Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Integrating Physics and Data-Driven Approaches: An Explainable and Uncertainty-Aware Hybrid Model for Wind Turbine Power Prediction

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

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

Pith's one-line read A residual neural network trained on the gap between a physics-based power model and observed output cuts wind power prediction error by 37% while keeping the model explainable.

desk verdict Competent hybrid power-curve paper whose headline 37% gain is likely inflated by random split and heavy filtering; fixable with a chronological split. read the letter →

arxiv 2502.07344 v1 pith:DSFAJDWS submitted 2025-02-11 cs.LG cs.AIcs.CE

Alfonso Gijón , Simone Eiraudo , Antonio Manjavacas , Daniele Salvatore Schiera , Miguel Molina-Solana , Juan Gómez-Romero This is my paper · ORCID
classification cs.LGcs.AIcs.CE
keywords hybridmodelswindturbinepowerpredictionresiduallearningSHAPconformaluncertaintyquantificationcoefficientSCADAdata
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 wind-turbine power can be predicted more accurately without giving up physical interpretability by adding a data-driven residual model on top of a physics-based one. On four turbines at the La Haute Borne farm, the hybrid reduces mean absolute percentage error from 3.71% for the physics-only model to 2.32%, a 37% relative improvement, and matches a purely data-driven model. The residual network is explained with SHAP, which also reveals which unmodeled variables matter, and conformalized quantile regression supplies prediction intervals with 86% coverage. If correct, this gives operators an accurate, trustworthy predictor that can flag anomalies and guide turbine control.

What carries the argument

The load-bearing object is the additive decomposition $\hat{P} = P_{\mathrm{phys}} + P_{\mathrm{res}}$. The physics term uses the wind-power equation $P = \tfrac{1}{2}C_p\rho A v^3$, but the power coefficient $C_p$ is not fixed to an empirical curve; it is learned by a neural network whose sigmoid output respects the Betz limit $C_p \leq 0.5926$, keeping the term interpretable and differentiable. The residual term is a second neural network trained on the residuals $y - P_{\mathrm{phys}}$ using eight inputs, including temperatures and angles, so it can absorb phenomena the equation misses. Explainability uses SHAP's additivity to split each prediction into physical and residual contributions, and uncertainty is produced by conformalized quantile regression, which trains lower and upper quantile models and calibrates interval widths on held-out data.

What would settle it

Re-run the comparison on a chronological hold-out, such as the last 20% of the 2013–2020 records, or on the discarded records with $C_p > 0.5926$, 3-sigma anomalies, and low-speed cutoffs, and check whether the hybrid model's MAPE stays near 2.3% against the physics-based model's 3.7%; if the gap shrinks, the reported 37% improvement is an artifact of the split.

Watch

Extended reading notes

Core claim

The paper's central claim is that the additive hybrid model $\hat{P} = P_{\mathrm{phys}}(v,\theta,\omega) + P_{\mathrm{res}}(v,\theta,\omega,T_{\mathrm{out}},T_n,T_r,\alpha_v,\alpha_w)$, with $P_{\mathrm{phys}}$ built from the kinetic-energy formula $P = \tfrac{1}{2}C_p\rho A v^3$ and $C_p$ learned by a bounded neural network, achieves MAE 11.73 kW, RMSE 24.48 kW, MAPE 2.32%, and R² 0.9976 on the test split, versus physics-only MAE 16.31 kW and MAPE 3.71%. The residual term is responsible for the gain and is most active at intermediate wind speeds between 7.5 and 12.5 m/s. SHAP values rank wind speed, rotor speed, and outdoor temperature as the residual model's main drivers, and the paper presents this as the first hybrid wind-power model to combine explainability with conformal uncertainty quantification.

Load-bearing premise

The reliability of the reported improvement rests on the assumption that the cleaned, randomly split 10-minute SCADA records fairly represent normal turbine operation and carry no temporal leakage.

Editorial extensions

If this is right

  • The physics-based term alone gives a usable benchmark, and adding the residual term brings error down to the level of a full black-box model, so operators do not sacrifice accuracy for interpretability.
  • Because the residual term is systematically nonzero in the 7.5 to 12.5 m/s range, the model identifies the wind-speed regime where the physics equation most needs correction.
  • The model is fully differentiable, so it can be embedded in pitch or rotor-speed control optimization rather than used only as a forecasting tool.
  • The same additive structure could transfer to a different turbine or farm by retraining the residual submodel while updating the physics-based parameters separately.

Reading between the lines

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

  • The random 80/20 split means adjacent 10-minute records from the same turbine can appear in both training and test sets; a chronological hold-out would be a stricter test of the 37% improvement.
  • A testable consequence of the SHAP analysis is that adding outdoor temperature as an explicit factor in the power equation should shrink the residual term; this could be checked by fitting $P_{\mathrm{phys}}$ with a temperature correction and measuring the remaining residual.
  • The residual model's near-zero output at low and high wind speeds suggests the physics equation is already adequate there, so future data collection could focus on the mid-speed range where the largest prediction intervals occur.
  • If the transferability claim holds, fine-tuning only the residual submodel on a new turbine should require substantially less data than training a black-box power curve from scratch.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes a hybrid semi-parametric model for wind turbine power prediction, combining a physics-inspired submodel Pphys = (1/2)CpρAv^3 with a neural-network residual submodel Pres, so that total power is P = Pphys + Pres (Eq. 5). The Cp in the physics submodel is itself learned by a neural network constrained by the Betz limit. The residual submodel uses a broader set of eight features. The authors report on a 10-minute SCADA dataset from four turbines at La Haute Borne, after removing roughly 30% of raw records via Cp > Betz, 3-sigma anomaly filtering, and a low-velocity cutoff, and after a random 80/20 train/test split. Table 2 shows the hybrid model reduces MAPE by 37% relative to the physics-based model and approximately matches a purely data-driven model. SHAP analysis is used to interpret the residual submodel, and a conformalized quantile regression (CQR) method provides prediction intervals with an empirical coverage of 86% and mean width of 49 kW.

Significance. If the reported improvements are robust, the additive hybrid architecture is a useful contribution to wind-power prediction, combining a partially interpretable physics component with a flexible residual correction, and the SHAP-based explainability plus CQR uncertainty quantification is a valuable template for real-world deployment. The paper also makes its code publicly available, which supports reproducibility. The central claim of a 37% MAPE improvement over the physics-based model is, however, contingent on the evaluation protocol: the random temporal split and aggressive preprocessing may inflate accuracy, and the empirical coverage below the nominal level weakens the uncertainty-quantification claim. These issues are addressable, but they are load-bearing for the headline results.

major comments (4)
  1. [Section 4.4] The random 80/20 split of a 10-minute SCADA time series creates a test set interleaved in time with the training set. Consecutive records in such series are strongly autocorrelated (reflected in the high R²=0.9953 of the physics model alone), so the flexible residual network in Eq. (5) can exploit temporal proximity to near-training samples rather than learning a transferable correction. The reported 37% MAPE improvement is therefore an optimistic estimate of out-of-sample performance. A chronological split (e.g., train on earlier years, test on later years) or a blocked/grouped cross-validation respecting time order is needed to support the abstract's claim; without it, the headline result is not reliable.
  2. [Section 4.2] The preprocessing removes approximately 30% of raw records: points with Cp > Betz limit, 3-sigma anomalies relative to a median power curve, and low-wind-speed data. These excluded records are precisely the regimes—low wind, transients, sensor noise—where a deployed power-prediction model must operate. Evaluation on the remaining filtered subset means Table 2 measures accuracy only on 'normal operation' samples. The paper should either report metrics on the raw unfiltered data (or a held-out chronological block before filtering) or explicitly scope the claim to the filtered regime. Without this, the 37% improvement may not generalize.
  3. [Section 4.3 and Eq. (5)] The 'physics-based' submodel is not a closed-form physical law: the power coefficient Cp is a neural network fitted to the same data, and the residual submodel is defined to fit r = y − Pphys. Consequently, the hybrid model's improvement over Pphys partially reflects increased model capacity and a two-stage fitting procedure, not an independent physical derivation. To substantiate the claim of physics grounding, the authors should compare against a monolithic neural network with the same total parameter count and input features, or otherwise demonstrate that the physics-based form (rather than the extra capacity) is responsible for the accuracy gain and interpretability.
  4. [Section 5.2] The CQR procedure reports an empirical coverage of 86% with a mean interval length of 49 kW, but the nominal confidence level is not explicitly stated. The text mentions training 'upper (95%) and lower (5%)' quantile models, which suggests a nominal 90% interval. A conformal method should give coverage close to the nominal level up to finite-sample slack; a 4-point shortfall indicates either a miscalibrated quantile model, a too-small calibration set, or a mismatch between the split procedure and the conformal correction. The authors should report the nominal α, the calibration set size, and analyze why coverage is below target; this is essential for the paper's reliability- and uncertainty-quantification claims.
minor comments (4)
  1. [Table 2] All metrics are reported from a single random split with no variance estimates or repeated runs. Given the sensitivity of neural networks to initialization and the stochastic split, reporting mean±std over several seeds (or at least over multiple splits) would strengthen the comparison.
  2. [Section 4.4] The text says hyperparameters were optimized with the hyperband algorithm and also mentions a grid search; please clarify whether a grid search was used to define the hyperband search space or as a separate procedure.
  3. [Figure 9] The caption says the data curve is 'nearly indistinguishable to the human eye' and lies beneath the model curve; the figure may benefit from a zoomed inset or a residual plot to visualize the difference and the uncertainty intervals more clearly.
  4. [Section 4.3] The sigmoid output layer constrains Cp to (0,1), but the text should state whether any additional constraint (e.g., a lower bound of zero) is imposed on the power output and how negative residual predictions are handled.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the residual submodel is fitted by definition, but the 37% gain is a held-out empirical comparison, not a construction-forced result.

full rationale

The paper's central result is an empirical model comparison on a held-out test set, not a derivation from first principles. The residual submodel is trained on r = y - fphys(x) by definition (Eq. 5 and Section 3.1), but this does not force the reported test-set improvement: a residual network could fail to improve held-out MAPE if it overfits or if the physics submodel already captured the signal. The reported metrics in Table 2 are evaluated on a 20% held-out split, so the 37% MAPE improvement is not an identity. The 'physics-based' submodel does use a neural-network power coefficient fitted to the same data (Section 4.3), which weakens the label 'physics-based' and is a legitimate modeling and baseline concern, but the paper explicitly discloses this and treats it as a physics-inspired architecture, not a closed-form physical law. Self-citations [52] and [56] are used only for qualitative consistency and as a prior version of this work; neither supplies a load-bearing premise or forbids alternatives. The main risks identified in the manuscript - random 80/20 split of autocorrelated 10-minute SCADA data and removal of about 30% of records via Betz-limit, 3-sigma, and low-velocity filters (Section 4.2) - concern experimental design and generalization, not circularity. The paper also explicitly cautions that SHAP correlations are not causal, avoiding an overclaim that would invite a circularity objection. Overall, no step reduces by construction to its own input.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The hybrid model's accuracy rests on two fitted neural networks (Cp and residual) plus several author-chosen thresholds and a random temporal split. No new physical entities are introduced, so the ledger is dominated by fitted parameters and domain assumptions rather than invented objects.

free parameters (6)
  • Cp neural network weights = not reported
    Fitted to map (v, theta, omega) to power coefficient in Section 4.3; the 'physics-based' submodel therefore depends on a learned function, not a fixed physical constant.
  • Residual neural network weights = not reported
    Trained on r = y - Pphys in Section 4.3 to predict residuals from eight features; these weights carry the 37% improvement.
  • Hyperparameters (activation, learning rate, layers, neurons) = selected by Hyperband grid search
    Section 4.4: hyperparameters were optimized with Keras Tuner; batch size 128 and MAE loss are reported, but final architecture values are not stated.
  • Data filtering thresholds = Cp <= 0.5926, 3-sigma anomaly, low-velocity cutoff
    Section 4.2: these author-chosen cutoffs remove 30% of the raw data and define the population on which all metrics are computed.
  • Feature selection correlation threshold = 0.9
    Section 5.1: variables with correlation below 0.9 were excluded, reducing 35 features to 9.
  • Conformal quantile levels = 5% and 95% (alpha not stated)
    Section 5.2: quantile models are trained at alpha/2 and 1-alpha/2; the achieved coverage is 86%, below the nominal 90% if alpha=0.1.
assumptions (5)
  • domain assumption Betz limit 0.5926 is the upper bound on Cp.
    Used to constrain the sigmoid output layer of the Cp network in Section 4.3.
  • domain assumption P = 0.5 Cp rho A v^3 is a valid approximate model for turbine power.
    Eq. 4 is the actuator disk relation that defines the physics submodel; rho and A are not separately calibrated in the paper.
  • domain assumption Cp has no known physical closed form and can be represented by a neural network.
    Section 4.1 states 'The power coefficient's functional form is not dictated by physical laws'; this is the premise that lets the physics submodel be partly learned.
  • ad hoc to paper Filtered records (30% of raw data) are non-physical or anomalous noise.
    Section 4.2 removes Cp > Betz, 3-sigma deviations, and low-velocity points; if these records include real operating states, the reported errors are conditional on a filtered subpopulation.
  • domain assumption Random 80/20 split yields independent training and test sets for time-series data.
    Section 4.4 applies a random split to 10-minute SCADA records; temporal autocorrelation could leak information and inflate test accuracy.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Integrating Physics and Data-Driven Approaches: An Explainable and Uncertainty-Aware Hybrid Model for Wind Turbine Power Prediction." pith.science (2026). https://pith.science/paper/DSFAJDWS

@misc{pith2026250207344,
  author       = {Pith},
  title        = {Pith review of: Integrating Physics and Data-Driven Approaches: An Explainable and Uncertainty-Aware Hybrid Model for Wind Turbine Power Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DSFAJDWS}},
  note         = {Machine review of arXiv:2502.07344}
}
read the original abstract

The rapid growth of the wind energy sector underscores the urgent need to optimize turbine operations and ensure effective maintenance through early fault detection systems. While traditional empirical and physics-based models offer approximate predictions of power generation based on wind speed, they often fail to capture the complex, non-linear relationships between other input variables and the resulting power output. Data-driven machine learning methods present a promising avenue for improving wind turbine modeling by leveraging large datasets, enhancing prediction accuracy but often at the cost of interpretability. In this study, we propose a hybrid semi-parametric model that combines the strengths of both approaches, applied to a dataset from a wind farm with four turbines. The model integrates a physics-inspired submodel, providing a reasonable approximation of power generation, with a non-parametric submodel that predicts the residuals. This non-parametric submodel is trained on a broader range of variables to account for phenomena not captured by the physics-based component. The hybrid model achieves a 37% improvement in prediction accuracy over the physics-based model. To enhance interpretability, SHAP values are used to analyze the influence of input features on the residual submodel's output. Additionally, prediction uncertainties are quantified using a conformalized quantile regression method. The combination of these techniques, alongside the physics grounding of the parametric submodel, provides a flexible, accurate, and reliable framework. Ultimately, this study opens the door for evaluating the impact of unmodeled variables on wind turbine power generation, offering a basis for potential optimization.

Figures

Figures reproduced from arXiv: 2502.07344 by the authors.

Figure 1
Figure 1. Box plots of wind speed, pitch angle, rotor speed, outdoor temperature and [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗
Figure 2
Figure 2. Average power curve derived from the data of the four wind turbines. [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Diagram of the hybrid model, according to the taxonomy presented in [ [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The Cp curves derived from the physics-based model are represented as a function of the tip speed ratio λ, for various pitch angles in different colors (measured in radians). The observed data are represented with solid lines, while the model predictions are shown with…
Figure 5
Figure 5. Figure 5: (a) Absolute residual between the true power and the prediction made by the [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: The contributions of the physics-based model (orange) and the residual model [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: (a) Ranking of mean absolute SHAP values of the hybrid model. (b) Sum of [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: SHAP value of the residual power model vs. associated variable for (a) wind [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]
Figure 9
Figure 9. Figure 9: Power curve comparing model predictions with uncertainty intervals (blue), [PITH_FULL_IMAGE:figures/full_fig_p019_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Right-sizing Recommendations (RSR): Cloud Workload Conformal Prediction for Virtual Machines in Data Center Operations

    cs.AI 2026-06 conditional novelty 4.0 of 10

    On long-lived Azure VM workloads, XGBoost and CatBoost with bootstrap conformal prediction produce the best-ranked upper-bound CPU utilization intervals for right-sizing recommendations.

Reference graph

Works this paper leans on

55 extracted references · 40 canonical work pages · cited by 1 Pith paper

  1. [1]

    International Renewable Energy Agency, World Energy Transitions Outlook 2023: 1.5 °C Pathway, Volume 1, International Renewable (2023)

    work page 2023

  2. [2]

    J. Wu, S. Guo, J. Li, D. Zeng, Big data meet green challenges: Big data toward green applications, IEEE Systems Journal 10 (3) (2016) 888–900. doi:10.1109/JSYST.2016.2550530

    work page arXiv 2016

  3. [3]

    Z. Yao, Y. Lum, A. Johnston, L. M. Mejia-Mendoza, X. Zhou, Y. Wen, A. Aspuru-Guzik, E. H. Sargent, Z. W. Seh, Machine learning for a sustainable energy future, Nature Reviews Materials 8 (3) (2023) 202–

    work page 2023

  4. [4]

    Munteanu, A

    I. Munteanu, A. I. Bratcu, E. Ceang ˘A, N.-A. Cutululis, Optimal control of wind energy systems: towards a global approach, Vol. 22, Springer, 2008

    work page 2008

  5. [5]

    O. T. Bindingsbø, M. Singh, K. Øvsthus, A. Keprate, Fault detection of a wind turbine generator bearing using interpretable machine learning, Frontiers in Energy Research 11 (2023). doi:10.3389/fenrg.2023. 1284676

    work page doi:10.3389/fenrg.2023 2023

  6. [7]

    Watson, L

    S. Watson, L. Landberg, J. Halliday, Application of wind speed forecast- ing to the integration of wind energy into a large scale power system, IEE Proceedings - Generation, Transmission and Distribution 141 (1994) 357–362. doi:10.1049/ip-gtd_19941215

    work page doi:10.1049/ip-gtd_19941215 1994

  7. [8]

    Benbouzid, T

    M. Benbouzid, T. Berghout, N. Sarma, S. Djurovi´ c, Y. Wu, X. Ma, Intelligent condition monitoring of wind power systems: State of the art review, Energies 14 (18) (2021). doi:10.3390/en14185967. 22

    work page doi:10.3390/en14185967 2021

  8. [10]

    Pujana, M

    A. Pujana, M. Esteras, E. Perea, E. Maqueda, P. Calvez, Hybrid-model- based digital twin of the drivetrain of a wind turbine and its application for failure synthetic data generation, Energies 16 (2) (2023)

    work page 2023

Show all 55 references
  1. [11]

    Fern´ andez de la Mata, A

    F. Fern´ andez de la Mata, A. Gij´ on, M. Molina-Solana, J. G´ omez- Romero, Physics-informed neural networks for data-driven simulation: Advantages, limitations, and opportunities, Physica A: Statistical Me- chanics and its Applications 610 (2023) 128415.doi:10.1016/j.physa. ...

    2023

  2. [12]

    von Stosch, R

    M. von Stosch, R. Oliveira, J. Peres, S. Feyo de Azevedo, Hybrid semi- parametric modeling in process systems engineering: Past, present and future, Computers & Chemical Engineering 60 (2014) 86–101. doi: 10.1016/j.compchemeng.2013.08.008

    2014 doi

  3. [13]

    Roscher, B

    R. Roscher, B. Bohn, M. F. Duarte, J. Garcke, Explainable machine learning for scientific insights and discoveries, IEEE Access 8 (2020) 42200–42216. doi:10.1109/ACCESS.2020.2976199

    2020

  4. [14]

    Abdar, F

    M. Abdar, F. Pourpanah, S. Hussain, D. Rezazadegan, L. Liu, M. Ghavamzadeh, P. Fieguth, X. Cao, A. Khosravi, U. R. Acharya, V. Makarenkov, S. Nahavandi, A review of uncertainty quantification in deep learning: Techniques, applications and challenges, Inf. Fusion 76 (C) (2021) ...

    2021 doi

  5. [15]

    C. Chen, H. Liu, Medium-term wind power forecasting based on multi- resolution multi-learner ensemble and adaptive model selection, En- ergy Conversion and Management 206 (2020) 112492. doi:10.1016/ j.enconman.2020.112492

    2020

  6. [16]

    Zehtabiyan-Rezaie, A

    N. Zehtabiyan-Rezaie, A. Iosifidis, M. Abkar, Data-driven fluid mechan- ics of wind farms: A review, Journal of Renewable and Sustainable En- ergy 14 (3) (2022) 032703

    2022

  7. [17]

    M. F. Howland, J. O. Dabiri, Wind farm modeling with interpretable physics-informed machine learning, Energies 12 (14) (2019). doi:10. 3390/en12142716. 23

    2019

  8. [18]

    Carpintero-Renteria, D

    M. Carpintero-Renteria, D. Santos-Martin, A. Lent, C. Ramos, Wind turbine power coefficient models based on neural networks and poly- nomial fitting, IET renewable power generation. 14 (11) (2020-08). doi:10.1049/iet-rpg.2019.1162

    2020

  9. [19]

    O. C. Castillo, V. R. Andrade, J. J. R. Rivas, R. O. Gonz´ alez, Compar- ison of power coefficients in wind turbines considering the tip speed ratio and blade pitch angle, Energies 16 (6) (2023). doi:10.3390/ en16062774

    2023

  10. [21]

    Liu, H.-Q

    H. Liu, H.-Q. Tian, C. Chen, Y. fei Li, A hybrid statistical method to predict wind speed and wind power, Renewable Energy 35 (8) (2010) 1857–1861. doi:10.1016/j.renene.2009.12.011

    2010 doi

  11. [22]

    Clifton, L

    A. Clifton, L. Kilcher, J. K. Lundquist, P. Fleming, Using machine learning to predict wind turbine power output, Environmental Research Letters 8 (2) (2013) 024009. doi:10.1088/1748-9326/8/2/024009

    2013 doi

  12. [23]

    Heinermann, O

    J. Heinermann, O. Kramer, Machine learning ensembles for wind power prediction, Renewable Energy 89 (2016) 671–679. doi:10.1016/j. renene.2015.11.073

    2016 doi

  13. [24]

    H. Sun, C. Qiu, L. Lu, X. Gao, J. Chen, H. Yang, Wind turbine power modelling and optimization using artificial neural network with wind field experimental data, Applied Energy 280 (2020) 115880. doi:10. 1016/j.apenergy.2020.115880

    2020

  14. [25]

    Demolli, A

    H. Demolli, A. S. Dokuz, A. Ecemis, M. Gokcek, Wind power fore- casting based on daily wind speed data using machine learning algo- rithms, Energy Conversion and Management 198 (2019) 111823. doi: 10.1016/j.enconman.2019.111823

    2019

  15. [26]

    C. Moss, R. Maulik, G. V. Iungo, Augmenting insights from wind tur- bine data through data-driven approaches, Applied Energy 376 (2024) 124116. doi:10.1016/j.apenergy.2024.124116. 24

    2024

  16. [27]

    Dhungana, A machine learning approach for wind turbine power forecasting for maintenance planning, Energy Informatics 8 (1) (2025)

    H. Dhungana, A machine learning approach for wind turbine power forecasting for maintenance planning, Energy Informatics 8 (1) (2025)

    2025

  17. [28]

    J. Wang, H. Li, H. Wang, S. J. Pan, X. Xie, Trustworthy machine learn- ing: Robustness, generalization, and interpretability, in: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD ’23, Association for Computing Machinery, New York, NY, U...

    2023

  18. [29]

    doi:10.1186/s42162-024-00459-4

    doi

  19. [30]

    Tascikaraoglu, M

    A. Tascikaraoglu, M. Uzunoglu, A review of combined approaches for prediction of short-term wind speed and power, Renewable and Sustain- able Energy Reviews 34 (2014) 243–254. doi:10.1016/j.rser.2014. 03.033

    2014 doi

  20. [31]

    Stoffel, C

    P. Stoffel, C. L¨ offler, S. Eser, A. K¨ umpel, D. M¨ uller, Combining data- driven and physics-based process models for hybrid model predictive control of building energy systems, in: 2022 30th Mediterranean Con- ference on Control and Automation (MED), 2022, pp. 121–126. doi...

    2022

  21. [32]

    J. Chen, H. Liu, C. Chen, Z. Duan, Wind speed forecasting using multi- scale feature adaptive extraction ensemble model with error regression correction, Expert Systems with Applications 207 (2022) 117358. doi: 10.1016/j.eswa.2022.117358

    2022

  22. [33]

    H. Liu, C. Yu, H. Wu, Z. Duan, G. Yan, A new hybrid ensemble deep reinforcement learning model for wind speed short term forecasting, En- ergy 202 (2020) 117794. doi:10.1016/j.energy.2020.117794

    2020

  23. [34]

    Linardatos, V

    P. Linardatos, V. Papastefanopoulos, S. Kotsiantis, Explainable ai: A review of machine learning interpretability methods, Entropy (Basel) 23 (1) (2020) 18. doi:10.3390/e23010018

    2020 doi

  24. [35]

    S. Yin, H. Liu, Wind power prediction based on outlier correction, ensemble reinforcement learning, and residual correction, Energy 250 (2022) 123857. doi:10.1016/j.energy.2022.123857

    2022

  25. [36]

    Aerts, L

    F. Aerts, L. Lanzilao, J. Meyers, Bayesian uncertainty quantification framework for wake model calibration and validation with historical wind farm power data, Wind Energy 26 (8) (2023) 786–802. doi: 10.1002/we.2841

    2023 doi

  26. [37]

    G. A. M. van Kuik, J. Peinke, R. Nijssen, D. Lekou, J. Mann, J. N. Sørensen, C. Ferreira, J. W. van Wingerden, D. Schlipf, P. Gebraad, 25 H. Polinder, A. Abrahamsen, G. J. W. van Bussel, J. D. Sørensen, P. Tavner, C. L. Bottasso, M. Muskulus, D. Matha, H. J. Linde- boom, S. De...

    2016

  27. [38]

    J. W. Pascal Richter, M. Frank, Uncertainty quantification of offshore wind farms using monte carlo and sparse grid, Energy Sources, Part B: Economics, Planning, and Policy 17 (1) (2022) 2000520. doi:10.1080/ 15567249.2021.2000520

    2022

  28. [39]

    J. H. Mclean, M. R. Jones, B. J. O’Connell, E. Maguire, T. J. Rogers, Physically meaningful uncertainty quantification in probabilis- tic wind turbine power curve models as a damage-sensitive feature, Structural Health Monitoring 22 (6) (2023) 3623–3636. doi:10.1177/ 14759217231155379

    2023

  29. [40]

    W. Liao, J. Fang, L. Ye, B. Bak-Jensen, Z. Yang, F. Porte-Agel, Can we trust explainable artificial intelligence in wind power forecasting?, Applied Energy 376 (2024) 124273. doi:10.1016/j.apenergy.2024. 124273

    2024 doi

  30. [41]

    doi:10.1016/j.eswa.2023.121464

    Data-driven interpretable ensemble learning methods for the prediction of wind turbine power incorporating shap analysis, Expert Systems with Applications 237 (2024) 121464. doi:10.1016/j.eswa.2023.121464

    2024

  31. [42]

    H. F. Mahmoud, Parametric versus semi and nonparametric regression models, arXiv preprint arXiv:1906.10221 (2019). doi:10.48550/arXiv. 1906.10221. 26

    2019 doi arXiv

  32. [43]

    Letzgus, K.-R

    S. Letzgus, K.-R. M¨ uller, An explainable ai framework for robust and transparent data-driven wind turbine power curve models, Energy and AI 15 (2024) 100328. doi:10.1016/j.egyai.2023.100328

    2024

  33. [44]

    A. N. Angelopoulos, S. Bates, et al., Conformal prediction: A gentle introduction, Foundations and Trends ® in Machine Learning 16 (4) (2023) 494–591. doi:10.48550/arXiv.2107.07511

    2023 doi arXiv

  34. [45]

    S. M. Lundberg, S.-I. Lee, A unified approach to interpreting model pre- dictions, in: I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (Eds.), Advances in Neural Information Processing Systems 30, Curran Associates, Inc., 2017, pp. 4765–4774

    2017

  35. [46]

    Cordier, V

    T. Cordier, V. Blot, L. Lacombe, T. Morzadec, A. Capitaine, N. Brunel, Flexible and Systematic Uncertainty Estimation with Conformal Predic- tion via the MAPIE library, in: Proceedings of the Twelfth Symposium on Conformal and Probabilistic Prediction with Applications, Vol. 2...

    2023

  36. [47]

    Romano, E

    Y. Romano, E. Patterson, E. Cand` es, Conformalized quantile regression, Conference on Neural Information Processing Systems, NeurIPS (2019)

    2019

  37. [48]

    Effenberger, N

    N. Effenberger, N. Ludwig, A collection and categorization of open- source wind and wind power datasets, Wind Energy 25 (10) (2022) 1659–

    2022

  38. [49]

    Aerodynamics of Horizontal Axis Wind Turbines, John Wiley & Sons, Ltd, 2011, Ch. 3, pp. 39–136. doi:10.1002/9781119992714.ch3

    2011 doi

  39. [50]

    van Bekkum, M

    M. van Bekkum, M. de Boer, F. van Harmelen, A. Meyer-Vitali, A. t. Teije, Modular design patterns for hybrid learning and reasoning sys- tems, Applied Intelligence 51 (9) (2021) 6528–6546. doi:10.1007/ s10489-021-02394-3

    2021

  40. [51]

    la haute borne

    R. Mandzhieva, R. Subhankulova, Data-driven applications for wind energy analysis and prediction: The case of “la haute borne” wind farm, Digital Chemical Engineering 4 (2022) 100048. doi:10.1016/j.dche. 2022.100048

    2022

  41. [52]

    Gij´ on, A

    A. Gij´ on, A. Pujana-Goitia, E. Perea, M. Molina-Solana, J. G´ omez- Romero, Prediction of wind turbines power with physics-informed neural networks and evidential uncertainty quantification, Arxiv: 2307.14675 (2023). arXiv:2307.14675, doi:10.48550/arXiv.2307.14675

    2023 doi arXiv

  42. [53]

    Dillon, J

    E. Dillon, J. LaRiviere, S. Lundberg, J. Roth, V. Syrgkanis, Be careful when interpreting predictive models in search of causal insights, SHAP documentation. Accessed: 2025-01-22 (2025). URL https://shap.readthedocs.io 27

    2025

  43. [54]

    L. C. Pagnini, M. Burlando, M. P. Repetto, Experimental power curve of small-size wind turbines in turbulent urban environment, Applied Energy 154 (2015) 112–121. doi:10.1016/j.apenergy.2015.04.117

    2015 doi

  44. [55]

    Mehrjoo, M

    M. Mehrjoo, M. Jafari Jozani, M. Pawlak, Wind turbine power curve modeling for reliable power prediction using monotonic regression, Re- newable Energy 147 (08 2019). doi:10.1016/j.renene.2019.08.060

    2019 doi

  45. [56]

    Gij´ on, S

    A. Gij´ on, S. Eiraudo, A. Manjavacas, L. Bottaccioli, A. Lanzini, M. Molina-Solana, J. G´ omez-Romero, Explainable hybrid semi- parametric model for prediction of power generated by wind turbines, in: L. Franco, C. de Mulatier, M. Paszynski, V. V. Krzhizhanovskaya, J. J. Dong...

    2024 doi

  46. [57]

    J. Yan, H. Zhang, Y. Liu, S. Han, L. Li, Uncertainty estimation for wind energy conversion by probabilistic wind turbine power curve modelling, Applied Energy 239 (2019) 1356–1370. doi:10.1016/j.apenergy. 2019.01.180

    2019 doi

  47. [215]

    doi:10.1038/s41578-022-00490-5

    doi

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.