A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data

Wind farm performance monitoring has traditionally relied on deterministic models, such as power curves or machine learning approaches, which often fail to account for farm-wide behavior and the uncertainty quantification necessary for the reliable detection of underperformance. To overcome these li...

Full description

Saved in:
Bibliographic Details
Main Authors: Francisco Javier Jara Ávila, Timothy Verstraeten, Pieter Jan Daems, Ann Nowé, Jan Helsen
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/18/14/3764
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849418228230520832
author Francisco Javier Jara Ávila
Timothy Verstraeten
Pieter Jan Daems
Ann Nowé
Jan Helsen
author_facet Francisco Javier Jara Ávila
Timothy Verstraeten
Pieter Jan Daems
Ann Nowé
Jan Helsen
author_sort Francisco Javier Jara Ávila
collection DOAJ
description Wind farm performance monitoring has traditionally relied on deterministic models, such as power curves or machine learning approaches, which often fail to account for farm-wide behavior and the uncertainty quantification necessary for the reliable detection of underperformance. To overcome these limitations, we propose a probabilistic methodology for turbine-level active power prediction and uncertainty estimation using high-frequency SCADA data and farm-wide autoregressive information. The method leverages a Stochastic Variational Gaussian Process with a Linear Model of Coregionalization, incorporating physical models like manufacturer power curves as mean functions and enabling flexible modeling of active power and its associated variance. The approach was validated on a wind farm in the Belgian North Sea comprising over 40 turbines, using only 15 days of data for training. The results demonstrate that the proposed method improves predictive accuracy over the manufacturer’s power curve, achieving a reduction in error measurements of around 1%. Improvements of around 5% were seen in dominant wind directions (200°–300°) using 2 and 3 Latent GPs, with similar improvements observed on the test set. The model also successfully reconstructs wake effects, with Energy Ratio estimates closely matching SCADA-derived values, and provides meaningful uncertainty estimates and posterior turbine correlations. These results demonstrate that the methodology enables interpretable, data-efficient, and uncertainty-aware turbine-level power predictions, suitable for advanced wind farm monitoring and control applications, enabling a more sensitive underperformance detection.
format Article
id doaj-art-10c8a21e6d63470bb5d1dfa0dee461ec
institution Kabale University
issn 1996-1073
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-10c8a21e6d63470bb5d1dfa0dee461ec2025-08-20T03:32:31ZengMDPI AGEnergies1996-10732025-07-011814376410.3390/en18143764A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency DataFrancisco Javier Jara Ávila0Timothy Verstraeten1Pieter Jan Daems2Ann Nowé3Jan Helsen4Acoustics and Vibrations Research Group, Vrije Universiteit Brussel, 1050 Brussels, BelgiumAcoustics and Vibrations Research Group, Vrije Universiteit Brussel, 1050 Brussels, BelgiumAcoustics and Vibrations Research Group, Vrije Universiteit Brussel, 1050 Brussels, BelgiumArtificial Intelligence Lab, Vrije Universiteit Brussel, 1050 Brussels, BelgiumAcoustics and Vibrations Research Group, Vrije Universiteit Brussel, 1050 Brussels, BelgiumWind farm performance monitoring has traditionally relied on deterministic models, such as power curves or machine learning approaches, which often fail to account for farm-wide behavior and the uncertainty quantification necessary for the reliable detection of underperformance. To overcome these limitations, we propose a probabilistic methodology for turbine-level active power prediction and uncertainty estimation using high-frequency SCADA data and farm-wide autoregressive information. The method leverages a Stochastic Variational Gaussian Process with a Linear Model of Coregionalization, incorporating physical models like manufacturer power curves as mean functions and enabling flexible modeling of active power and its associated variance. The approach was validated on a wind farm in the Belgian North Sea comprising over 40 turbines, using only 15 days of data for training. The results demonstrate that the proposed method improves predictive accuracy over the manufacturer’s power curve, achieving a reduction in error measurements of around 1%. Improvements of around 5% were seen in dominant wind directions (200°–300°) using 2 and 3 Latent GPs, with similar improvements observed on the test set. The model also successfully reconstructs wake effects, with Energy Ratio estimates closely matching SCADA-derived values, and provides meaningful uncertainty estimates and posterior turbine correlations. These results demonstrate that the methodology enables interpretable, data-efficient, and uncertainty-aware turbine-level power predictions, suitable for advanced wind farm monitoring and control applications, enabling a more sensitive underperformance detection.https://www.mdpi.com/1996-1073/18/14/3764wind farmSCADApossible powerkernelsuncertainty
spellingShingle Francisco Javier Jara Ávila
Timothy Verstraeten
Pieter Jan Daems
Ann Nowé
Jan Helsen
A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
Energies
wind farm
SCADA
possible power
kernels
uncertainty
title A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
title_full A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
title_fullStr A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
title_full_unstemmed A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
title_short A Methodology for Turbine-Level Possible Power Prediction and Uncertainty Estimations Using Farm-Wide Autoregressive Information on High-Frequency Data
title_sort methodology for turbine level possible power prediction and uncertainty estimations using farm wide autoregressive information on high frequency data
topic wind farm
SCADA
possible power
kernels
uncertainty
url https://www.mdpi.com/1996-1073/18/14/3764
work_keys_str_mv AT franciscojavierjaraavila amethodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT timothyverstraeten amethodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT pieterjandaems amethodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT annnowe amethodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT janhelsen amethodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT franciscojavierjaraavila methodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT timothyverstraeten methodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT pieterjandaems methodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT annnowe methodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata
AT janhelsen methodologyforturbinelevelpossiblepowerpredictionanduncertaintyestimationsusingfarmwideautoregressiveinformationonhighfrequencydata