Transfer Learning With Neural Networks for Bearing Fault Diagnosis in Changing Working Conditions
Open accessThe DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.
Abstract
Traditional machine learning algorithms have made great achievements in data-driven fault diagnosis. However, they assume that all the data must be in the same working condition and have the same distribution and feature space. They are not applicable for real-world working conditions, which often change with time, so the data are hard to obtain. In order to utilize data in different working conditions to improve the performance, this paper presents a transfer learning approach for fault diagnosis with neural networks. First, it learns characteristics from massive source data and adjusts the parameters of neural networks accordingly. Second, the structure of neural networks alters for the change of data distribution. In the same time, some parameters are transferred from source task to target task. Finally, the new model is trained by a small amount of target data in another working condition. The Case Western Reserve University bearing data set is used to validate the performance of the proposed transfer learning approach. Experimental results show that the proposed transfer learning approach can improve the classification accuracy and reduce the training time comparing with the conventional neural network method when there are only a small amount of target data.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Compared with the diagnosis model just trained with a small number of target-domain samples, the fine-tuned diagnosis model presented the faster convergence rate and higher diagnosis accuracy.”
From Applications of machine learning to machine fault diagnosis: A review and roadmap (2020) · cites this paper as [438]
Citations
Cites
Huang, N. E., Shen, Z., Long, S. R., Wu, M. C., Shih, H. H., Zheng, Q., Yen, N.-C., Tung, C. C., & Liu, H. H. (1998) · Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rspa.1998.0193 · Cited by 23,564 · cited 2×
“Lei et al. [28] proposed a mechanical fault diagnosis model based on WPD and EMD.”
MethodPan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · cited 1×
“It has made great progresses in image, audio and text recognition [16]–[18].”
BackgroundWeiss, K., Khoshgoftaar, T. M., & Wang, D. (2016) · Journal of Big Data · doi:10.1186/s40537-016-0043-6 · Cited by 6,260
Robust Model-Based Fault Diagnosis for Dynamic Systems
Chen, J., & Patton, R. J. (1999) · The International Series on Asian Studies in Computer and Information Science · doi:10.1007/978-1-4615-5149-2 · Cited by 4,262 · cited 1×
“These fault diagnosis methods can be divided into model driven methods [1], [2] and data driven methods.”
MethodLearning and Transferring Mid-level Image Representations Using Convolutional Neural Networks
Oquab, M., Bottou, L., Laptev, I., & Sivic, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.222 · Cited by 3,150 · cited 1×
“[20] combined convolutional neural networks with transfer learning to learn and transfer mid-level image representations.”
BackgroundGao, Z., Cecati, C., & Ding, S. X. (2015) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2015.2417501 · Cited by 3,135 · cited 1×
“These fault diagnosis methods can be divided into model driven methods [1], [2] and data driven methods.”
MethodBoosting for transfer learning
Dai, W., Yang, Q., Xue, G.-R., & Yu, Y. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273521 · Cited by 1,724 · cited 1×
“[22] proposed a transfer learning framework called TrAdaBoost to select most useful and different samples as additional training data to improve the performance.”
BackgroundJia, F., Lei, Y., Lin, J., Zhou, X., & Lu, N. (2016) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2015.10.025 · Cited by 1,710 · cited 1×
“[9] used deep neural networks to recognize faults by training the model with frequency spectra extracted from vibration signals.”
Self-taught learning: transfer learning from unlabeled data
Raina, R., Battle, A., Lee, H., Packer, B., & Ng, A. Y. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273592 · Cited by 1,510 · cited 1×
“[21] designed selftaught learning to learn from unlabeled data and it improves classification accuracies of given image, audio and text classification tasks, respectively.”
BackgroundTransfer Learning for Reinforcement Learning Domains: A Survey
Taylor, M. E., & Stone, P. (2009) · doi:10.5555/1577069.1755839 · Cited by 1,488 · cited 1×
“The aim of transfer is to take advantage of experience learnt in one task and improve the performance in a similar but different task [24].”
Background
Research works printed but not linked: 1
- 0, Bearing Data Center Seeded Fault Test Data
Cited by
Items 201–300 of 379
Tong, J., Liu, C., Zheng, J., Pan, H., Wang, X., & Bao, J. (2022) · Measurement Science and Technology · doi:10.1088/1361-6501/ac6f46 · Cited by 16
“However, due to the complex and variable working states in actual industrial production, it is a huge challenge to meet the above two prerequisites, which leads to a drastic performance degradation of deep learning based fault diagnosis models [9, 10].”
Singh, M. T. (2025) · Results in Engineering · doi:10.1016/j.rineng.2025.106566 · Cited by 16
Visual perception-based fault diagnosis in froth flotation using statistical approaches
Zhang, J., Tang, Z., Xie, Y., Ai, M., & Gui, W. (2020) · Tsinghua Science & Technology · doi:10.26599/tst.2019.9010071 · Cited by 16
“Key words: fault diagnosis; Gabor; copula model; isolation forest; rejection sampling 1 Introduction Fault detection and diagnosis are important and challenging problems in many disciplines, such as hot strip mill rolling[1], bearing in rotary machinery systems[2, 3], minerals processing[4], and wireless sensor networ…”
BackgroundA fault diagnosis method for rolling bearings based on RDDAN under multivariable working conditions
Shi, H., Gan, C., Zhang, X., Meng, W., & Huang, C. (2022) · Measurement Science and Technology · doi:10.1088/1361-6501/ac9854 · Cited by 16
“Yet, most algorithms account for only a single variable, such as load or speed, overlooking the effect of noise that cannot be avoided in live production [23–25].”
Wu, J., Lin, M., Lv, Y., & Cheng, Y. (2022) · Quality Engineering · doi:10.1080/08982112.2022.2140436 · Cited by 16
Fault Diagnosis for TE Process Using RBF Neural Network
Liu, X., & He, H. (2021) · IEEE Access · doi:10.1109/access.2021.3107360 · Cited by 15
Soylu, E., Gül, S., Koca, K. A., Türkoğlu, M., Terzi, M., & Şengür, A. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110558 · Cited by 15
Aduwenye, P., Nutakor, C., Roininen, L., & Sopanen, J. (2025) · IEEE Access · doi:10.1109/access.2025.3548549 · Cited by 16
“Since the vibration data from the simulation are one-dimensional and lengthy, the sliding frame technique [51], [52] is used to extract samples.”
Farahani, H. S., Fatehi, A., & Shoorehdeli, M. A. (2020) · International Conference on Machine Learning and Applications · doi:10.1109/icmla51294.2020.00182 · Cited by 15
“Nevertheless, most of these researches are related to fault detection in the rotary machines [15]–[18].”
BackgroundSinha, S., Franciosa, P., & Ceglarek, D. J. (2021) · IEEE Access · doi:10.1109/access.2021.3068867 · Cited by 15
“Successful applications of transfer learning across multiple domains [7], [13]–[16] have enabled scalability in a sustainable manner that does not require exhaustive training data and computation capabilities.”
BackgroundImproved bilayer convolution transfer learning neural network for industrial fault detection
Wang, J., Zhang, W., Wu, H., & Zhou, J. (2021) · The Canadian Journal of Chemical Engineering · doi:10.1002/cjce.24281 · Cited by 15
Shul, Y., Yi, W., Choi, J., Kang, D.-S., & Choi, J. (2023) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2023.110102 · Cited by 14
Distance-guided domain adaptation for bearing fault diagnosis under variable operating conditions
Hei, Z., Shi, Q., Fan, X., Qian, F., Kumar, A., Zhong, M., & Zhou, Y. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad3c5e · Cited by 14
“Zhang et al proposed a neural network weight-sharing-based approach for solving the identification of bearing faults under variable operating conditions [15].”
Liu, J. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128398 · Cited by 14
Open Set Recognition for Machinery Fault Diagnosis
Xu, J., Kovatsch, M., & Lucia, S. (2021) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin45523.2021.9557572 · Cited by 17
Intelligent Fault Diagnosis for Machinery Based on Enhanced Transfer Convolutional Neural Network
Chen, Z., Qi, Z., Huang, R., Yixiao, L., Jipu, L., & Li, W. (2021) · Journal of Mechanical Engineering · doi:10.3901/jme.2021.21.096 · Cited by 22
“…不需一致,能够在彼此不同但相关的两个域间挖掘 领域不变的本质结构和特征,使得标注数据等有监 督信息可在领域间实现迁移和复用 [8] 。在故障诊断 领域,一些学者采用时频域特征和迁移成分分析 [9] 、 奇异值分解和 TrAdaBoost 迁移算法 [10] 以及神经网 络辅助训练分类器 [11] 等传统方法开展了一些迁移学 习智能诊断方法研究。此外,将深度神经网络和迁 移学习技术相结合,研究基于特征表示的深度迁移 学习方法,在解决小样本学习和域偏移等诊断问题 上也有了较多的探索和应用 [12] 。例如在解决小样本 学习方面,邵海东等 [13] 提出了一种提升深度迁移自…”
柴, 铮., 汪, 嘉., Zhao, C., 丁, 进., & 孙, 优. (2023) · Scientia Sinica Informationis · doi:10.1360/ssi-2022-0328 · Cited by 15
Sepulveda, N. F. E., & Sinha, J. K. (2024) · Structural Health Monitoring · doi:10.1177/14759217241249055 · Cited by 14
“8 Zhang et al. 9 studied bearing defects on changing working conditions through the ANN.”
Dang, Z., Lv, Y., Li, Y., & Wei, G. (2019) · Shock and Vibration · doi:10.1155/2019/3262818 · Cited by 15
“1 of BPNN training: 200, Test: 200 [71] Transfer learning CWRU, six Neural networks 91.”
BackgroundKumar, P., Hati, A. S., Padmanaban, S., Leonowicz, Z. M., & Chakrabarti, P. (2020) · IEEE International Conference on Environment and Electrical Engineering and IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) · doi:10.1109/eeeic/icpseurope49358.2020.9160712 · Cited by 13
Lv, Y., Zhang, X., Cheng, Y., & Lee, C. K. M. (2024) · Quality and Reliability Engineering International · doi:10.1002/qre.3597 · Cited by 13
Similarity-based Multi-source Transfer Learning Approach for Time Series Classification
Senanayaka, A., Mamun, A. A., Bond, G., Tian, W., Wang, H., Fuller, S., Falls, T. C., Rahimi, S., & Bian, L. (2022) · International Journal of Prognostics and Health Management · doi:10.36001/ijphm.2022.v13i2.3267 · Cited by 16
“Well-known classifier models, SVM and ANN without transfer, and transfer learning method TLNN (R. A. N. Zhang et al., 2017) are utilized to explore the unknown state prediction of the target domain given multi-sources (with the same conditions and features extracted from the vibration signals).”
MethodRuan, D., Zhang, F., Zhang, L., & Yan, J. (2023) · Journal of Vibration Engineering & Technologies · doi:10.1007/s42417-023-01106-0 · Cited by 13
“For example, a bearing test bench usually works under various conditions and probably even under a transient cycle [12, 13].”
Zhang, Z., Shao, M., Wang, L., Shao, S., & Ma, C. (2021) · Sensors · doi:10.3390/s21103382 · Cited by 13
“Then, the proposed approach with varying values of λ ∈ [1, 10] is executed when k = 50 and γ = 10.”
BackgroundMethodInfluentialDeep Transfer Learning Network for Fault Diagnosis Under Variable Working Conditions
An, Y., Zhang, K., Liu, Q., Chai, Y., & Huang, X. (2021) · CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS) · doi:10.1109/safeprocess52771.2021.9693606 · Cited by 15
“[17] proposed a fault diagnosis method based on a transfer learning network, which data form different working conditions is utilized to validate the performance.”
MethodAnomaly Detection of a Reciprocating Compressor using Autoencoders
Charoenchitt, C., & Tangamchit, P. (2021) · International Symposium on Instrumentation, Control, Artificial Intelligence, and Robotics (ICA-SYMP) · doi:10.1109/ica-symp50206.2021.9358453 · Cited by 12
“In the study of RAN et al [6], the transfer learning technique has been combined with the deep neural network to predict the machine faults.”
MethodPrediction of photovoltaic power generation based on LSTM and transfer learning digital twin
Yang, H. L., & Wang, W. (2023) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2467/1/012015 · Cited by 12
“Image classification [14], computer vision recognition [15], defect diagnosis [16-17], and certain prediction problems [18-19] are only some of the many applications that make extensive use of transfer learning now that the theory of transfer learning has been developed.”
BackgroundMulti‐Channel Domain Adaptation Deep Transfer Learning for Bridge Structure Damage Diagnosis
Xiao, H., Ogai, H., & Wang, W. (2022) · IEEJ Transactions on Electrical and Electronic Engineering · doi:10.1002/tee.23671 · Cited by 12
“Zhang et al. [12] applied transfer learning methods to neural networks and constructed a fault diagnosis network based on transfer learning theory.”
BackgroundMethodZhang, Z., Peng, L., Dai, G., Wang, M., Bai, J., Lei, Z., & Li, J. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3291736 · Cited by 11
“As a subtype of transfer learning [18], DA is capable of extracting features that are more akin between different domains, while simultaneously ensuring good classification performance on source data [19], [20], [21].”
BackgroundLiao, R., Wang, C.-G., Peng, F., Liang, W., Zhang, Y., & Zhang, X. (2024) · IEEE Access · doi:10.1109/access.2024.3351935 · Cited by 11
“It is very clear that the accuracy of DTM-Bearing is much better than that of traditional machine learning methods (ANN [5] and SVM [2]) in the fault detection of inner race fault (IRF), ball fault (BF), and outer race fault (ORF).”
BackgroundMethodResultInfluentialLiu, W. Y., & Ren, H. (2020) · International Journal of Energy Research · doi:10.1002/er.5560 · Cited by 11
Xiao, H., Wang, W., Ogai, H., Wang, M., & Shen, R. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3192307 · Cited by 11
“Based on the theory of domain learning discussed in [14] and Equation 2, the loss functions between the predicted domain label and the ground truth in the domain classifier for the i th sample of source and target domains are expressed in Equations 3 and 4.”
BackgroundRobust Bayesian Transfer Learning Between Kalman Filters
Papež, M., & Quinn, A. (2019) · Machine learning for signal processing ... · doi:10.1109/mlsp.2019.8918783 · Cited by 11
“Transfer learning has mostly been used to enhance traditional machine learning [2, 3] and reinforcement [4] learning algorithms, and has been widely deployed in various statistical signal processing applications, including genomics [5], cross-language speech recognition [6], fault diagnosis [7], video analysis [8], et…”
MethodGas Path Fault Diagnosis of Turboshaft Engine Based on Novel Transfer Learning Methods
Zhao, Y.-P., Jin, H.-J., & Liu, H. (2024) · Journal of Dynamic Systems Measurement and Control · doi:10.1115/1.4064846 · Cited by 11
Towards a Standard Benchmarking Framework for Domain Adaptation in Intelligent Fault Diagnosis
Farag, M. M. (2025) · IEEE Access · doi:10.1109/access.2025.3537817 · Cited by 11
Tran, K., Pham, L. D., Nguyen, V.-R., & Nguyen, H.-S.-H. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-03848-8 · Cited by 11
“For instance, pre-trained ResNet-50 models have been fine-tuned for bearing fault detection tasks, and transfer learning has also been applied to multilayer perceptron architectures (X. Wang et al., 2020; R. Zhang, Tao, Wu, & Guan, 2017).”
MethodSenanayaka, A., Tian, W., Falls, T. C., & Bian, L. (2023) · Journal of Manufacturing Science and Engineering · doi:10.1115/1.4057052 · Cited by 10
Wang, H., Chun, Z., Zhang, N., Chen, Y.-T., & Chen, Y. (2019) · Prognostics and System Health Management Conference · doi:10.1109/phm-qingdao46334.2019.8943008 · Cited by 13
“Reference [17] trains the neural networks by working condition 1 which has enough data, then retrains the networks by a small amount of data in working condition 2.”
Transfer Learning with Unsupervised Domain Adaptation Method for Bearing Fault Diagnosis
Chen, X., Yang, R., Wen, H., & Guan, S. (2021) · CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS) · doi:10.1109/safeprocess52771.2021.9693742 · Cited by 12
Domain Specific Learning for Sentiment Classification and Activity Recognition
Wang, H., Xue, Y., Zhen, X., & Tu, X. (2018) · IEEE Access · doi:10.1109/access.2018.2871349 · Cited by 10
“It has been applied and popularized in many fields such as wireless positioning [1], text classification [2], spam filtering [3], visual recognition [4], bearing fault diagnosis [5], humanoid robot learning [6] and so on.”
MethodXiao, H., Ogai, H., & Wang, W. (2024) · Structure and Infrastructure Engineering · doi:10.1080/15732479.2023.2167214 · Cited by 9
“In recent years, TL has been successfully applied in various fields such as image recognition, speech recognition (Zhuang, Luo, He, & Shi, 2015), and fault diagnosis (Zhang, Tao, Wu, & Guan, 2017; Han, Liu, & Fan, 2018; Chen, Yang, & Zhou, 2018).”
Zabin, M., Choi, H., Kabir, M. K., Kabir, A. N. B., & Uddin, J. (2024) · Electronics · doi:10.3390/electronics13163112 · Cited by 9
Wen, T., Chen, R., & Tang, L. (2021) · IEEE Access · doi:10.1109/access.2021.3069884 · Cited by 10
“Zhang et al. [20] proposed a transfer learning method for fault diagnosis based on the neural network so as to improve the diagnostic performance through the data of different working conditions.”
MethodBearing Fault Diagnosis Based on Small Sample Learning of Maml–Triplet
Cheng, Q., He, Z., Zhang, T., Li, Y., Liu, Z., & Zhang, Z. (2022) · Applied Sciences · doi:10.3390/app122110723 · Cited by 10
Cui, Z., Cao, H., Ai, Z., & Wang, J. (2023) · Applied Sciences · doi:10.3390/app131910606 · Cited by 9
Transfer Learning Based Method for Frequency Response Model Updating with Insufficient Data
Deng, Z., Zhang, X., & Zhao, Y. (2020) · Sensors · doi:10.3390/s20195615 · Cited by 9
“In those aforementioned TL research, the accessory source domain data are usually similar with target domain samples, such as the vibration signal obtained from the same experiment system under different working load [24] or the same machine signal in real-cases and laboratory [29].”
BackgroundLei, P., Shen, C., Zhu, J., Wang, D., Jiang, X., & Zhu, Z. (2020) · Measurement Science and Technology · doi:10.1088/1361-6501/abc38b · Cited by 9
On Model-Based Transfer Learning Method for the Detection of Inter-Turn Short Circuit Faults in PMSM
Wang, M., Song, Q., & Lai, W. (2023) · Sensors · doi:10.3390/s23229145 · Cited by 8
“Zhang et al. transferred the parameters and modified the structure of a shallow ANN trained by sufficient source data to a similar task, in which only a limited number of labeled samples are available [20].”
Artificial Intelligence in Fault Diagnosis of Industrial Machinery: A Comprehensive Review
Feisa, T. T., Gebremedhen, H. S., Kibrete, F., Woldemichael, D. E., & Enyew, G. G. (2025) · Structural Control and Health Monitoring · doi:10.1155/stc/4640227 · Cited by 10
Li, Z., Zhong, Z., Zhang, Z., Mao, W., & Zhang, W. (2025) · Lubricants · doi:10.3390/lubricants13030116 · Cited by 8
Li, Q. '., & Liang, S. Y. (2018) · Symmetry · doi:10.3390/sym10070243 · Cited by 8
“Some methods are used via signal adaptive decomposition, such as empirical mode decomposition (EMD) [11,12], local mean decomposition (LMD) [13,14], and variational mode decomposition (VMD) [15,16]; in addition, some methods are used by intelligent supervised learning techniques, such as artificial neural network (ANN…”
MethodRuan, D., Zhang, F., & Yan, J. (2021) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2021.08.141 · Cited by 8
Research on Rolling Bearing Fault Diagnosis Based on DRS Frequency Spectrum Image and Deep Learning
Zhuoxian, L. I., Wang, H., Chen, J., Zhou, Z., & Chen, W. (2023) · The International Journal of Acoustics and Vibration · doi:10.20855/ijav.2023.28.21942 · Cited by 8
Simani, S., Farsoni, S., & Castaldi, P. (2023) · Energies · doi:10.3390/en16093644 · Cited by 9
RetractedZhang, Z., Chen, H., Li, S., & An, Z. (2020) · Measurement Science and Technology · doi:10.1088/1361-6501/ab78c4 · Cited by 8
Yu, Z., Shan, D., & Sun, R. (2025) · Measurement · doi:10.1016/j.measurement.2025.120008 · Cited by 7
Huynh, H. X., Duong‐Trung, N., Nguyen, T. N. Q., Le, B., & Le, T. H. (2021) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2021.0120697 · Cited by 7
“Further details on CNN and its next-generation investigating transfer learning [23], [24], [25], [26], [27] leaves to interesting readers.”
BackgroundZhang, Z., Tang, A., & Zhang, T. (2023) · Sensors · doi:10.3390/s23198207 · Cited by 7
“Break through the limitation of a single type of signal for pump fault diagnosis by using multiple types of signals for multi-signal fusion.”
BackgroundMulti-feature fusion-based TCA-WKNN cross-sensor fault diagnosis method for dynamic weighing
Liang, W., Chen, Z., Zhong, J., Liao, H., & Shuncong Zhong, A. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/ad02b2 · Cited by 7
“This practice departs from traditional machine-learning assumptions that require independent co-distributions between training and testing samples, as well as substantial training data for effective classification [21].”
Yang, Y., He, Z., Yao, H., Wang, Y., Feng, J., & Wu, Y. (2023) · Actuators · doi:10.3390/act12120466 · Cited by 7
An audio‐based risky flight detection framework for quadrotors
Liu, W., Liu, C., Sajedi, S. O., Su, H., Liang, X. P., & Zheng, M. (2024) · IET Cyber-Systems and Robotics · doi:10.1049/csy2.12105 · Cited by 7
“Zhang et al. [26] employed transfer learning to guarantee the bearing fault diagnosis performance under different working conditions.”
Fault Classification of Induction Motor Bearing Using Adaptive Neuro Fuzzy Inference System
Gowthami, K., & Kalaivani, L. (2019) · International Conference on Electrical Energy Systems (ICEES) · doi:10.1109/icees.2019.8719244 · Cited by 7
Research on the Rapid Diagnostic Method of Rolling Bearing Fault Based on Cloud–Edge Collaboration
Tang, X., Xu, L., & Chen, G. (2022) · Entropy · doi:10.3390/e24091277 · Cited by 8
“Zhang [13] proposed a transfer learning method based on CNNs for fault diagnosis in order to utilize data under different working conditions to improve performance.”
MethodKumar, P., Hati, A. S., & Kumar, P. (2023) · International Journal of Adaptive Control and Signal Processing · doi:10.1002/acs.3643 · Cited by 7
Open Set Recognition Methods for Fault Diagnosis: A Review
Rehman, A. U., Jiao, W., Sun, J., Pan, H., & Yan, T. (2023) · International Conference on Advanced Computational Intelligence (ICACI) · doi:10.1109/icaci58115.2023.10146128 · Cited by 13
“These methods include adjusting the network's parameters and structure based on a large amount of source data and transferring some parameters to the target task for training on a small amount of target data [24].”
MethodTransfer learning: a cross domain LSTM way towards sustainable power predictive analytics
Garg, S., & Krishnamurthi, R. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17635-5 · Cited by 8
“As researched in [29] and implemented in [30] this is application to incorporate the changed input/output space of target task than source task.”
Usas: A Sustainable Continuous-Learning´ Framework for Edge Servers
Mishra, C. S., Sampson, J., Kandemir, M., Narayanan, V., & Das, C. R. (2024) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca57654.2024.00073 · Cited by 14
“Our workloads included Audio [23], [35](speech classification), 3D Point Clouds [14], [24](object classification) and Inertial Measurement Unit sensor data [9], [106](fault and activity detection).”
BackgroundDiversity maximization-based transfer diagnosis approach of rotating machinery
She, D., Chen, J., Yan, X., Zhao, X., & PECHT, M. G. (2023) · Structural Health Monitoring · doi:10.1177/14759217231164921 · Cited by 7
“12 Kang et al. 13 introduced a transfer fault diagnosis method by multifea-ture construction and variable mode decomposition.”
MethodData-level transfer learning for degradation modeling and prognosis
Fallahdizcheh, A., & Wang, C. (2022) · Journal of Quality Technology · doi:10.1080/00224065.2022.2081103 · Cited by 7
“This assumption, however, limits the impacts and applications of data-driven prognosis (Zhang et al. 2017).”
BackgroundSignal recognition based on transfer learning for Φ-OTDR fiber optic distributed disturbance sensor
Gan, J., & Xiao, Y. (2024) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2822/1/012086 · Cited by 6
“In the field of mechanical fault diagnosis, Zhang et al. showed that neural networks can learn features from massive data and train new models with a small amount of target data in another environment[19].”
Kambale, W. V., Salem, M., Benarbia, T., Machot, F. A., & Kyamakya, K. (2024) · Symmetry · doi:10.3390/sym16020241 · Cited by 6
“Task adaptation is another aspect of fine-tuning [30].”
Transfer learning for crash design
Colella, G., Lange, V. A., & Duddeck, F. M. E. (2024) · Discover Mechanical Engineering · doi:10.1007/s44245-024-00059-2 · Cited by 7
“In [24], the authors propose a TL approach to enhance the fault types classification of rolling bearings in different working conditions.”
Bouharrouti, N. E., Saberi, A. N., Khan, M. T., Kudelina, K., Naseer, M. U., & Belahcen, A. (2025) · IET Electric Power Applications · doi:10.1049/elp2.70074 · Cited by 6
Dharmadhikari, S., Raut, R., Bhattacharya, C., Ray, A., & Basak, A. (2022) · Metals · doi:10.3390/met12111849 · Cited by 7
Hwang, J., Lustig, N., Jung, M., & Lee, J. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e18086 · Cited by 6
Flotation Fault Diagnosis Method Using Statistical Approaches
Zhang, J., Tang, Z., Xie, Y., Ai, M., & Gui, W. (2019) · International Conference on Advanced Cloud and Big Data · doi:10.1109/cbd.2019.00055 · Cited by 6
“INTRODUCTION Fault detection and diagnosis is an important and challenging problem in many disciplines such as hot strip mill rolling [1], bearing in rotary machinery systems [2, 3], minerals processing [4], and wireless sensor networks [5].”
BackgroundTian, W., Guo, X., Xu, M., & Zhang, X. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3586458 · Cited by 5
“In comparison with traditional machine learning methods, Zhang et al. [41] conducted fault diagnosis research based on transfer learning and demonstrated its remarkable performance, which is comparable to large-scale convolutional neural networks (CNNs), even only with a small-sample dataset.”
Ma, C., Nan, S., Zhang, K., XiangFeng, Z., Chen, J., & Xu, Y. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad9f87 · Cited by 5
“Zhang proposed a neural network-based transfer learning method that aligns fault features of bearings under different operating conditions, achieving accurate fault diagnosis in variable conditions [23].”
Davoodabadi, A., Behzad, M., Arghand, H. A., Mohammadi, S., & Gelman, L. (2025) · Machines · doi:10.3390/machines13050351 · Cited by 5
Man, Q., Yu, H., Feng, K., Olofsson, T., & Lu, W. (2024) · Energy and Buildings · doi:10.1016/j.enbuild.2024.114041 · Cited by 5
Comparing the Effectiveness of Two Convolutional Neural Networks Methods on Fault Diagnosis
Liu, Y., & Li, J. (2020) · IEEE Data Driven Control and Learning Systems Conference (DDCLS) · doi:10.1109/ddcls49620.2020.9275096 · Cited by 5
“Other ML methods also had been used to fault diagnosis , such as K-nearest neighbors (KNN) [6], neural fuzzy network [7], Bayesian networks [8], transfer learning [9], random forest [10], ensemble learning [11].”
MethodZhu, Y. M., Liang, X., Wang, T., Xie, J., & Yang, J. N. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3151933 · Cited by 10
Liu, Z., Xiao, H., Zhang, T., & Li, G. (2025) · Machines · doi:10.3390/machines13080698 · Cited by 4
Fault Diagnosis of Rolling Bearings Based on an Ascending-Dimension Convolutional Neural Network
Bai, X. D., Zhong, X., Liu, Y., Zhang, K. M., Meng, W., Li, J., & Zhang, X. (2026) · Machines · doi:10.3390/machines14030302 · Cited by 4
“Currently, a number of deep learning methods are commonly used, including transfer learning [18,19], deep residual networks (ResNets) [20,21], generating confrontation networks [22], deep confidence networks [23], convolutional neural networks [24] and so on.”
Bearing Fault Diagnosis Using Machine Learning Models
Chandrvanshi, S., Sharma, S. K., Singh, M. P., & Singh, R. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-99-9562-2_18 · Cited by 4
Lu, O. H. T. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124297 · Cited by 4
Bearing Fault Diagnosis Based on Improved DBN Combining Attention Mechanism
Zhang, X., Geng, Y., Zhao, J., & Jiang, W. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892543 · Cited by 4
Rigatos, G. G., Zervos, N., Abbaszadeh, M., Siano, P., Serpanos, D. N., & Siadimas, V. (2020) · International Journal of Systems Science · doi:10.1080/00207721.2020.1792579 · Cited by 4
Towards the use of IoT and AI for Pervasive Exergames
Woodward, K., Kanjo, E., & Parker, W. (2021) · IEEE Global Conference on Artificial Intelligence and Internet of Things (GCAIoT) · doi:10.1109/gcaiot53516.2021.9692952 · Cited by 4
Deep multi feature dynamic adversarial diagnosis approach of rotating machinery
She, D., Chen, J., Yan, X., Hu, W., Zhang, H., & PECHT, M. G. (2022) · Measurement Science and Technology · doi:10.1088/1361-6501/ac7a94 · Cited by 4
“In the field of fault diagnosis, some scholars have carried out research on transfer diagnosis methods by using traditional methods such as transfer component analysis [8], singular value decomposition and TrAdaBoost [9].”
A case study of fault diagnosis based on deep autoencoder
Ma, H., & Ma, Z. (2023) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-224077 · Cited by 5
Data–Driven Wake Steering Control for a Simulated Wind Farm Model
Simani, S., Farsoni, S., & Castaldi, P. (2023) · International Journal of Robotics and Automation Technology · doi:10.31875/2409-9694.2023.10.02 · Cited by 4
“A support vector machine classifier is then used to predict problems [25].”
MethodTraganitis, P. A., & Strangas, E. G. (2023) · IEEE International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives (SDEMPED) · doi:10.1109/sdemped54949.2023.10271469 · Cited by 4
“Among those [37], [38] combined parameter transfer learning with”
BackgroundCai, Z., Liu, Y., Zhang, W., & Zhao, T. (2023) · International Conference on Robotics, Control and Automation Engineering (RCAE) · doi:10.1109/rcae59706.2023.10398828 · Cited by 4
“…model determines its performance in a big way, usually, the deeper the number of layers of the model, the better its feature learning ability [7][8], but due to the small size of the fault dataset, the CNN model used for bearing fault diagnosis can only superimpose at most 5 hidden layers, and…”
BackgroundTransfer Learning with 2D Vibration Images for Fault Diagnosis of Bearings Under Variable Speed
Ahmad, Z., Hasan, M. J., & Kim, J.-M. (2022) · Lecture notes in networks and systems · doi:10.1007/978-3-030-96308-8_14 · Cited by 4
Li, Q., Ji, Y., & Yang, T. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/ade90c · Cited by 3
“Current research (e.g. [30–32]) predominantly focuses on applying TL models to fault diagnosis in traditional machinery (e.g. gearboxes, rolling bearings), where the core task involves analyzing structured vibration signals with relatively uniform data patterns and well-defined fault categories.”
Li, J., Xiaofeng, Y., Tao, Z., Xiaoyu, R., Ma, J., & Huang, X. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adf90d · Cited by 3
“Transfer learning enables knowledge transfer from a data-rich source domain to a data-scarce target domain [35].”
An, Y., Hou, N., & Lv, Z. (2026) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.70084 · Cited by 4
“…into practical problems such as lung pattern analysis (Christodoulidis et al. 2017), multi-variable working condition bearing fault diagnosis (R. Zhang et al. 2017), and pavement breakage detection (Gopalakrishnan et al. 2017) after entering the third stage of the development history, which…”
Domain-augmented open-set generalization framework for fault diagnosis in rotating machinery
Ren, Z., & Wei, Y. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.115134 · Cited by 3
Transfer Learning with CLIP for Bearing Fault Diagnosis
He, D., He, M., & Taffari, A. (2024) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero58975.2024.10521337 · Cited by 3
Metadata
- Authors
- Zhang, R., Tao, H., Wu, L., & Guan, Y.
- Affiliations
- Capital Normal University (CN)
- Venue
- IEEE Access
- Published
- 2017
- Volume
- 5
- Pages
- 14347-14357
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W2731372149
- Cited by
- 378
- References
- 27
- Citations by year
- 2026: 23 · 2025: 29 · 2024: 37 · 2023: 48 · 2022: 62 · 2021: 62 · 2020: 58 · 2019: 44 · 2018: 14 · 2017: 1
- FWCI
- 20.10
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Natural Science Foundation of China (61202027)
- Beijing Natural Science Foundation of China (4122015)
- Project of Construction of Innovative Teams and Teacher Career Development for Universities and Colleges under Beijing Municipality (IDHT20150507)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Machine Fault Diagnosis Techniques
- Physical Sciences › Engineering › Control and Systems Engineering › Fault Detection and Control Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- transfer learning, bearing fault diagnosis, artificial neural networks, changing working conditions, domain adaptation, distribution shift, vibration signal classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.