基于跨维度特征融合的航空发动机寿命预测

doi:10.3785/j.issn.1008-973X.2025.07.018

浙江大学学报(工学版)

2025, Vol. 59

Issue (7): 1504-1513 DOI: 10.3785/j.issn.1008-973X.2025.07.018

机械与能源工程

基于跨维度特征融合的航空发动机寿命预测

章东平1(

),王大为1,何数技1,汤斯亮2,刘志勇3,4,刘中秋5

1. 中国计量大学信息工程学院，浙江杭州 310018
2. 浙江大学计算机科学与技术学院，浙江杭州 310058
3. 浙江大学工程师学院，浙江杭州 310015
4. 浙江中控技术股份有限公司，浙江杭州 310053
5. 浙江中正智能科技有限公司，浙江杭州 310052

Remaining useful life prediction of aircraft engines based on cross-dimensional feature fusion

Dongping ZHANG1(

),Dawei WANG1,Shuji HE1,Siliang TANG2,Zhiyong LIU3,4,Zhongqiu LIU5

1. College of Information Engineering, China Jiliang University, Hangzhou 310018, China
2. College of Computer Science and Technology, Zhejiang University, Hangzhou 310058, China
3. Polytechnic Institute of Zhejiang University, Hangzhou 310015, China
4. Supcon Technology Limited Company, Hangzhou 310053, China
5. Miaxis Biometrics Limited Company, Hangzhou 310052, China

全文: PDF(1410 KB) HTML

摘要：

针对多传感器变量数据下变工况航空发动机剩余使用寿命（RUL）预测精度低、泛化能力弱及无法充分挖掘数据特征信息的问题，提出基于跨维度多尺度卷积网络（CDMSCN）的RUL预测方法. 利用通道注意力机制建模不同传感器信号对RUL的交叉影响；构建多尺度门控卷积层，捕捉发动机操作过程中不同时间粒度的潜在故障演变模式；通过通道依赖模块挖掘隐藏的通道特征信息. 通过挖掘并整合发动机传感器数据中的多维特征，有效实现对发动机RUL的动态表征和预测. 在涡扇发动机的公开数据集上开展实验，结果表明，所提方法能够有效提高基于复杂数据集的RUL预测精度.

关键词： 航空发动机; 剩余使用寿命（RUL）; 通道注意力; 多维特征; 深度学习

Abstract:

To address the issues of low prediction accuracy, weak generalization ability, and insufficient feature extraction in remaining useful life (RUL) prediction under varying conditions for aircraft engines with multi-sensor variable data, a method based on cross-dimensional multi-scale convolutional network (CDMSCN) was proposed. The cross-influence of different sensor signals on RUL was modeled using a channel attention mechanism. Multi-scale gated convolutional layers were constructed to capture potential fault evolution patterns at different time granularities during engine operation. Hidden channel feature information was extracted through a channel-dependent module. By mining and integrating multi-dimensional features in the engine sensor data, the dynamic characterization and prediction of engine RUL were effectively achieved. Experiments on a public dataset of turbofan engines were carried out, and results showed that the proposed method significantly improved the prediction accuracy of RUL based on complex datasets.

Key words: aircraft engine remaining useful life (RUL) channel attention multi-dimensional feature deep learning

收稿日期: 2024-05-17 出版日期: 2025-07-25

CLC:

TH 17

基金资助: 浙江省重点研发计划项目（2024C01028，2024C01108，2022C01144）.

作者简介: 章东平（1970—），男，教授，从事深度学习与设备预测性维护研究. orcid.org/0000-0001-6743-8945. E-mail：06a0303103@cjlu.edu.cn

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	章东平
	王大为
	何数技
	汤斯亮
	刘志勇
	刘中秋

引用本文:

章东平,王大为,何数技,汤斯亮,刘志勇,刘中秋. 基于跨维度特征融合的航空发动机寿命预测[J]. 浙江大学学报(工学版), 2025, 59(7): 1504-1513.

Dongping ZHANG,Dawei WANG,Shuji HE,Siliang TANG,Zhiyong LIU,Zhongqiu LIU. Remaining useful life prediction of aircraft engines based on cross-dimensional feature fusion. Journal of ZheJiang University (Engineering Science), 2025, 59(7): 1504-1513.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2025.07.018 或 https://www.zjujournals.com/eng/CN/Y2025/V59/I7/1504

图 1 剩余使用寿命预测框架

图 2 时间序列嵌入模块

图 3 通道注意力模块

图 4 多尺度时间依赖模块

图 5 门控单元

图 6 航空发动机示意图

表 1 商用模块化航空推进系统仿真数据集的主要参数

表 2 商用模块化航空推进系统仿真数据集中的传感器信息

图 7 剩余使用寿命的分段线性退化函数

表 3 跨维度多尺度卷积网络的超参数

图 8 时间窗口大小对预测方法性能的影响

图 9 通道维度大小对预测方法性能的影响

图 10 所提方法在4个子数据集中的剩余使用寿命预测结果

表 4 不同预测方法在4个子数据集中的评价指标对比结果

表 5 跨维度多尺度卷积网络及其变体

表 6 跨维度多尺度卷积网络及其变体的性能对比

图 11 跨维度多尺度卷积网络及其变体的剩余使用寿命预测结果

图 12 不同退化阶段的注意力分数可视化

1	DE PATER I, REIJNS A, MITICI M Alarm-based predictive maintenance scheduling for aircraft engines with imperfect remaining useful life prognostics[J]. Reliability Engineering and System Safety, 2022, 221: 108341 doi: 10.1016/j.ress.2022.108341
2	郭俊锋, 刘国华, 刘国伟基于长序列的航空发动机剩余使用寿命预测方法[J]. 北京航空航天大学学报, 2024, 50 (3): 774- 784 GUO Junfeng, LIU Guohua, LIU Guowei Prediction method of remaining useful life of aero-engine based on long sequence[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024, 50 (3): 774- 784
3	PRAKASH O, SAMANTARAY A K Prognosis of dynamical system components with varying degradation patterns using model–data–fusion[J]. Reliability Engineering and System Safety, 2021, 213: 107683 doi: 10.1016/j.ress.2021.107683
4	XIANG S, QIN Y, LUO J, et al Multicellular LSTM-based deep learning model for aero-engine remaining useful life prediction[J]. Reliability Engineering and System Safety, 2021, 216: 107927 doi: 10.1016/j.ress.2021.107927
5	GE R, ZHAI Q, WANG H, et al Wiener degradation models with scale-mixture normal distributed measurement errors for RUL prediction[J]. Mechanical Systems and Signal Processing, 2022, 173: 109029 doi: 10.1016/j.ymssp.2022.109029
6	LIU K, ZOU T J, XIN M C, et al RUL prediction based on two-phase Wiener process[J]. Quality and Reliability Engineering International, 2022, 38 (7): 3829- 3843 doi: 10.1002/qre.3177
7	WANG H, NI G, CHEN J, et al Research on rolling bearing state health monitoring and life prediction based on PCA and Internet of Things with multi-sensor[J]. Measurement, 2020, 157: 107657 doi: 10.1016/j.measurement.2020.107657
8	NGUYEN K T P, MEDJAHER K, GOGU C Probabilistic deep learning methodology for uncertainty quantification of remaining useful lifetime of multi-component systems[J]. Reliability Engineering and System Safety, 2022, 222: 108383 doi: 10.1016/j.ress.2022.108383
9	HUANG C G, YIN X, HUANG H Z, et al An enhanced deep learning-based fusion prognostic method for RUL prediction[J]. IEEE Transactions on Reliability, 2020, 69 (3): 1097- 1109 doi: 10.1109/TR.2019.2948705
10	LIU X, LEI Y, LI N, et al RUL prediction of machinery using convolutional-vector fusion network through multi-feature dynamic weighting[J]. Mechanical Systems and Signal Processing, 2023, 185: 109788 doi: 10.1016/j.ymssp.2022.109788
11	HONG S, QIN C, LAI X, et al State-of-health estimation and remaining useful life prediction for lithium-ion batteries based on an improved particle filter algorithm[J]. Journal of Energy Storage, 2023, 64: 107179 doi: 10.1016/j.est.2023.107179
12	XU X, TANG S, YU C, et al Remaining useful life prediction of lithium-ion batteries based on Wiener process under time-varying temperature condition[J]. Reliability Engineering and System Safety, 2021, 214: 107675 doi: 10.1016/j.ress.2021.107675
13	CHEN Y, LIU D, DING X, et al Variational encoding based on factorized temporal-channel fusion and feature fusion for interpretable remaining useful life prediction[J]. Advanced Engineering Informatics, 2024, 59: 102316 doi: 10.1016/j.aei.2023.102316
14	莫仁鹏, 司小胜, 李天梅, 等基于多尺度特征与注意力机制的轴承寿命预测[J]. 浙江大学学报: 工学版, 2022, 56 (7): 1447- 1456 MO Renpeng, SI Xiaosheng, LI Tianmei, et al Bearing life prediction based on multi-scale features and attention mechanism[J]. Journal of Zhejiang University: Engineering Science, 2022, 56 (7): 1447- 1456
15	ZHANG Q, LIU Q, YE Q An attention-based temporal convolutional network method for predicting remaining useful life of aero-engine[J]. Engineering Applications of Artificial Intelligence, 2024, 127: 107241 doi: 10.1016/j.engappai.2023.107241
16	JIN R, ZHOU D, WU M, et al An adaptive and dynamical neural network for machine remaining useful life prediction[J]. IEEE Transactions on Industrial Informatics, 2024, 20 (2): 1093- 1102 doi: 10.1109/TII.2023.3254656
17	LIU L, SONG X, ZHOU Z Aircraft engine remaining useful life estimation via a double attention-based data-driven architecture[J]. Reliability Engineering and System Safety, 2022, 221: 108330 doi: 10.1016/j.ress.2022.108330
18	LI H, ZHAO W, ZHANG Y, et al Remaining useful life prediction using multi-scale deep convolutional neural network[J]. Applied Soft Computing, 2020, 89: 106113 doi: 10.1016/j.asoc.2020.106113
19	ZHU J, MA J, WU J A regularized constrained two-stream convolution augmented Transformer for aircraft engine remaining useful life prediction[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108161 doi: 10.1016/j.engappai.2024.108161
20	ZHANG Y, SU C, WU J, et al Trend-augmented and temporal-featured Transformer network with multi-sensor signals for remaining useful life prediction[J]. Reliability Engineering and System Safety, 2024, 241: 109662 doi: 10.1016/j.ress.2023.109662
21	LIN L, WU J, FU S, et al Channel attention & temporal attention based temporal convolutional network: a dual attention framework for remaining useful life prediction of the aircraft engines[J]. Advanced Engineering Informatics, 2024, 60: 102372 doi: 10.1016/j.aei.2024.102372
22	袁烨, 黄虹, 程骋, 等基于特征注意力机制的GRU-GAN航空发动机剩余寿命预测[J]. 中国科学: 技术科学, 2022, 52 (1): 198- 212 YUAN Ye, HUANG Hong, CHENG Cheng, et al Remaining useful life prediction of the aircraft engine based on the GRU-GAN network with a feature attention mechanism[J]. Scientia Sinica: Technologica, 2022, 52 (1): 198- 212 doi: 10.1360/SST-2021-0434
23	SONG Y, GAO S, LI Y, et al Distributed attention-based temporal convolutional network for remaining useful life prediction[J]. IEEE Internet of Things Journal, 2021, 8 (12): 9594- 9602 doi: 10.1109/JIOT.2020.3004452
24	CHOLLET F. Xception: deep learning with depthwise separable convolutions [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1800–1807.
25	MO Y, WU Q, LI X, et al Remaining useful life estimation via transformer encoder enhanced by a gated convolutional unit[J]. Journal of Intelligent Manufacturing, 2021, 32 (7): 1997- 2006 doi: 10.1007/s10845-021-01750-x
26	XU Q, CHEN Z, WU K, et al KDnet-RUL: a knowledge distillation framework to compress deep neural networks for machine remaining useful life prediction[J]. IEEE Transactions on Industrial Electronics, 2022, 69 (2): 2022- 2032 doi: 10.1109/TIE.2021.3057030
27	HUANG C G, HUANG H Z, LI Y F A bidirectional LSTM prognostics method under multiple operational conditions[J]. IEEE Transactions on Industrial Electronics, 2019, 66 (11): 8792- 8802 doi: 10.1109/TIE.2019.2891463
28	CHEN W, LIU C, CHEN Q, et al Multi-scale memory-enhanced method for predicting the remaining useful life of aircraft engines[J]. Neural Computing and Applications, 2023, 35 (3): 2225- 2241 doi: 10.1007/s00521-022-07378-z

[1]	王圣举,张赞. 基于加速扩散模型的缺失值插补算法[J]. 浙江大学学报(工学版), 2025, 59(7): 1471-1480.
[2]	蔡永青,韩成,权巍,陈兀迪. 基于注意力机制的视觉诱导晕动症评估模型[J]. 浙江大学学报(工学版), 2025, 59(6): 1110-1118.
[3]	王立红,刘新倩,李静,冯志全. 基于联邦学习和时空特征融合的网络入侵检测方法[J]. 浙江大学学报(工学版), 2025, 59(6): 1201-1210.
[4]	徐慧智,王秀青. 基于车辆图像特征的前车距离与速度感知[J]. 浙江大学学报(工学版), 2025, 59(6): 1219-1232.
[5]	陈赞,李冉,冯远静,李永强. 基于时间维超分辨率的视频快照压缩成像重构[J]. 浙江大学学报(工学版), 2025, 59(5): 956-963.
[6]	马莉,王永顺,胡瑶,范磊. 预训练长短时空交错Transformer在交通流预测中的应用[J]. 浙江大学学报(工学版), 2025, 59(4): 669-678.
[7]	陈巧红,郭孟浩,方贤,孙麒. 基于跨模态级联扩散模型的图像描述方法[J]. 浙江大学学报(工学版), 2025, 59(4): 787-794.
[8]	顾正宇,赖菲菲,耿辰,王希明,戴亚康. 基于知识引导的缺血性脑卒中梗死区分割方法[J]. 浙江大学学报(工学版), 2025, 59(4): 814-820.
[9]	姚明辉,王悦燕,吴启亮,牛燕,王聪. 基于小样本人体运动行为识别的孪生网络算法[J]. 浙江大学学报(工学版), 2025, 59(3): 504-511.
[10]	梁礼明,龙鹏威,金家新,李仁杰,曾璐. 基于改进YOLOv8s的钢材表面缺陷检测算法[J]. 浙江大学学报(工学版), 2025, 59(3): 512-522.
[11]	杨凯博,钟铭恩,谭佳威,邓智颖,周梦丽,肖子佶. 基于半监督学习的多场景火灾小规模稀薄烟雾检测[J]. 浙江大学学报(工学版), 2025, 59(3): 546-556.
[12]	陈智超,杨杰,李凡,冯志成. 基于深度学习的列车运行环境感知关键算法研究综述[J]. 浙江大学学报(工学版), 2025, 59(1): 1-17.
[13]	刘登峰,陈世海,郭文静,柴志雷. 基于轻量残差网络的高效半色调算法[J]. 浙江大学学报(工学版), 2025, 59(1): 62-69.
[14]	赵顗,安醇,李铭浩,马健霄,怀硕. 城市快速路互通交织区车辆的换道持续距离选择[J]. 浙江大学学报(工学版), 2025, 59(1): 205-212.
[15]	李凡,杨杰,冯志成,陈智超,付云骁. 基于图像识别的弓网接触点检测方法[J]. 浙江大学学报(工学版), 2024, 58(9): 1801-1810.

Viewed

Full text

Abstract

Cited

Shared

Discussed