Prediction of wind turbines power with physics-informed neural networks and evidential uncertainty quantification
File(s)
Accepted version
Author(s)
Gijon, Alfonso
Pujana-Goitia, Ainhoa
Perea, Eugenio
Molina-Solana, Miguel
Gomez-Romero, Juan
Type
Journal Article
Abstract
The ever-growing use of wind energy requires the optimization of turbine operations through pitch angle controllers and early fault detection. Accurate and robust models that replicate turbine behavior are essential, particularly for predicting generated power from wind speed. Existing empirical and physics-based models often fail to capture the complex relationships between input variables and power output, aggravated by wind variability. Data-driven methods offer promising alternatives by improving model accuracy and scalability with large datasets. In this study, we use physics-informed neural networks to model historical data from four turbines in a wind farm, embedding physical constraints into the learning process. The proposed models predict power, torque, and power coefficient with high accuracy for both the data and the governing physical laws. Notably, neural networks improve the prediction of the power coefficient by an order of magnitude over empirical models. Physics-informed neural networks also show higher robustness than standard networks under limited data, maintaining accuracy even when trained on reduced datasets. Finally, the inclusion of an evidential layer provides uncertainty estimations that align with absolute errors and allow confidence intervals on the power curve, ensuring consistency with both observed data and manufacturer specifications.
Date Issued
2026-01-15
Date Acceptance
2025-11-23
Citation
Engineering Applications of Artificial Intelligence, 2026, 164 (Part B)
ISSN
0952-1976
Publisher
Elsevier
Journal / Book Title
Engineering Applications of Artificial Intelligence
Volume
164
Issue
Part B
Copyright Statement
Copyright © 2025 Elsevier Ltd. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
License URL
Subjects
Automation & Control Systems
BIG DATA
Computer Science
Computer Science, Artificial Intelligence
CURVE
ENERGY
Engineering
Engineering, Electrical & Electronic
Engineering, Multidisciplinary
Physics-informed neural networks
Power prediction
Science & Technology
SYSTEMS
Technology
Uncertainty quantification
Wind energy
Wind turbine modeling
Publication Status
Published
Article Number
113331
Date Publish Online
2025-11-30