Pith. sign in

REVIEW 4 major objections 5 minor 91 references

Explainable AI guided unsupervised fault diagnostics for high-voltage circuit breakers

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper shows that a healthy-data-only autoencoder, clustering, and Integrated-Gradients explanations can detect breaker faults, separate fault types, and indicate likely faulty mechanisms without any fault labels.

desk verdict Solid unsupervised detection and clustering on real CB data; the XAI 'diagnostics' claim is the soft spot because it is validated only against the pseudo-label classifier, not against the physical fault causes. read the letter →

arxiv 2507.19168 v1 pith:GFPISXZ6 submitted 2025-07-25 cs.LG eess.SP

Chi-Ching Hsu , Gaëtan Frusque , Florent Forest , Felipe Macedo , Christian M. Franck , Olga Fink This is my paper · ORCID
classification cs.LGeess.SP
keywords high-voltagecircuitbreakerconditionmonitoringunsupervisedfaultdetectionsegmentationdiagnosticsconvolutionalautoencodervibrationsignalstime-frequencyspectrogramsexplainableartificialintelligence
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

High-voltage circuit breakers fail rarely but with costly consequences, and most monitoring today relies on physical parameters such as gas pressure with preset thresholds, which miss many mechanical faults. This paper argues that an unsupervised pipeline can replace supervised fault diagnostics: a convolutional autoencoder trained only on healthy vibration and acoustic recordings detects deviations, clustering in the latent space separates fault types, and Integrated Gradients attributions through a classifier trained on cluster pseudo-labels show experts which time-frequency regions drive each cluster. On an experimental spring-drive breaker with four artificially induced spring/damper faults, the healthy-only detector catches 98.21% of faulty samples and K-means clustering reaches an adjusted Rand index of 0.9172. If this transfers to real substations, utilities could continuously monitor breakers and investigate emerging fault signatures without ever collecting a labeled failure dataset.

What carries the argument

The key machinery is a pseudo-label bridge that turns an unsupervised clustering result into a supervised explanation target. In step one, the convolutional autoencoder compresses each spectrogram into an 80-dimensional latent vector and the reconstruction residual with threshold $τ=μ+3σ$ is the fault detector. In step two, clustering on latent vectors produces pseudo-labels for every test sample, with K-means, OPTICS, and the self-organizing map as interchangeable instantiations. In step three, a one-layer softmax classifier is trained on the frozen latent features with those pseudo-labels as targets, which makes Integrated Gradients applicable; max-pooling the attribution maps yields the low-resolution diagnostics matrix. The pseudo-label bridge is what carries the argument: it converts the paper's unsupervised fault segmentation into a form that supervised XAI can interrogate, and it works regardless of which clustering algorithm produced the labels.

What would settle it

Collect healthy vibration and acoustic recordings from breakers in service across different seasons, gas pressures, and interrupted-current levels and feed them through the same healthy-trained CAE; if the mean residual distribution or its spread shifts so that a substantial fraction of healthy operations exceed $μ+3σ$, the fixed-threshold detection claim fails. A laboratory version would vary chamber gas pressure and temperature across healthy operations and measure how often the healthy-only threshold is crossed.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central finding is that fault diagnostics for high-voltage circuit breakers can be decomposed into three stages that require no fault labels. A convolutional autoencoder is trained on log-Mel spectrograms of size $128×100×4$ (three accelerometer directions plus a microphone) using only healthy opening operations, and the mean absolute reconstruction error of a test sample is compared with $τ=μ+3σ$ computed from the healthy training residuals; on the experimental data this detects 98.21% of faulty samples. The encoder's 80-dimensional latent vectors are then clustered with K-means (K=5), OPTICS, or a self-organizing map, separating healthy and fault conditions; K-means reaches an adjusted Rand index of 0.9172. To make the clusters explainable, the paper trains a one-layer softmax classifier on the frozen latent features using cluster assignments as pseudo-labels, then applies Integrated Gradients to this classifier; the resulting attributions are max-pooled into a 4×5 diagnostics matrix that localizes cluster-driving signal content in time and frequency. An occlusion-based faithfulness check shows that removing the most attributed spectrogram regions changes the classifier output more than removing random regions, supporting the claim that the explanations track the features that actually determine cluster membership, and the matrices show physically coherent signatures for spring-tension and damper-viscosity faults.

Load-bearing premise

The detection threshold assumes that healthy vibration and acoustic behavior stays stable over time and across operating conditions, so any deviation beyond a fixed cutoff is a fault; if ordinary healthy variation from temperature, gas pressure, or interrupted current is large enough, the detector would either miss faults or flood operators with false alarms.

Editorial extensions

If this is right

  • Fault monitoring can begin with only healthy recordings from an in-service breaker; no artificially induced faults or ground-truth labels are needed to train the detector.
  • Unknown or previously unseen fault types still get separated into clusters, and XAI explanation maps can indicate whether two clusters share a common culprit, such as low spring tension with different damper states.
  • The pseudo-label classifier is clustering-agnostic, so practitioners can choose K-means, OPTICS, or a self-organizing map and still obtain attributions for the clusters.
  • A single axial-direction accelerometer gives clustering performance close to the full four-sensor set, which lowers installation cost for real deployments.
  • Using all four sensors gives the best separation, with all clustering metrics above 0.9, so sensor fusion remains the reference configuration.

Reading between the lines

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

  • Editorial inference: the explanations are only as trustworthy as the clusters that generate the pseudo-labels; when clustering merges two fault subtypes, as low-spring samples with different damper viscosities do here, the attribution maps may highlight what separates clusters rather than what isolates the physical defect.
  • Editorial inference: the paper's own observation that healthy samples split into subclusters by experimental day points to a direct upgrade: replacing the fixed threshold with a healthy model conditioned on measured temperature, gas pressure, and interrupted current, so the detector adapts to normal environmental drift.
  • Editorial inference: because the pipeline needs only healthy recordings, a natural next test is fleet-level deployment in which one healthy model built from several same-type breakers serves as the baseline and any breaker whose latent trajectory drifts is flagged for inspection.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes an unsupervised condition-monitoring framework for high-voltage circuit breakers. A convolutional autoencoder (CAE) is trained on healthy-only time-frequency spectrograms; faulty samples are detected when the mean absolute reconstruction residual exceeds a fixed threshold tau = mean + 3 sigma. Faulty samples are then segmented by clustering CAE latent features with K-means, OPTICS, or a self-organizing map, yielding pseudo-labels. A shallow classifier is trained on those pseudo-labels and Integrated Gradients is applied to produce time-frequency 'diagnostics matrices' intended to explain cluster assignments and support fault diagnostics. The framework is evaluated on laboratory data from one CB under healthy, spring-tension, and damper-viscosity fault conditions. The reported fault detection rate for faulty samples is 98.21%, K-means clustering achieves an adjusted Rand index of 0.9172, and the XAI faithfulness experiment shows attribution-based occlusion changes classifier outputs more than random occlusion.

Significance. If the results hold, the framework would be a useful step toward monitoring CBs without requiring fault labels during training, which is a real operational constraint. The manuscript has concrete strengths: fault detection and segmentation are evaluated against external ground-truth labels with multiple metrics; three clustering algorithms are compared in both offline and online settings; the sensor ablation study is informative; and the online clustering experiments demonstrate that the pipeline can operate incrementally. The central caveat is that the XAI component is validated only as being faithful to a classifier trained on pseudo-labels, not as identifying physical fault causes. That gap, plus the unquantified stability of the healthy distribution and an inconsistency between two reported clustering scores, currently limits the strength of the contribution.

major comments (4)
  1. [Section 3.2, Eq. (4); Section 5.1, Figure 6] The fault detection claim relies on a fixed threshold tau = mean + 3 sigma computed on healthy training residuals, with the stated assumption that the healthy condition remains stable over time and that healthy-to-healthy deviations are smaller than healthy-to-faulty deviations. This assumption is load-bearing for the real-world detection claim, but it is not demonstrated with data. The manuscript itself notes that seasonal temperature, gas pressure, and interrupted current levels can affect healthy signals. Please provide a quantitative analysis of healthy-to-healthy residual variability across operating or environmental conditions, and report the false-positive rate on the healthy test samples, so that the detection threshold can be assessed.
  2. [Section 3.3, Section 4.2, Eq. (13), Figure 12] The XAI validation establishes only faithfulness of the Integrated Gradients attributions to the classifier C_theta_c, which is trained on pseudo-labels obtained from the same clustering step. It therefore does not demonstrate that the highlighted time-frequency regions correspond to the physical effects of the seeded faults, such as spring tension or damper viscosity. The qualitative comparison in Figure 11 is shown for one representative sample per condition and is not quantitatively evaluated. Moreover, the conclusion in Section 6 that the approach achieves diagnostics 'even if the fault type has not yet been observed' is not supported by any leave-one-fault-out experiment. Please add a quantitative link between attributions and known physical fault signatures, or a leave-one-fault-out study, and otherwise temper the diagnostics claims to cluster explanation.
  3. [Section 5.2, Table 3 and Section 5.5, Table 5] Tables 3 and 5 report different adjusted Rand index values for what appears to be the same configuration: K-means with K = 5 using all four sensors gives ARI = 0.9172 in Table 3 and ARI = 0.9045 in Table 5. No explanation is given for this discrepancy. Please clarify whether the sensor influence study uses a different data split, a different random initialization, or a different evaluation protocol; this is needed for reproducibility.
  4. [Section 5.1, Figure 6] The paper reports that approximately 98.21% of faulty samples are detected, i.e., a false-negative rate of 1.79%, but it does not report how many healthy test samples exceed the threshold. Since the test set includes healthy samples and the detection threshold is derived from healthy training residuals, the false-positive rate is essential for interpreting the detection performance. Please report the confusion between healthy and faulty samples at the chosen threshold.
minor comments (5)
  1. [Section 5.2, Figure 8] For the SOM clustering result, only the homogeneity score is reported among the external clustering metrics, while Table 3 reports ARI, completeness, and v-measure for K-means and OPTICS. Please report the full set of external metrics for SOM to enable a direct comparison.
  2. [Section 2.3] There is a typo: 'back-box model' should be 'black-box model'.
  3. [Section 5.4, Figure 12] The faithfulness plot would be easier to interpret with error bars or confidence intervals, since the averaging is over a small number of samples (approximately 340 operations in total).
  4. [Section 4.3] The manuscript states that the architecture is selected from a grid search but does not report the search space or the selection criterion. Adding this information would improve reproducibility.
  5. [Section 5.5, Table 5] The sentence 'The direction of the accelerometer installation does not show a significant difference based on this experimental dataset' is contradicted in part by the reported metrics, where the vertical and axial accelerometers outperform the horizontal one by approximately 0.03-0.04 in ARI. Please rephrase to describe the observed differences accurately.

Circularity Check

1 steps flagged · score 6.0 of 10

The XAI diagnostics validation is closed-loop: attributions are computed for a classifier fitted to the clustering pseudo-labels, so faithfulness only re-describes cluster separation rather than fault causes.

  1. fitted input called prediction [Section 3.3 (XAI-guided Fault Diagnostics) and Section 4.2 (Evaluation Metrics), Eq. (13)]
    "we create pseudo-labels for the test dataset, which contains both healthy samples and various fault types, based on the clustering results. ... The classifier Cθc(·) is trained using samples from the test dataset Dtest as inputs and the one-hot encoded cluster pseudo-labels y∈{ 0, 1}K assigned from the clustering results. ... IG is applied to this classifier to obtain feature attribution explanations for each test sample. ... faithfulness evaluation involves measuring the change in the classifierCθc(·) output when occluding the features selected by an explanation"

    The pseudo-labels are the output of the unsupervised clustering step, and the classifier is trained to reproduce exactly those pseudo-labels. Integrated Gradients is then applied to this classifier, and the only quantitative validation reported is faithfulness, Eq. (13), which measures how much occluding high-attribution pixels changes the output of the same pseudo-label classifier. By construction, this loop can only show that the attributions track the decision boundary of a model fitted to the clusters; it cannot show that the highlighted spectrogram regions correspond to aged or faulty components.

full rationale

The fault-detection and fault-segmentation components are not circular: the CAE is trained only on healthy data and its residuals are thresholded against an independent test set, and the clustering is scored against external ground-truth labels (ARI, homogeneity, completeness, v-measure). The circularity is confined to the XAI diagnostics claim. Section 3.3 generates pseudo-labels from the unsupervised clustering, trains the classifier on those exact pseudo-labels, and then applies Integrated Gradients to that classifier. Section 4.2 defines faithfulness as the change in this same classifier's output under occlusion of high-attribution pixels. Consequently the quantitative support for the 'diagnostics' is closed-loop: it demonstrates that the attributions track the decision boundary of a model fitted to the clusters, which is true by construction, and it does not demonstrate that the highlighted time-frequency regions correspond to spring or damper faults. The abstract's 'potential indications of the aged or faulty components' therefore rests on the qualitative pixel-wise comparisons in Figure 11 rather than on the reported faithfulness curves. No self-citation chain is load-bearing; the issue is the pseudo-label classifier loop. Overall score reflects partial circularity: detection and segmentation are independently validated, while the XAI diagnostics claim reduces, in its quantitative validation, to explaining the clustering output.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The framework introduces no new physical entities. It relies on standard ML components and a handcrafted diagnostics matrix. The main contributions are architectural combinations and an application to CB vibration data.

free parameters (5)
  • Detection threshold multiplier = 3 (threshold = healthy residual mean + 3 standard deviations)
    The 3-sigma multiplier is chosen by hand in Eq. (4); the threshold is computed from the healthy training residuals and directly controls the false negative and false positive trade-off.
  • Number of clusters K for K-means = 5
    K is set to 5 using elbow curve analysis in Section 4.3; the dataset has four fault types plus healthy, so K=5 includes the healthy cluster. The choice affects all segmentation metrics.
  • CAE architecture and latent dimension = convolutional depths 16, 8, 1; latent 16x5 (80 dims)
    Selected via grid search in Section 4.3; the latent space feeds clustering, so its geometry directly determines segmentation quality.
  • Max pooling size for diagnostics matrix = (32, 20) giving 5 time intervals and 4 frequency bands
    Chosen by hand in Section 4.3 to create a human-readable diagnostics matrix; affects the resolution and interpretation of the XAI maps.
  • SOM grid and parameters = 10x10 grid, spread sigma 5, learning rate 0.05
    Set in Section 4.3 for the SOM variant; not load-bearing for the main K-means results but reported for completeness.
assumptions (4)
  • domain assumption Fixed-threshold detection assumes the healthy data distribution is stable over time.
    Section 3.2 states a fixed threshold is used because CBs are operated infrequently and the healthy condition remains stable; seasonal temperature and gas pressure changes are acknowledged but not modeled. If healthy variation exceeds the 3-sigma threshold, false alarms or missed faults would result in the field.
  • domain assumption The domain gap between CBs of the same type is small compared to healthy-versus-faulty differences.
    Section 5 states this assumption to justify transferring results from a single CB to other units of the same type; it is untested and load-bearing for real-world deployment.
  • domain assumption Clusters from the CAE latent space correspond to distinct fault types.
    The clustering is performed on latent features learned only from healthy data; the paper evaluates against ground-truth labels and finds most clusters align, but the mapping is not guaranteed for unseen fault types.
  • ad hoc to paper Pseudo-labels from clustering are valid targets for the explanation classifier.
    The classifier is trained to reproduce cluster assignments, so the Integrated Gradients explanations describe the clustering boundary, not necessarily the physical fault mechanism. This is the basis of the XAI diagnostics step.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explainable AI guided unsupervised fault diagnostics for high-voltage circuit breakers." pith.science (2026). https://pith.science/paper/GFPISXZ6

@misc{pith2026250719168,
  author       = {Pith},
  title        = {Pith review of: Explainable AI guided unsupervised fault diagnostics for high-voltage circuit breakers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GFPISXZ6}},
  note         = {Machine review of arXiv:2507.19168}
}
read the original abstract

Commercial high-voltage circuit breaker (CB) condition monitoring systems rely on directly observable physical parameters such as gas filling pressure with pre-defined thresholds. While these parameters are crucial, they only cover a small subset of malfunctioning mechanisms and usually can be monitored only if the CB is disconnected from the grid. To facilitate online condition monitoring while CBs remain connected, non-intrusive measurement techniques such as vibration or acoustic signals are necessary. Currently, CB condition monitoring studies using these signals typically utilize supervised methods for fault diagnostics, where ground-truth fault types are known due to artificially introduced faults in laboratory settings. This supervised approach is however not feasible in real-world applications, where fault labels are unavailable. In this work, we propose a novel unsupervised fault detection and segmentation framework for CBs based on vibration and acoustic signals. This framework can detect deviations from the healthy state. The explainable artificial intelligence (XAI) approach is applied to the detected faults for fault diagnostics. The specific contributions are: (1) we propose an integrated unsupervised fault detection and segmentation framework that is capable of detecting faults and clustering different faults with only healthy data required during training (2) we provide an unsupervised explainability-guided fault diagnostics approach using XAI to offer domain experts potential indications of the aged or faulty components, achieving fault diagnostics without the prerequisite of ground-truth fault labels. These contributions are validated using an experimental dataset from a high-voltage CB under healthy and artificially introduced fault conditions, contributing to more reliable CB system operation.

Figures

Figures reproduced from arXiv: 2507.19168 by the authors.

Figure 1
Figure 1. Scope of fault detection, segmentation, and diagnostics, modified from [24]. Fault detection aims [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Overall proposed framework. In step 1 (fault detection), a convolutional autoencoder (CAE) is [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Experimental setup with the test CB, the installation location of microphone, vibration sensors, [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Installation locations of three vibration sensors (piezoelectric accelerometers) as depicted in Ta￾ble 1, adhesive mounted with LOCTITE in three directions with respect to the drive [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: Example of vibration signals in the vertical direction for all five conditions (upper row) and their [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: The fault detection results based on the CAE residuals with healthy (green) and faulty samples [PITH_FULL_IMAGE:figures/full_fig_p018_6.png]
Figure 7
Figure 7. Figure 7: Offline clustering results using K-means (left) and OPTICS (right) with the CAE latent features, visualized in 2D using t-SNE. Colors are assigned from the clustering results, while markers are the ground￾truth conditions as defined in [PITH_FULL_IMAGE:figures/full_fi…
Figure 8
Figure 8. Figure 8: Offline clustering results using SOM with grid size (10, 10), colored by the ground-truth labels. [PITH_FULL_IMAGE:figures/full_fig_p020_8.png]
Figure 9
Figure 9. Figure 9: Clustering results using OPTICS with the CAE latent features over an incremental data stream [PITH_FULL_IMAGE:figures/full_fig_p021_9.png]
Figure 10
Figure 10. Figure 10: Clustering results using SOM with the CAE latent features over an incremental data stream [PITH_FULL_IMAGE:figures/full_fig_p021_10.png]
Figure 11
Figure 11. Figure 11: The representative instances for each condition are selected as the closest samples to each cluster centroid in the CAE latent space. Each column corresponds to different condition: (a) Condition #1 nSnD, (b) Condition #2 nSdD, (c) Condition #3 hSnD, (d) Condition #4 …
Figure 12
Figure 12. Figure 12: Faithfulness of the explanations evaluated by representing the change in classifier prediction [PITH_FULL_IMAGE:figures/full_fig_p025_12.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

91 extracted references · 67 canonical work pages

  1. [1]

    A general model, estimation, and procedure for modeling recurrent failure process of high-voltage circuit breakers considering multivariate impacts,

    W. Hu, P. Westerlund, P. Hilber, C. Chen, and Z. Yang, “A general model, estimation, and procedure for modeling recurrent failure process of high-voltage circuit breakers considering multivariate impacts,” Reliability engineering & system safety , vol. 220, p. 108276, 2022. 26

    2022

  2. [2]

    Condition monitoring of high voltage circuit breakers: Past to future,

    A. A. Razi-Kazemi and K. Niayesh, “Condition monitoring of high voltage circuit breakers: Past to future,” IEEE Transactions on Power Delivery , vol. 36, no. 2, pp. 740–750, 2020

    2020

  3. [3]

    Circuit breaker condition assessment through a fuzzy-probabilistic analysis of actuating coil’s current,

    A. A. Razi-Kazemi, “Circuit breaker condition assessment through a fuzzy-probabilistic analysis of actuating coil’s current,” IET Generation, Transmission & Distribution , vol. 10, no. 1, pp. 48–56, 2016

    2016

  4. [4]

    An approach for hvcb mechanical fault diag- nosis based on a deep belief network and a transfer learning strategy,

    Y. Pan, F. Mei, H. Miao, J. Zheng, K. Zhu, and H. Sha, “An approach for hvcb mechanical fault diag- nosis based on a deep belief network and a transfer learning strategy,”Journal of Electrical Engineering & Technology, vol. 14, pp. 407–419, 2019

    2019

  5. [5]

    Applicability of auxiliary contacts in circuit breaker online condition assessment,

    A. Razi-Kazemi, “Applicability of auxiliary contacts in circuit breaker online condition assessment,” Electric power systems research, vol. 128, pp. 53–59, 2015

    2015

  6. [6]

    Fault analysis of high-voltage circuit breakers based on coil current and contact travel waveforms through modified svm classifier,

    F. N. Rudsari, A. A. Razi-Kazemi, and M. A. Shoorehdeli, “Fault analysis of high-voltage circuit breakers based on coil current and contact travel waveforms through modified svm classifier,” IEEE Transactions on Power Delivery , vol. 34, no. 4, pp. 1608–1618, 2019

    2019

  7. [7]

    Intelligent failure diagnosis for gas circuit breakers based on dynamic resistance measurements,

    M. Abdollah and A. A. Razi-Kazemi, “Intelligent failure diagnosis for gas circuit breakers based on dynamic resistance measurements,” IEEE Transactions on Instrumentation and Measurement , vol. 68, no. 9, pp. 3066–3077, 2018

    2018

  8. [8]

    Prediction of the dynamic contact resistance of circuit breaker based on the kernel partial least squares,

    Y. Liu, G. Zhang, H. Qin, Y. Geng, J. Wang, J. Yang, and K. Zhao, “Prediction of the dynamic contact resistance of circuit breaker based on the kernel partial least squares,” IET Generation, Transmission & Distribution, vol. 12, no. 8, pp. 1815–1821, 2018

    2018

Show all 91 references
  1. [9]

    Timings of high voltage circuit-breaker,

    B. Rusek, G. Balzer, M. Holstein, and M.-S. Claessens, “Timings of high voltage circuit-breaker,” Electric power systems research, vol. 78, no. 12, pp. 2011–2016, 2008

    2011

  2. [10]

    Study of condition monitoring by operating sound diagnosis for circuit breaker in substation,

    T. Sugimoto and M. Shimizu, “Study of condition monitoring by operating sound diagnosis for circuit breaker in substation,” in 2019 5th International Conference on Electric Power Equipment-Switching Technology (ICEPE-ST). IEEE, 2019, pp. 622–625

    2019

  3. [11]

    Development of acoustic diagnostics for opening and closing operations of gas circuit breakers,

    T. Iwata, T. Endo, J. Nukaga, Y. Takahashi, and T. Nishimura, “Development of acoustic diagnostics for opening and closing operations of gas circuit breakers,” in 2022 6th International Conference on Electric Power Equipment-Switching Technology (ICEPE-ST) . IEEE, 2022, pp. 67–71

    2022

  4. [12]

    Field circuit breaker inspection using machine learning and data analytics on sound recognition,

    S. Darnsomboon and W. Boon-Nontae, “Field circuit breaker inspection using machine learning and data analytics on sound recognition,” CIGRE 2022 Session , no. D2-PS1-10991, 2022

    2022

  5. [13]

    A novel u-net and capsule network for few-shot high-voltage circuit breaker mechanical fault diagnosis,

    X. Ye, J. Yan, Y. Wang, J. Wang, and Y. Geng, “A novel u-net and capsule network for few-shot high-voltage circuit breaker mechanical fault diagnosis,” Measurement, vol. 199, p. 111527, 2022

    2022

  6. [14]

    Continuous monitoring of circuit breakers using vibration analysis,

    H. Hoidalen and M. Runde, “Continuous monitoring of circuit breakers using vibration analysis,” IEEE Transactions on Power Delivery , vol. 20, no. 4, pp. 2458–2465, 2005

    2005

  7. [15]

    Mechanical fault diagnosis of a high voltage circuit breaker based on high-efficiency time-domain feature extraction with entropy features,

    J. Qi, X. Gao, and N. Huang, “Mechanical fault diagnosis of a high voltage circuit breaker based on high-efficiency time-domain feature extraction with entropy features,” Entropy, vol. 22, no. 4, p. 478, 2020

    2020

  8. [16]

    Review of digital vibration signal analysis techniques for fault diagnosis of high-voltage circuit breakers,

    Y. Tan, E. Hu, Y. Liu, J. Li, and W. Chen, “Review of digital vibration signal analysis techniques for fault diagnosis of high-voltage circuit breakers,” IEEE Transactions on Dielectrics and Electrical Insulation, 2023

    2023

  9. [17]

    A systematic review for switchgear asset management in power grids: Condition monitoring, health assessment, and maintenance strategy,

    N. Zhou, Y. Xu, S. Cho, and C. T. Wee, “A systematic review for switchgear asset management in power grids: Condition monitoring, health assessment, and maintenance strategy,” IEEE Transactions on Power Delivery , 2023

    2023

  10. [18]

    Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models,

    M. Arias Chao, C. Kulkarni, K. Goebel, and O. Fink, “Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models,”International Journal of Prognostics and Health Management , 2019

    2019

  11. [19]

    Domain adaptation via alignment of operation profile for remaining useful lifetime prediction,

    I. Nejjar, F. Geissmann, M. Zhao, C. Taal, and O. Fink, “Domain adaptation via alignment of operation profile for remaining useful lifetime prediction,” Reliability Engineering & System Safety , vol. 242, p. 109718, 2024

    2024

  12. [20]

    Reliability analysis of detecting false alarms that employ neural networks: A real case study on wind turbines,

    A. P. Marug´ an, A. M. P. Chac´ on, and F. P. G. M´ arquez, “Reliability analysis of detecting false alarms that employ neural networks: A real case study on wind turbines,” Reliability Engineering & System Safety, vol. 191, p. 106574, 2019. 27

    2019

  13. [21]

    Dirt and mud detection and diagnosis on a wind turbine blade employing guided waves and supervised learning classifiers,

    A. A. Jim´ enez, C. Q. G. Mu˜ noz, and F. P. G. M´ arquez, “Dirt and mud detection and diagnosis on a wind turbine blade employing guided waves and supervised learning classifiers,” Reliability Engineering & System Safety , vol. 184, pp. 2–12, 2019

    2019

  14. [22]

    Non-contact sensing for anomaly detection in wind turbine blades: A focus-svdd with complex-valued auto-encoder approach,

    G. Frusque, D. Mitchell, J. Blanche, D. Flynn, and O. Fink, “Non-contact sensing for anomaly detection in wind turbine blades: A focus-svdd with complex-valued auto-encoder approach,”Mechanical Systems and Signal Processing, vol. 208, p. 111022, 2024

    2024

  15. [23]

    Condition assessment of power circuit breakers based on machine learning algorithms,

    K. Obarcanin, D. Skulj, and B. Lacevic, “Condition assessment of power circuit breakers based on machine learning algorithms,” IEEE Transactions on Power Delivery , 2023

    2023

  16. [24]

    Implicit supervision for fault detection and segmentation of emerging fault types with deep variational autoencoders,

    M. A. Chao, B. T. Adey, and O. Fink, “Implicit supervision for fault detection and segmentation of emerging fault types with deep variational autoencoders,” Neurocomputing, vol. 454, pp. 324–338, 2021

    2021

  17. [25]

    A comparison of residual-based methods on fault detection,

    C.-C. Hsu, G. Frusque, and O. Fink, “A comparison of residual-based methods on fault detection,” in Annual Conference of the PHM Society , vol. 15, no. 1, 2023

    2023

  18. [26]

    Fault diagnosis of circuit breaker energy storage mechanism based on current- vibration entropy weight characteristic and grey wolf optimization–support vector machine,

    S. Zhao and E. Wang, “Fault diagnosis of circuit breaker energy storage mechanism based on current- vibration entropy weight characteristic and grey wolf optimization–support vector machine,” Ieee Ac- cess, vol. 7, pp. 86 798–86 809, 2019

    2019

  19. [27]

    Prognostics and health management (phm): Where are we and where do we (need to) go in theory and practice,

    E. Zio, “Prognostics and health management (phm): Where are we and where do we (need to) go in theory and practice,” Reliability Engineering & System Safety , vol. 218, p. 108119, 2022

    2022

  20. [28]

    Sensitivity analysis by differential importance measure for unsupervised fault diagnostics,

    G. Floreale, P. Baraldi, X. Lu, P. Rossetti, and E. Zio, “Sensitivity analysis by differential importance measure for unsupervised fault diagnostics,” Reliability Engineering & System Safety , vol. 243, p. 109846, 2024

    2024

  21. [29]

    Integration of novel sensors and machine learning for predictive main- tenance in medium voltage switchgear to enable the energy and mobility revolutions,

    M. W. Hoffmann, S. Wildermuth, R. Gitzel, A. Boyaci, J. Gebhardt, H. Kaul, I. Amihai, B. Forg, M. Suriyah, T. Leibfried et al., “Integration of novel sensors and machine learning for predictive main- tenance in medium voltage switchgear to enable the energy and mobility revolu...

    2020

  22. [30]

    Deep learning for anomaly detection: A review,

    G. Pang, C. Shen, L. Cao, and A. V. D. Hengel, “Deep learning for anomaly detection: A review,” ACM computing surveys (CSUR) , vol. 54, no. 2, pp. 1–38, 2021

    2021

  23. [31]

    Auto-encoding variational bayes,

    D. P. Kingma and M. Welling, “Auto-encoding variational bayes,” arXiv preprint arXiv:1312.6114 , 2013

    2013 arXiv

  24. [32]

    Anomaly detection and diagnosis for wind turbines using long short- term memory-based stacked denoising autoencoders and xgboost,

    C. Zhang, D. Hu, and T. Yang, “Anomaly detection and diagnosis for wind turbines using long short- term memory-based stacked denoising autoencoders and xgboost,” Reliability Engineering & System Safety, vol. 222, p. 108445, 2022

    2022

  25. [33]

    A method for fault detection in multi-component systems based on sparse autoencoder-based deep neural networks,

    Z. Yang, P. Baraldi, and E. Zio, “A method for fault detection in multi-component systems based on sparse autoencoder-based deep neural networks,” Reliability Engineering & System Safety , vol. 220, p. 108278, 2022

    2022

  26. [34]

    Estimating the support of a high-dimensional distribution,

    B. Sch¨ olkopf, J. C. Platt, J. Shawe-Taylor, A. J. Smola, and R. C. Williamson, “Estimating the support of a high-dimensional distribution,” Neural computation, vol. 13, no. 7, pp. 1443–1471, 2001

    2001

  27. [35]

    Support vector data description,

    D. M. Tax and R. P. Duin, “Support vector data description,” Machine learning, vol. 54, pp. 45–66, 2004

    2004

  28. [36]

    Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings,

    P. Bergmann, M. Fauser, D. Sattlegger, and C. Steger, “Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings,” in Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition , 2020, pp. 4183–4192

    2020

  29. [37]

    Multiresolution knowledge distillation for anomaly detection,

    M. Salehi, N. Sadjadi, S. Baselizadeh, M. H. Rohban, and H. R. Rabiee, “Multiresolution knowledge distillation for anomaly detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 14 902–14 912

    2021

  30. [38]

    Anomaly detection via reverse distillation from one-class embedding,

    H. Deng and X. Li, “Anomaly detection via reverse distillation from one-class embedding,” in Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 9737–9746

    2022

  31. [39]

    A unifying review of deep and shallow anomaly detection,

    L. Ruff, J. R. Kauffmann, R. A. Vandermeulen, G. Montavon, W. Samek, M. Kloft, T. G. Dietterich, and K.-R. M¨ uller, “A unifying review of deep and shallow anomaly detection,”Proceedings of the IEEE, vol. 109, no. 5, pp. 756–795, 2021

    2021

  32. [40]

    Improving unsupervised defect seg- mentation by applying structural similarity to autoencoders,

    P. Bergmann, S. L¨ owe, M. Fauser, D. Sattlegger, and C. Steger, “Improving unsupervised defect seg- mentation by applying structural similarity to autoencoders,” arXiv preprint arXiv:1807.02011 , 2018. 28

    2018 arXiv

  33. [41]

    Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection,

    P. Bergmann, M. Fauser, D. Sattlegger, and C. Steger, “Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 9592–9600

    2019

  34. [42]

    A novel method for fault diagnosis of fluid end of drilling pump under complex working conditions,

    G. Li, J. Hu, Y. Ding, A. Tang, J. Ao, D. Hu, and Y. Liu, “A novel method for fault diagnosis of fluid end of drilling pump under complex working conditions,” Reliability Engineering & System Safety , vol. 248, p. 110145, 2024

    2024

  35. [43]

    Attention-based multiscale denoising residual convolutional neural networks for fault diagnosis of rotating machinery,

    Y. Xu, X. Yan, K. Feng, X. Sheng, B. Sun, and Z. Liu, “Attention-based multiscale denoising residual convolutional neural networks for fault diagnosis of rotating machinery,” Reliability Engineering & System Safety, vol. 226, p. 108714, 2022

    2022

  36. [44]

    Mechanical fault diagnosis of high voltage circuit breakers based on wavelet time-frequency entropy and one-class support vector machine,

    N. Huang, H. Chen, S. Zhang, G. Cai, W. Li, D. Xu, and L. Fang, “Mechanical fault diagnosis of high voltage circuit breakers based on wavelet time-frequency entropy and one-class support vector machine,” Entropy, vol. 18, no. 1, p. 7, 2015

    2015

  37. [45]

    Condition evaluation for opening damper of spring operated high-voltage circuit breaker using vibration time-frequency image,

    Q. Yang, J. Ruan, Z. Zhuang, and D. Huang, “Condition evaluation for opening damper of spring operated high-voltage circuit breaker using vibration time-frequency image,” IEEE Sensors Journal , vol. 19, no. 18, pp. 8116–8126, 2019

    2019

  38. [46]

    Small-sample fault diagnosis method for high-voltage circuit breakers via data augmentation and deep learning,

    Q. Yang, Z. Wang, J. Ruan, and Z. Zhuang, “Small-sample fault diagnosis method for high-voltage circuit breakers via data augmentation and deep learning,” IEEE Transactions on Instrumentation and Measurement, 2024

    2024

  39. [47]

    Few-shot transfer learning with attention mechanism for high-voltage circuit breaker fault diagnosis,

    Y. Wang, J. Yan, X. Ye, Q. Jing, J. Wang, and Y. Geng, “Few-shot transfer learning with attention mechanism for high-voltage circuit breaker fault diagnosis,” IEEE Transactions on Industry Applica- tions, vol. 58, no. 3, pp. 3353–3360, 2022

    2022

  40. [48]

    Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI,

    A. B. Arrieta, N. D´ ıaz-Rodr´ ıguez, J. Del Ser, A. Bennetot, S. Tabik, A. Barbado, S. Garc´ ıa, S. Gil-L´ opez, D. Molina, R. Benjamins, R. Chatila, and F. Herrera, “Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsi...

    2019 arXiv

  41. [49]

    Explainable Predictive Maintenance,

    S. Pashami, S. Nowaczyk, Y. Fan, J. Jakubowski, N. Paiva, N. Davari, S. Bobek, S. Jamshidi, H. Sarmadi, A. Alabdallah, R. P. Ribeiro, B. Veloso, M. Sayed-Mouchaweh, L. Rajaoarisoa, G. J. Nalepa, and J. Gama, “Explainable Predictive Maintenance,” Jun. 2023, arXiv:2306.05120 [cs...

    2023 arXiv

  42. [50]

    Explainable Predictive Maintenance: A Survey of Current Methods, Challenges and Opportunities,

    L. Cummins, A. Sommers, S. B. Ramezani, S. Mittal, J. Jabour, M. Seale, and S. Rahimi, “Explainable Predictive Maintenance: A Survey of Current Methods, Challenges and Opportunities,” Jan. 2024, arXiv:2401.07871 [cs]. [Online]. Available: http://arxiv.org/abs/2401.07871

    2024 arXiv

  43. [51]

    Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications,

    W. Samek, G. Montavon, S. Lapuschkin, C. J. Anders, and K.-R. M¨ uller, “Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications,” Proceedings of the IEEE, vol. 109, no. 3, pp. 247–278, Mar. 2021, proceedings of the IEEE. [Online]. Available: https://i...

    2021

  44. [52]

    Visualizing and understanding convolutional networks,

    M. D. Zeiler and R. Fergus, “Visualizing and understanding convolutional networks,” in Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceed- ings, Part I 13 . Springer, 2014, pp. 818–833

    2014

  45. [53]

    How to explain individual classification decisions,

    D. Baehrens, T. Schroeter, S. Harmeling, M. Kawanabe, K. Hansen, and K.-R. M¨ uller, “How to explain individual classification decisions,” Journal of Machine Learning Research, vol. 11, no. 61, pp. 1803–1831, 2010. [Online]. Available: http://jmlr.org/papers/v11/baehrens10a.html

    2010

  46. [54]

    Axiomatic attribution for deep networks,

    M. Sundararajan, A. Taly, and Q. Yan, “Axiomatic attribution for deep networks,” in International conference on machine learning. PMLR, 2017, pp. 3319–3328

    2017

  47. [55]

    Layer-wise relevance propaga- tion: an overview,

    G. Montavon, A. Binder, S. Lapuschkin, W. Samek, and K.-R. M¨ uller, “Layer-wise relevance propaga- tion: an overview,” Explainable AI: interpreting, explaining and visualizing deep learning , pp. 193–209, 2019

    2019

  48. [56]

    On pixel-wise expla- nations for non-linear classifier decisions by layer-wise relevance propagation,

    S. Bach, A. Binder, G. Montavon, F. Klauschen, K.-R. M¨ uller, and W. Samek, “On pixel-wise expla- nations for non-linear classifier decisions by layer-wise relevance propagation,” PloS one, vol. 10, no. 7, p. e0130140, 2015

    2015

  49. [57]

    An explainable one-dimensional 29 convolutional neural networks based fault diagnosis method for building heating, ventilation and air conditioning systems,

    G. Li, Q. Yao, C. Fan, C. Zhou, G. Wu, Z. Zhou, and X. Fang, “An explainable one-dimensional 29 convolutional neural networks based fault diagnosis method for building heating, ventilation and air conditioning systems,” Building and Environment , vol. 203, p. 108057, Oct. 2021...

    2021

  50. [58]

    Decoupled Feature-Temporal CNN: Explaining Deep Learning-Based Machine Health Monitoring,

    C. Zhu, Z. Chen, R. Zhao, J. Wang, and R. Yan, “Decoupled Feature-Temporal CNN: Explaining Deep Learning-Based Machine Health Monitoring,” IEEE Transactions on Instrumentation and Measurement , vol. 70, pp. 1–13, 2021. [Online]. Available: https://ieeexplore.ieee.org/abstract/...

    2021

  51. [59]

    Explainable Deep Ensemble Model for Bearing Fault Diagnosis Under Variable Conditions,

    Z. Chen, W. Qin, G. He, J. Li, R. Huang, G. Jin, and W. Li, “Explainable Deep Ensemble Model for Bearing Fault Diagnosis Under Variable Conditions,” IEEE Sensors Journal , vol. 23, no. 15, pp. 17 737–17 750, Aug. 2023. [Online]. Available: https://ieeexplore.ieee.org/abstract/...

    2023

  52. [60]

    On the explainability of convolutional neural networks processing ultrasonic guided waves for damage diagnosis,

    L. Lomazzi, S. Fabiano, M. Parziale, M. Giglio, and F. Cadini, “On the explainability of convolutional neural networks processing ultrasonic guided waves for damage diagnosis,” Mechanical Systems and Signal Processing, vol. 183, p. 109642, Jan. 2023, publisher: Academic Press....

    2023

  53. [61]

    Vibration Signals Analysis by Explainable Artificial In- telligence (XAI) Approach: Application on Bearing Faults Diagnosis,

    H.-Y. Chen and C.-H. Lee, “Vibration Signals Analysis by Explainable Artificial In- telligence (XAI) Approach: Application on Bearing Faults Diagnosis,” IEEE Access , vol. 8, pp. 134 246–134 256, 2020, conference Name: IEEE Access. [Online]. Available: https://ieeexplore.ieee....

    2020

  54. [62]

    An Explainable Neural Network for Fault Diagnosis With a Frequency Activation Map,

    M. S. Kim, J. P. Yun, and P. Park, “An Explainable Neural Network for Fault Diagnosis With a Frequency Activation Map,” IEEE Access, vol. 9, pp. 98 962–98 972, 2021, conference Name: IEEE Access. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/9477565

    2021

  55. [63]

    Does Your Model Think Like an Engineer? Explainable AI for Bearing Fault Detection with Deep Learning,

    T. Decker, M. Lebacher, and V. Tresp, “Does Your Model Think Like an Engineer? Explainable AI for Bearing Fault Detection with Deep Learning,” in ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , Jun. 2023, pp. 1–5. [Online]...

    2023

  56. [64]

    Explainable Convolutional Neural Network for Gearbox Fault Diagnosis,

    J. Grezmak, P. Wang, C. Sun, and R. X. Gao, “Explainable Convolutional Neural Network for Gearbox Fault Diagnosis,” Procedia CIRP, vol. 80, pp. 476–481, Jan. 2019. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2212827118312873

    2019

  57. [65]

    A Study on the Effectiveness of Current Data in Motor Mechanical Fault Diagnosis Using XAI,

    J.-H. Han, S.-U. Park, and S.-K. Hong, “A Study on the Effectiveness of Current Data in Motor Mechanical Fault Diagnosis Using XAI,” Journal of Electrical Engineering & Technology , vol. 17, no. 6, pp. 3329–3335, Nov. 2022. [Online]. Available: https://doi.org/10.1007/s42835-0...

    2022 doi

  58. [66]

    Explainable AI for Bearing Fault Prognosis Using Deep Learning Techniques,

    D. C. Sanakkayala, V. Varadarajan, N. Kumar, Karan, G. Soni, P. Kamat, S. Kumar, S. Patil, and K. Kotecha, “Explainable AI for Bearing Fault Prognosis Using Deep Learning Techniques,” Micromachines, vol. 13, no. 9, p. 1471, Sep. 2022. [Online]. Available: https://www.mdpi.com/...

    2022

  59. [67]

    Fault feature assessment method for high-voltage circuit breakers based on explainable image recognition,

    Y. Tan, J. Gong, S. Li, J. Li, and W. Chen, “Fault feature assessment method for high-voltage circuit breakers based on explainable image recognition,” IEEE Transactions on Dielectrics and Electrical Insulation, 2024

    2024

  60. [68]

    Explaining the predictions of unsupervised learning models,

    G. Montavon, J. Kauffmann, W. Samek, and K.-R. M¨ uller, “Explaining the predictions of unsupervised learning models,” in International Workshop on Extending Explainable AI Beyond Deep Models and Classifiers. Springer, 2020, pp. 117–138

    2020

  61. [69]

    From fault detection to anomaly explanation: A case study on predictive maintenance,

    J. Gama, R. P. Ribeiro, S. Mastelini, N. Davari, and B. Veloso, “From fault detection to anomaly explanation: A case study on predictive maintenance,” Journal of Web Semantics , vol. 81, p. 100821, Jul. 2024. [Online]. Available: https://www.sciencedirect.com/science/article/p...

    2024

  62. [70]

    From clustering to cluster explanations via neural networks,

    J. Kauffmann, M. Esders, L. Ruff, G. Montavon, W. Samek, and K.-R. M¨ uller, “From clustering to cluster explanations via neural networks,” IEEE Transactions on Neural Networks and Learning Systems, 2022

    2022

  63. [71]

    Concept Bottleneck Models,

    P. W. Koh, T. Nguyen, Y. S. Tang, S. Mussmann, E. Pierson, B. Kim, and P. Liang, “Concept Bottleneck Models,” in Proceedings of the 37 th International Conference on Machine Learning , 2020

    2020

  64. [72]

    Interpretable Prognostics with Concept Bottleneck Models,

    F. Forest, K. Rombach, and O. Fink, “Interpretable Prognostics with Concept Bottleneck Models,” May 2024, arXiv:2405.17575 [cs, eess, stat]. [Online]. Available: http://arxiv.org/abs/2405.17575 30

    2024 arXiv

  65. [73]

    Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictions,

    O. Li, H. Liu, C. Chen, and C. Rudin, “Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictions,” Nov. 2017, arXiv:1710.04806 [cs, stat]. [Online]. Available: http://arxiv.org/abs/1710.04806

    2017 arXiv

  66. [74]

    This Looks Like That: Deep Learning for Interpretable Image Recognition,

    C. Chen, O. Li, D. Tao, A. Barnett, C. Rudin, and J. K. Su, “This Looks Like That: Deep Learning for Interpretable Image Recognition,” in Advances in Neural In- formation Processing Systems , vol. 32. Curran Associates, Inc., 2019. [Online]. Available: https://proceedings.neur...

    2019

  67. [75]

    Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review,

    S. Verma, V. Boonsanong, M. Hoang, K. E. Hines, J. P. Dickerson, and C. Shah, “Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review,” Nov. 2022, arXiv:2010.10596 [cs, stat]. [Online]. Available: http://arxiv.org/abs/2010.10596

    2022 arXiv

  68. [76]

    Scf-net: A sparse counterfactual generation network for interpretable fault diagnosis,

    J. F. Barraza, E. L. Droguett, and M. R. Martins, “Scf-net: A sparse counterfactual generation network for interpretable fault diagnosis,” Reliability Engineering & System Safety , vol. 250, p. 110285, 2024

    2024

  69. [77]

    Optics: Ordering points to identify the clustering structure,

    M. Ankerst, M. M. Breunig, H.-P. Kriegel, and J. Sander, “Optics: Ordering points to identify the clustering structure,” ACM Sigmod record, vol. 28, no. 2, pp. 49–60, 1999

    1999

  70. [78]

    The self-organizing map,

    T. Kohonen, “The self-organizing map,” Proceedings of the IEEE, vol. 78, no. 9, pp. 1464–1480, 1990

    1990

  71. [79]

    Large-scale Vibration Monitoring of Aircraft Engines from Operational Data using Self-organized Models,

    F. Forest, Q. Cochard, C. Noyer, M. Joncour, J. Lacaille, M. Lebbah, and H. Azzag, “Large-scale Vibration Monitoring of Aircraft Engines from Operational Data using Self-organized Models,” in Annual Conference of the PHM Society , vol. 12, Nov. 2020, p. 11. [Online]. Available...

    2020

  72. [80]

    From classification to segmentation with explainable ai: A study on crack detection and growth monitoring,

    F. Forest, H. Porta, D. Tuia, and O. Fink, “From classification to segmentation with explainable ai: A study on crack detection and growth monitoring,” Automation in Construction , vol. 165, p. 105497, 2024

    2024

  73. [81]

    Carvalho, M

    A. Carvalho, M. L. Cormenzana, H. Furuta, W. Grieshaber, A. Hyrczak, D. Kopejtkova, J. Krone, M. Kudoke, D. Makareinis, J. Martins, K. Mestrovic, I. Ohno, J. Ostlund, K. Park, J. Patel, C. Protze, M. Runde, J. Schmid, J. Skog, C. Solver, B. Sweeney, and F. Waite, “Cigre techni...

    2004

  74. [82]

    Diagnostics of high voltage circuit breakers by monitoring of vibration signals,

    F. Macedo, T. Christen, and A. Garyfallos, “Diagnostics of high voltage circuit breakers by monitoring of vibration signals,” CIGRE Condition Monitoring, Diagnosis and Maintenance, CMDM2023 , 2023

    2023

  75. [83]

    Mimii dataset: Sound dataset for malfunctioning industrial machine investigation and inspection,

    H. Purohit, R. Tanabe, K. Ichige, T. Endo, Y. Nikaido, K. Suefusa, and Y. Kawaguchi, “Mimii dataset: Sound dataset for malfunctioning industrial machine investigation and inspection,” arXiv preprint arXiv:1909.09347, 2019

    1909 arXiv

  76. [84]

    Speech emotion recognition from 3d log-mel spectrograms with deep learning network,

    H. Meng, T. Yan, F. Yuan, and H. Wei, “Speech emotion recognition from 3d log-mel spectrograms with deep learning network,” IEEE access, vol. 7, pp. 125 868–125 881, 2019

    2019

  77. [85]

    Comparing partitions,

    L. Hubert and P. Arabie, “Comparing partitions,” Journal of classification , vol. 2, pp. 193–218, 1985

    1985

  78. [86]

    V-measure: A conditional entropy-based external cluster evalua- tion measure,

    A. Rosenberg and J. Hirschberg, “V-measure: A conditional entropy-based external cluster evalua- tion measure,” in Proceedings of the 2007 joint conference on empirical methods in natural language processing and computational natural language learning (EMNLP-CoNLL) , 2007, pp. 410–420

    2007

  79. [87]

    A Survey and Implementation of Performance Metrics for Self-Organized Maps,

    F. Forest, M. Lebbah, H. Azzag, and J. Lacaille, “A Survey and Implementation of Performance Metrics for Self-Organized Maps,” Nov. 2020. [Online]. Available: http://arxiv.org/abs/2011.05847

    2020 arXiv

  80. [88]

    Towards robust interpretability with self-explaining neural net- works,

    D. Alvarez Melis and T. Jaakkola, “Towards robust interpretability with self-explaining neural net- works,” Advances in neural information processing systems , vol. 31, 2018

    2018

  81. [89]

    Adam: A method for stochastic optimization,

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  82. [90]

    Minisom: minimalistic and numpy-based implementation of the self organizing map,

    G. Vettigli, “Minisom: minimalistic and numpy-based implementation of the self organizing map,”

  83. [2018]

    Available: https://github.com/JustGlowing/minisom/ 31

    [Online]. Available: https://github.com/JustGlowing/minisom/ 31

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.