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE Press， 2015： 1-9.
9	LI H. Remaining useful life prediction using multi-scale deep convolutional neural network［J］. Applied Soft Computing， 2020， 89： 106113.
10	LIM P， GOH C K， TAN K C， et al. Estimation of remaining useful life based on switching Kalman filter neural network ensemble［J］. Annual Conference of the Prognostics and Health Management Society， 2014， 6（1）： 1-9.
11	ZHENG S， RISTOVSKI K， FARAHAT A， et al. Long short-term memory network for remaining useful life estimation［C］∥ 2017 IEEE International Conference on Prognostics and Health Management. Piscataway： IEEE Press， 2017： 88-95.
12	GUO L， LI N， JIA F， et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings［J］. Neurocomputing， 2017， 240： 98-109.
13	HOCHREITER S， SCHMIDHUBER J. Long short-term memory［J］. Neural Computation， 1997， 9（8）： 1735-1780.
14	CHO K， VAN MERRIENBOER B， GULCEHRE C， et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation［DB/OL］. arXiv preprint： 1406.1078， 2014.
15	CHEN J L， JING HJ， CHANG Y H， et al. Gated recurrent unit based recurrent neural network for remaining useful life prediction of nonlinear deterioration process［J］. Reliability Engineering & System Safety， 2019， 185： 372-382.
16	MA M， MAO Z. Deep-convolution-based LSTM network for remaining useful life prediction［J］. IEEE Transactions on Industrial Informatics， 2021， 17（3）： 1658-1667.
17	刘畅，陈雯柏. 一种基于MSDCNN-LSTM的设备RUL预测方法［J］. 西北工业大学学报， 2021， 39（2）： 407-413.
	LIU C， CHEN W B. A RUL prediction method of equipments based on MSDCNN-LSTM［J］. Journal of Northwestern Polytechnical University， 2021， 39（2）： 407-413 （in Chinese）.
18	KHAN S， YAIRI T. A review on the application of deep learning in system health management［J］. Mechanical Systems and Signal Processing， 2018， 107： 241-265.
19	PILLAI S， VADAKKEPAT P. Two stage deep learning for prognostics using multi-loss encoder and convolutional composite features［J］. Expert Systems with Applications， 2021， 171： 114569.
20	HEIMES F O. Recurrent neural networks for remaining useful life estimation［C］∥ 2008 International Conference on Prognostics and Health Management. Piscataway： IEEE Press， 2008： 10423512.
21	ZHU H G， ZENG H， LIU J， et al. Logish： A new nonlinear nonmonotonic activation function for convolutional neural network［J］. Neurocomputing， 2021， 458： 490-499.
22	WANG Q L， WU B G， ZHU P F， et al. ECA-net： Efficient channel attention for deep convolutional neural networks［C］∥ 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2020： 11531-11539.
23	KINGMA D P， BA J L. Adam： A method for stochastic optimization［DB/OL］. arXiv preprint： 1412.6980， 2014.
24	WANG B， LEI Y， LI N， et al. Deep separable convolutional network for remaining useful life prediction of machinery［J］. Mechanical Systems and Signal Processing， 2019， 134： 106330.

数据集	数量
数据集	FD001	FD002	FD003	FD004
训练集的发动机	100	260	100	249
测试集的发动机	100	259	100	248
总训练样本	20 631	53 579	24 720	61 249
总测试样本	13 096	33 991	16 596	41 214
失效模式	1	1	2	2
运行条件	1	6	1	6

数据集	K个传感器数据	抛弃指标的索引
FD001	14	2， 3， 4， 5， 9， 10， 14， 20， 22， 23
FD002	15	2， 3， 4， 5， 9， 10， 20， 22， 23
FD003	21	2， 3， 4
FD004	21	2， 3， 4

数据集	输入样本大小	训练样本数	验证样本数
FD001	40×14	11 711	5 020
FD002	40×15	30 533	13 086
FD003	50×21	15 273	6 547
FD004	50×21	34 333	14 715

参数	数值
3条支路上卷积层的核大小	10， 15， 20
扩张卷积的卷积核大小	2
扩张卷积的扩张率	2
两个全连接层卷积滤波器的个数	80， 60
时间窗口大小	40， 40， 50， 50
3条支路的卷积核个数	10， 10， 10
Dropout率	0.2

方法	RMSE
方法	FD001	FD002	FD003	FD004
CNN^［5］	18.45	30.29	19.82	29.16
DLSTM^［11］	16.14	24.49	16.18	28.17
DCNN^［6］	12.61	22.36	12.64	23.31
DSCN^［24］	10.95	20.47	10.62	22.64
MSCNN^［9］	11.44	19.35	11.67	22.22
MLE+CCF^［19］	11.57	18.84	11.83	20.78
MAAN	10.80	14.79	10.93	16.78

Remaining useful life prediction based on multi-scale adaptive attention network

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 19

References 24

Related Articles 15

Recommended Articles

Metrics

Comments

数据集	RMSE
数据集	线性	非线性
FD001	13.20±0.30	10.80±0.15
FD002	17.38±0.83	14.79±0.36
FD003	13.63±0.54	10.93±0.44
FD004	19.63±0.60	16.78±0.80

[1]	Guixian QU, Dongyang LIU, Xu YANG, Tian QIU, Chuankai LIU, Shuiting DING, Shuzheng YUAN, Kan GUO. Remaining useful life prediction method based on temporal information enhancement of sensors [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(17): 231634-231634.
[2]	Jiayu CHEN, Qinhua LU, Xuhang WANG, Zhilong SHI, Hongjuan GE, Min XIE. Intelligent health evaluation and degradation warning of aircraft air-conditioning systems: A method fusing mission profile and PCA-MD [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(15): 231604-231604.
[3]	Biyao QIANG, Kaining SHI, Junxue REN, Yaoyao SHI. Instance transfer for tool remaining useful life prediction cross working conditions [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(13): 629038-629038.
[4]	Tianmei LI, Xiaosheng SI, Jianxun ZHANG. Data-model interactive remaining useful life prediction method for multi-sensor monitored linear stochastic degrading devices [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(8): 227190-227190.
[5]	Chao REN, Huiqin LI, Tianmei LI, Jianxun ZHANG, Xiaosheng SI. Equipment remaining useful life prediction method with dynamic calibration of degradation model [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(19): 228345-228345.
[6]	Jiaxin YANG, Shengjin TANG, Liang LI, Xiaoyan SUN, Shuai QI, Xiaosheng SI. Remaining useful life prediction of implicit nonlinear Wiener degradation process based on multi-source information [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(12): 227662-227662.
[7]	Baokui YANG, Jianxun ZHANG, Huiqin LI, Xiaosheng SI. Remaining useful life prediction method for nonlinear degrading equipment based on Box-Cox transformation and random coefficient regression model [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(11): 227660-227660.
[8]	DONG Qing, ZHENG Jianfei, HU Changhua, YU Tonghui, MU Hanxiao. Remaining useful life prediction for adaptive Wiener process method with random shock [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(9): 225914-225914.
[9]	ZHANG Shengfei, LI Tianmei, HU Changhua, DU Dangbo, SI Xiaosheng. Missing data generation method and its application in remaining useful life prediction based on deep convolutional generative adversarial network [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(8): 225708-225708.
[10]	MU Hanxiao, ZHENG Jianfei, HU Changhua, ZHAO Ruixing, DONG Qing. Remaining useful life prediction of multivariate degradation equipment based on CDBN and BiLSTM [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(7): 325403-325403.
[11]	XUE Xiaofeng, TIAN Jing, HE Shuming, FENG Yunwen. Nonlinear degradation assessment of aircraft components monitored by multi-sensors [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(5): 524342-524342.
[12]	REN Ziqiang, SI Xiaosheng, HU Changhua, WANG Xi. Remaining useful life prediction method for engine combining multi-sensors data [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2019, 40(12): 223312-223312.
[13]	ZHANG Jianxun, HU Changhua, ZHOU Zhijie, SI Xiaosheng, DU Dangbo. Multiple Degradation Variables Modeling for Remaining Useful Life Estimation of Gyros Based on Copula Function [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2014, 35(4): 1111-1121.
[14]	XU Yuguo, QIU Jing, LIU Guanjun. Joint Optimization of Maintenance and Inventory in Equipment Autonomic Logistics [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2013, 34(8): 1864-1873.
[15]	SHANG Yongshuang, LI Wenhai, LIU Changjie, SHENG Pei. Prediction of Remaining Useful Life for Equipment with Partially Observed Information [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2012, (5): 848-854.

方法	评分函数值
方法	FD001	FD002	FD003	FD004
CNN^［5］	1 286.7	13 570	1 596.2	7 886.4
DLSTM^［11］	338	4 450	852	5 550
DCNN^［6］	273.7	10 412	284.1	12 466
DSCN^［24］	260.67	4 367.56	246.55	5 168.64
MSCNN^［9］	196.22	3 747	241.89	4 844
MLE+CCF^［19］	208	1 606	262	2 081
MAAN	183.77	901.71	214.17	1 189.11

数据集	评分函数值
数据集	线性	非线性
FD001	303.13±18.98	183.77±7.54
FD002	1 708.45±474.15	901.71±122.36
FD003	380.06±87.52	214.17±26.65
FD004	1 854.25±415.17	1 189.11±107.43