Predicting the Remaining Useful Life of Non-ideal Data Through a Joint Framework of Sequential Contrastive Learning and Attention Mechanisms

doi:10.3901/JME.2025.18.027

Abstract

Abstract: In the context of monitoring big data, the acquisition of ideal data is often constrained, resulting in a scarcity of high-value data. Consequently, traditional intelligent prediction models are characterized by weak generalizability and low accuracy. To address this issue, a sequential contrastive attention network (SCAN) is proposed to effectively utilize fragmentary, non-ideal data without complete lifespan labels for predicting the remaining useful life of equipment. Initially, a sequential contrastive encoder is constructed to automatically annotate the deep degradation features of non-ideal data without requiring lifespan labels, enhancing the temporal autocorrelation of the extracted features. Subsequently, a lifespan prediction attention decoder is designed to dynamically allocate weights to the encoded degradation features and perform parallel computation fitting. On this basis, an encoding-decoding training optimization mode based on elastic weight sharing is introduced to achieve an organic combination and efficient interaction between the encoder and decoder. The effectiveness of the proposed method is verified using full lifespan test instances of bearings and urban rail transit train wheels. Results demonstrate that the proposed method achieves superior prediction accuracy compared to other approaches, with an average accuracy improvement of approximately 34% in both application scenarios.

Key words: remaining useful life prediction, sequential contrastive learning, attention mechanism, bearings, urban rail transit train wheelset

CLC Number:

TG156

LIN Tianjiao, SONG Liuyang, CUI Lingli, WANG Huaqing. Predicting the Remaining Useful Life of Non-ideal Data Through a Joint Framework of Sequential Contrastive Learning and Attention Mechanisms[J]. Journal of Mechanical Engineering, 2025, 61(18): 27-37.

References

[1] GUO Wei，LI Xiang，WAN Xuecheng. A novel approach to bearing prognostics based on impulse-driven measures，improved morphological filter and practical health indicator construction[J]. Reliability Engineering & System Safety，2023，238：109451.
[2] 李乃鹏，蔡潇，雷亚国，等. 一种融合多传感器数据的数模联动机械剩余寿命预测方法[J]. 机械工程学报，2021，57(20)：29-37，46. LI Naipeng，CAI Xiao，LEI Yaguo，et al. A model-data-fusion remaining useful life prediction method withmulti-sensor fusion for machinery[J]. Journal of Mechanical Engineering，2021，57(20)：29-37，46.
[3] HE Rui，TIAN Zhigang，ZUO Mingjian. A transferable neural network method for remaining useful life prediction[J]. Mechanical Systems and Signal Processing，2023，183：109608.
[4] 庞哲楠，裴洪，李天梅，等. 考虑不完美维修的随机退化设备剩余寿命自适应预测方法[J]. 机械工程学报，2023，59(2)：14-29. PANG Zhenan，PEI Hong，LI Tianmei，et al. Remaining useful lifetime prognostic approach for stochastic degradation equipment considering imperfect maintenance activities[J]. Journal of Mechanical Engineering，2023，59(2)：14-29.
[5] DING Yifei，DING Peng，JIA Minping. A novel remaining useful life prediction method of rolling bearings based on deep transfer auto-encoder[J]. IEEE Transactions on Instrumentation and Measurement，2021，70：1-12.
[6] ZENG Lingquan，ZHENG Junhua，YAO Le，et al. Dynamic bayesian networks for feature learning and transfer applications in remaining useful life estimation[J]. IEEE Transactions on Instrumentation and Measurement，2023，72：1-12.
[7] XIA Pengcheng，HUANG Yixiang，LI Peng，et al. Fault knowledge transfer assisted ensemble method for remaining useful life prediction[J]. IEEE Transactions on Industrial Informatics，2022，18(3)：1758-1769.
[8] NGUYEN K T P，MEDJAHER K，TRAN D T. A review of artificial intelligence methods for engineering prognostics and health management with implementation guidelines[J]. Artificial Intelligence Review，2023，56(4)：3659-3709.
[9] 沈天浩，丁康，黎杰，等. 图结构联合时序数据驱动的装备剩余使用寿命预测方法[J]. 机械工程学报，2023，59(12)：183-194. SHEN Tianhao，DING Kang，LI Jie，et al. Graph structure and temporal data driven remaining useful life prediction method for machinery[J]. Journal of Mechanical Engineering，2023，59(12)：183-194.
[10] 王宇，刘秋发，彭一真. 非均匀监测条件下滚动轴承剩余寿命预测方法[J]. 机械工程学报，2023，59(23)：96-104. WANG Yu，LIU Qiufa，PENG Yizhen. A remaining useful life prediction approach with nonuniform monitoring conditions for rolling bearings [J]. Journal of Mechanical Engineering，2023，59(23)：96-104.
[11] VOS K，PENG Zhongxiao，JENKINS C，et al. Vibration- based anomaly detection using LSTM-SVM approaches[J]. Mechanical Systems and Signal Processing，2022，169：108752.
[12] 林亮行，马国政，孙建芳，等. 基于VMD-EEMD-LSTM的涂层型关节轴承剩余使用寿命预测方法[J]. 机械工程学报，2023，59(9)：125-136. LIN Liangxing，MA Guozheng，SUN Jianfang，et al. Remaining useful life prediction method of coated spherical plain bearing based on VMD-EEMD-LSTM [J]. Journal of Mechanical Engineering，2023，59(9)：125-136.
[13] SUN Shuguang，LIU Jinfa，WANG Jingqin，et al. Remaining useful life prediction for ac contactor based on MMPE and LSTM with dual attention mechanism[J]. IEEE Transactions on Instrumentation and Measurement，2022，71：1-13.
[14] 杨超颖，刘颉，周凯波. 基于路图注意力网络的轴承剩余寿命预测方法[J]. 机械工程学报，2023，59(12)：195-201. YANG Chaoying，LIU Jie，ZHOU Kaibo. Path graph attention network-based bearing remaining useful life prediction method[J]. Journal of Mechanical Engineering，2023，59(12)：195-201.
[15] CHEN Jiaxian，HUANG Ruyi，CHEN Zhuyun，et al. Transfer learning algorithms for bearing remaining useful life prediction：A comprehensive review from an industrial application perspective[J]. Mechanical Systems and Signal Processing，2023，193：110239.
[16] 刘华开，丁康，何国林，等. 联合故障机理和卷积神经网络的齿轮剩余使用寿命预测方法研究[J]. 机械工程学报，2024，60(12)：116-125. LIU Huakai，DING Kang，HE Guolin，et al. Research on the prediction method of remaining useful life of gears by combining fault mechanism and convolutional neural network[J]. Journal of Mechanical Engineering，2024，60(12)：116-125.
[17] CAO Yudong，DING Yifei，JIA Minping，et al. A novel temporal convolutional network with residual self- attention mechanism for remaining useful life prediction of rolling bearings[J]. Reliability Engineering and System Safety，2021，215：107813.
[18] SHI Zunya，CHEHADE A. A dual-LSTM framework combining change point detection and remaining useful life prediction[J]. Reliability Engineering and System Safety，2021，205：107257.
[19] CAI Li，YIN Hongpeng，LIN Jingdong，et al. An update-strategy-based gaussian process regression method for aeroengines fault prediction[J]. IEEE Transactions on Industrial Informatics，2024，20(2)：1941-1951.
[20] MEI Yingjie，LIU Yongmeng，LI Jiahao，et al. Digital twin of large-scale coaxiality measuring instrument with six dimensions：Realizing the unification of aeroengine rotors measurement and assembly[J]. IEEE Transactions on Industrial Informatics，2024，20(3)：4504-4516.
[21] MAO Wentao，LIU Keying，ZHANG Yanna，et al. Self-supervised deep tensor domain-adversarial regression adaptation for online remaining useful life prediction across machines[J]. IEEE Transactions on Instrumentation and Measurement，2023，72：1-16.
[22] WANG Yilin，SHEN Lei，ZHANG Yuxuan，et al. Self-supervised health representation decomposition based on contrast learning[J]. Reliability Engineering and System Safety，2023，239：109455.
[23] MAO Wentao，CHRN Jiaxian，LIU Jing，et al. Self- supervised deep domain-adversarial regression adaptation for online remaining useful life prediction of rolling bearing under unknown working condition[J]. IEEE Transactions on Industrial Informatics，2023，19(2)：27-37.
[24] LI Yunfeng，LI Ming，YAN Zhong，et al. Application of life cycle of aeroengine mainshaft bearing based on digital twin[J]. Processes，2023，11(6)：1768.
[25] 张菀，张泰瑀，贾民平，等. 多传感器融合和MHA-LSTM的电机轴承剩余寿命预测[J]. 仪器仪表学报，2024，45(3)：84-93. ZHANG Wan，ZHANG Taiyu，JIA Minping，et al. Prediction of residual useful life of motor bearings based on multi-sensor fusion and MHA-LSTM[J]. Chinese Journal of Scientifitic Instrument，2024：45(3)：84-93.
[26] HARRER S. Attention is not all you need：The complicated case of ethically using large language models in healthcare and medicine[J]. Ebiomedicine，2023，90：104512.
[27] ZHANG Jiusi，LI Xiang，TIAN Jilun，et al. An integrated multi-head dual sparse self-attention network for remaining useful life prediction[J]. Reliability Engineering & System Safety，2023，233：109096.
[28] HAO Wei，LI Zhixuan，QIN Guohao，et al. A novel prediction method based on bi-channel hierarchical vision transformer for rolling bearings’ remaining useful life[J]. Processes，2023，11(4)：1153.
[29] JIANG Li，ZHANG Tianao，LEI Wei，et al. A new convolutional dual-channel Transformer network with time window concatenation for remaining useful life prediction of rolling bearings[J]. Advanced Engineering Informatics，2023，56：101966.
[30] LI Naipeng，LEI Yaguo，LIN Jing，et al. An improved exponential model for predicting remaining useful life of rolling element bearings[J]. IEEE Transactions on Industrial Electronics，2015，62(12)：7762-7773.
[31] WANG Biao，LEI Yaguo，LI Naipeng，et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings[J]. IEEE Transactions on Reliability，2020，69(1)：401-412.
[32] SMITH W A，RANDALL R B. Rolling element bearing diagnostics using the case western reserve university data：A benchmark study[J]. Mechanical Systems and Signal Processing，2015，64-65：100-131.
[33] HUANG Chenggeng，HUANG Hongzhong，LI Yanfeng. A bidirectional LSTM prognostics method under multiple operational conditions[J]. IEEE Transactions on Industrial Electronics，2019，66(11)：8792-8802.
[34] HAN Yunlong，LI Conghui，ZHENG Linfeng，et al. Remaining useful life prediction of lithium-ion batteries by using a denoising transformer-based neural network[J]. Energies，2023，16(17)：6328.
[35] ZHU Jun，CHEN Nan，PENG Weiwen. Estimation of bearing remaining useful life based on multiscale convolutional neural network[J]. IEEE Transactions on Industrial Electronics，2019，66(4)：3208-3216.