Pith. sign in

REVIEW 4 major objections 5 minor 47 references

Data-Driven Structural State Estimation via Multi-Fidelity Gaussian Process Models

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Simulated guided-wave data can replace costly experiments in structural damage estimation when fused through a two-level Gaussian process.

desk verdict A solid applied demonstration that multi-fidelity GPRM helps when experimental data are scarce, but the second test case's 'simulated' data are calibrated on the same experiments, so that part of the evidence is weaker than it looks. read the letter →

arxiv 2505.01666 v1 pith:X2FGEU6K submitted 2025-05-03 eess.SP

Yiming Fan , Fotis Kopsaftopoulos This is my paper · ORCID
classification eess.SP
keywords multi-fidelityGaussianprocessregressionstructuralhealthmonitoringdamageindexguidedwavesactivelearningco-krigingstateestimationuncertaintyquantification
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

This paper aims to show that structural damage states can be estimated accurately even when real experimental data are scarce, by training a two-level Gaussian process on both experimental and simulated guided-wave features. Across two test cases—crack size on an aluminum coupon and applied load on another coupon—the multi-fidelity model repeatedly achieves lower RMSE and higher $R^2$ than a standard Gaussian process regression model trained on the same experimental data alone. The practical promise is that cheap simulation can substitute for costly experiments without sacrificing accuracy, and that active learning makes the data-efficient procedure stable.

What carries the argument

The load-bearing object is the two-level auto-regressive co-kriging identity $f_2(x)=\rho f_1(x)+\delta(x)$, in which the experimental mapping $f_2$ is written as a scaled version of the simulated mapping $f_1$ plus a Gaussian-process discrepancy term $\delta$. This turns the problem of fusing experiment and simulation into one joint Gaussian process with a squared-exponential kernel, a cross-correlation parameter $\rho$, and a variance lower-bound constraint that keeps predicted uncertainty no smaller than the scatter in the experimental data. The same Gaussian process posterior then supplies both the mean damage-state estimate and the acquisition function values (L2 loss, max variance, upper confidence bound, expected improvement) used for active learning.

What would settle it

Take a dataset where the experimental DI trend is known at many states and the simulated DI deviates from it by a bias that changes sign along the state axis. Train the multi-fidelity model on two experimental states plus simulated points located only inside the sign-change region, and check whether test RMSE on the held-out experimental states stays below the standard-Gaussian-process baseline; if it does not, the constant-scaling premise is the cause.

Watch

Extended reading notes

Core claim

The central claim is that auto-regressive co-kriging, applied to damage indices extracted from guided waves, lets a small set of experimental DI values (as few as two or three states) be augmented with simulated DI values to produce mean predictions and confidence bounds that follow the experimental trend better than a standard Gaussian process using only the experimental points. In the fixed-high-fidelity task, adding simulated points to a fixed experimental set lowers RMSE and raises $R^2$ once enough points are included, although the first added points can initially increase error. In the constant-total-states task, replacing experimental states with simulated states degrades accuracy more slowly than removing the experimental data outright. In the load case, combining the multi-fidelity model with active learning gives faster and more stable RMSE convergence than randomly selected simulated points, and can outperform a standard Gaussian process trained on more experimental states.

Load-bearing premise

The load-bearing assumption is that the difference between simulated and experimental damage indices is a fixed scaling factor plus a stable discrepancy; if the simulation's bias changes across damage states, the extra simulated points can pull the estimate away from the real trend—the paper itself reports RMSE rising before it falls.

Editorial extensions

If this is right

  • Simulated guided-wave data can substitute for a large fraction of experimental states: in the load test, models with two experimental sets plus simulated points beat a standard Gaussian process with three or four experimental sets.
  • The multi-fidelity model can be used to fill data-sparse regions of the damage-state axis, where experiments are impractical or prohibitively costly.
  • An optimal amount of low-fidelity data exists: RMSE first rises and then falls as simulated points are added, so a sweep of added points can identify the best low-fidelity budget.
  • Active-learning acquisition functions, especially upper confidence bound and expected improvement, make convergence faster and less dependent on the random order of data addition.

Reading between the lines

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

  • The same co-kriging structure should transfer to other damage-sensitive features, such as spectral or time-frequency features, as long as the simulation-to-experiment relationship is approximately affine; a head-to-head comparison of DI types under identical experimental budgets would test this.
  • The reported initial RMSE rise suggests a practical diagnostic: monitoring the derivative of RMSE with respect to added simulated points could let an operator stop augmentation before simulation bias dominates the estimate.
  • Because the paper applies a hard variance lower bound, a fully Bayesian treatment that treats the bound as a prior on noise levels would reveal how much of the gain comes from the constraint versus the multi-fidelity fusion itself.
  • For deployment on real structures, the main unknown is whether the cross-correlation parameter learned on one sensor path or coupon generalizes to another; a cross-validation across sensor paths would settle that.
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 applies multi-fidelity Gaussian process regression (MF-GPRM), based on the autoregressive co-kriging model of Kennedy and O'Hagan, to damage state estimation in two aluminum plate test cases. Damage indices (DIs) extracted from experimental guided-wave signals are treated as high-fidelity data, and DIs from FEM simulations (test case 1) or a physics-based load-compensation model (test case 2) as low-fidelity data. The authors compare MF-GPRM with standard GPRM trained on the same experimental data, reporting lower RMSE and higher R^2 for MF-GPRM across several tasks, and additionally investigate active-learning acquisition rules for selecting simulated data points. The central claim is that multi-fidelity fusion of cheap simulated data with a small experimental set improves probabilistic damage-state estimation.

Significance. If the claims hold, the paper offers a practical route to reducing experimental data collection costs in guided-wave SHM by incorporating lower-fidelity simulated data into a probabilistic regression framework. The main strengths are the use of two different data-source types (FEM and physics-based reconstruction), the integration of active learning with multiple acquisition criteria, and the availability of a first test case that relies on an independent FEM simulation. However, the second test case's low-fidelity data are not independent of the experimental data, which substantially weakens the generalizability claim. Overall, the paper is an incremental but potentially useful application of an established method rather than a methodological breakthrough.

major comments (4)
  1. [Sec. 4.2, Sec. 4.3] The low-fidelity dataset in the second test case is not an independent simulation source. The constants A, B in Eq. (19) and K_phase in Eq. (20) are computed from experimental signals collected at the five loads 0, 5, 10, 15, 20 kN, and the simulated signals are then generated at 0.5 kN increments. The simulated DIs are therefore effectively a dense interpolation of the same experimental response, rather than predictions from an independently calibrated physics model. As a result, the large RMSE/R^2 gains in Figs. 14 and 15 and the numbers in Table 4 cannot be interpreted as evidence that inexpensive independent simulations can substitute for experiments; at most they demonstrate that a physics-based interpolation of the experimental response can be exploited. This compromises the universal-superiority conclusion in Sec. 5. The authors should either redo the second test case with a genuinely independent simulation (e.g., an FEM model of the loaded plate) or explicitly reframe the test case as physics-based data augmentation and correspondingly limit the claims in the abstract and conclusion.
  2. [Sec. 3.3.1, Appendix A.2] The variance lower-bound constraint is a free parameter whose value appears to be selected based on the results. Section 3.3.1 states that "three times the largest variance of the experimental data was chosen as the lower constraint" for sigma_1^2 and sigma_2^2, while Appendix A.2 shows outputs for multipliers 1, 5, 10, and 15 but gives no criterion for choosing 3. The sensitivity of the main RMSE/R^2 comparisons to this multiplier is not reported. Because the uncertainty bounds are a central claimed advantage of the method, the choice should be justified by a principled rule such as cross-validation, or a sensitivity analysis over the multiplier range should be reported for at least the main comparisons.
  3. [Sec. 3.3, Figs. 8 and 11] The experimental training sets are formed by randomly drawing 15 of 20 realizations, but no repeated-split statistics are reported. RMSE and R^2 values are presented for a single split per configuration, so the observed differences between standard GPRM and multi-fidelity GPRM may be within the sampling variability of the experimental realizations. The authors should repeat the random splitting (e.g., 10 or more seeds) and report mean and standard error of the performance metrics to support the claim that the improvements are statistically meaningful.
  4. [Sec. 2.1.1, Eq. (2)] The model assumes a single state-independent correlation coefficient rho between the low- and high-fidelity maps, and the paper acknowledges that "the actual value can change at different locations" without providing any diagnostic for this assumption. The empirical observation in Sec. 3.3.1 and Fig. 8 that RMSE initially increases when simulated points are added could be a symptom of rho being misestimated in some regions of the state domain. Please provide evidence for the adequacy of the constant-rho assumption, for example by estimating rho over a moving window of the damage-state domain or by comparing against a model with a spatially varying rho; at minimum, report the fitted values of rho for the studied paths and discuss whether they are consistent across the domain.
minor comments (5)
  1. [Sec. 3.3.1, Fig. 7(d)] The text states that RMSE decreased from 0.032593 to 0.030838, "approximately halving the original value from GPRM," but this is a reduction of about 5%, not a halving. Please correct the description to reflect the actual magnitude of the improvement.
  2. [Appendix A.2, Figs. A.4 and A.5] The captions for these figures list panel labels incorrectly, repeating "(a)" and "(b)" instead of "(a)" through "(d)". Please correct the panel designations so that the text and figures are unambiguous.
  3. [Sec. 4.2] The text says "an increment of 0.5N, which is one-tenth of the experimental state increment"; since the experimental increments are in kN, this should be "0.5 kN."
  4. [Sec. 3.3.1, Sec. 4] There are several typos: "the experimental dat" should be "the experimental data" in Sec. 3.3.1, and "demonstrat" should be "demonstrate" in the introductory paragraph of Sec. 4. Please proofread the final text.
  5. [Sec. 4.3.3, Fig. 17(b)] The active-learning RMSE curves appear to be single-run results, whereas the random-selection comparison uses 10 seeds. If the active-learning criteria were not repeated across multiple trials, please state this explicitly and discuss the potential variability of the active-learning results.

Circularity Check

1 steps flagged · score 6.0 of 10

Test case 2's 'simulated' DIs come from a load-compensation model calibrated on the same experimental load states, so those multi-fidelity gains are partly forced; test case 1's FEM data keeps the claim partially independent.

  1. fitted input called prediction [Section 4.2 (Physics-based Load Compensation Model), used in Sections 4.3.1 and 4.3.2]
    ""Five static loads—0, 5, 10, 15, and 20 kN—were applied to produce the experimental signals." ... "A and B are unknown model constants which can be calculated from experimental signals collected at various loading conditions." ... "K_phase is an unknown constant and can be estimated from experimental measurements along with strain distribution." ... "In this test case, the simulated signals were generated using this physics-based model. ... For this test, an increment of 0.5N, which is one-tenth of the experimental state increment, was selected to ensure sufficient resolution.""

    By the paper's own description, the constants A, B, and K_phase of the load-compensation model are fit to experimental signals from the same five load states (0, 5, 10, 15, 20 kN) that constitute the high-fidelity dataset. The simulated DIs at 0.5 kN increments are then produced from this calibrated model and fed into the multi-fidelity GPRM as the low-fidelity source. In Task 1 (Sec. 4.3.1), the experimental training sets are 0 and 20 kN (or 0, 10, 20 kN), and the added 'simulated' points at 5, 10, and 15 kN are generated by a model that has already seen experimental signals at exactly those states.

full rationale

The multi-fidelity Gaussian process formulation in Section 2.1 is a standard auto-regressive co-kriging model trained by marginal likelihood maximization; there is no circularity in the mathematics itself, and test case 1 uses independent FEM simulations, so the approach is not inherently circular. The identified circularity is confined to test case 2: the physics-based load-compensation model used as the low-fidelity data source is calibrated on experimental signals from the same load states, so adding its outputs as 'simulated' points in Tasks 1–3 partially re-injects the high-fidelity data that the comparison is supposed to save. This makes the test case 2 numerical gains (e.g., RMSE decreasing from about 0.0075 to 0.0025 in Section 4.3.1) overstate the value of independent simulations. Because one full validation test case relies on this leakage while the other is independent, the circularity is partial rather than total; a score of 6 reflects 'one or more predictions reduce by construction, partial circularity.'

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

The central claim rests on the standard auto-regressive GP link (Eq. 2), the smoothness of DI evolution, and the validity of simulated data as a lower-fidelity source. The variance lower-bound multiplier is the main hand-tuned quantity; UCB/EI parameters are unstated. No new physical entities are introduced.

free parameters (4)
  • Variance lower-bound multiplier c = 3
    Lower bound for both sigma_1^2 and sigma_2^2 set to c times the largest experimental variance; Section 3.3.1 selects c=3 after Appendix A.2 compares c=1,5,10,15.
  • UCB lambda = not reported
    Equation (16) defines the UCB acquisition function with lambda; the paper does not state the value used in Section 4.3.3.
  • EI xi = not reported
    Equation (17) defines expected improvement with xi; the paper does not state the value used in Section 4.3.3.
  • DI choice = RMSD
    The RMSD DI is used for all reported regressions because it 'displays a wider range' and is 'more sensitive' (Sec. 3.3); this selection affects every result.
assumptions (5)
  • domain assumption f2(x) = rho f1(x) + delta(x) with constant rho
    Invoked in Eq. 2 (Sec. 2.1.1); the whole multi-fidelity correction rests on this linear, state-independent link.
  • domain assumption Damage index varies smoothly with damage state
    GP regression over state assumes a continuous mapping; the paper notes DIs 'do not increase uniformly' (Sec. 3.3), so smoothness is an assumption.
  • domain assumption Simulated data is a valid lower-fidelity source
    Used in both test cases; in test case 2 the simulated signals are reconstructed from parameters fit to the same experimental data (Sec. 4.2), which weakens this assumption.
  • standard math Linear combination of Gaussian processes is Gaussian
    Used to write Eq. (3) and the joint GP in Eq. (4) (Sec. 2.1.1).
  • ad hoc to paper Prediction variance lower bound sigma_*^2 >= sigma_2^2
    Constraint imposed during training (Sec. 2.1.2) to prevent overconfidence; not derived from data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Data-Driven Structural State Estimation via Multi-Fidelity Gaussian Process Models." pith.science (2026). https://pith.science/paper/X2FGEU6K

@misc{pith2026250501666,
  author       = {Pith},
  title        = {Pith review of: Data-Driven Structural State Estimation via Multi-Fidelity Gaussian Process Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X2FGEU6K}},
  note         = {Machine review of arXiv:2505.01666}
}
read the original abstract

Guided wave-based techniques have been used extensively in Structural Health Monitoring (SHM). Models using guided waves can provide information from both time and frequency domains to make themselves accurate and robust. Probabilistic SHM models, which have the ability to account for uncertainties, are developed when decision confidence intervals are of interest. Most active-sensing guided-wave methods rely on the assumption that a large dataset can be collected, making them impractical when data collection is constrained by time or environmental factors. Meanwhile, although simulation results may lack the accuracy of real-world data, they are easier to obtain. In this context, models that integrate data from multiple sources have the potential to combine the accuracy of experimental data with the convenience of simulated data, without requiring large and potentially costly experimental datasets. The goal of this work is to introduce and assess a probabilistic multi-fidelity Gaussian process regression framework for damage state estimation via the use of both experimental and simulated guided waves. The main differences from previous works include the integration of damage-sensitive features (damage indices, DIs) extracted from both experimental and numerical sources, as well as the use of a relatively small amount of real-world data. The proposed model was validated by two test cases where multiple data sources exist. For each test case, experimental data were collected from a piezoelectric sensor network attached to an aluminum plate with various structural conditions, while simulated data were generated using either multiphysics finite element model (FEM) or physics-based signal reconstruction approaches under the same conditions.

Figures

Figures reproduced from arXiv: 2505.01666 by the authors.

Figure 1
Figure 1. The flowchart that demonstrates the main steps of guided-wave based multi-fidelity GPRM [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Left: The notched aluminum coupon used in the first test case, featuring a 20-mm notch [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Sample signals collected in test case 1: Panel a and b display the experimental and [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (15 more)
Figure 4
Figure 4. Figure 4: The indicative evolution of two types of introduced Damage Indices (DIs) with respect [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Indicative RMSD DI evolution of two data sources applied in this study with respect to [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: DI regression for path 2-6 from GPRM and multi-fidelity GPRM: (a) prediction using 3 [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: DI regression for path 3-4 from GPRM and multi-fidelity GPRM: (a) prediction using [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: RMSE and 𝑅 2 comparison between GPRM and multi-fidelity GPRM for path 2-6 and 3-4. the multi-fidelity model can produce more accurate mean estimates and reduce uncertainty. Based on this conclusion, a second test case, where simulated data is more abundant, was conduct…
Figure 9
Figure 9. Figure 9: DI regression for path 2-6 from GPRM and multi-fidelity GPRM: (a) prediction using [PITH_FULL_IMAGE:figures/full_fig_p016_9.png]
Figure 10
Figure 10. Figure 10: DI regression for path 3-5 from GPRM and multi-fidelity GPRM: (a) to (c) predictions [PITH_FULL_IMAGE:figures/full_fig_p017_10.png]
Figure 11
Figure 11. Figure 11: RMSE and 𝑅 2 comparison between GPRM and multi-fidelity GPRM for path 2-6 and 3-5. 17 [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]
Figure 12
Figure 12. Figure 12: The Al coupon used in the second test case with four 3-gm weights and 6 PZT sensors. [PITH_FULL_IMAGE:figures/full_fig_p018_12.png]
Figure 13
Figure 13. Figure 13: The first wave package of both experimental and the physics-based reconstructed signals [PITH_FULL_IMAGE:figures/full_fig_p019_13.png]
Figure 14
Figure 14. Figure 14: Task 1 results of the 2nd test case. Panels (a) and (d): baseline DI regression for path [PITH_FULL_IMAGE:figures/full_fig_p020_14.png]
Figure 15
Figure 15. Figure 15: RMSE and 𝑅 2 comparison between GPRM and multi-fidelity GPRM with batch learning using random selection for path 1-6 : (a) RMSE comparison for path 1-6 with loads at 0 and 20 kN; (b) RMSE comparison for path 1-6 with loads at 0, 10 and 20 kN; (c) 𝑅 2 comparison for pa…
Figure 16
Figure 16. Figure 16: Task 2 of the 2nd test case. DI regression for path 1-6 from GPRM and multi-fidelity [PITH_FULL_IMAGE:figures/full_fig_p022_16.png]
Figure 17
Figure 17. Figure 17: RMSE trend curves from multi-fidelity models using (a) random selection and (b) four [PITH_FULL_IMAGE:figures/full_fig_p023_17.png]
Figure 18
Figure 18. Figure 18: The first 6 iterations of multi-fidelity GPRM results using UCB as acquisition function. [PITH_FULL_IMAGE:figures/full_fig_p023_18.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 45 canonical work pages

  1. [1]

    An introduction to structural health monitoring,

    Farrar, C. R. and Worden, K., “An introduction to structural health monitoring,”Philosoph- ical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 365, No. 1851, 2007, pp. 303–315

    work page 2007

  2. [2]

    Structural health monitoring techniques for aircraft composite structures,

    Diamanti, K. and Soutis, C., “Structural health monitoring techniques for aircraft composite structures,” Progress in Aerospace Sciences, Vol. 46, No. 8, 2010, pp. 342–352. 24

    work page 2010

  3. [3]

    State-of-the-art technologies for UAV inspections,

    Jordan, S., Moore, J., Hovet, S., Box, J., Perry, J., Kirsche, K., Lewis, D., and Tse, Z. T. H., “State-of-the-art technologies for UAV inspections,”IET Radar, Sonar & Navigation, Vol. 12, No. 2, 2018, pp. 151–164

    work page 2018

  4. [4]

    Life-cyclereliability-basedoptimizationofcivilandaerospace structures,

    Frangopol, D.M.andMaute, K., “Life-cyclereliability-basedoptimizationofcivilandaerospace structures,” Structure and Infrastructure Engineering, Vol. 81, No. 1, 2003, pp. 397–410

    work page 2003

  5. [5]

    Structural health monitoring: Closing the gap between research and industrial deployment,

    Cawley, P., “Structural health monitoring: Closing the gap between research and industrial deployment,” Structural Health Monitoring, Vol. 17, No. 5, 2018, pp. 1225–1244

    work page 2018

  6. [6]

    Probabilistic active sensing acousto-ultrasound SHM based on non-parametric stochastic representations,

    Amer, A. and Kopsaftopoulos, F. P., “Probabilistic active sensing acousto-ultrasound SHM based on non-parametric stochastic representations,”Proceedings of the Vertical Flight Society 75th Annual Forum & Technology Display, Philadelphia, PA, USA, May 2019

    work page 2019

  7. [7]

    Towards Unified Probabilistic Rotorcraft Damage De- tection and Quantification via Non-parametric Time Series and Gaussian Process Regression Models,

    Amer, A. and Kopsaftopoulos, F. P., “Towards Unified Probabilistic Rotorcraft Damage De- tection and Quantification via Non-parametric Time Series and Gaussian Process Regression Models,” Proceedings of the Vertical Flight Society 76th Annual Forum & Technology Display, Virginia Beach, VA, USA, October 2020

    work page 2020

  8. [8]

    Damage detection sen- sitivity characterization of acousto-ultrasound-based structural health monitoring techniques,

    Janapati, V., Kopsaftopoulos, F., Li, F., Lee, S., and Chang, F.-K., “Damage detection sen- sitivity characterization of acousto-ultrasound-based structural health monitoring techniques,” StructurRasmussen-Williams06 al Health Monitoring, Vol. 15, No. 2, 2016, pp. 143–161

    work page 2016

Show all 47 references
  1. [9]

    A probabilistic crack size quantification method using in-situ Lamb wave test and Bayesian updating,

    Yang, J., He, J., Guan, X., Wang, D., Chen, H., Zhang, W., and Liu, Y., “A probabilistic crack size quantification method using in-situ Lamb wave test and Bayesian updating,”Mechanical Systems and Signal Processing, Vol. 78, 2016, pp. 118–133

    2016

  2. [10]

    A Lamb wave based fatigue crack length estimation method using finite element simulations,

    He, J., Ran, Y., Liu, B., Yang, J., and Guan, X., “A Lamb wave based fatigue crack length estimation method using finite element simulations,”The 9th International Symposium on NDT in Aerospace, Xiamen, China, Xiamen, China, November 2017

    2017

  3. [11]

    Monitoring of fatigue crack propagation by damage index of ultrasonic guided waves calculated by various acoustic features,

    Jin, H., Yan, J., Li, W., and Qing, X., “Monitoring of fatigue crack propagation by damage index of ultrasonic guided waves calculated by various acoustic features,”Applied Sciences, Vol. 9, 2019, pp. 4254

    2019

  4. [12]

    Active interface debonding detection of a concrete- filled steel tube with piezoelectric technologies using wavelet packet analysis,

    Xu, B., Zhang, T., Song, G., and Gu, H., “Active interface debonding detection of a concrete- filled steel tube with piezoelectric technologies using wavelet packet analysis,”Mechanical Sys- tems and Signal Processing, Vol. 36, 2013, pp. 7–17

    2013

  5. [13]

    Piezoelectric Wafer Active Sensors for Structural Health Monitoring of Com- posite Structures Using Tuned Guided Waves,

    Giurgiutiu, V., “Piezoelectric Wafer Active Sensors for Structural Health Monitoring of Com- posite Structures Using Tuned Guided Waves,”Journal of Engineering Materials and Technol- ogy, Vol. 133, No. 4, 2011, pp. 041012

    2011

  6. [14]

    Multimodal structural health monitoring based on active and passive sensing,

    Nasrollahi, A., Deng, W., Ma, Z., and Rizzo, P., “Multimodal structural health monitoring based on active and passive sensing,” Structural Health Monitoring, Vol. 17, No. 2, 2018, pp. 395–409

    2018

  7. [15]

    Monitoring fatigue crack growth in narrow structural components using Lamb wave technique,

    Lim, S. I., Cui, L., Liu, Y., and Soh, C. K., “Monitoring fatigue crack growth in narrow structural components using Lamb wave technique,” Proc. SPIE 7981, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems, San Diego, CA, USA, April 2011. 25

    2011

  8. [16]

    Comparative study of deterioration of com- posite due to moisture using strain, electro-mechanical impedence, and guided waves,

    Soman, R., Malinowski, P., and Ostachowicz, W., “Comparative study of deterioration of com- posite due to moisture using strain, electro-mechanical impedence, and guided waves,”Proc. SPIE 10600, Health Monitoring of Structural and Biological Systems XII, Denver, CO, USA, March 2018

    2018

  9. [17]

    In-situ acousto-ultrasonic monitoring of crack propagation in AL2024 alloy,

    Vanniamparambil, P., Bartoli, I., Hazeli, K., Caudra, J., Schwartz, E., Saralaya, R., and Kontsos, A., “In-situ acousto-ultrasonic monitoring of crack propagation in AL2024 alloy,” Proc. SPIE 8348, Health Monitoring of Structural and Biological Systems, San Diego, CA, USA, April 2012

    2012

  10. [18]

    Uncertainty quantification of guided waves propagation for active sensing structural health monitoring,

    Ahmed, S. and Kopsaftopoulos, F., “Uncertainty quantification of guided waves propagation for active sensing structural health monitoring,”Proceedings of the vertical flight society 75th annual forum & technology display, Philadelphia, PA, USA, 2019, pp. 13–16

    2019

  11. [19]

    A stochastic global identification framework for aerospace structures operating under varying flight states,

    Kopsaftopoulos, F., Nardari, R., Li, Y.-H., and Chang, F.-K., “A stochastic global identification framework for aerospace structures operating under varying flight states,”Mechanical Systems and Signal Processing, Vol. 98, 2018, pp. 425–447

    2018

  12. [20]

    Support-vector networks,

    Cortes, C. and Vapnik, V., “Support-vector networks,”Machine learning, Vol. 20, No. 3, 1995, pp. 273–297

    1995

  13. [21]

    Cristianini, N., Shawe-Taylor, J., et al.,An introduction to support vector machines and other kernel-based learning methods, Cambridge university press, 2000

    2000

  14. [22]

    Physics-constrained deep learn- ing for high-dimensional surrogate modeling and uncertainty quantification without labeled data,

    Zhu, Y., Zabaras, N., Koutsourelakis, P.-S., and Perdikaris, P., “Physics-constrained deep learn- ing for high-dimensional surrogate modeling and uncertainty quantification without labeled data,” Journal of Computational Physics, Vol. 394, 2019, pp. 56–81

    2019

  15. [23]

    Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification,

    Tripathy, R. K. and Bilionis, I., “Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification,”Journal of computational physics, Vol. 375, 2018, pp. 565–588

    2018

  16. [24]

    Machine learning approach to model order reduction of nonlinear systems via autoencoder and lstm networks,

    Simpson, T., Dervilis, N., and Chatzi, E., “Machine learning approach to model order reduction of nonlinear systems via autoencoder and lstm networks,”arXiv preprint arXiv:2109.11213, 2021

    2021 arXiv

  17. [25]

    Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow,

    Nakamura, T., Fukami, K., Hasegawa, K., Nabae, Y., and Fukagata, K., “Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow,”Physics of Fluids, Vol. 33, No. 2, 2021, pp. 025116

    2021

  18. [26]

    Numerical Gaussian processes for time- dependentandnonlinearpartialdifferentialequations,

    Raissi, M., Perdikaris, P., and Karniadakis, G. E., “Numerical Gaussian processes for time- dependentandnonlinearpartialdifferentialequations,” SIAM Journal on Scientific Computing, Vol. 40, No. 1, 2018, pp. A172–A198

    2018

  19. [27]

    Uncertainty propagation using infinite mixture of gaus- sian processes and variational bayesian inference,

    Chen, P., Zabaras, N., and Bilionis, I., “Uncertainty propagation using infinite mixture of gaus- sian processes and variational bayesian inference,”Journal of computational physics, Vol. 284, 2015, pp. 291–333

    2015

  20. [28]

    Multi-output separable Gaussian pro- cess: Towards an efficient, fully Bayesian paradigm for uncertainty quantification,

    Bilionis, I., Zabaras, N., Konomi, B. A., and Lin, G., “Multi-output separable Gaussian pro- cess: Towards an efficient, fully Bayesian paradigm for uncertainty quantification,”Journal of Computational Physics, Vol. 241, 2013, pp. 212–239. 26

    2013

  21. [29]

    Gaussian processes in machine learning,

    Rasmussen, C. E., “Gaussian processes in machine learning,”Summer school on machine learn- ing, Springer, 2003, pp. 63–71

    2003

  22. [30]

    Statistical Time Series Meth- ods for Multicopter Fault Detection and Identification,

    Dutta, A., McKay, M., Kopsaftopoulos, F., and Gandhi, F., “Statistical Time Series Meth- ods for Multicopter Fault Detection and Identification,”Vertical Flight Society International Powered Lift Conference, San Jose, CA, AHS, Jan 2020

    2020

  23. [31]

    Aircraft parametric structural load monitoring using gaussian process regression,

    Fuentes, R., Cross, E., Halfpenny, A., Worden, K., and Barthorpe, R. J., “Aircraft parametric structural load monitoring using gaussian process regression,”EWSHM-7th European workshop on structural health monitoring, 2014

    2014

  24. [32]

    Probabilistic Damage Quantificationvia the Integration of Non-parametric Time-series and Gaussian Process Regression Models,

    Amer, A. and Kopsaftopoulos, F. P., “Probabilistic Damage Quantificationvia the Integration of Non-parametric Time-series and Gaussian Process Regression Models,”Proceedings of the 12th International Workshop on Structural Health Monitoring (IWSHM 2019), Palo Alto, CA, USA, Se...

    2019

  25. [33]

    An information theoretic approach to use high-fidelity codes to calibrate low-fidelity codes,

    Lewis, A., Smith, R., Williams, B., and Figueroa, V., “An information theoretic approach to use high-fidelity codes to calibrate low-fidelity codes,”Journal of Computational Physics, Vol. 324, 2016, pp. 24–43

    2016

  26. [34]

    Multi-fidelity optimization via surrogate modelling,

    Forrester, A. I., Sóbester, A., and Keane, A. J., “Multi-fidelity optimization via surrogate modelling,” Proceedings of the royal society a: mathematical, physical and engineering sciences, Vol. 463, No. 2088, 2007, pp. 3251–3269

    2007

  27. [35]

    Recursive co-kriging model for design of computer experiments with multiple levels of fidelity,

    Le Gratiet, L. and Garnier, J., “Recursive co-kriging model for design of computer experiments with multiple levels of fidelity,”International Journal for Uncertainty Quantification, Vol. 4, No. 5, 2014

    2014

  28. [36]

    Multi-fidelity modelling via recursive co-kriging and Gaussian–Markov random fields,

    Perdikaris, P., Venturi, D., Royset, J. O., and Karniadakis, G. E., “Multi-fidelity modelling via recursive co-kriging and Gaussian–Markov random fields,”Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 471, No. 2179, 2015, pp. 20150018

    2015

  29. [37]

    Com- bining experimental data and computer simulations, with an application to flyer plate experi- ments,

    Gattiker, J., Higdon, D., Keller-McNulty, S., McKay, M., Moore, L., and Williams, B., “Com- bining experimental data and computer simulations, with an application to flyer plate experi- ments,” Bayesian Analysis, Vol. 1, No. 4, 2006, pp. 765–792

    2006

  30. [38]

    Computer model calibration using high-dimensional output,

    Higdon, D., Gattiker, J., Williams, B., and Rightley, M., “Computer model calibration using high-dimensional output,” Journal of the American Statistical Association, Vol. 103, No. 482, 2008, pp. 570–583

    2008

  31. [39]

    Deep multi-fidelity Gaussian processes,

    Raissi, M. and Karniadakis, G., “Deep multi-fidelity Gaussian processes,” arXiv preprint arXiv:1604.07484, 2016

    2016 arXiv

  32. [40]

    Inferring solutions of differential equations using noisy multi-fidelity data,

    Raissi, M., Perdikaris, P., and Karniadakis, G. E., “Inferring solutions of differential equations using noisy multi-fidelity data,”Journal of Computational Physics, Vol. 335, 2017, pp. 736–746

    2017

  33. [41]

    Machine learning of linear differential equa- tionsusingGaussianprocesses,

    Raissi, M., Perdikaris, P., and Karniadakis, G. E., “Machine learning of linear differential equa- tionsusingGaussianprocesses,” Journal of Computational Physics, Vol.348, 2017, pp.683–693

    2017

  34. [42]

    Multi-fidelity Bayesian neural networks: Al- gorithms and applications,

    Meng, X., Babaee, H., and Karniadakis, G. E., “Multi-fidelity Bayesian neural networks: Al- gorithms and applications,”Journal of Computational Physics, Vol. 438, 2021, pp. 110361. 27

    2021

  35. [43]

    A deep neural network, multi-fidelity surrogate model approach for Bayesian model updating in shm,

    Torzoni, M., Manzoni, A., and Mariani, S., “A deep neural network, multi-fidelity surrogate model approach for Bayesian model updating in shm,”European Workshop on Structural Health Monitoring, Springer, 2023, pp. 1076–1086

    2023

  36. [44]

    Damage State Estimation via Multi-fidelity Gaussian Process Regression Models for Active-Sensing Structure Health Monitoring,

    Fan, Y. and Kopsaftopoulos, F., “Damage State Estimation via Multi-fidelity Gaussian Process Regression Models for Active-Sensing Structure Health Monitoring,”European Workshop on Structural Health Monitoring, Springer, 2023, pp. 267–276

    2023

  37. [45]

    Loadmonitoringandcompensationstrategiesforguided- waves based structural health monitoring using piezoelectric transducers,

    Roy, S., Ladpli, P., andChang, F.-K., “Loadmonitoringandcompensationstrategiesforguided- waves based structural health monitoring using piezoelectric transducers,”Journal of Sound and Vibration, Vol. 351, 2015, pp. 206–220

    2015

  38. [46]

    A novel physics-based temperature compensation model for structural health monitoring using ultrasonic guided waves,

    Roy, S., Lonkar, K., Janapati, V., and Chang, F.-K., “A novel physics-based temperature compensation model for structural health monitoring using ultrasonic guided waves,”Structural Health Monitoring, Vol. 13, No. 3, 2014, pp. 321–342

    2014

  39. [47]

    GaussianProcessRegressionforActiveSensingProbabilistic Structural Health Monitoring: Experimental Assessment Across Multiple Damage and Loading Scenarios,

    Amer, A.andKopsaftopoulos, F., “GaussianProcessRegressionforActiveSensingProbabilistic Structural Health Monitoring: Experimental Assessment Across Multiple Damage and Loading Scenarios,” arXiv preprint arXiv:2106.14841, 2021. 28 Appendices A First Test Case A.1 Task1 a b c d ...

    2021 arXiv

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.