Pith. sign in

REVIEW 5 major objections 5 minor 38 references

Transformer-Based Bearing Fault Detection using Temporal Decomposition Attention Mechanism

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A Transformer that splits attention into trend and seasonal branches achieves 98.1% accuracy on the CWRU bearing fault benchmark.

desk verdict The paper's TDA and HEMA components are underspecified to the point of vacuity, and the 98.1% accuracy claim rests on an unreported protocol. read the letter →

arxiv 2412.11245 v1 pith:WR2FSPMY submitted 2024-12-15 cs.LG cs.AIeess.SP

Marzieh Mirzaeibonehkhater , Mohammad Ali Labbaf-Khaniki , Mohammad Manthouri This is my paper · ORCID
classification cs.LGcs.AIeess.SP
keywords bearingfaultdetectiontemporaldecompositionattentionTransformerHullexponentialmovingaverageCWRUdatasettimeseriesclassificationpredictivemaintenancemechanism
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes a Transformer variant for bearing fault detection that splits the self-attention computation into two branches, one for slow-moving trends and one for repeating seasonal patterns. Each branch applies its own temporal bias and value stream, and the two attention outputs are added together before classification. The authors claim that on the Case Western Reserve University (CWRU) bearing dataset this design, combined with a Hull Exponential Moving Average (HEMA) front-end, reaches 98.1% accuracy and beats plain Transformers plus several CNN baselines. If correct, the contribution would be a fault classifier that can attribute its decisions to trend-like or seasonal behavior in vibration signals.

What carries the argument

The load-bearing object is the Temporal Decomposition Attention (TDA) mechanism: two softmax attention branches over the same query-key product, one using a trend value matrix $V_{\mathrm{trend}}$ and one using a seasonal value matrix $V_{\mathrm{season}}$, each modulated by a learnable temporal bias before the softmax, then added together to form the final attention output. The second component is HEMA, a Hull moving average computed with exponential moving averages instead of weighted ones, used as a noise-reducing feature extractor on the residual data. Together they are intended to let the model attend separately to long-term drift and periodic fluctuations, and the additive recombination is what carries the claimed accuracy gain.

What would settle it

Re-run the same CWRU classification with the TDA branch replaced by a single standard softmax attention over one value matrix (so $V_{\mathrm{trend}}=V_{\mathrm{season}}=V$ and both temporal biases are fixed to 1); if accuracy stays at 98.1%, the temporal decomposition is not responsible for the reported gain.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that temporal bias encoding plus seasonal-trend decomposition inside the attention mechanism improves bearing fault classification. The value stream is split into a trend component and a seasonal component, each attended to with a softmax over the same query-key product but scaled by a learnable temporal factor ($\alpha_t^{\mathrm{trend}}$ and $\alpha_t^{\mathrm{season}}$), and the two attention outputs are recombined additively. With HEMA smoothing applied to the residual data first, the resulting HEMA-Transformer-TDA model reports 98.1% accuracy on the nine-class CWRU task. The authors interpret the gap over plain Transformer attention and the CNN baselines as evidence that separately modeling long-term drift and short-term periodicities is the active ingredient.

Load-bearing premise

The empirical result depends on the assumption that the trend and seasonal branches of the attention mechanism are computed from the data in some definite way; the paper never specifies how those branches are built, so the 98.1% cannot yet be attributed to the mechanism.

Editorial extensions

If this is right

  • The same additive trend-plus-seasonal attention could be applied to other time series tasks with trends or periodic patterns, as the abstract anticipates.
  • On the paper's CWRU results, HEMA pre-processing alone improves over the plain Transformer, and adding TDA improves further, so both components are claimed to contribute to the 98.1%.
  • The model reports near-zero false alarm rates and low missed alarm rates for most fault classes, including the challenging Ball_014_1 class.
  • The trend/seasonal split gives a concrete route to interpretability: an operator could inspect which branch's attention weights are active when a fault is flagged.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the two-branch design works as described, the same TDA module could be inserted into any Transformer-based forecasting or anomaly-detection model where trend/season separation is meaningful, not just fault classifiers.
  • Because HEMA is a pure smoothing operator, it could be tested as a generic front-end for other classifiers on vibration data, independent of the Transformer.
  • CWRU is heavily benchmarked, so a sharper test of the state-of-the-art claim would be a comparison with matched parameter counts on a less saturated dataset or under different loads and speeds.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper proposes a Transformer-based method for bearing fault detection on the CWRU dataset. The two claimed contributions are a Temporal Decomposition Attention (TDA) mechanism that separately models trend and seasonal components, and a Hull Exponential Moving Average (HEMA) feature-extraction module. The authors report an accuracy of 98.1% for their HEMA-Transformer-TDA model, claiming it outperforms AlexNet, GoogleNet, ResNet, Wavelet-Attention, a base Transformer, and HEMA-Transformer. The methodology sections provide equations for HEMA and TDA, but key quantities and procedures are not defined, and the experimental section gives no train/test split, hyperparameters, or run-to-run variance.

Significance. If fully specified and verified, the proposed approach could be a competitive method on a saturated benchmark, but the contribution as written is not assessable. The paper does not ship machine-checked proofs, code, or a precise specification of the proposed mechanism; the central accuracy claim is unsupported. The attempt to bring temporal decomposition into attention is a reasonable direction, but the manuscript does not yet provide enough detail to attribute the reported result to the proposed components.

major comments (5)
  1. [3.2, Eqs. (12)-(14)] The TDA mechanism is underspecified to the point of being unimplementable. V_trend, V_season, alpha_t^trend, and alpha_t^season are never defined, nor is any procedure given for constructing them from the input. If one assumes V_trend = V_season = V and a scalar alpha, Eq. (14) reduces to standard attention with a learnable temperature, so the claimed seasonal-trend decomposition is vacuous. Additionally, Eq. (14) calls Attention_trend(Q,K,V_trend) without the alpha argument that appears in Eq. (12), making the equations internally inconsistent.
  2. [3.3] The proposed pipeline depends on 'identify[ing] and remov[ing]' trend and seasonality components of the CWRU dataset, but no decomposition method is specified: there is no description of whether STL, moving-average detrending, or another algorithm is used, no seasonal period or window size is given, and no relationship between the removed components and V_trend/V_season is stated. Without this step the model is not defined.
  3. [3.1, Eqs. (2)-(5)] The HEMA definition is incorrect as written. WMA1(n) and WMA2(n) are both set to WMA(n/2), so Diff = 2 * WMA1 - WMA2 reduces to WMA(n/2); moreover, WMA(n, Diff) in Eq. (5) is never defined as a function of two arguments. The standard Hull moving average uses WMA(2 * WMA(n/2)) - WMA(n), or an equivalent two-step calculation, but the equations here do not implement that or any other well-defined filter.
  4. [4.1] The central claim of 98.1% accuracy is not verifiable. The manuscript does not report the train/test split, data preprocessing, tokenization, sequence length, embedding dimension, number of heads, learning rate, loss function, training epochs, or any measure of run-to-run variance. Figure 2 shows single-point accuracy values with no error bars, and the baselines (e.g., 'Transformer' and 'Wavelet-Attention') are not described. Without these details the reported improvement cannot be reproduced or checked.
  5. [4.3] The per-class performance claims are stated in the text and in Figures 6-10, but no numerical table of precision, recall, F1, FAR, and MAR values is provided. Phrases like 'around 0.88-0.93' and 'F1-score of 1 for several classes' cannot be checked against figures that are not readable in the text, especially without axis labels and class legends. This weakens the paper's assertion of 'exceptional precision, recall, and F1-scores.'
minor comments (5)
  1. [Abstract] There is a typo in the abstract: 'no vel attention mechanism' should be 'novel attention mechanism.'
  2. [4.1] The text refers to 'Figure 4' but the displayed figure is labeled Fig. 2; later 'Figs. (6-10)' is used for Figures 6-10. Please renumber figures and references consistently.
  3. [Table 1] The table's abbreviation column uses file names such as 'IR_007_1' but does not state the load condition (1 HP, 1772 rpm) or sampling frequency; adding these to the header would make the experimental conditions explicit.
  4. [3.3] The paper says 'statistical measures like skewness and kurtosis are calculated' but never states whether these features are used as model inputs or how they affect the classifier; please clarify their role in the architecture.
  5. [2.1] Reference [16] on universal adversarial perturbations appears unrelated to the fault-detection literature survey; either connect it to the discussion or remove it.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the 98.1% accuracy is an empirical measurement, and no fitted value or self-citation is repackaged as a derivation.

full rationale

The paper makes no first-principles derivation; its central claim is a measured accuracy on the CWRU dataset. I checked the main candidates for circularity. (i) HEMA: Eqs. (2) and (3) set WMA1(n)=WMA2(n)=WMA(n/2), so Eq. (4) makes Diff=WMA(n/2) and Eq. (5) becomes WMA(n, WMA(n/2)); this is a degenerate or ill-defined definition, but it does not make any reported prediction equivalent to an input. (ii) TDA: Eqs. (12)-(14) leave V_trend, V_season, alpha_t^trend, and alpha_t^season undefined, and the paper never specifies the trend/season decomposition algorithm described in Section 3.3. This is a serious reproducibility and correctness defect, and it means the accuracy cannot be attributed to the proposed mechanism; however, it is not circularity because the 98.1% figure is not derived from those equations. (iii) There are no self-citations: the references are external (CWRU, AlexNet, etc.), and no uniqueness theorem or prior author result is imported as a load-bearing premise. (iv) The paper does not report whether accuracy is computed on a held-out test set, which is a missing experimental detail; without evidence that the evaluation set overlaps the training set, a claim of circularity would be speculation. For these reasons, no circular step is exhibited, and the score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim depends on learnable attention scaling factors, a hand-chosen smoothing window, a smoothing parameter, and an unspecified seasonal period. The paper also assumes that signal decomposition into trend and seasonal components is meaningful and that the additive attention combination is valid. No new physical entities are introduced.

free parameters (4)
  • temporal decay factor alpha_t^trend and alpha_t^season = not reported
    Learnable scalars multiplying QK^T in Eqs. 12-13; no values or training details are given, and the central claim depends on them.
  • HEMA window size n = not reported
    The window length in Eqs. 1-10 is chosen by hand and never reported; it controls the smoothing and thus affects all features.
  • smoothing factor alpha in EMA = 0.2 (stated 'typically')
    Eq. 6 uses alpha, said to typically be 0.2, but the actual value used in experiments is not given.
  • seasonal period for decomposition = not reported
    The trend/seasonal decomposition in Section 3.3 requires choosing a seasonal period; the paper does not state how this was set.
assumptions (4)
  • domain assumption Vibration signals can be cleanly separated into trend and seasonal components plus residual.
    Section 3.3 states the trend and seasonality are 'identified and removed' to isolate residuals, but no decomposition method or validation is provided.
  • ad hoc to paper Additive combination of two softmax attentions (Eq. 14) is a valid attention mechanism.
    The paper simply sums Attention_trend and Attention_season without deriving why this is appropriate or normalizing.
  • ad hoc to paper A scalar multiplier on attention logits (alpha_t) can capture temporal dependencies.
    No explanation is given for how a scalar decay factor, which is the same for all features, can represent time-varying importance.
  • domain assumption HEMA reduces noise while preserving fault-relevant information.
    Claimed in Sections 3.1 and 5 without measurements or comparison to other filters.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Transformer-Based Bearing Fault Detection using Temporal Decomposition Attention Mechanism." pith.science (2026). https://pith.science/paper/WR2FSPMY

@misc{pith2026241211245,
  author       = {Pith},
  title        = {Pith review of: Transformer-Based Bearing Fault Detection using Temporal Decomposition Attention Mechanism},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WR2FSPMY}},
  note         = {Machine review of arXiv:2412.11245}
}
read the original abstract

Bearing fault detection is a critical task in predictive maintenance, where accurate and timely fault identification can prevent costly downtime and equipment damage. Traditional attention mechanisms in Transformer neural networks often struggle to capture the complex temporal patterns in bearing vibration data, leading to suboptimal performance. To address this limitation, we propose a novel attention mechanism, Temporal Decomposition Attention (TDA), which combines temporal bias encoding with seasonal-trend decomposition to capture both long-term dependencies and periodic fluctuations in time series data. Additionally, we incorporate the Hull Exponential Moving Average (HEMA) for feature extraction, enabling the model to effectively capture meaningful characteristics from the data while reducing noise. Our approach integrates TDA into the Transformer architecture, allowing the model to focus separately on the trend and seasonal components of the data. Experimental results on the Case Western Reserve University (CWRU) bearing fault detection dataset demonstrate that our approach outperforms traditional attention mechanisms and achieves state-of-the-art performance in terms of accuracy and interpretability. The HEMA-Transformer-TDA model achieves an accuracy of 98.1%, with exceptional precision, recall, and F1-scores, demonstrating its effectiveness in bearing fault detection and its potential for application in other time series tasks with seasonal patterns or trends.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 36 canonical work pages

  1. [1]

    Introduction Industrial systems and plants are the backbone of modern economies, providing essential goods and services to societies worldwide. However, these complex systems are prone to faults and failures, which can have catastrophic consequences, including equipment damage, production downtime, and even loss of li fe [1]. The timely detection of fault...

    work page

  2. [2]

    The CWRU dataset is a prominent benchmark for assessing fault detection techniques

    Bearing Fault Detection Fault detection is critical for maintaining the efficiency and safety of industrial operations, as it helps identify issues early and prevents costly failures. The CWRU dataset is a prominent benchmark for assessing fault detection techniques. This section provides an overview of current research on fault detection and presents the...

    work page

  3. [3]

    It begins with an explanation of the HEMA for feature extraction, followed by a detailed description of the TDA mechanism

    Methodology This section outlines the key components and the proposed framework for bearing fault detection. It begins with an explanation of the HEMA for feature extraction, followed by a detailed description of the TDA mechanism. Finally, the integration of these components into the proposed HEMA-Transformer-TDA model is presented, highlighting the inno...

    work page 2005

  4. [4]

    Each subsection offers a detailed evaluation of the models' efficiency, dependability, and capability to handle the challenges of bearing fault detection

    Performance Evaluation This section assesses the performance of the proposed models through accuracy analysis, confusion matrices, and essential fault detection metrics. Each subsection offers a detailed evaluation of the models' efficiency, dependability, and capability to handle the challenges of bearing fault detection. Table 1 outlines the specifics o...

    work page

  5. [5]

    Conclusion This study presents a novel approach for bearing fault detection in rotating machinery, combining a Transformer neural network with TDA and HEMA feature extraction. The TDA mechanism effectively captures both long-term trends and seasonal patterns within the time series data, enabling the model to learn intricate temporal dependencies and impro...

    work page

  6. [6]

    Fault diagnosis of hydro -turbine via the incorporation of bayesian algorithm optimized CNN-LSTM neural network,

    F. Dao, Y. Zeng, and J. Qian, “Fault diagnosis of hydro -turbine via the incorporation of bayesian algorithm optimized CNN-LSTM neural network,” Energy, vol. 290, p. 130326, 2024

    work page 2024

  7. [7]

    Towards better benchmarking using the CWRU bearing fault dataset,

    J. Hendriks, P. Dumond, and D. A. Knox, “Towards better benchmarking using the CWRU bearing fault dataset,” Mech. Syst. Signal Process., vol. 169, p. 108732, 2022

    work page 2022

  8. [8]

    Bearing fault detection and diagnosis using case western reserve university dataset with deep learning approaches: A review,

    D. Neupane and J. Seok, “Bearing fault detection and diagnosis using case western reserve university dataset with deep learning approaches: A review,” Ieee Access, vol. 8, pp. 93155–93178, 2020

    work page 2020

Show all 38 references
  1. [9]

    Extracting features from time series,

    C. Herff and D. J. Krusienski, “Extracting features from time series,” Fundam. Clin. data Sci., pp. 85–100, 2019

    2019

  2. [10]

    Time domain synchronous moving average and its application to gear fault detection,

    L. Zhang and N. Hu, “Time domain synchronous moving average and its application to gear fault detection,” IEEE Access, vol. 7, pp. 93035–93048, 2019

    2019

  3. [11]

    End -to-end CNN+ LSTM deep learning approach for bearing fault diagnosis,

    A. Khorram, M. Khalooei, and M. Rezghi, “End -to-end CNN+ LSTM deep learning approach for bearing fault diagnosis,” Appl. Intell., vol. 51, no. 2, pp. 736–751, 2021

    2021

  4. [12]

    A study on the evaluation of tokenizer performance in natural language processing,

    S. Choo and W. Kim, “A study on the evaluation of tokenizer performance in natural language processing,” Appl. Artif. Intell., vol. 37, no. 1, p. 2175112, 2023

    2023

  5. [13]

    Fault diagnosis using variational autoencoder GAN and focal loss CNN under unbalanced data,

    W. Li et al. , “Fault diagnosis using variational autoencoder GAN and focal loss CNN under unbalanced data,” Struct. Heal. Monit., p. 14759217241254120, 2024

    2024

  6. [14]

    Battery fault diagnosis and failure prognosis for electric vehicles using spatio-temporal transformer networks,

    J. Zhao, X. Feng, J. Wang, Y. Lian, M. Ouyang, and A. F. Burke, “Battery fault diagnosis and failure prognosis for electric vehicles using spatio-temporal transformer networks,” Appl. Energy, vol. 352, p. 121949, 2023

    2023

  7. [15]

    Twins transformer: Cross-attention based two- branch transformer network for rotating bearing fault diagnosis,

    J. Li, Y. Bao, W. Liu, P. Ji, L. Wang, and Z. Wang, “Twins transformer: Cross-attention based two- branch transformer network for rotating bearing fault diagnosis,” Measurement, vol. 223, p. 113687, 2023

    2023

  8. [16]

    Compound fault diagnosis for industrial robots based on dual-transformer networks,

    C. Chen, C. Liu, T. Wang, A. Zhang, W. Wu, and L. Cheng, “Compound fault diagnosis for industrial robots based on dual-transformer networks,” J. Manuf. Syst., vol. 66, pp. 163–178, 2023

    2023

  9. [17]

    Deep learning attention mechanism in medical image analysis: Basics and beyonds,

    X. Li et al., “Deep learning attention mechanism in medical image analysis: Basics and beyonds,” Int. J. Netw. Dyn. Intell., pp. 93–116, 2023

    2023

  10. [18]

    presents a fault diagnosis approach for rotating machinery using RNNs, demonstrating the effectiveness of RNNs in learning temporal patterns and relationships in vibration signals to accurately identify and classify faults in rotating machinery. [19] proposes a novel fault dia...

  11. [19]

    Vision Transformer Based Tokenization for Enhanced Breast Cancer Histopathological Images Classification,

    M. L. Abimouloud, K. Bensid, M. Elleuch, O. Aiadi, and M. Kherallah, “Vision Transformer Based Tokenization for Enhanced Breast Cancer Histopathological Images Classification,” in IFIP International Conference on Artificial Intelligence Applications and Innovations , Springer,...

    2024

  12. [20]

    Starnet: Sensor trustworthiness and anomaly recognition via approximated likelihood regret for robust edge autonomy,

    N. Darabi, S. Tayebati, S. Ravi, T. Tulabandhula, and A. R. Trivedi, “Starnet: Sensor trustworthiness and anomaly recognition via approximated likelihood regret for robust edge autonomy,” arXiv Prepr. arXiv2309.11006, 2023, doi: 10.48550/arXiv.2309.11006

    2023 doi arXiv

  13. [21]

    Deep prediction on financial market sequence for enhancing economic policies,

    S. Salahshour, M. Salimi, K. Tehranian, N. Erfanibehrouz, M. Ferrara, and A. Ahmadian, “Deep prediction on financial market sequence for enhancing economic policies,” Decis. Econ. Financ., pp. 1–20, 2024, doi: 10.1007/s10203-024-00488-4

    2024 doi

  14. [22]

    Exploring Cross -model Neuronal Correlations in the Context of Predicting Model Performance and Generalizability,

    H. E. Oskouie, L. Levine, and M. Sarrafzadeh, “Exploring Cross -model Neuronal Correlations in the Context of Predicting Model Performance and Generalizability,” arXiv Prepr. arXiv2408.08448, 2024

    2024

  15. [23]

    Application of recurrent neural network to mechanical fault diagnosis: a review,

    J. Zhu, Q. Jiang, Y. Shen, C. Qian, F. Xu, and Q. Zhu, “Application of recurrent neural network to mechanical fault diagnosis: a review,” J. Mech. Sci. Technol., vol. 36, no. 2, pp. 527–542, 2022, doi: 10.1007/s12206-022-0102-1

    2022 doi

  16. [24]

    Fault diagnosis of rotating machinery based on recurrent neural networks,

    Y. Zhang, T. Zhou, X. Huang, L. Cao, and Q. Zhou, “Fault diagnosis of rotating machinery based on recurrent neural networks,” Measurement, vol. 171, p. 108774, 2021

    2021

  17. [25]

    A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems,

    T. Huang, Q. Zhang, X. Tang, S. Zhao, and X. Lu, “A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems,” Artif. Intell. Rev., vol. 55, no. 2, pp. 1289–1315, 2022

    2022

  18. [26]

    A power transformer fault prediction method through temporal convolutional network on dissolved gas chromatography data,

    M. Xing, W. Ding, H. Li, and T. Zhang, “A power transformer fault prediction method through temporal convolutional network on dissolved gas chromatography data,” Secur. Commun. Networks, vol. 2022, no. 1, p. 5357412, 2022

    2022

  19. [27]

    An innovative transformer neural network for fault detection and classification for photovoltaic modules,

    E. A. Ramadan, N. M. Moawad, B. A. Abouzalm, A. A. Sakr, W. F. Abouzaid, and G. M. El-Banby, “An innovative transformer neural network for fault detection and classification for photovoltaic modules,” Energy Convers. Manag., vol. 314, p. 118718, 2024

    2024

  20. [28]

    Variational attention-based interpretable transformer network for rotary machine fault diagnosis,

    Y. Li, Z. Zhou, C. Sun, X. Chen, and R. Yan, “Variational attention-based interpretable transformer network for rotary machine fault diagnosis,” IEEE Trans. neural networks Learn. Syst., 2022

    2022

  21. [29]

    CNN -based transformer model for fault detection in power system networks,

    J. B. Thomas, S. G. Chaudhari, K. V Shihabudheen, and N. K. Verma, “CNN -based transformer model for fault detection in power system networks,” IEEE Trans. Instrum. Meas., vol. 72, pp. 1 – 10, 2023

    2023

  22. [30]

    A planetary gearbox fault diagnosis method based on time-series imaging feature fusion and a transformer model,

    R. Wu, C. Liu, T. Han, J. Yao, and D. Jiang, “A planetary gearbox fault diagnosis method based on time-series imaging feature fusion and a transformer model,” Meas. Sci. Technol., vol. 34, no. 2, p. 24006, 2022

    2022

  23. [31]

    Bearing fault detection and diagnosis using case western reserve university dataset with deep learning approaches: A review,

    D. Neupane and J. Seok, “Bearing fault detection and diagnosis using case western reserve university dataset with deep learning approaches: A review,” IEEE Access, vol. 8, pp. 93155–93178, 2020, doi: 10.1109/ACCESS.2020.2990528

    2020

  24. [32]

    Machine learning based bearing fault diagnosis using the case western reserve university data: A review,

    X. Zhang, B. Zhao, and Y. Lin, “Machine learning based bearing fault diagnosis using the case western reserve university data: A review,” Ieee Access, vol. 9, pp. 155598–155608, 2021

    2021

  25. [33]

    Enhanced Fault Detection in Bearings Using Machine Learning and Raw Accelerometer Data: A Case Study Using the Case Western Reserve University Dataset,

    K. K. Raj, S. Kumar, R. R. Kumar, and M. Andriollo, “Enhanced Fault Detection in Bearings Using Machine Learning and Raw Accelerometer Data: A Case Study Using the Case Western Reserve University Dataset,” Information, vol. 15, no. 5, p. 259, 2024

    2024

  26. [34]

    Hull, Active Investing

    A. Hull, Active Investing. Australia, Wrightbooks, 2001

    2001

  27. [35]

    Moving convolutional neural networks to embedded systems: the alexnet and VGG -16 case,

    C. Alippi, S. Disabato, and M. Roveri, “Moving convolutional neural networks to embedded systems: the alexnet and VGG -16 case,” in 2018 17th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), IEEE, 2018, pp. 212–223

    2018

  28. [36]

    A feature transferring fault diagnosis based on WPDR, FSWT and GoogLeNet,

    G. Cao, K. Zhang, K. Zhou, H. Pan, Y. Xu, and J. Liu, “A feature transferring fault diagnosis based on WPDR, FSWT and GoogLeNet,” in 2020 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), IEEE, 2020, pp. 1–6

    2020

  29. [37]

    A transfer convolutional neural network for fault diagnosis based on ResNet-50,

    L. Wen, X. Li, and L. Gao, “A transfer convolutional neural network for fault diagnosis based on ResNet-50,” Neural Comput. Appl., vol. 32, no. 10, pp. 6111–6124, 2020

    2020

  30. [38]

    Motor fault diagnosis algorithm based on wavelet and attention mechanism,

    Y. Yan, Q. Liu, and X. qin Gao, “Motor fault diagnosis algorithm based on wavelet and attention mechanism,” J. Sensors, vol. 2021, no. 1, p. 3782446, 2021

    2021

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.