旋转机械是现代工业生产的重要设备，广泛应用于航天、船舶、能源、化工、交通等行业。随着旋转机械向大型化、复杂化、精密化、智能化方向发展，恶劣的运行环境和复杂的工况等因素增加了其性能退化失效及发生故障的概率。针对旋转机械安全运行和高效维护难题，故障预测与健康管理（Prognostic and Health Management, PHM）技术为旋转机械预测性维护与运维管理提供了更为科学的策略。健康指标构建（Health Indicator, HI）和剩余使用寿命（Remaining Useful Life, RUL）预测作为PHM的重要组成部分，可以准确掌握旋转机械运行状态、发现其潜在故障，并为主动采取应对措施提供依据。因此，开展旋转机械HI构建和RUL预测的共性理论和方法研究，对于提高旋转机械运行的可靠性具有重大意义。本文针对工业复杂场景下旋转机械HI构建难度大及综合性能差、RUL预测准确度低等问题，基于深度学习理论和技术，开展了深入的研究，完成的主要工作如下：

（1）针对多点监测条件下退化特征难以完备、有效表达，导致HI无监督构建难度大的问题，深入研究了多域混合特征时空融合的旋转机械HI无监督构建方法。首先，研究基于多域混合特征与加权融合评价准则的多点监测振动信号特征提取与选择方法，实现退化特征的准确表达，提高退化特征的完备性及有效性。其次，建立集成自注意力机制、长短期记忆神经网络、卷积自编码器的无监督HI，实现了无标签样本情况下兼具多时空退化信息的HI模型构建。最后，采用Savitzky-Golay滤波方法提高HI的平滑性。以标准轴承退化数据集和所搭建的减速器全寿命周期退化实验平台数据集为例，验证了所提方法，结果表明该方法实现了旋转机械退化过程的有效表征，在两个数据集上早期状态退化点发现时间相比于其它方法分别平均提前了8.6min和22.9min。

（2）针对时空关系非完备条件下样本数据的退化特征难以提取，导致HI性能差的问题，深入研究了深层时空退化特征自提取的旋转机械HI自动构建方法。首先，利用参数自寻优变分模态分解将降噪后的振动信号分解为多个子模态分量并将模态分量进行傅里叶变换，将振动信号、振动频域信号、振动信号子模态及其频域信号耦合组成二维矩阵输入网络，提高样本表达能力。其次，构建融合多尺度卷积、卷积长短期记忆网络、注意力机制的卷积自编码器，用于自动提取样本深层时空退化特征以提高HI的非线性量化能力。最后，引入一个二次函数约束损失项来完成HI的形状约束，提高HI的整体性能。以标准轴承退化数据集和所搭建的减速器全寿命周期退化实验平台数据集为例，验证了所提方法，结果表明该方法构建的HI平均综合得分在两个数据集上相比于其它方法分别平均提升了0.1542和0.2341，同时在早期故障点和状态退化点、HI可预测性等方面均有提升。

（3）针对变工况及个体异质条件下样本数据分布规律差异大，导致HI建模过程中退化共性特征难以提取、不同工况下HI性能差异大的问题，深入研究了时空共性退化特征自适应提取的旋转机械HI构建方法。首先，建立集成金字塔卷积及Transformer的卷积自编码器，实现时间、空间两个维度时空退化特征的深层提取。其次，提出一种基于工况领域对齐及个体退化对齐的特征同质性约束方法，以提取不同工况下样本同质性特征，减少样本个体差异性及领域差异性。最后，设计一种自适应权重向量自编码器结构，通过权重向量的自适应调整使模型自动捕获具有退化趋势的时空特征，引入二次函数退化规律约束模型输出并构造HI，提升模型的自适应能力。以标准轴承退化数据集和所搭建的减速器全寿命周期退化实验平台数据集为例，验证了所提方法，结果表明该方法减少了不同工况和个体间特征差异的影响，HI平均综合得分在两个数据集上相比于其它方法分别平均提升了0.2026和0.1664，同时在早期故障点和状态退化点、HI可预测性等方面均有提升。

（4）针对复杂工况下旋转机械HI退化速率具有时变性，导致剩余使用寿命预测方法准确度低的问题，深入研究了多尺度双向门控循环网络及Transformer的旋转机械RUL预测方法。首先，构造基于不同时间步长的多尺度HI样本，引入多尺度输入来表达设备不同退化速率下退化状态之间的关联信息，防止有用信息丢失并抑制不确定干扰。其次，建立集成双向门控循环网络及Transformer的寿命预测模型，通过多个并行双向门控循环神经网络提取多尺度HI样本下的退化信息，避免不同时间尺度内的退化趋势信息对剩余使用寿命预测的影响，基于Transformer模型进一步提取HI退化特征，提升模型长序列预测精度。最后，以标准轴承退化数据集和所搭建的减速器全寿命周期退化实验平台数据集构建的HI为例，验证了所提方法，结果表明该方法与其它方法相比有效提升了预测准确度。

Rotating machinery is an important equipment in modern industrial production, and it is widely used in aerospace, shipping, energy, chemical industry, transportation, and machine tools. With the large scale, complexity, precision, and intelligence of rotating machinery, the adverse operating environment and complex working conditions increase the probability of performance degradation and failure. In view of the difficult problem of safe operation and efficient maintenance of rotating machinery, the Prognostic and Health Management (PHM) technology provides a more scientific strategy for the predictive maintenance and operation management of rotating machinery. As a crucial part of PHM, the construction of health indicator (HI) and the prediction of remaining useful life (RUL) can accurately grasp the operating state of rotating machinery, discover its potential faults, and provide a basis for taking proactive measures. Therefore, it is of great significance to study the common theory and method of HI construction and RUL prediction for rotating machinery. Based on deep learning theory and technology, this paper conducted in-depth research on the difficulties of HI construction, poor overall performance of HI, and low prediction accuracy of RUL for rotating machinery in complex industrial scenarios. The main works are as follows:

(1)Aiming at the difficulty of unsupervised construction of HI due to the incomplete and ineffective expression of degraded features under multi-point monitoring conditions, an unsupervised HI construction method based on the multi-domain mixed features and spatiotemporal fusion method for constructing HI of rotating machinery is deeply studied. Firstly, the feature extraction and selection method of multi-point monitoring vibration signal based on multi-domain mixed feature and weighted fusion evaluation criteria is studied to achieve the accurate expression of the degradation features and improve the completeness and effectiveness of the degradation features. Secondly, an unsupervised HI integrating the self-attention mechanism, long short-term memory network, and convolutional autoencoder is established to construct the HI with multiple spatiotemporal degradation information in the case of unlabeled samples. Finally, Savitzky-Golay filtering is employed to improve the HI's smoothness. The standard bearing degradation dataset and the whole life cycle degradation experimental platform dataset of the reducer are used as examples to verify the proposed method. The results show that this method can effectively represent the degradation process of rotating machinery, and the detection time of the early state degradation point on the two datasets is 8.6 min and 22.9 min earlier than other methods.

(2)To address the issue of the sample degradation features are difficult to extract under the condition of incomplete spatiotemporal relationship, which leads to poor HI performance, an automatic HI construction method with deep spatiotemporal degradation feature self-extraction of rotating machinery is studied. Firstly, parameter self-optimization variational mode decomposition is used to decompose the denoised vibration signal into multiple sub-modal components and perform the Fourier transform on the modal components. The vibration signals, frequency domain signals, sub-modal signals, and their frequency domain representations are coupled to form a two-dimensional matrix and input into the network to enhance the sample expressiveness. Secondly, a convolutional autoencoder network is proposed that integrates multi-scale convolution, convolutional long short-term memory network, and attention mechanism to automatically extract deep spatiotemporal degradation features of samples to improve the nonlinear quantization ability of HI. Finally, a quadratic function constraint loss term is introduced to constrain the HI's shape and improve the HI's overall performance. The standard bearing degradation dataset and the whole life cycle degradation experimental platform dataset of the reducer are used as examples to verify the proposed method. The results show that the average comprehensive score of HI constructed by the proposed method increased by 0.1542 and 0.2341 compared with other methods on the two datasets. At the same time, it can detect early fault points and state degradation points earlier, and also improve the HI predictability.

(3)Aiming at the problem that degradation common features are difficult to extract in the HI modeling process and HI performance varies greatly under different working conditions due to the large difference of sample data distribution law under varying working conditions and individual heterogeneity, the HI construction method of rotating machinery with spatiotemporal common features adaptive extraction is deeply studied. Firstly, a convolutional autoencoder network integrating pyramid convolution and Transformer is established to achieve deep extraction of spatiotemporal degradation features in both time and space dimensions. Secondly, a feature homogeneity constraint method based on domain alignment and individual degradation alignment is proposed to extract sample homogeneity features under different working conditions and reduce the sample's individual and domain differences. Finally, an adaptive weight vector autoencoder structure is constructed, which automatically captures spatiotemporal features with degradation trends through adaptive adjustment of weight vectors. The quadratic function degradation law is introduced to constrain the model output and improve the adaptive ability of the model. The standard bearing degradation dataset and the whole life cycle degradation experimental platform dataset of the reducer are used as examples to verify the proposed method. The results show that the proposed method can reduce the feature discrepancy between different working conditions and individuals. The average comprehensive score of HI constructed by the proposed method increased by 0.2026 and 0.1664 compared with other methods on the two datasets. At the same time, the performance of detecting early fault points, early state degradation points, and HI predictability has been improved.

Aiming at the low accuracy of the RUL prediction method due to the time-varying HI degradation rate of rotating machinery under complex working conditions, a RUL prediction method of rotating machinery based on multi-scale bidirectional gated recurrent network and Transformer is deeply studied. Firstly, multi-scale HI samples based on different time steps are constructed, and multi-scale inputs are introduced to express the correlation information between degradation states at different degradation rates, so as to prevent the loss of useful information and suppress uncertain interference. Secondly, a RUL prediction model integrating the bidirectional gated recurrent unit network and Transformer is established. The multiple parallel bidirectional gated recurrent unit networks are used to extract the degradation information under multi-scale HI samples to avoid the influence of degradation trend information in different time scales on the RUL prediction. Based on the Transformer model, HI degradation features are further extracted to improve the accuracy of the model's long series prediction. Finally, the HI curves constructed based on the standard bearing degradation dataset and the whole life cycle degradation dataset of the reducer are taken as an example to verify the above method, and the results show that the prediction accuracy of the proposed method is effectively improved compared with other methods.

[1]陈雪峰, 訾艳阳. 智能运维与健康管理[M]. 北京: 机械工业出版社, 2018.

[2]乔卉卉. 基于深度学习的复杂条件下旋转机械健康状态识别与预测[D]. 天津: 天津大学, 2020.

[3]Kumar A, Gandhi C P, Zhou Y, et al. Latest developments in gear defect diagnosis and prognosis: A review[J]. Measurement, 2020, 158: 107735.

[4]刘东东. 旋转机械故障信息挖掘及状态评估方法研究[D]. 北京: 北京交通大学, 2021.

[5]Researchnester. Rotating Equipment Market Size & Share, Growth Forecasts 2036[EB/OL]. https://www.researchnester.com/reports/rotating-equipment-market/5353. 2023-10-31.

[6]Meng H X, Li Y F. A review on prognostics and health management (PHM) methods of lithium-ion batteries[J]. Renewable and Sustainable Energy Reviews, 2019, 116: 109405.

[7]Fink O, Wang Q, Svensen M, et al. Potential, challenges and future directions for deep learning in prognostics and health management applications[J]. Engineering Applications of Artificial Intelligence, 2020, 92: 103678.

[8]Chen Y H, Peng G L, Zhu Z Y, et al. A novel deep learning method based on attention mechanism for bearing remaining useful life prediction[J]. Applied Soft Computing, 2020, 86: 105919.

[9]Hoang D T, Kang H J. A survey on Deep Learning based bearing fault diagnosis[J]. Neurocomputing, 2019, 335: 327-335.

[10]Tamuradov V, Medjaher K, Camci F, et al. Machine Health Indicator Construction Framework for Failure Diagnostics and Prognostics[J]. Journal of Signal Processing Systems, 2020, 92(6): 591-609.

[11]Li N P, Lei Y G, Lin J, 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.

[12]Huang Z Y, Xu Z G, Ke X J, et al. Remaining useful life prediction for an adaptive skew-Wiener process model[J]. Mechanical Systems and Signal Processing, 2017, 87: 294-306.

[13]Wang G, Xiang J W. Remain useful life prediction of rolling bearings based on exponential model optimized by gradient method[J]. Measurement, 2021, 176: 109161.

[14]Kosasih B Y, Caesarendra W, Tieu K, et al. Degradation trend estimation and prognosis of large low speed slewing bearing lifetime[J]. Applied mechanics and materials, 2014, 493: 343-348.

[15]Benkedjouh T, Medjaher K, Zerhouni N, et al. Health assessment and life prediction of cutting tools based on support vector regression[J]. Journal of Intelligent Manufacturing, 2015, 26(2): 213-223.

[16]Tse P W, Wang D. Enhancing the abilities in assessing slurry pumps’ performance degradation and estimating their remaining useful lives by using captured vibration signals[J]. Journal of Vibration and Control, 2017, 23(12): 1925-1937.

[17]Wang Y X, Xiang J W, Markert R, et al. Spectral kurtosis for fault detection, diagnosis and prognostics of rotating machines: A review with applications[J]. Mechanical Systems and Signal Processing, 2016, 66: 679-698.

[18]Bozchalooi I S, Liang M. A smoothness index-guided approach to wavelet parameter selection in signal de-noising and fault detection[J]. Journal of Sound and Vibration, 2007, 308(1-2): 246-267.

[19]Wang D. Some further thoughts about spectral kurtosis, spectral L2/L1 norm, spectral smoothness index and spectral Gini index for characterizing repetitive transients[J]. Mechanical Systems and Signal Processing, 2018, 108: 360-368.

[20]Singh J, Darpe A K, Singh S P. Bearing damage assessment using Jensen-Rényi Divergence based on EEMD[J]. Mechanical Systems and Signal Processing, 2017, 87: 307-339.

[21]Qin A S, Zhang Q H, Hu Q, et al. Remaining Useful Life Prediction for Rotating Machinery Based on Optimal Degradation Indicator[J]. Shock and Vibration, 2017, 2017: 1-12.

[22]Liu T, Chen J, Dong G M. Zero crossing and coupled hidden Markov model for a rolling bearing performance degradation assessment[J]. Journal of Vibration and Control, 2014, 20(16): 2487-2500.

[23]Li F, Chen Y, Wang J X, et al. A reinforcement learning unit matching recurrent neural network for the state trend prediction of rolling bearings[J]. Measurement, 2019, 145: 191-203.

[24]Qian Y N, Yan R Q, Hu S J. Bearing Degradation Evaluation Using Recurrence Quantification Analysis and Kalman Filter[J]. IEEE Transactions on Instrumentation and Measurement, 2014, 63(11): 2599-2610.

[25]Kong X F, Yang J. Remaining Useful Life Prediction of Rolling Bearings Based on RMS-MAVE and Dynamic Exponential Regression Model[J]. IEEE Access, 2019, 7: 169705-169714.

[26]Inturi V, Balaji S V, Gyanam P, et al. An integrated condition monitoring scheme for health state identification of a multi-stage gearbox through Hurst exponent estimates[J]. Structural Health Monitoring, 2023, 22(1): 730-745.

[27]He M, Guo W. An integrated approach for bearing health indicator and stage division using improved Gaussian mixture model and confidence value[J]. IEEE Transactions on Industrial Informatics, 2021, 18(8): 5219-5230.

[28]Huang G J, Li H K, Ou J Y, et al. A performance degradation assessment method using one-dimensional sparse representation self-learning dictionary[J]. Measurement Science and Technology, 2021, 32(11): 115103.

[29]Nistane V. Optimum prediction model of remaining useful life for rolling element bearing based on integrating optimize health indicator (OHI) and machine learning algorithm[J]. World Journal of Engineering, 2024, 21(1): 170-185.

[30]Yan T T, Wang D, Zheng M M, et al. Fisher’s discriminant ratio based health indicator for locating informative frequency bands for machine performance degradation assessment[J]. Mechanical Systems and Signal Processing, 2022, 162: 108053.

[31]钱门贵, 陈涛, 于耀翔, 等. 一种改进基尼指数加权的轴承健康指标构建方法[J]. 中国机械工程, 2023, 34(15): 1813-1819+1855.

[32]Hong S, Zhou Z, Zio E, et al. An adaptive method for health trend prediction of rotating bearings[J]. Digital Signal Processing, 2014, 35: 117-123.

[33]王奉涛, 陈旭涛, 柳晨曦, 等. 基于KPCA和WPHM的滚动轴承可靠性评估与寿命预测[J]. 振动、测试与诊断, 2017, 37(3): 476-483.

[34]Lu C, Chen J, Hong R J, et al. Degradation trend estimation of slewing bearing based on LSSVM model[J]. Mechanical Systems and Signal Processing, 2016, 76-77: 353-366.

[35]Mithun Praveen H, Shah D, Pandey K D, et al. PCA based health indicator for remaining useful life prediction of wind turbine gearbox[J]. Vibroengineering Procedia, 2019, 29: 31-36.

[36]Rai A, Kim J M. A novel health indicator based on the Lyapunov exponent, a probabilistic self-organizing map, and the Gini-Simpson index for calculating the RUL of bearings[J]. Measurement, 2020, 164: 108002.

[37]Liao L X, Jin W J, Pavel R. Enhanced Restricted Boltzmann Machine With Prognosability Regularization for Prognostics and Health Assessment[J]. IEEE Transactions on Industrial Electronics, 2016, 63(11): 7076-7083.

[38]Hong S, Zhou Z, Zio E, et al. Condition assessment for the performance degradation of bearing based on a combinatorial feature extraction method[J]. Digital Signal Processing, 2014, 27: 159-166.

[39]Wang Y, Peng Y Z, Zi Y Y, et al. A Two-Stage Data-Driven-Based Prognostic Approach for Bearing Degradation Problem[J]. IEEE Transactions on Industrial Informatics, 2016, 12(3): 924-932.

[40]Yang C Y, Ma J, Wang X D, et al. A novel based-performance degradation indicator RUL prediction model and its application in rolling bearing[J]. ISA transactions, 2022, 121: 349-364.

[41]Zhang J, Jiang N, Li H K, et al. Online health assessment of wind turbine based on operational condition recognition[J]. Transactions of the Institute of Measurement and Control, 2019, 41(10): 2970-2981.

[42]Wen J, Gao H. Remaining useful life prediction of the ball screw system based on weighted Mahalanobis distance and an exponential model[J]. Journal of Vibroengineering, 2018, 20(4): 1691-1707.

[43]Rai A, Upadhyay S H. The use of MD-CUMSUM and NARX neural network for anticipating the remaining useful life of bearings[J]. Measurement, 2017, 111: 397-410.

[44]Li Q, Yan C F, Wang W, et al. Health Indicator Construction Based on MD-CUMSUM With Multi-Domain Features Selection for Rolling Element Bearing Fault Diagnosis[J]. IEEE Access, 2019, 7: 138528-138540.

[45]Kim S, Park H J, Seo Y H, et al. A Robust Health Indicator for Rotating Machinery Under Time-Varying Operating Conditions[J]. IEEE Access, 2022, 10: 4993-5001.

[46]刘晨辉, 温广瑞, 苏宇, 等. 基于混合域相对特征和FOA-XGBoost滚动轴承退化评估[J]. 振动、测试与诊断, 2021, 41(05): 880-887+1031.

[47]Liu K B, Gebraeel N Z, Shi J J. A data-level fusion model for developing composite health indices for degradation modeling and prognostic analysis[J]. IEEE Transactions on Automation Science and Engineering, 2013, 10(3): 652-664.

[48]Xu F, Huang Z L, Yang F F, et al. Constructing a health indicator for roller bearings by using a stacked auto-encoder with an exponential function to eliminate concussion[J]. Applied Soft Computing, 2020, 89: 106119.

[49]Hou B C, Wang D, Wang Y, et al. Adaptive weighted signal preprocessing technique for machine health monitoring[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-11.

[50]Yan X A, Jia M P. A novel optimized SVM classification algorithm with multi-domain feature and its application to fault diagnosis of rolling bearing[J]. Neurocomputing, 2018, 313: 47-64.

[51]赵靖, 廖英英, 杨绍普, 等. 基于无迹卡尔曼滤波的动态贝叶斯小波变换在轴承故障诊断中的应用[J]. 振动与冲击, 2020, 39(11): 53-62.

[52]Guo S, Zhang B, Yang T, et al. Multitask convolutional neural network with information fusion for bearing fault diagnosis and localization[J]. IEEE Transactions on Industrial Electronics, 2020, 67(9): 8005-8015.

[53]Khodja M E A, Aimer A F, Boudinar A H, et al. Bearing fault diagnosis of a PWM inverter fed-induction motor using an improved short time fourier transform[J]. Journal of Electrical Engineering & Technology, 2019, 14(3): 1201-1210.

[54]Gourisaria M K, Agrawal R, Sahni M, et al. Comparative analysis of audio classification with MFCC and STFT features using machine learning techniques[J]. Discover Internet of Things, 2024, 4(1): 1.

[55]Guo J C, Liu Y, Li J H, et al. Rotating machinery fault detection using a new version of intrinsic time-scale decomposition[J]. IEEE Sensors Journal, 2024, 24(2): 1905-1918.

[56]Li M, Zhan C R, Lv Y G, et al. Miniature mass spectrometer signal processing based on ensemble empirical mode decomposition feature enhancement[J]. Rapid Communications in Mass Spectrometry, 2023, 37(15): e9536.

[57]彭丹丹, 刘志亮, 靳亚强, 等. 基于软筛分停止准则的改进经验模态分解及其在旋转机械故障诊断中的应用[J]. 机械工程学报, 2019, 55(10): 122-132.

[58]Zhan X B, Bai H J, Yan H, et al. Diesel Engine Fault Diagnosis Method Based on Optimized VMD and Improved CNN[J]. Processes, 2022, 10(11): 2162.

[59]Liu Z C, Li S Y, Wang R C, et al. Research on Fault Feature Extraction Method of Rolling Bearing Based on SSA-VMD-MCKD[J]. Electronics, 2022, 11(20): 3404.

[60]Tian J, Liu L, Zhang F, et al. Multi-domain entropy-random forest method for the fusion diagnosis of inter-shaft bearing faults with acoustic emission signals[J]. Entropy, 2019, 22(1): 57.

[61]Huo Z Q, Martínez-García M, Zhang Y, et al. Entropy measures in machine fault diagnosis: Insights and applications[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(6): 2607-2620.

[62]Li Q C, Ding X X, Huang W B, et al. Transient feature self-enhancement via shift-invariant manifold sparse learning for rolling bearing health diagnosis[J]. Measurement, 2019, 148: 1-12.

[63]Zeng M, Zhang W M, Chen Z. Group-based K-SVD denoising for bearing fault diagnosis[J]. IEEE Sensors Journal, 2019, 19(5): 6335-6343.

[64]Hu C F, Wang Y X. Multidimensional denoising of rotating machine based on tensor factorization[J]. Mechanical Systems and Signal Processing, 2019, 122: 273-289.

[65]Jin X H, Fan J C, Chow T W S. Fault detection for rolling-element bearings using multivariate statistical process control methods[J]. IEEE Transactions on Instrumentation and Measurement, 2018, 68(9): 3128-3136.

[66]Garcia-Bracamonte J E, Ramirez-Cortes J M, Rangel-Magdaleno J D J, et al. An approach on MCSA-based fault detection using independent component analysis and neural networks[J]. IEEE Transactions on Instrumentation and Measurement, 2019, 68(5): 1353-1361.

[67]Liu Y H, Hu Z B, Zhang Y S. Bearing feature extraction using multi-structure locally linear embedding[J]. Neurocomputing, 2021, 428: 280-290.

[68]Li X Q, Jiang H K, Xiong X, et al. Rolling bearing health prognosis using a modified health index based hierarchical gated recurrent unit network[J]. Mechanism and Machine Theory, 2019, 13: 229-249.

[69]Miao J, Yang T, Sun L, et al. Graph regularized locally linear embedding for unsupervised feature selection[J]. Pattern Recognition, 2022, 122: 108299.

[70]Hu Z B, Yin H S, Liu Y H. Locally linear embedding vote: A novel filter method for feature selection[J]. Measurement, 2022, 190: 110535.

[71]曾大懿, 蒋雨良, 邹益胜, 等. 一种新的轴承寿命预测特征评价指标构建与验证[J]. 振动与冲击, 2021, 40(22): 18-27.

[72]Wang F T, Liu X F, Deng G, et al. Remaining life prediction method for rolling bearing based on the long short-term memory network[J]. Neural Processing Letters, 2019, 50(3): 2437-2454.

[73]Liu P, Li G D, Liu X G, et al. A novel non-uniform control vector parameterization approach with time grid refinement for flight level tracking optimal control problems[J]. ISA Transactions, 2018, 73: 66-78.

[74]杨创艳, 马军, 王晓东, 等. 特征融合与灰色回归的滚动轴承性能退化评估[J]. 电子学报, 2022, 50(01): 106-115.

[75]Zhu K H, Chen L, Hu X. Rolling element bearing fault diagnosis based on multi-scale global fuzzy entropy, multiple class feature selection and support vector machine[J]. Transactions of the Institute of Measurement and Control, 2019, 41(14): 4013-4022.

[76]王望望, 邓林峰, 赵荣珍, 等. 集成KPCA与t-SNE的滚动轴承故障特征提取方法[J]. 振动工程学报, 2021, 34(02): 431-440.

[77]Jain D, Singh V. An efficient hybrid feature selection model for dimensionality reduction[J]. Procedia Computer Science, 2018, 132: 333-341.

[78]Zhang X H, Xu Y, He Y L, et al. Novel manifold learning based virtual sample generation for optimizing soft sensor with small data[J]. ISA transactions, 2021, 109: 229-241.

[79]张鑫, 郭顺生, 李益兵, 等. 基于拉普拉斯特征映射和深度置信网络的半监督故障识别[J]. 机械工程学报, 2020, 56(01): 69-81.

[80]Yousaf M, Khan M S S, Rehman T U, et al. Nric: A noise removal approach for nonlinear isomap method[J]. Neural Processing Letters, 2021, 53(3): 2277-2304.

[81]佘道明. 基于深度学习的滚动轴承健康评估与剩余寿命预测研究[D]. 南京: 东南大学, 2020.

[82]Guo L, Li N P, Jia F, et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings[J]. Neurocomputing, 2017, 240: 98-109.

[83]Hasani R M, Wang G D, Grosu R. An automated auto-encoder correlation-based health-monitoring and prognostic method for machine bearings[J]. arXiv preprint arXiv: 1703.06272, 2017.

[84]Hu Y, Palme T, Fink O. Deep health indicator extraction: A method based on autoencoders and extreme learning machines[C]//Annual Conference of the PHM Society. 2016, 8(1): 446-452.

[85]She D M, Jia M P, Pecht M G. Sparse auto-encoder with regularization method for health indicator construction and remaining useful life prediction of rolling bearing. Measurement Science and Technology, 2020, 31(10): 105005.

[86]Abid A, Khan M T, Khan M S. Multidomain features-based GA optimized artificial immune system for bearing fault detection[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2020, 50(1): 348-359.

[87]Lei Z H, Wen G R, Dong S Z, et al. An intelligent fault diagnosis method based on domain adaptation and its application for bearings under polytropic working conditions[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-14.

[88]Zhou Z T, Chen J L, Zi Y Y, et al. A modified SOM method based on nonlinear neural weight updating for bearing fault identification in variable speed condition[J]. Journal of Mechanical Science and Technology, 2020, 34(5): 1901-1912.

[89]Hu C H, Pei H, Si X S, et al. A prognostic model based on DBN and diffusion process for degrading bearing[J]. IEEE Transactions on Industrial Electronics, 2019, 67(10): 8767-8777.

[90]Chang Y H, Chen J L, Lv H X, et al. Heterogeneous bi-directional recurrent neural network combining fusion health indicator for predictive analytics of rotating machinery[J]. ISA transactions, 2022, 122: 409-423.

[91]Chen Z W, Guo R J, Lin Z, et al. A data-driven health monitoring method using multiobjective optimization and stacked autoencoder based health indicator[J]. IEEE Transactions on Industrial Informatics, 2020, 17(9): 6379-6389.

[92]Li J, Zi Y Y, Wang Y, et al. Health indicator construction method of bearings based on Wasserstein dual-domain adversarial networks under normal data only[J]. IEEE Transactions on Industrial Electronics, 2022, 69(10): 10615-10624.

[93]Guo L, Lei Y G, Li N P, et al. Machinery health indicator construction based on convolutional neural networks considering trend burr[J]. Neurocomputing, 2018, 292: 142-150.

[94]She D M, Jia M P. Wear indicator construction of rolling bearings based on multi-channel deep convolutional neural network with exponentially decaying learning rate[J]. Measurement, 2019, 135: 368-375.

[95]Cheng Y W, Hu K, Wu J, et al. A convolutional neural network based degradation indicator construction and health prognosis using bidirectional long short-term memory network for rolling bearings[J]. Advanced Engineering Informatics, 2021, 48: 101247.

[96]Chen D, Qin Y, Wang Y, et al. Health indicator construction by quadratic function-based deep convolutional auto-encoder and its application into bearing RUL prediction[J]. ISA transactions, 2021, 114: 44-56.

[97]Xu F, Shu X, Li X, et al. Health indicator construction for roller bearing based on an unsupervised deep belief network with a novel sigmoid zero local minimum point model[J]. Structural Health Monitoring, 2021, 20(4): 2110-2123.

[98]Ping G, Chen J L, Pan T Y, et al. Degradation feature extraction using multi-source monitoring data via logarithmic normal distribution based variational auto-encoder[J]. Computers in Industry, 2019, 109: 72-82.

[99]Hemmer M, Klausen A, Van Khang H, et al. Health indicator for low-speed axial bearings using variational autoencoders[J]. IEEE Access, 2020, 8: 35842-35852.

[100]Chen D L, Qin Y, Wang Y, et al. Health indicator construction by quadratic function-based deep convolutional auto-encoder and its application into bearing RUL prediction[J]. ISA transactions, 2021, 114: 44-56.

[101]Ma W P, Guo L, Gao H L, et al. A health indicator construction method based on self-attention convolutional autoencoder for rotating machine performance assessment[J]. Measurement, 2022, 204: 112108.

[102]Luo J H, Zhang X. Convolutional neural network based on attention mechanism and Bi-LSTM for bearing remaining life prediction[J]. Applied Intelligence, 2022, 52(1): 1076-1091.

[103]Ye Z J, Zhang Q, Shao S Y, et al. Rolling bearing health indicator extraction and RUL prediction based on multi-scale convolutional autoencoder[J]. Applied Sciences, 2022, 12(11): 5747.

[104]岳研, 刘畅, 刘韬. 基于深度融合神经网络的轴承健康指标构建[J]. 电子测量与仪器学报, 2021, 35(07): 44-52.

[105]文井辉, 伍荣森, 李帅永, 等. 基于DRSN和优化BiLSTM的轴承剩余寿命预测方法[J]. 计算机集成制造系统, 2024, 30(05): 1877-1888.

[106]Wu C Z, Feng F Z, Wu S J, et al. A method for constructing rolling bearing lifetime health indicator based on multi-scale convolutional neural networks[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2019, 41(11): 526.

[107]Xu F, Huang Z L, Yang F F, et al. Constructing a health indicator for roller bearings by using a stacked auto-encoder with an exponential function to eliminate concussion[J]. Applied Soft Computing, 2020, 89: 106119.

[108]Kulevome D K B, Wang H, Wang X G. Deep neural network based classification of rolling element bearings and health degradation through comprehensive vibration signal analysis[J]. Journal of Systems Engineering and Electronics, 2022, 33(1): 233-246.

[109]Zhang G, Liang W G, She B, et al. Rotating machinery remaining useful life prediction scheme using deep-learning-based health indicator and a new RVM[J]. Shock and Vibration, 2021, 2021(1): 8815241.

[110]郝勇, 刘尚宗, 吴文辉. 振动图像结合CNN的轴承振动信号分析方法研究[J]. 机械科学与技术, 2022, 41(12): 1943-1949.

[111]Schmidt S, Heyns P S, De Villiers J P. A tacholess order tracking methodology based on a probabilistic approach to incorporate angular acceleration information into the maxima tracking process[J]. Mechanical Systems and Signal Processing, 2018, 100: 630-646.

[112]Lu S L, Qin Y, Hang J, et al. Adaptively estimating rotation speed from DC motor current ripple for order tracking and fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2018, 68(3): 741-753.

[113]Huang H, Baddour N, Liang M. Bearing fault diagnosis under unknown time-varying rotational speed conditions via multiple time-frequency curve extraction[J]. Journal of Sound and Vibration, 2018, 414: 43-60.

[114]Tabrizi A A, Al-Bugharbee H, Trendafilova I, et al. A cointegration-based monitoring method for rolling bearings working in time-varying operational conditions[J]. Meccanica, 2017, 52(4): 1201-1217.

[115]赵川, 冯志鹏, 张颖琳, 等. 基于双对抗编码的时变工况下行星齿轮箱智能故障诊断[J]. 振动与冲击, 2021, 40(20): 158-167.

[116]Wang T Y, Zhang L, Qiao H H, et al. Fault diagnosis of rotating machinery under time-varying speed based on order tracking and deep learning[J]. Journal of Vibroengineering, 2020, 22(2): 366-382.

[117]Zhang W, Li C H, Peng G L, et al. A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load[J]. Mechanical systems and signal processing, 2018, 100: 439-453.

[118]Li X, Zhang W, Ding Q. A robust intelligent fault diagnosis method for rolling element bearings based on deep distance metric learning[J]. Neurocomputing, 2018, 310: 77-95.

[119]Li X, Zhang W, Ding Q. Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics, 2018, 66(7): 5525-5534.

[120]Mao W T, Chen J X, Chen Y J, et al. Construction of health indicators for rotating machinery using deep transfer learning with multiscale feature representation[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-13.

[121]Chen Z P, Zhu H P, Wu J, et al. Health indicator construction for degradation assessment by embedded LSTM-CNN autoencoder and growing self-organized map[J]. Knowledge-Based Systems, 2022, 252: 109399.

[122]Chen D L, Qin Y, Qian Q, et al. Transfer life prediction of gears by cross-domain health indicator construction and multi-hierarchical long-term memory augmented network[J]. Reliability Engineering & System Safety, 2023, 230: 108916.

[123]Qian Y N, Yan R Q, Gao R X. A multi-time scale approach to remaining useful life prediction in rolling bearing[J]. Mechanical Systems and Signal Processing, 2017, 83: 549-567.

[124]Sanhen N E, Braun M. Fatigue life prediction of welded joints at sub-zero temperatures using modified Paris-Erdogan parameters[J]. Theoretical and Applied Fracture Mechanics, 2023, 127: 104088.

[125]Lei Y G, Li N P, Gontarz S, et al. A Model-Based Method for Remaining Useful Life Prediction of Machinery[J]. IEEE Transactions on Reliability, 2016, 65(3): 1314-1326.

[126]Sheng Y Y, Liu H Y, Li J B. Bearing performance degradation assessment and remaining useful life prediction based on data-driven and physical model[J]. Measurement Science and Technology, 2023, 34(5): 055002.

[127]Wang J J, Gao R X, Yuan Z, et al. A joint particle filter and expectation maximization approach to machine condition prognosis[J]. Journal of Intelligent Manufacturing, 2019, 30(2): 605-621.

[128]Cui L L, Li W J, Wang X, et al. Comprehensive remaining useful life prediction for rolling element bearings based on time-varying particle filtering[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-10.

[129]Luo P, Hu J, Zhang L, et al. Research on remaining useful life prediction method of rolling bearing based on health indicator extraction and trajectory enhanced particle filter[J]. Journal of Dynamics, Monitoring and Diagnostics, 2022: 66-83.

[130]Cai H S, Feng J S, Li W Z, et al. Similarity-based particle filter for remaining useful life prediction with enhanced performance[J]. Applied Soft Computing, 2020, 94: 106474.

[131]尹建程. 基于相似轨迹的多工况下设备寿命预测方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2020.

[132]Pestaña-Melero F L, Haff G G, Rojas F J, et al. Reliability of the load-velocity relationship obtained through linear and polynomial regression models to predict the 1-repetition maximum load[J]. Journal of Applied Biomechanics, 2018, 34(3): 184-190.

[133]Huang Z L, Xu F, Yang F F. State of health prediction of lithium-ion batteries based on autoregression with exogenous variables model[J]. Energy, 2023, 262: 125497.

[134]Xiahou T F, Zeng Z G, Liu Y. Remaining useful life prediction by fusing expert knowledge and condition monitoring information[J]. IEEE Transactions on Industrial Informatics, 2020, 17(4): 2653-2663.

[135]Hu J W, Chen P. Predictive maintenance of systems subject to hard failure based on proportional hazards model[J]. Reliability Engineering & System Safety, 2020, 196: 106707.

[136]Guan Q L, Wei X K, Zhang H X, et al. Remaining useful life prediction for degradation processes based on the Wiener process considering parameter dependence[J]. Quality and Reliability Engineering International, 2024, 40(3): 1221-1245.

[137]Giorgio M, Guida M, Pulcini G. The transformed gamma process for degradation phenomena in presence of unexplained forms of unit-to-unit variability[J]. Quality and Reliability Engineering International, 2018, 34(4): 543-562.

[138]Zhang N, Jiang Z P, Sun Y B, et al. Model-data hybrid driven approach for remaining useful life prediction of cutting tool based on improved inverse Gaussian process[J]. Journal of Manufacturing Processes, 2024, 124: 604-620.

[139]胡楚叶. 基于人工神经网络的电力变压器健康状态评价与预测方法研究[D]. 武汉: 华中科技大学, 2021.

[140]Hou L Y, Li Y Y, Yao W, et al. Research on degradation prediction of rolling bearing based on adaptive multi-GA-BP[J]. Measurement and Control, 2022, 55(5-6): 491-501.

[141]Yao J C, Lu B C, Zhang J L. Multi-step-ahead tool state monitoring using clustering feature-based recurrent fuzzy neural networks[J]. IEEE Access, 2021, 9: 113443-113453.

[142]Chelmiah E T, McLoone V I, Kavanagh D F. Low Complexity Non-Linear Spectral Features and Wear State Models for Remaining Useful Life Estimation of Bearings[J]. Energies, 2023, 16(14): 5312.

[143]Wang Z Y, Shangguan W, Cai B G, et al. A weight-allocation-based ensemble remaining useful life prediction approach for a single device[J]. Measurement, 2024, 224: 113885.

[144]Zan T, Liu Z H, Wang H, et al. Prediction of performance deterioration of rolling bearing based on JADE and PSO-SVM[J]. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2021, 235(9): 1684-1697.

[145]陈佳鲜, 毛文涛, 刘京, 等. 基于深度时序特征迁移的轴承剩余寿命预测方法[J]. 控制与决策, 2021, 36(07): 1699-1706.

[146]Wang Q B, Zhao B, Ma H B, et al. A method for rapidly evaluating reliability and predicting remaining useful life using two-dimensional convolutional neural network with signal conversion[J]. Journal of Mechanical Science and Technology, 2019, 33(6): 2561-2571.

[147]Li X, Zhang W, Ding Q. Deep learning-based remaining useful life estimation of bearings using multi-scale feature extraction[J]. Reliability engineering & system safety, 2019, 182: 208-218.

[148]Xu X W, Li X, Ming W W, et al. A novel multi-scale CNN and attention mechanism method with multi-sensor signal for remaining useful life prediction[J]. Computers & Industrial Engineering, 2022, 169: 108204.

[149]Du X J, Jia W C, Yu P, et al. RUL prediction based on GAM-CNN for rotating machinery[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2023, 45(3): 142.

[150]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(07): 1447-1456.

[151]Cheng H, Kong X G, Chen G G, et al. Transferable convolutional neural network based remaining useful life prediction of bearing under multiple failure behaviors[J]. Measurement, 2021, 168: 108286.

[152]Zhang Y X, Li Y X, Wang Y L, et al. Adaptive spatio-temporal graph information fusion for remaining useful life prediction[J]. IEEE Sensors Journal, 2021, 22(4): 3334-3347.

[153]Pan Y N, Cheng D L, Wei T T, et al. Rolling bearing performance degradation assessment based on deep belief network and improved support vector data description[J]. Mechanical Systems and Signal Processing, 2022, 181: 109458.

[154]Gai J B, Zhong K Y, Du X J, et al. Detection of gear fault severity based on parameter-optimized deep belief network using sparrow search algorithm[J]. Measurement, 2021, 185: 110079.

[155]Pan Y B, Hong R J, Chen J, et al. A hybrid DBN-SOM-PF-based prognostic approach of remaining useful life for wind turbine gearbox[J]. Renewable Energy, 2020, 152: 138-154.

[156]Wang B, Lei Y G, Yan T, et al. Recurrent convolutional neural network: A new framework for remaining useful life prediction of machinery[J]. Neurocomputing, 2020, 379: 117-129.

[157]Cheng Y W, Wang C, Wu J, et al. Multi-dimensional recurrent neural network for remaining useful life prediction under variable operating conditions and multiple fault modes[J]. Applied Soft Computing, 2022, 118: 108507.

[158]Yu W N, Kim I Y, Mechefske C. Remaining useful life estimation using a bidirectional recurrent neural network based autoencoder scheme[J]. Mechanical Systems and Signal Processing, 2019, 129: 764-780.

[159]Wu Y T, Yuan M, Dong S P, et al. Remaining useful life estimation of engineered systems using vanilla LSTM neural networks[J]. Neurocomputing, 2018, 275: 167-179.

[160]葛阳, 郭兰中, 牛曙光, 等. 基于t-SNE和LSTM的旋转机械剩余寿命预测[J]. 振动与冲击, 2020, 39(07): 223-231+273.

[161]Wu J, Hu K, Cheng Y W, et al. Data-driven remaining useful life prediction via multiple sensor signals and deep long short-term memory neural network[J]. ISA transactions, 2020, 97: 241-250.

[162]Xiang S, Qin Y, Zhu C C, et al. Long short-term memory neural network with weight amplification and its application into gear remaining useful life prediction[J]. Engineering Applications of Artificial Intelligence, 2020, 91: 103587.

[163]Zhai Y M, Deng A D, Li J, et al. Remaining Useful Life Prediction of Rolling Bearings Based on Recurrent Neural Network[J]. Journal of Artificial Intelligence, 2019, 1(1): 19-27.

[164]蒋全胜, 许伟洋, 朱俊俊, 等. 基于动态加权卷积长短时记忆网络的滚动轴承剩余寿命预测方法[J]. 振动与冲击, 2022, 41(17): 282-291.

[165]Cao L X, Zhang H Y, Meng Z, et al. A parallel GRU with dual-stage attention mechanism model integrating uncertainty quantification for probabilistic RUL prediction of wind turbine bearings[J]. Reliability Engineering & System Safety, 2023, 235: 109197.

[166]姚德臣, 李博阳, 刘恒畅, 等. 基于注意力GRU算法的滚动轴承剩余寿命预测[J]. 振动与冲击, 2021, 40(17): 116-123.

[167]Zhou J H, Qin Y, Chen D L, et al. Remaining useful life prediction of bearings by a new reinforced memory GRU network[J]. Advanced Engineering Informatics, 2022, 53: 101682.

[168]Xu W Y, Jiang Q S, Shen Y H, et al. RUL prediction for rolling bearings based on Convolutional Autoencoder and status degradation model[J]. Applied Soft Computing, 2022, 130: 109686.

[169]Ma J, Su H, Zhao W L, et al. Predicting the Remaining Useful Life of an Aircraft Engine Using a Stacked Sparse Autoencoder with Multilayer Self-Learning[J]. Complexity, 2018, 2018(1): 3813029.

[170]张继冬, 邹益胜, 蒋雨良, 等. 基于全卷积变分自编码网络FCVAE的轴承剩余寿命预测方法[J]. 振动与冲击, 2020, 39(19): 13-18+25.

[171]龚众望. 旋转机械部件的健康状态评估和退化趋势预测联合研究[D]. 武汉: 武汉科技大学, 2022.

[172]Li J, Jia Y J, Niu M B, et al. Remaining useful life prediction of turbofan engines using CNN-LSTM-SAM approach[J]. IEEE Sensors Journal, 2023, 23(9): 10241-10251.

[173]Ma P, Li G F, Zhang H L, et al. Prediction of remaining useful life of rolling bearings based on multiscale efficient channel attention CNN and bidirectional GRU[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73: 1-13.

[174]Al-Dulaimi A, Zabihi S, Asif A, et al. A multimodal and hybrid deep neural network model for Remaining Useful Life estimation[J]. Computers in Industry, 2019, 108: 186-196.

[175]Sun C, Ma M, Zhao Z B, et al. Deep Transfer Learning Based on Sparse Autoencoder for Remaining Useful Life Prediction of Tool in Manufacturing[J]. IEEE Transactions on Industrial Informatics, 2019, 15(4): 2416-2425.

[176]Ren L, Cheng X J, Wang X K, et al. Multi-scale Dense Gate Recurrent Unit Networks for bearing remaining useful life prediction[J]. Future Generation Computer Systems, 2019, 94: 601-609.

[177]康守强, 周月, 王玉静, 等. 基于改进SAE和双向LSTM的滚动轴承RUL预测方法[J]. 自动化学报, 2022, 48 (09): 2327-2336.

[178]刘业峰, 王帅, 刘晶晶, 等. 基于TCN-SA和Bi-GRU的轴承剩余寿命预测[J/OL]. 计算机集成制造系统, 1-11[2024-09-12].

[179]Mo Y, Wu Q H, 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.

[180]Zhang L J, Wang B, Yuan X M, et al. Remaining useful life prediction via improved CNN, GRU and residual attention mechanism with soft thresholding[J]. IEEE Sensors Journal, 2022, 22(15): 15178-15190.

[181]Guo J C, Si Z T, Xiang J W. A compound fault diagnosis method of rolling bearing based on wavelet scattering transform and improved soft threshold denoising algorithm, Measurement, 2022, 196: 111276.

[182]Yu Y, Shang Q, Xie T. A Hybrid Model for Short-Term Traffic Flow Prediction Based on Variational Mode Decomposition, Wavelet Threshold Denoising, and Long Short-Term Memory Neural Network[J]. Complexity, 2021, 2021(1): 7756299.

[183]Abid A, Khan M T, Khan M S. Multidomain features-based GA optimized artificial immune system for bearing fault detection[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2017, 50(1): 348-359.

[184]Chen F F, Cheng M T, Tang B P, et al. A novel optimized multi-kernel relevance vector machine with selected sensitive features and its application in early fault diagnosis for rolling bearings[J]. Measurement, 2020, 156: 107583.

[185]Rajabi S, Azari M S, Santini S, et al. Fault diagnosis in industrial rotating equipment based on permutation entropy, signal processing and multi-output neuro-fuzzy classifier[J]. Expert systems with applications, 2022, 206: 117754.

[186]王冬, 侯炳昌, 王玉婷, 等. 稀疏测度和复杂性测度及其在设备健康监测中的研究进展[J/OL]. 机械工程学报, 1-17[2024-09-12].

[187]Nguyen K T P, Medjaher K. An automated health indicator construction methodology for prognostics based on multi-criteria optimization[J]. ISA transactions, 2021, 113: 81-96.

[188]John A, Sadasivan J, Seelamantula C S. Adaptive Savitzky-Golay filtering in non-Gaussian noise[J]. IEEE Transactions on Signal Processing, 2021, 69: 5021-5036.

[189]雷亚国, 韩天宇, 王彪, 等. XJTU-SY滚动轴承加速寿命试验数据集解读[J]. 机械工程学报, 2019, 55 (16): 1-6.

[190]Wang B, Lei Y G, Li N P, et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings[J]. IEEE Transactions on Reliability, 2018, 69(1): 401-412.

[191]Li Z, Wu J G, Yue X W. A shape-constrained neural data fusion network for health index construction and residual life prediction[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 32(11): 5022-5033.

[192]Xu F, Shu X, Li X, et al. Health indicator construction for roller bearing based on an unsupervised deep belief network with a novel sigmoid zero local minimum point model[J]. Structural Health Monitoring, 2021, 20(4): 2110-2123.

[193]Li Q, Yan C F, Wang W, et al. Health indicator construction based on MD-CUMSUM with multi-domain features selection for rolling element bearing fault diagnosis[J]. IEEE Access, 2019, 7: 138528-138540.

[194]Yan T T, Wang D, Kong J Z, et al. Definition of signal-to-noise ratio of health indicators and its analytic optimization for machine performance degradation assessment[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-16.

[195]姚家琪, 荆华, 赵春晖. 一种面向噪声环境中旋转机械故障诊断的多模态耦合输入神经网络[J]. 控制与决策, 2023, 38 (07): 1918-1926.

[196]Lei Z H, Wen G R, Dong S Z, et al. An intelligent fault diagnosis method based on domain adaptation and its application for bearings under polytropic working conditions[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-14.

[197]Xue J K, Shen B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems science & control engineering, 2020, 8(1): 22-34.

[198]Jana D, Patil J, Herkal S, et al. CNN and Convolutional Autoencoder (CAE) based real-time sensor fault detection, localization, and correction[J]. Mechanical Systems and Signal Processing, 2022, 169: 108723.

[199]Wen Y X, Wu J G, Das D, et al. Degradation modeling and RUL prediction using Wiener process subject to multiple change points and unit heterogeneity[J]. Reliability Engineering & System Safety, 2018, 176: 113-124.

[200]Wen Y X, Wu J G, Zhou Q, et al. Multiple-change-point modeling and exact Bayesian inference of degradation signal for prognostic improvement[J]. IEEE Transactions on Automation Science and Engineering, 2018, 16(2): 613-628.

[201]Tsui K L, Chen N, Zhou Q, et al. Prognostics and health management: A review on data driven approaches[J]. Mathematical Problems in Engineering, 2015, 2015(1): 793161.

[202]Si X S, Wang W B, Hu C H, et al. Remaining useful life estimation based on a nonlinear diffusion degradation process[J]. IEEE Transactions on reliability, 2012, 61(1): 50-67.

[203]Wu W F, Ni C C. A study of stochastic fatigue crack growth modeling through experimental data[J]. Probabilistic Engineering Mechanics, 2003, 18(2): 107-118.

[204]Zhang W, Peng G L, Li C H, et al. A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals[J]. Sensors, 2017, 17(2): 425.

[205]Li X, Zhang W, Ding Q. Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics, 2018, 66(7): 5525-5534.

[206]Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//30st Conference on Neural Information Processing Systems. 2017: 1-11.

[207]姜宇迪, 胡晖, 殷跃红. 基于无监督迁移学习的电梯制动器剩余寿命预测[J]. 上海交通大学学报, 2021, 55 (11): 1408-1416.

[208]赵冬冬. 基于深度学习的动车组轴承状态评估与剩余使用寿命预测方法研究[D]. 北京: 北京交通大学, 2022.

[209]Qin Y, Yang J H, Zhou J H, et al. A new supervised multi-head self-attention autoencoder for health indicator construction and similarity-based machinery RUL prediction[J]. Advanced Engineering Informatics, 2023, 56: 101973.