Articles | Volume 9, issue 1
https://doi.org/10.5194/wes-9-181-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/wes-9-181-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Active trailing edge flap system fault detection via machine learning
Andrea Gamberini
CORRESPONDING AUTHOR
Siemens Gamesa Renewable Energy A/S, Brande, Denmark
Department of Wind and Energy Systems, DTU, Roskilde, Denmark
Imad Abdallah
Chair of Structural Mechanics and Monitoring, ETH Zurich, Zurich, Switzerland
Related authors
Andrea Gamberini, Thanasis Barlas, Alejandro Gomez Gonzalez, and Helge Aagaard Madsen
Wind Energ. Sci., 9, 1229–1249, https://doi.org/10.5194/wes-9-1229-2024, https://doi.org/10.5194/wes-9-1229-2024, 2024
Short summary
Short summary
Movable surfaces on wind turbine (WT) blades, called active flaps, can reduce the cost of wind energy. However, they still need extensive testing. This study shows that the computer model used to design a WT with flaps aligns well with measurements obtained from a 3month test on a commercial WT featuring a prototype flap. Particularly during flap actuation, there were minimal differences between simulated and measured data. These findings assure the reliability of WT designs incorporating flaps.
Philip Imanuel Franz, Imad Abdallah, Gregory Duthé, Julien Deparday, Ali Jafarabadi, Alexander Popp, Sarah Barber, and Eleni Chatzi
Wind Energ. Sci. Discuss., https://doi.org/10.5194/wes-2025-26, https://doi.org/10.5194/wes-2025-26, 2025
Preprint under review for WES
Short summary
Short summary
New designs of large wind turbine blades have become increasingly flexible, and thus need cost-efficient monitoring solutions. Hence, we investigate if aerodynamic pressure measurements from a low-cost sensing system can be used to detect structural damage. Our research is based on a wind tunnel study, emulating a simplified wind turbine blade under various conditions. We show that using a convolutional neural network-based method, structural damage can indeed be detected and its severity rated.
Andrea Gamberini, Thanasis Barlas, Alejandro Gomez Gonzalez, and Helge Aagaard Madsen
Wind Energ. Sci., 9, 1229–1249, https://doi.org/10.5194/wes-9-1229-2024, https://doi.org/10.5194/wes-9-1229-2024, 2024
Short summary
Short summary
Movable surfaces on wind turbine (WT) blades, called active flaps, can reduce the cost of wind energy. However, they still need extensive testing. This study shows that the computer model used to design a WT with flaps aligns well with measurements obtained from a 3month test on a commercial WT featuring a prototype flap. Particularly during flap actuation, there were minimal differences between simulated and measured data. These findings assure the reliability of WT designs incorporating flaps.
Yuriy Marykovskiy, Thomas Clark, Justin Day, Marcus Wiens, Charles Henderson, Julian Quick, Imad Abdallah, Anna Maria Sempreviva, Jean-Paul Calbimonte, Eleni Chatzi, and Sarah Barber
Wind Energ. Sci., 9, 883–917, https://doi.org/10.5194/wes-9-883-2024, https://doi.org/10.5194/wes-9-883-2024, 2024
Short summary
Short summary
This paper delves into the crucial task of transforming raw data into actionable knowledge which can be used by advanced artificial intelligence systems – a challenge that spans various domains, industries, and scientific fields amid their digital transformation journey. This article underscores the significance of cross-industry collaboration and learning, drawing insights from sectors leading in digitalisation, and provides strategic guidance for further development in this area.
Sarah Barber, Julien Deparday, Yuriy Marykovskiy, Eleni Chatzi, Imad Abdallah, Gregory Duthé, Michele Magno, Tommaso Polonelli, Raphael Fischer, and Hanna Müller
Wind Energ. Sci., 7, 1383–1398, https://doi.org/10.5194/wes-7-1383-2022, https://doi.org/10.5194/wes-7-1383-2022, 2022
Short summary
Short summary
Aerodynamic and acoustic field measurements on operating large-scale wind turbines are key for the further reduction in the costs of wind energy. In this work, a novel cost-effective MEMS (micro-electromechanical systems)-based aerodynamic and acoustic wireless measurement system that is thin, non-intrusive, easy to install, low power and self-sustaining is designed and tested.
Related subject area
Thematic area: Materials and operation | Topic: Operation and maintenance, condition monitoring, reliability
On the modeling errors of digital twins for load monitoring and fatigue assessment in wind turbine drivetrains
Machine-learning-based virtual load sensors for mooring lines using simulated motion and lidar measurements
Unsupervised anomaly detection of permanent-magnet offshore wind generators through electrical and electromagnetic measurements
Full-scale wind turbine performance assessment using the turbine performance integral (TPI) method: a study of aerodynamic degradation and operational influences
A machine learning-based approach for active monitoring of blades pitch misalignment in wind turbines
Operation and maintenance cost comparison between 15 MW direct-drive and medium-speed offshore wind turbines
Sensitivity of fatigue reliability in wind turbines: effects of design turbulence and the Wöhler exponent
Grand challenges in the digitalisation of wind energy
Overview of normal behavior modeling approaches for SCADA-based wind turbine condition monitoring demonstrated on data from operational wind farms
Assessing the rotor blade deformation and tower–blade tip clearance of a 3.4 MW wind turbine with terrestrial laser scanning
Introducing a data-driven approach to predict site-specific leading-edge erosion from mesoscale weather simulations
Population Based Structural Health Monitoring: Homogeneous Offshore Wind Model Development
Wind turbine main-bearing lubrication – Part 2: Simulation-based results for a double-row spherical roller main bearing in a 1.5 MW wind turbine
Reduction of wind-turbine-generated seismic noise with structural measures
Very low frequency IEPE accelerometer calibration and application to a wind energy structure
Felix C. Mehlan and Amir R. Nejad
Wind Energ. Sci., 10, 417–433, https://doi.org/10.5194/wes-10-417-2025, https://doi.org/10.5194/wes-10-417-2025, 2025
Short summary
Short summary
A digital twin is a virtual representation that mirrors the wind turbine's real behavior through simulation models and sensor measurements and can assist in making key decisions such as planning the replacement of parts. These models and measurements are, of course, not perfect and only give an incomplete picture of the real behavior. This study investigates how large the uncertainty of such models and measurements is and to what extent it affects the decision-making process.
Moritz Gräfe, Vasilis Pettas, Nikolay Dimitrov, and Po Wen Cheng
Wind Energ. Sci., 9, 2175–2193, https://doi.org/10.5194/wes-9-2175-2024, https://doi.org/10.5194/wes-9-2175-2024, 2024
Short summary
Short summary
This study explores a methodology using floater motion and nacelle-based lidar wind speed measurements to estimate the tension and damage equivalent loads (DELs) on floating offshore wind turbines' mooring lines. Results indicate that fairlead tension time series and DELs can be accurately estimated from floater motion time series. Using lidar measurements as model inputs for DEL predictions leads to similar accuracies as using displacement measurements of the floater.
Ali Dibaj, Mostafa Valavi, and Amir R. Nejad
Wind Energ. Sci., 9, 2063–2086, https://doi.org/10.5194/wes-9-2063-2024, https://doi.org/10.5194/wes-9-2063-2024, 2024
Short summary
Short summary
This study emphasizes the need for effective condition monitoring in permanent magnet offshore wind generators to tackle issues like demagnetization and eccentricity. Utilizing a machine learning model and high-resolution measurements, we explore methods of early fault detection. Our findings indicate that flux monitoring with affordable, easy-to-install stray flux sensors with frequency information offers a promising fault detection strategy for large megawatt-scale offshore wind generators.
Tahir H. Malik and Christian Bak
Wind Energ. Sci., 9, 2017–2037, https://doi.org/10.5194/wes-9-2017-2024, https://doi.org/10.5194/wes-9-2017-2024, 2024
Short summary
Short summary
We explore the effect of blade modifications on offshore wind turbines' performance through a detailed analysis of 12 turbines over 12 years. Introducing the turbine performance integral method, which utilises time-series decomposition that combines various data sources, we uncover how blade wear, repairs and software updates impact efficiency. The findings offer valuable insights into improving wind turbine operations, contributing to the enhancement of renewable energy technologies.
Sabrina Milani, Jessica Leoni, Stefano Cacciola, Alessandro Croce, and Mara Tanelli
Wind Energ. Sci. Discuss., https://doi.org/10.5194/wes-2024-100, https://doi.org/10.5194/wes-2024-100, 2024
Revised manuscript accepted for WES
Short summary
Short summary
In this paper, we propose a novel machine-learning framework pitch misalignment detection in wind turbines. By using a minimal set of standard sensors, our method detects misalignments as small as 0.1deg and localizes the affected blades. It combines signal processing with a hierarchical classification structure and linear regression for precise severity quantification. Evaluation results validate the approach, showing notable accuracy in misalignment classification, regression, and localization
Orla Donnelly, Fraser Anderson, and James Carroll
Wind Energ. Sci., 9, 1345–1362, https://doi.org/10.5194/wes-9-1345-2024, https://doi.org/10.5194/wes-9-1345-2024, 2024
Short summary
Short summary
We collate the latest reliability data in operations and maintenance (O&M) for offshore wind turbines, specifically large turbines of 15 MW. We use these data to model O&M of an offshore wind farm at three different sites. We compare two industry-dominant drivetrain configurations in terms of O&M cost for 15 MW turbines and determine if previous results for smaller turbines still hold true. Comparisons between drivetrains are topical within industry, and we produce cost comparisons for them.
Shadan Mozafari, Paul Veers, Jennifer Rinker, and Katherine Dykes
Wind Energ. Sci., 9, 799–820, https://doi.org/10.5194/wes-9-799-2024, https://doi.org/10.5194/wes-9-799-2024, 2024
Short summary
Short summary
Turbulence is one of the main drivers of fatigue in wind turbines. There is some debate on how to model the turbulence in normal wind conditions in the design phase. To address such debates, we study the fatigue load distribution and reliability following different models of the International Electrotechnical Commission 61400-1 standard. The results show the lesser importance of load uncertainty due to turbulence distribution compared to the uncertainty of material resistance and Miner’s rule.
Andrew Clifton, Sarah Barber, Andrew Bray, Peter Enevoldsen, Jason Fields, Anna Maria Sempreviva, Lindy Williams, Julian Quick, Mike Purdue, Philip Totaro, and Yu Ding
Wind Energ. Sci., 8, 947–974, https://doi.org/10.5194/wes-8-947-2023, https://doi.org/10.5194/wes-8-947-2023, 2023
Short summary
Short summary
Wind energy creates huge amounts of data, which can be used to improve plant design, raise efficiency, reduce operating costs, and ease integration. These all contribute to cheaper and more predictable energy from wind. But realising the value of data requires a digital transformation that brings
grand challengesaround data, culture, and coopetition. This paper describes how the wind energy industry could work with R&D organisations, funding agencies, and others to overcome them.
Xavier Chesterman, Timothy Verstraeten, Pieter-Jan Daems, Ann Nowé, and Jan Helsen
Wind Energ. Sci., 8, 893–924, https://doi.org/10.5194/wes-8-893-2023, https://doi.org/10.5194/wes-8-893-2023, 2023
Short summary
Short summary
This paper reviews and implements several techniques that can be used for condition monitoring and failure prediction for wind turbines using SCADA data. The focus lies on techniques that respond to requirements of the industry, e.g., robustness, transparency, computational efficiency, and maintainability. The end result of this research is a pipeline that can accurately detect three types of failures, i.e., generator bearing failures, generator fan failures, and generator stator failures.
Paula Helming, Alex Intemann, Klaus-Peter Webersinke, Axel von Freyberg, Michael Sorg, and Andreas Fischer
Wind Energ. Sci., 8, 421–431, https://doi.org/10.5194/wes-8-421-2023, https://doi.org/10.5194/wes-8-421-2023, 2023
Short summary
Short summary
Using renewable energy such as wind energy is vital. To optimize the energy yield from wind turbines, they have increased in size, leading to large blade deformations. This paper measures these deformations for different wind loads and the distance between the blade and the tower from 170 m away from the wind turbine. The paper proves that the blade deformation increases in the wind direction with increasing wind speed, while the distance between the blade and the tower decreases.
Jens Visbech, Tuhfe Göçmen, Charlotte Bay Hasager, Hristo Shkalov, Morten Handberg, and Kristian Pagh Nielsen
Wind Energ. Sci., 8, 173–191, https://doi.org/10.5194/wes-8-173-2023, https://doi.org/10.5194/wes-8-173-2023, 2023
Short summary
Short summary
This paper presents a data-driven framework for modeling erosion damage based on real blade inspections and mesoscale weather data. The outcome of the framework is a machine-learning-based model that can predict and/or forecast leading-edge erosion damage based on weather data and user-specified wind turbine characteristics. The model output fits directly into the damage terminology used by the industry and can therefore support site-specific maintenance planning and scheduling of repairs.
Innes Murdo Black, Moritz Werther Häckell, and Athanasios Kolios
Wind Energ. Sci. Discuss., https://doi.org/10.5194/wes-2022-93, https://doi.org/10.5194/wes-2022-93, 2022
Revised manuscript accepted for WES
Short summary
Short summary
Population based structural health monitoring is a low-cost monitoring campaign. The cost reduction from this type of digital enabled asset management tool is manifested by sharing information, in this case a wind farm foundation, within the population. By sharing the information in the wind farm this reduces the amount of sensors and physical model updating, reducing the cost of the monitoring campaign.
Edward Hart, Elisha de Mello, and Rob Dwyer-Joyce
Wind Energ. Sci., 7, 1533–1550, https://doi.org/10.5194/wes-7-1533-2022, https://doi.org/10.5194/wes-7-1533-2022, 2022
Short summary
Short summary
This paper is the second in a two-part study on lubrication in wind turbine main bearings. Investigations are conducted concerning lubrication in the double-row spherical roller main bearing of a 1.5 MW wind turbine. This includes effects relating to temperature, starvation, grease-thickener interactions and possible non-steady EHL effects. Results predict that the modelled main bearing would be expected to operate under mixed lubrication conditions for a non-negligible proportion of its life.
Rafael Abreu, Daniel Peter, and Christine Thomas
Wind Energ. Sci., 7, 1227–1239, https://doi.org/10.5194/wes-7-1227-2022, https://doi.org/10.5194/wes-7-1227-2022, 2022
Short summary
Short summary
In order to find consensus between wind energy producers and seismologists, we study the possibility of reducing wind turbine noise recorded at seismological stations. We find that drilling half-circular holes in front of the wind turbines helps to reduce the seismic noise. We also study the influence of topographic effects on seismic noise reduction.
Clemens Jonscher, Benedikt Hofmeister, Tanja Grießmann, and Raimund Rolfes
Wind Energ. Sci., 7, 1053–1067, https://doi.org/10.5194/wes-7-1053-2022, https://doi.org/10.5194/wes-7-1053-2022, 2022
Short summary
Short summary
This work presents a method to use low-noise IEPE sensors in the low-frequency range down to 0.05 Hz. In order to achieve phase and amplitude accuracy with this type of sensor in the low-frequency range, a new calibration procedure for this frequency range was developed. The calibration enables the use of the low-noise IEPE sensors for large structures, such as wind turbines. The calibrated sensors can be used for wind turbine monitoring, such as fatigue monitoring.
Cited articles
Abdallah, Imad Chatzi, E.: Probabilistic fault diagnostics using ensemble time-varying decision tree learning, Zenodo, https://doi.org/10.5281/zenodo.3474633, 2019. a, b
Badihi, H., Zhang, Y., Jiang, B., Pillay, P., and Rakheja, S.: A Comprehensive Review on Signal-Based and Model-Based Condition Monitoring of Wind Turbines: Fault Diagnosis and Lifetime Prognosis, Proc. IEEE, 110, 754–806, https://doi.org/10.1109/JPROC.2022.3171691, 2022. a
Barlas, T., Pettas, V., Gertz, D., and Madsen, H. A.: Extreme load alleviation using industrial implementation of active trailing edge flaps in a full design load basis, J. Phys.: Conf. Ser., 753, 17426596, https://doi.org/10.1088/1742-6596/753/4/042001, 2016. a
Bir, G.: Multi-blade coordinate transformation and its application to wind turbine analysis, in: 46th AIAA Aerospace Sciences Meeting and Exhibit, 7 January 2008–10 January 2008 Reno, Nevada, 2008–1300, https://doi.org/10.2514/6.2008-1300, 2008. a
Breiman, L.: Bagging predictors, Mach. Learn., 24, 123–140, 1996. a
Carrasco Kind, M. and Brunner, R.: TPZ: Photometric redshift PDFs and ancillary information by using prediction trees and random forests, Mon. Notices Roy. Astron. Soc., 432, 1483–1501, 2013. a
Cho, S., Gao, Z., and Moan, T.: Model-based fault detection, fault isolation and fault-tolerant control of a blade pitch system in floating wind turbines, Renewa. Energy, 120, 306–321, 2018. a
de Azevedo, H. D. M., Araújo, A. M., and Bouchonneau, N.: A review of wind turbine bearing condition monitoring: State of the art and challenges, Renew. Sustain. Energ. Rev., 56, 368–379, 2016. a
Dempster, A., Schmidt, D. F., and Webb, G. I.: Minirocket: A very fast (almost) deterministic transform for time series classification, in: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, Virtual Event Singapore, August 2021, 248–257, https://doi.org/10.1145/3447548.3467231, 2021. a, b
Dimitrov, N., Natarajan, A., and Kelly, M.: Model of wind shear conditional on turbulence and its impact on wind turbine loads, Wind Energy, 18, 1917–1931, https://doi.org/10.1002/we.1797, 2015. a
Duthé, G., Abdallah, I., Barber, S., and Chatzi, E.: Modeling and Monitoring Erosion of the Leading Edge of Wind Turbine Blades, Energies, 14, https://doi.org/10.3390/en14217262, 2021. a
Fisker Skjoldan, P.: Aeroelastic modal dynamics of wind turbines including anisotropic effects, Risø National Laboratory, ISBN 9788755038486, 2011. a
Gálvez-Carrillo, M. and Kinnaert, M.: Sensor fault detection and isolation in doubly-fed induction generators accounting for parameter variations, Renew. Energy, 36, 1447–1457, https://doi.org/10.1016/j.renene.2010.10.021, 2011. a
Gamberini, A., Gomez Gonzalez, A., and Barlas, T.: Aeroelastic model validation of an Active Trailing Edge Flap System tested on a 4.3 MW wind turbine, J. Phys.: Conf. Ser., 2265, 032014, https://doi.org/10.1088/1742-6596/2265/3/032014, 2022. a, b
Gao, Z. and Liu, X.: An overview on fault diagnosis, prognosis and resilient control for wind turbine systems, Processes, 9, https://doi.org/10.3390/pr9020300, 2021. a
García Márquez, F. P. and Peinado Gonzalo, A.: A Comprehensive Review of Artificial Intelligence and Wind Energy, Arch. Comput. Meth. Eng., 29, 2935–2958, https://doi.org/10.1007/s11831-021-09678-4, 2022. a
Gomez Gonzalez, A., Enevoldsen, P. B., Madsen, H. A., and Barlas, A.: Test of an active flap system on a 4.3 MW wind turbine, TORQUE 2022, https://doi.org/10.1088/1742-6596/2265/3/032016, 2022. a, b
Hossain, M. L., Abu-Siada, A., and Muyeen, S.: Methods for advanced wind turbine condition monitoring and early diagnosis: A literature review, Energies, 11, 1309, https://doi.org/10.3390/en11051309, 2018. a
IEC: Standard IEC 61400-1: 2019, Wind Energy Generation System – Part 1: Design Requirements, 2019. a
Ismail Fawaz, H., Lucas, B., Forestier, G., Pelletier, C., Schmidt, D. F., Weber, J., Webb, G. I., Idoumghar, L., Muller, P.-A., and Petitjean, F.: Inceptiontime: Finding alexnet for time series classification, Data Min. Knowl. Discov., 34, 1936–1962, 2020. a
Lubba, C. H., Sethi, S. S., Knaute, P., Schultz, S. R., Fulcher, B. D., and Jones, N. S.: catch22: CAnonical Time-series CHaracteristics: Selected through highly comparative time-series analysis, Data Min. Knowl. Discov., 33, 1821–1852, https://doi.org/10.1007/s10618-019-00647-x, 2019. a, b
Malekloo, A., Ozer, E., AlHamaydeh, M., and Girolami, M.: Machine learning and structural health monitoring overview with emerging technology and high-dimensional data source highlights, Struct. Health Monit., 21, 1906–1955, 2022. a
Oguiza, I.: tsai – A state-of-the-art deep learning library for time series and sequential data, Github [code], https://github.com/timeseriesAI/tsai (last access: 19 April 2022), 2022. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.: Scikit-learn: Machine Learning in Python, J. Mach. Learn. Res., 12, 2825–2830, 2011. a
Pettas, V., Barlas, T., Gertz, D., and Madsen, H. A.: Power performance optimization and loads alleviation with active flaps using individual flap control, J. Phys.: Conf. Ser., 749,012010, https://doi.org/10.1088/1742-6596/749/1/012010, 2016. a
Stetco, A., Dinmohammadi, F., Zhao, X., Robu, V., Flynn, D., Barnes, M., Keane, J., and Nenadic, G.: Machine learning methods for wind turbine condition monitoring: A review, Renew. Energy, 133, 620–635, https://doi.org/10.1016/j.renene.2018.10.047, 2019. a
Wang, Z., Yan, W., and Oates, T.: Time series classification from scratch with deep neural networks: A strong baseline, in: IEEE 2017 International joint conference on neural networks (IJCNN), 14–19 May 2017, Anchorage, AK, USA, 1578–1585, https://doi.org/10.1109/IJCNN.2017.7966039, 2017. a
Zappalá, D., Tavner, P. J., Crabtree, C. J., and Sheng, S.: Side-band algorithm for automatic wind turbine gearbox fault detection and diagnosis, IET Renew. Power Generat., 8, 380–389, 2014. a
Short summary
Active trailing edge flaps can potentially reduce wind turbine (WT) loads. To monitor their performance, we present two methods based on machine learning that identify flap health states, including degraded performance, in normal power production and idling condition. Both methods rely only on sensors commonly available on WTs. One approach properly detects all the flap states if a fault occurs on only one blade. The other approach can identify two specific flap states in all fault scenarios.
Active trailing edge flaps can potentially reduce wind turbine (WT) loads. To monitor their...
Altmetrics
Final-revised paper
Preprint