Pith. sign in

REVIEW 3 major objections 3 minor 35 references

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

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Feeding deep-learning diagnostics with calibrated physics-model variables yields near-perfect fault detection and isolates faults to a single root-cause signal.

desk verdict The detection win from feeding calibrated model parameters into a one-class autoencoder is credible and worth attention, but the isolation result is largely by construction because the injected fault is the same parameter the network flags. read the letter →

arxiv 1908.01529 v2 pith:SMBUGAMF submitted 2019-08-05 eess.SY cs.SY

Manuel Arias Chao , Chetan Kulkarni , Kai Goebel , Olga Fink This is my paper · ORCID
classification eess.SYcs.SY
keywords faultdetectionisolationhybriddiagnosticsdeeplearningone-classclassificationvariationalautoencoderphysics-basedperformancemodelC-MAPSS
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 tries to establish that a hybrid diagnostic pipeline—calibrating a physics-based performance model and feeding its inferred variables to a deep one-class model—beats pure data-driven fault detection and isolation, especially when operating conditions vary widely. It argues that unmeasured process variables and model tuning parameters carry fault information that raw sensor signals mask, and that adding them to the input space gives the network a representation in which a fault concentrates in one meaningful variable. On a simulated turbofan dataset with four high-pressure-compressor efficiency faults, the hybrid inputs raise detection accuracy from roughly 25 percent for the best pure data-driven deep model to 97.5–100 percent, and isolation flags only the responsible efficiency parameter instead of a smeared set of correlated sensors. If correct, the method offers a practical route to diagnostics for safety-critical assets where faulty samples are rare and labeled fault classes are unavailable.

What carries the argument

The central mechanism is the calibrated input augmentation: the system model $S(w,\theta)$ is calibrated with an Unscented Kalman Filter to infer the health-parameter vector $\theta$ and the virtual sensors, producing the augmented input $x=[w,\hat{x}_s,\hat{x}_v,\theta]$ that the deep-learning FDI model consumes. The load-bearing property is that each tested fault mode is encoded as a change in one component of $\theta$, so the reconstruction-error isolation score can attribute the fault to a single variable instead of a smeared set of correlated measurements. The same augmented input also feeds the one-class network that performs detection, with a percentile-based threshold on the reconstruction error deciding healthy versus faulty.

What would settle it

Inject a fault that is not captured by any of the ten calibrated health parameters and run the same pipeline; if the isolation step flags an unaffected parameter or reverts to a smeared set of residuals, the central claim of precise isolation is not supported.

Watch

Extended reading notes

Core claim

The central claim is that the fault source can be made visible to a learning algorithm by calibrating a system performance model and using the calibrated model's inferred quantities—reconstructed sensor readings, virtual sensors, and component health parameters—as inputs alongside the raw measurements. In the proposed architecture an Unscented Kalman Filter adjusts model parameters so that the model reproduces the observed sensor data; the augmented representation $x=[w,\hat{x}_s,\hat{x}_v,\theta]$ is then fed to an autoencoder-based one-class network for detection and to a reconstruction-similarity index for isolation. The authors report that this calibration-based hybrid input space yields detection accuracies of 97.5 to 100 percent on their simulated C-MAPSS turbofan case, versus 24.5 percent for the best pure data-driven deep model, and that the isolation step flags exactly variable 40, the high-pressure-compressor efficiency modifier, for all four tested faults. They present this precise isolation as removing the smearing effect, in which correlated sensors outside the faulty subsystem also show large reconstruction errors.

Load-bearing premise

The isolation result depends on the real fault being representable as a change in one of the calibrated model parameters that are fed to the network, because the tested faults were injected exactly by changing such a parameter.

Editorial extensions

If this is right

  • For systems operated across widely changing conditions, the hybrid calibration-based inputs can turn a one-class deep model into a near-perfect fault detector without requiring any faulty training samples.
  • When a fault mode is represented in the model's parameter set, fault isolation can pinpoint the affected component instead of spreading over correlated sensors.
  • Detection performance can stay high even when calibration is noisy, because the fault signature also appears in the inferred variables; accurate parameter estimates matter most for isolation rather than detection.
  • The framework can be extended by concatenating residuals to the healthy model response with the calibrated variables, which the paper identifies as a natural combined input option.

Reading between the lines

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

  • A natural next test would apply the same calibrated-input recipe to fault modes that are not in the model's parameter list, such as sensor bias or unmodelled wear; the paper itself expects isolation quality to drop in that setting.
  • If the pattern transfers, the method could reduce the need for labeled fault data in other safety-critical assets with available performance models, such as power turbines or rotating machinery, though the calibration step would have to be rebuilt for each system.
  • The sharp isolation result is partly a consequence of the synthetic setup, where the injected fault is exactly one of the calibrated parameters; a real-world evaluation would need to confirm that the reported single-variable isolation survives model mismatch and sensor noise.
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

3 major / 3 minor

Summary. The paper proposes a hybrid fault detection and isolation framework for complex systems under varying operating conditions. A physics-based performance model is calibrated to measured data with an Unscented Kalman Filter, and the resulting estimates of the measured signals, virtual sensors, and model health parameters are concatenated with the operating conditions to form the input of a one-class autoencoder-based diagnostics model. Fault detection is carried out by a one-class network operating on a latent embedding, and fault isolation is performed by examining per-input reconstruction deviations. The method is evaluated on a synthetic C-MAPSS turbofan dataset with four increasing HPC efficiency degradation faults, and it is compared against feed-forward, AE, VAE, HELM, and OC-SVM baselines with raw, residual, and hybrid inputs. The paper reports near-perfect detection with the hybrid calibrated inputs and states that these inputs isolate only the HPC efficiency parameter, concluding that the physics-enriched inputs outperform pure data-driven approaches and overcome the smearing effect in fault isolation.

Significance. If fully supported, the paper would provide a useful demonstration that adding inferred model states and health parameters to the input space can transform a difficult one-class fault detection problem under transient operating conditions, with the reported detection accuracy rising from roughly 25 percent on raw measurements to near 100 percent on calibrated inputs. The paper is systematic in comparing several deep architectures and includes a noise-robustness analysis of the calibrated features, which is a positive element. However, the evidence for the headline isolation claim is substantially weaker than the abstract suggests: the isolation outcome is largely built into the experimental design, the main experiment uses no sensor noise and no model mismatch, and fault isolation is not scored with a quantitative metric. The contribution is best assessed as a promising architecture and a carefully described case study, not yet as a validated general claim of precise fault isolation.

major comments (3)
  1. [Section 6.2; Tables 1, 5, and 9] The fault-isolation evaluation is circular with respect to the fault-injection procedure. The four faults are all generated by modifying the HPC efficiency modifier in C-MAPSS (Table 1), the hybrid input includes HPC_eff_mod as variable 40 (Table 5), and the autoencoder flags exactly variable 40 for all four faults (Table 9). Because the same model generates the data and provides the calibration, the isolation result is close to reading out the injected parameter rather than an independent diagnostic inference. This is not an internal inconsistency, but the evaluation does not support the general 'precise isolation' claim in the abstract. The manuscript itself concedes in Sections 7 and 8 that isolation capability will decrease when the calibrated model does not represent the fault mode and leaves faults not represented in the system model to future work. To support the claim, the authors should either evaluate fault modes that are not represented in theta, for example a sensor bias or a fault in a component not captured by the ten modifiers, or explicitly rescope the claim to faults representable in the calibrated parameter set.
  2. [Section 5.1 and Section 6.2] The validation is conducted in a favorable setting that does not exercise the model-mismatch conditions emphasized in the introduction. No additional sensor noise is used in the main experiment (Section 5.1), and the same C-MAPSS model both generates the condition monitoring data and is used for calibration, so the UKF sees exactly the model class that generated the observations. The robustness experiment in Section 6.1 contaminates the calibration factors after estimation rather than the sensor measurements, and it does not perturb the model structure. The discussion in Section 4.2 acknowledges that model discrepancy, represented as delta(w), remains an open problem. As a result, the paper does not yet demonstrate the claimed benefit 'particularly for systems with a high variability of operating conditions' under realistic mismatch; an experiment with sensor noise, a deliberately misspecified model, or data generated under a different health-parameter evolution would materially strengthen the claim.
  3. [Section 5.5 and Table 9] Fault isolation is not evaluated with a quantitative metric. Table 9 reports only the indices of variables with dI > 1 for the AE and VAE models, and the text concludes that the hybrid models provide 'perfect isolation,' but there is no measure of isolation accuracy, rank, false-positive rate, or variability across the 10 runs used for detection. Without a defined rule for when isolation is correct and a statistic over runs and fault magnitudes, the reader cannot assess whether variable 40 is flagged overwhelmingly or only marginally above the threshold. Please define an isolation metric, for example the top-1 hit rate for the true parameter or the margin dI_40 minus the maximum dI over the other inputs, and report it with confidence intervals.
minor comments (3)
  1. [Table 5] Row 45 is assigned the symbol LPT_flow_mod but its description reads 'HPT flow modifier'; this should be corrected to avoid ambiguity with row 43.
  2. [Section 5.2] The text refers to 'A validation set ST subset DL comprising 6% of the labelled healthy data,' but the training subset was previously also denoted ST and the validation set is SV in Section 4.3 and Figure 9; the notation should be made consistent.
  3. [Section 5.3] The OC-SVM hyperparameters are selected by maximizing the F1 score on the test set, while the neural-network hyperparameters are selected on the validation set. This oracle choice favors the baseline; the authors should state explicitly that it gives an upper bound on the OC-SVM's realistic performance.

Circularity Check

2 steps flagged · score 6.0 of 10

Isolation to variable 40 is a readout of the injected HPC_eff_mod parameter, so the flagship precision-isolation claim reduces to the fault-injection design.

  1. fitted input called prediction [Section 6.2 (Fault Isolation), with Tables 1, 5, and 9]
    "The four faults present in the combined test setsDT+ are rooted in a HPC efficiency deficit. However, not all the models have an input space where the compressor efficiency is represented. Concretely, only the calibration-based hybrid model with inputs [W, ˆXs, ˆXv, ˆθ] has a representation of the HPC efficiency through the estimated model correcting parameters ˆθ. … Finally, hybrid models based on calibrated models with input signals [W, ˆXs, ˆXv, ˆθ] encode the fault signature in only variable 40; which corresponds to the component of θ representing the correction of the HPC efficiency."

    The fault mode is defined as HPC efficiency degradation (Table 1), and variable 40 of the hybrid input is HPC_eff_mod (Table 5). The 'perfect isolation' result is therefore the autoencoder flagging the same parameter that was perturbed to create the fault. The UKF-calibrated theta_40 is an input feature, not an independent diagnostic inference; the isolation output is a readout of the fault-injection ground truth. The paper explicitly notes that only the calibrated input has a representation of HPC efficiency, so the result is forced by the experimental design.

  2. self definitional [Section 3.1 (Calibration-Based Hybrid Diagnostics)]
    "The model parameters θ are indeed model tuning of the system components and hence a deteriorated behaviour of a sub component is precisely encoded in only one component of θ (i.e. θk) while it is at the same time manifested in the condition monitoring data and virtual sensors. As it will be shown in the case study (Section 5), this feature avoids the smearing characteristic of data-driven diagnostics models."

    This asserts as a property of calibration that a sub-component fault is 'precisely encoded' in a single theta component. Combined with the case study's fault definition (HPC efficiency, i.e., theta_40), the promised avoidance of smearing and the later 'perfect isolation' are consequences of the assumed one-to-one fault-to-parameter encoding rather than an empirical discovery.

full rationale

The paper's most distinctive claim is that the hybrid input [W, Xhat_s, Xhat_v, theta] isolates the fault source precisely, overcoming smearing. In the experiment, however, all four faults are the same mode: HPC efficiency degradation (Table 1), and the hybrid input includes the model correcting parameter HPC_eff_mod as variable 40 (Table 5). Section 6.2 states that only the calibrated input has a representation of HPC efficiency and that any such model 'provides perfect isolation' by flagging variable 40. The isolation result is therefore the autoencoder detecting the very parameter that was perturbed to create the fault; it is a calibrated-input readout rather than an independent inference of an unknown root cause. The paper itself acknowledges in Sections 7 and 8 that isolation capability will decrease when the calibrated model does not represent the fault mode, and leaves faults not represented in the system model to future work. This makes the claimed precise isolation conditional on the fault being in theta, i.e., by construction in this evaluation. No self-citation chain or ansatz-smuggling is present; the detection benchmark and network comparisons are self-contained. The score is 6 because the central isolation result reduces to the fault-injection design, while the broader detection comparison retains some independent empirical content.

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

The central claim rests on the calibrated model being a good representation of the process and on the fault mode being one of the calibrated parameters. The latter is satisfied by construction in the dataset and makes the isolation result less informative. No new physical entities are postulated beyond model outputs that already exist in C-MAPSS.

free parameters (5)
  • gamma (detection safety margin) = 1.5
    Eq. 8 multiplies the 99.9th percentile validation error to set the detection threshold; chosen by hand and affects all detection results.
  • beta (detection threshold) = P99.9 of validation error times 1.5
    Eqs. 7-8 define the similarity score threshold from healthy validation data; effectively a fitted constant.
  • nu_k (per-signal isolation thresholds) = P99.9 of validation reconstruction error per input component
    Eqs. 10-11 use these thresholds to decide which signals are anomalous.
  • OC-SVM hyperparameters = nu=0.001, gamma=0.1
    Chosen to maximize F1 on the test set (Section 5.3), which is an unusual selection procedure.
  • Network architecture and training hyperparameters = one-class [20,100,1]; autoencoder [n,20,8,20,n]; lr 0.001; epochs 500/2000; batch 16/512
    Fixed by hand across methods to enable fair comparison; not fitted to data but they influence results.
assumptions (4)
  • domain assumption The C-MAPSS system model is a sufficiently accurate representation of the real physical process (or of the data-generating process).
    Stated in Section 4.2; the experiments guarantee it by using C-MAPSS for both data generation and calibration, but real engines would violate it.
  • ad hoc to paper The fault mode is representable as a change in one calibrated model parameter (HPC_eff_mod).
    The dataset injects HPC efficiency degradation and includes HPC_eff_mod in theta (Tables 1 and 5), making isolation of variable 40 nearly tautological.
  • domain assumption Test operating conditions are within the distribution of the healthy training set.
    Section 5.1 states the test set contains a subset of flight conditions experienced during training; this may not hold in general operation.
  • domain assumption The Unscented Kalman Filter yields sufficiently accurate point estimates of theta from the measurements.
    Calibration quality is assumed; the paper tests sensitivity to noise in theta but not to systematic model error.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models." pith.science (2026). https://pith.science/paper/SMBUGAMF

@misc{pith2026190801529,
  author       = {Pith},
  title        = {Pith review of: Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SMBUGAMF}},
  note         = {Machine review of arXiv:1908.01529}
}
read the original abstract

With the increased availability of condition monitoring data and the increased complexity of explicit system physics-based models, the application of data-driven approaches for fault detection and isolation has recently grown. While detection accuracy of such approaches is generally good, their performance on fault isolation often suffers from the fact that fault conditions affect a large portion of the measured signals thereby masking the fault source. To overcome this limitation and enable a more accurate fault detection, we propose a hybrid approach combining physical performance models with deep learning algorithms. Unobserved process variables are inferred with a physics-based performance model to enhance the input space of a data-driven diagnostics model. To validate the effectiveness of the proposed method, we generate a condition monitoring dataset of an advanced gas turbine during flight conditions under healthy and four faulty operative conditions based on the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dynamical model. We evaluate the performance of the proposed method in combination with two different deep learning algorithms: feed forward neural networks and Variational Autoencoders, both of which demonstrate a significant improvement when applied within the hybrid fault detection and diagnostics framework. The proposed method is able to outperform pure data-driven solutions, particularly for systems with a high variability of operating conditions. It provides superior results both for fault detection as well as for fault isolation. For fault isolation, it overcomes the smearing effect that is observed in pure data-driven approaches and enables a precise isolation of the affected signal. We also demonstrate that deep learning algorithms provide a better performance on fault detection compared to the traditional machine learning algorithms.

Figures

Figures reproduced from arXiv: 1908.01529 by the authors.

Figure 1
Figure 1. Overall architecture of the hybrid diagnostics approach in [ [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Overall architecture of the parallel hybrid diagnostics framework in [ [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Overall architecture of the residual-based hybrid diagnostics in [ [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (17 more)
Figure 4
Figure 4. Figure 4: Overall architecture of the residual-based diagnostics approach. The deep learning diagnostics model receives [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 7
Figure 7. Figure 7: The proposed algorithm uses as input the extended representation provided by the calibrated system model [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]
Figure 5
Figure 5. Figure 5: Overall architecture of the physics-informed recurrent neural network in [18] [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Overall architecture of the calibration-based hybrid diagnostics framework. The deep learning diagnostics [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Block diagram of the proposed fault detection and isolation algorithm within the proposes hybrid diagnostics [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Schematic representation of the problem. Training dataset [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Subset of 10 flight envelopes given by the traces of altitude (top), flight Mach number (middle) and [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: Network architecture for the defined learning problem with an autoencoder (encoder-decoder) and the [PITH_FULL_IMAGE:figures/full_fig_p013_10.png]
Figure 11
Figure 11. Figure 11: Evolution of the accuracy with dataset complexity for AE model based on [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]
Figure 12
Figure 12. Figure 12: Similarity index for four HPC efficiency faults of different intensities with AE model based on [PITH_FULL_IMAGE:figures/full_fig_p015_12.png]
Figure 13
Figure 13. Figure 13: Fault detection accuracy as function of the noise levels for AE and OC-SVM model. [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: Noisy calibration factors for a noise level of [PITH_FULL_IMAGE:figures/full_fig_p017_14.png]
Figure 15
Figure 15. Figure 15: Pairwise scatter plot the first two components of the latent space [PITH_FULL_IMAGE:figures/full_fig_p018_15.png]
Figure 16
Figure 16. Figure 16: Pairwise scatter plot the first two components of the latent space [PITH_FULL_IMAGE:figures/full_fig_p018_16.png]
Figure 17
Figure 17. Figure 17: Pairwise scatter plot the first two components of the latent space [PITH_FULL_IMAGE:figures/full_fig_p019_17.png]
Figure 18
Figure 18. Figure 18: Simplified diagram of the turbo fan engine model in C-MAPSS with Fan, Combustor, High-pressure [PITH_FULL_IMAGE:figures/full_fig_p022_18.png]
Figure 19
Figure 19. Figure 19: Variational autoencoder network 25 [PITH_FULL_IMAGE:figures/full_fig_p025_19.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 34 canonical work pages

  1. [1]

    Deep Learning and Its Applications to Machine Health Monitoring: A Survey

    Rui Zhao, Ruqiang Yan, Zhenghua Chen, Kezhi Mao, Peng Wang, and Robert X Gao. Deep Learning and Its Applications to Machine Health Monitoring: A Survey. Technical report, dec 2016

    work page 2016

  2. [2]

    A review on the application of deep learning in system health management, jul 2018

    Samir Khan and Takehisa Yairi. A review on the application of deep learning in system health management, jul 2018

    work page 2018

  3. [3]

    Online class imbalance learning and its applications in fault detection

    Shuo Wang, Leandro L Minku, and Xin Yao. Online class imbalance learning and its applications in fault detection. International Journal of Computational Intelligence and Applications, 12(04):1340001, 2013

    work page 2013

  4. [4]

    Power distribution fault cause identification with imbalanced data using the data mining-based fuzzy classificatione-algorithm

    Le Xu, Mo-Yuen Chow, and Leroy S Taylor. Power distribution fault cause identification with imbalanced data using the data mining-based fuzzy classificatione-algorithm. IEEE Transactions on Power Systems, 22(1):164–171, 2007

    work page 2007

  5. [5]

    Imbalanced data fault diagnosis of rotating machinery using synthetic oversampling and feature learning

    Yuyan Zhang, Xinyu Li, Liang Gao, Lihui Wang, and Long Wen. Imbalanced data fault diagnosis of rotating machinery using synthetic oversampling and feature learning. Journal of manufacturing systems, 48:34–50, 2018

    work page 2018

  6. [6]

    Robust signal reconstruction for condition monitoring of industrial components via a modified auto associative kernel regression method

    Piero Baraldi, Francesco Di Maio, Pietro Turati, and Enrico Zio. Robust signal reconstruction for condition monitoring of industrial components via a modified auto associative kernel regression method. Mechanical Systems and Signal Processing, 60:29–44, 2015

    work page 2015

  7. [7]

    Fault detection based on signal reconstruction with auto-associative extreme learning machines

    Yang Hu, Thomas Palmé, and Olga Fink. Fault detection based on signal reconstruction with auto-associative extreme learning machines. Engineering Applications of Artificial Intelligence, 57:105–117, 2017

    work page 2017

  8. [8]

    Deep feature learning network for fault detection and isolation

    Gabriel Michau, Thomas Palmé, and Olga Fink. Deep feature learning network for fault detection and isolation. In PHM 2017, St. Petersburg, USA, 2-5 October 2017, pages 108–118. PHM Society, 2017

    work page 2017

Show all 35 references
  1. [9]

    Domain adaptive transfer learning for fault diagnosis

    Qin Wang, Gabriel Michau, and Olga Fink. Domain adaptive transfer learning for fault diagnosis. arXiv preprint arXiv:1905.06004, 2019

    1905 arXiv

  2. [10]

    User’s Guide for the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS)

    Dean K Frederick, Jonathan A Decastro, and Jonathan S Litt. User’s Guide for the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS). Technical report, 2007

    2007

  3. [11]

    DASHlink - Flight Data For Tail 687, 2012

    2012

  4. [12]

    Support vector method for novelty detection

    Bernhard Schölkopf, Robert Williamson, Alex Smola, John Shawe-Taylor, and John Piatt. Support vector method for novelty detection. In Advances in Neural Information Processing Systems, pages 582–588, 2000

    2000

  5. [13]

    Hybrid Model-Based and Data-Driven Fault Detection and Diagnostics for Commercial Buildings

    Stephen Frank, Mike Heaney, Xin Jin, Joseph Robertson, Howard Cheung, Ryan Elmore, and Gregor P Henze. Hybrid Model-Based and Data-Driven Fault Detection and Diagnostics for Commercial Buildings. Proceedings of the ACEEE Summer Study on Energy Efficiency in Buildings, Aug 21-26...

    2016

  6. [14]

    Hybrid Physics-Based and Data-Driven Phm

    H Hanachi, W Yu, I Y Kim, and C K Mechefske. Hybrid Physics-Based and Data-Driven Phm. In Canadian Machinery Vibration Association (CMVA) Annual Conference, Edmonton, Alberta, Canada, 2017

    2017

  7. [15]

    Rausch, Kai F

    Randal T. Rausch, Kai F. Goebel, Neil H. Eklund, and Brent J. Brunell. Integrated In-Flight Fault Detection and Accommodation: A Model-Based Study. In Volume 1: Turbo Expo 2005, pages 561–569. ASME, 2005

    2005

  8. [16]

    Variations on the Kalman Filter for Enhanced Performance Monitoring of Gas Turbine Engines

    S.J Borguet. Variations on the Kalman Filter for Enhanced Performance Monitoring of Gas Turbine Engines. Phd thesis, Université de Liège, 2012

    2012

  9. [17]

    Physics Guided Recurrent Neural Networks For Modeling Dynamical Systems: Application to Monitoring Water Temperature And Quality In Lakes

    Xiaowei Jia, Anuj Karpatne, Jared Willard, Michael Steinbach, Jordan Read, Paul C Hanson, Hilary A Dugan, and Vipin Kumar. Physics Guided Recurrent Neural Networks For Modeling Dynamical Systems: Application to Monitoring Water Temperature And Quality In Lakes. Technical repor...

    2018

  10. [18]

    Renato Giorgiani Nascimento and Felipe A.C. Viana. Fleet prognosis with physics-informed recurrent neural networks. In Structural Health Monitoring 2019: Enabling Intelligent Life-Cycle Health Management for Industry Internet of Things (IIOT) - Proceedings of the 12th Internat...

    2019

  11. [19]

    Arinan Dourado and Felipe A. C. Viana. Physics-Informed Neural Networks for Corrosion-Fatigue Prognosis. Proceedings of the Annual Conference of the PHM Society, 11(1), sep 2019

    2019

  12. [20]

    Wind Turbine Main Bearing Fatigue Life Estimation with Physics- informed Neural Networks

    Yigit Anil Yucesan and Felipe A C Viana. Wind Turbine Main Bearing Fatigue Life Estimation with Physics- informed Neural Networks. In Phm 2019, volume 11, pages 1–14, 2019

    2019

  13. [21]

    Brent J.; Brunell, Jr. Harry K. Mathews, and Aditya Kumar. United States Patent Design of an Adaptive Model-Based Control for Controlling a Gas Turbine, 2004

    2004

  14. [22]

    Lilley, Peter Mathé, and V olker Schloßhauer

    Manuel Arias Chao, Darrel S. Lilley, Peter Mathé, and V olker Schloßhauer. Calibration and Uncertainty Quan- tification of Gas Turbine Performance Models. In Proceedings of the ASME Turbo Expo , volume 7A, page V07AT29A001, 2015

    2015

  15. [23]

    W. J. Scheirer, A. de Rezende Rocha, A. Sapkota, and T. E. Boult. Toward Open Set Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(7):1757–1772, 2013

    2013

  16. [24]

    New extension of the Kalman filter to nonlinear systems

    Simon J Julier and Jeffrey K Uhlmann. New extension of the Kalman filter to nonlinear systems. In Signal Processing, Sensor Fusion, and Target Recognition VI, volume 3068, page 182, 1997

    1997

  17. [25]

    Moya and Don R

    Mary M. Moya and Don R. Hush. Network constraints and multi-objective optimization for one-class classification. Neural Networks, 9(3):463–474, 1996

    1996

  18. [26]

    Feature Learning for Fault Detection in High- Dimensional Condition-Monitoring Signals

    Gabriel Michau, Yang Hu, Thomas Palmé, and Olga Fink. Feature Learning for Fault Detection in High- Dimensional Condition-Monitoring Signals. Submitted for a possible publication in IEEE Transactions on Cybenetics, 2017

    2017

  19. [27]

    Hierarchical Extreme Learning Machine for unsupervised representation learning

    Wentao Zhu, Jun Miao, Laiyun Qing, and Guang-Bin Huang. Hierarchical Extreme Learning Machine for unsupervised representation learning. In 2015 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE, 2015

    2015

  20. [28]

    Auto-encoding variational bayes

    Diederik P Kingma and Max Welling. Auto-encoding variational bayes. In 2nd International Conference on Learning Representations, ICLR 2014 - Conference Track Proceedings, 2014

    2014

  21. [29]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V . Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in Python. Journal of Machine Learni...

    2011

  22. [30]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Lei Ba. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, 2015

    2015

  23. [31]

    Understanding the difficulty of training deep feedforward neural networks

    Xavier Glorot and Yoshua Bengio. Understanding the difficulty of training deep feedforward neural networks. Technical report, 2010

    2010

  24. [32]

    LeCun, Léon Bottou, Genevieve B

    Yann A. LeCun, Léon Bottou, Genevieve B. Orr, and Klaus Robert Müller. Efficient backprop. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7700 LECTU:9–48, 2012

    2012

  25. [33]

    Hands-on transfer learning with Python : implement advanced deep learning and neural network models using TensorFlow and Keras

    Dipanjan Sarkar, Raghav Bali, and Tamoghna Ghosh. Hands-on transfer learning with Python : implement advanced deep learning and neural network models using TensorFlow and Keras. 2018

    2018

  26. [34]

    InfoV AE: Balancing Learning and Inference in Variational Autoencoders

    Shengjia Zhao, Jiaming Song, and Stefano Ermon. InfoV AE: Balancing Learning and Inference in Variational Autoencoders. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 5885–5892, jul 2019

    2019

  27. [35]

    Tutorial on Variational Autoencoders

    Carl Doersch. Tutorial on Variational Autoencoders. Technical report, jun 2016. 21 A PREPRINT - JANUARY 1, 2020 Figure 18: Simplified diagram of the turbo fan engine model in C-MAPSS with Fan, Combustor, High-pressure compressor (HPC); high-pressure turbine (HPT); Low-pressure ...

    2016

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.