Articles | Volume 11, issue 5
https://doi.org/10.5194/wes-11-1791-2026
© Author(s) 2026. 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-11-1791-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
20 May 2026
Research article |
| 20 May 2026
| 20 May 2026
Uniform blade pitch misalignment in wind turbines: a learning-based detection and classification approach
Sabrina Milani
CORRESPONDING AUTHOR
Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB), Politecnico di Milano, Piazza L. Da Vinci 32, Milan, 20133, Italy
Jessica Leoni
CORRESPONDING AUTHOR
Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB), Politecnico di Milano, Piazza L. Da Vinci 32, Milan, 20133, Italy
Stefano Cacciola
CORRESPONDING AUTHOR
Dipartimento di Scienze e Tecnologie Aerospaziali (DAER), Politecnico di Milano, Politecnico di Milano, Via La Masa, 34, Milan 20156, Italy
Alessandro Croce
CORRESPONDING AUTHOR
Dipartimento di Scienze e Tecnologie Aerospaziali (DAER), Politecnico di Milano, Politecnico di Milano, Via La Masa, 34, Milan 20156, Italy
Mara Tanelli
CORRESPONDING AUTHOR
Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB), Politecnico di Milano, Piazza L. Da Vinci 32, Milan, 20133, Italy
Related authors
Sabrina Milani, Jessica Leoni, Stefano Cacciola, Alessandro Croce, and Mara Tanelli
Wind Energ. Sci., 10, 497–510, https://doi.org/10.5194/wes-10-497-2025, https://doi.org/10.5194/wes-10-497-2025, 2025
Short summary
Short summary
In this paper, we propose a novel machine-learning framework for pitch misalignment detection in wind turbines. Using a minimal set of standard sensors, our method detects misalignments as small as 0.1° 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.
Marinos Manolesos, Sandrine Aubrun, Christian Bak, Paolo Bettini, Filippo Campagnolo, Alessandro Croce, Arslan Salim Dar, Michael Hölling, Stefan Ivanell, Konstantinos Kellaris, Miguel Alfonso Mendez, Pierluigi Montinari, Franz Mühle, George Papadakis, Fernardo Porté-Agel, Daniele Ragni, Sebastiano Randino, Andrea Sciacchitano, Lorenzo Schena, Gerard Schepers, Antonio Segalini, Wei Yu, and Carlo Luigi Bottasso
Wind Energ. Sci. Discuss., https://doi.org/10.5194/wes-2026-51, https://doi.org/10.5194/wes-2026-51, 2026
Preprint under review for WES
Short summary
Short summary
As wind turbines grow to unprecedented sizes, ensuring their reliability requires advanced testing. We reviewed the current state of wind tunnel experiments, examining how scaled physical models are evaluated for wind flow, flexibility, and noise. We conclude that while laboratory testing remains essential, it is most powerful when integrated with computer simulations and real-world data. This combined approach will ultimately drive the design of more efficient and sustainable wind energy.
Sabrina Milani, Jessica Leoni, Stefano Cacciola, Alessandro Croce, and Mara Tanelli
Wind Energ. Sci., 10, 497–510, https://doi.org/10.5194/wes-10-497-2025, https://doi.org/10.5194/wes-10-497-2025, 2025
Short summary
Short summary
In this paper, we propose a novel machine-learning framework for pitch misalignment detection in wind turbines. Using a minimal set of standard sensors, our method detects misalignments as small as 0.1° 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.
Alessandro Croce, Stefano Cacciola, and Federico Isella
Wind Energ. Sci., 9, 1211–1227, https://doi.org/10.5194/wes-9-1211-2024, https://doi.org/10.5194/wes-9-1211-2024, 2024
Short summary
Short summary
For a few years now, various techniques have been studied to maximize the energy production of a wind farm, that is, from a system consisting of several wind turbines. These wind farm controller techniques are often analyzed individually and can generate loads higher than the design ones on the individual wind turbine. In this paper we study the simultaneous use of two different techniques with the goal of finding the optimal combination that at the same time preserves the design loads.
Filippo Trevisi, Carlo E. D. Riboldi, and Alessandro Croce
Wind Energ. Sci., 8, 1639–1650, https://doi.org/10.5194/wes-8-1639-2023, https://doi.org/10.5194/wes-8-1639-2023, 2023
Short summary
Short summary
The power equations of crosswind Ground-Gen and Fly-Gen airborne wind energy systems (AWESs) are refined to include the contribution from the aerodynamic wake. A novel power coefficient is defined by normalizing the aerodynamic power with the wind power passing through a disk with a radius equal to the AWES wingspan, allowing us to compare systems with different wingspans. Ground-Gen and Fly-Gen AWESs are compared in terms of their aerodynamic power potential.
Filippo Trevisi, Carlo E. D. Riboldi, and Alessandro Croce
Wind Energ. Sci., 8, 999–1016, https://doi.org/10.5194/wes-8-999-2023, https://doi.org/10.5194/wes-8-999-2023, 2023
Short summary
Short summary
Modeling the aerodynamic wake of airborne wind energy systems (AWESs) is crucial to properly estimating power production and to designing such systems. The velocities induced at the AWES from its own wake are studied with a model for the near wake and one for the far wake, using vortex methods. The model is validated with the lifting-line free-vortex wake method implemented in QBlade.
Roger Bergua, Amy Robertson, Jason Jonkman, Emmanuel Branlard, Alessandro Fontanella, Marco Belloli, Paolo Schito, Alberto Zasso, Giacomo Persico, Andrea Sanvito, Ervin Amet, Cédric Brun, Guillén Campaña-Alonso, Raquel Martín-San-Román, Ruolin Cai, Jifeng Cai, Quan Qian, Wen Maoshi, Alec Beardsell, Georg Pirrung, Néstor Ramos-García, Wei Shi, Jie Fu, Rémi Corniglion, Anaïs Lovera, Josean Galván, Tor Anders Nygaard, Carlos Renan dos Santos, Philippe Gilbert, Pierre-Antoine Joulin, Frédéric Blondel, Eelco Frickel, Peng Chen, Zhiqiang Hu, Ronan Boisard, Kutay Yilmazlar, Alessandro Croce, Violette Harnois, Lijun Zhang, Ye Li, Ander Aristondo, Iñigo Mendikoa Alonso, Simone Mancini, Koen Boorsma, Feike Savenije, David Marten, Rodrigo Soto-Valle, Christian W. Schulz, Stefan Netzband, Alessandro Bianchini, Francesco Papi, Stefano Cioni, Pau Trubat, Daniel Alarcon, Climent Molins, Marion Cormier, Konstantin Brüker, Thorsten Lutz, Qing Xiao, Zhongsheng Deng, Florence Haudin, and Akhilesh Goveas
Wind Energ. Sci., 8, 465–485, https://doi.org/10.5194/wes-8-465-2023, https://doi.org/10.5194/wes-8-465-2023, 2023
Short summary
Short summary
This work examines if the motion experienced by an offshore floating wind turbine can significantly affect the rotor performance. It was observed that the system motion results in variations in the load, but these variations are not critical, and the current simulation tools capture the physics properly. Interestingly, variations in the rotor speed or the blade pitch angle can have a larger impact than the system motion itself.
Koen Boorsma, Gerard Schepers, Helge Aagard Madsen, Georg Pirrung, Niels Sørensen, Galih Bangga, Manfred Imiela, Christian Grinderslev, Alexander Meyer Forsting, Wen Zhong Shen, Alessandro Croce, Stefano Cacciola, Alois Peter Schaffarczyk, Brandon Lobo, Frederic Blondel, Philippe Gilbert, Ronan Boisard, Leo Höning, Luca Greco, Claudio Testa, Emmanuel Branlard, Jason Jonkman, and Ganesh Vijayakumar
Wind Energ. Sci., 8, 211–230, https://doi.org/10.5194/wes-8-211-2023, https://doi.org/10.5194/wes-8-211-2023, 2023
Short summary
Short summary
Within the framework of the fourth phase of the International Energy Agency's (IEA) Wind Task 29, a large comparison exercise between measurements and aeroelastic simulations has been carried out. Results were obtained from more than 19 simulation tools of various fidelity, originating from 12 institutes and compared to state-of-the-art field measurements. The result is a unique insight into the current status and accuracy of rotor aerodynamic modeling.
Filippo Trevisi, Iván Castro-Fernández, Gregorio Pasquinelli, Carlo Emanuele Dionigi Riboldi, and Alessandro Croce
Wind Energ. Sci., 7, 2039–2058, https://doi.org/10.5194/wes-7-2039-2022, https://doi.org/10.5194/wes-7-2039-2022, 2022
Short summary
Short summary
The optimal control problem for the flight trajectories of Fly-Gen AWESs is expressed with a novel methodology in the frequency domain through a harmonic balance formulation. The solution gives the optimal trajectory and the optimal control inputs. Optimal trajectories have a circular shape squashed along the vertical direction, and the optimal control inputs can be modeled with only one or two harmonics. Analytical approximations for optimal trajectory characteristics are also given.
Alessandro Bianchini, Galih Bangga, Ian Baring-Gould, Alessandro Croce, José Ignacio Cruz, Rick Damiani, Gareth Erfort, Carlos Simao Ferreira, David Infield, Christian Navid Nayeri, George Pechlivanoglou, Mark Runacres, Gerard Schepers, Brent Summerville, David Wood, and Alice Orrell
Wind Energ. Sci., 7, 2003–2037, https://doi.org/10.5194/wes-7-2003-2022, https://doi.org/10.5194/wes-7-2003-2022, 2022
Short summary
Short summary
The paper is part of the Grand Challenges Papers for Wind Energy. It provides a status of small wind turbine technology in terms of technical maturity, diffusion, and cost. Then, five grand challenges that are thought to be key to fostering the development of the technology are proposed. To tackle these challenges, a series of unknowns and gaps are first identified and discussed. Improvement areas are highlighted, within which 10 key enabling actions are finally proposed to the wind community.
Alessandro Croce, Stefano Cacciola, and Luca Sartori
Wind Energ. Sci., 7, 1–17, https://doi.org/10.5194/wes-7-1-2022, https://doi.org/10.5194/wes-7-1-2022, 2022
Short summary
Short summary
In recent years, research has focused on the development of wind farm controllers with the aim of minimizing interactions between machines and thus improving the production of the wind farm.
In this work we have analyzed the effects of these recent technologies on a single wind turbine, with the aim of understanding the impact of these controllers on the design of the machine itself.
The analyses have shown there are non-negligible effects on some components of the wind turbine.
Cited articles
Al Lahham, E., Kanaan, L., Murad, Z., Khalid, H. M., Hussain, G. A., and Muyeen, S. M.: Online condition monitoring and fault diagnosis in wind turbines: A comprehensive review on structure, failures, health monitoring techniques, and signal processing methods, Energy Rep., 10, 1–23, https://doi.org/10.1016/j.egyr.2024.01.080, 2024. a
Bertelè, M., Bottasso, C. L., and Cacciola, S.: Automatic detection and correction of pitch misalignment in wind turbine rotors, Wind Energ. Sci., 3, 791–803, https://doi.org/10.5194/wes-3-791-2018, 2018. a
Bodson, M., Chiasson, J., and Novotnak, R. T.: Maximizing wind turbine energy capture using multivariable extremum seeking control, Wind Energy, 17, 589–600, https://doi.org/10.1002/we.1585, 2014. a
Bottasso, C. L. and Croce, A.: Cp-Lambda user manual, Tech. Rep., Dipartimento di Scienze e Tecnologie Aerospaziali, Politecnico di Milano, Milano, Italy, 2009–2018. a
Breiman, L.: Random forests, Mach. Learn., 45, 5–32, https://doi.org/10.1023/A:1010933404324, 2001. a
Cacciola, S. and Riboldi, C. E. D.: Equalizing Aerodynamic Blade Loads Through Individual Pitch Control Via Multiblade Multilag Transformation, J. Sol. Energy Eng., 139, 061008, https://doi.org/10.1115/1.4037744, 2017. a
Cacciola, S., Munduate Agud, I., and Bottasso, C. L.: Detection of rotor imbalance, including root cause, severity and location, J. Phys. Conf. Ser., 753, 072003, https://doi.org/10.1088/1742-6596/753/7/072003, 2016. a, b
Cacciola, S., Riboldi, C. E. D., and Croce, A.: Monitoring rotor aerodynamic and mass imbalances through a self-balancing control, J. Phys. Conf. Ser., 1037, 032041, https://doi.org/10.1088/1742-6596/1037/3/032041, 2018. a
Ceriani, L. and Verme, P.: The origins of the Gini index: extracts from Variabilità e Mutabilità (1912) by Corrado Gini, J. Econ. Inequal., 10, 421–443, https://doi.org/10.1007/s10888-011-9188-x, 2012. a
Dalsgaard, S., Blanke, M., and Brath, P.: Diagnosis of pitch and load effects, International Patent Number WO 2009/059606 A3, US7954372B2, Diagnosis of pitch and load defects, Google Patents, 2009. a
Desheng, S., Qian, L., Xin, Z., and Minrong, Z.: Wind turbine variable pitch system fault monitoring method and system, Patent Number CN113586366B, 2021. a
IEC requirements book Wind Turbines: Part 1: Design Requirements – DocsLib, Garrad Hassan and Partners Ltd, 3rd Edn., 2004. a
Johnson, K. E., Pao, L. Y., Balas, M. J., and Fingersh, L. J.: Experimental evaluation of extremum seeking based region-2 control for a wind turbine, IEEE Trans. Control Syst. Technol., 24, 1084–1091, https://doi.org/10.2514/6.2016-1737, 2016. a
Jonkman, J., Butterfield, S., Musial, W., and Scott, G.: Definition of a 5-MW Reference Wind Turbine for Offshore System Development, Tech. Rep. nREL/TP-500-38060, National Renewable Energy Laboratory (NREL), United States, https://www.osti.gov/biblio/947422 (last access: February 2009), 2009. a
Krstić, M. and Wang, H.-H.: Stability of extremum seeking feedback for general nonlinear dynamic systems, Automatica, 36, 595–601, https://doi.org/10.1016/S0005-1098(99)00183-1, 2000. a
Kusiak, A. and Verma, A.: A Data-Driven Approach for Monitoring Blade Pitch Faults in Wind Turbines, IEEE Trans. Sustain. Energy, 2, 87–96, https://doi.org/10.1109/TSTE.2010.2066585, 2011. a, b
Li, Y. and Wang, J.: Dynamic fault monitoring of wind turbine pitch system using SCADA data, Energies, 12, 3256, https://doi.org/10.3390/en12173256, 2019. a
Milani, S., Leoni, J., Cacciola, S., Croce, A., and Tanelli, M.: Automatic Detection and Intensity Classification of Pitch Misalignment of Wind Turbine Blades: a Learning-based Approach, J. Phys. Conf. Ser., 2767, 032010, https://doi.org/10.1088/1742-6596/2767/3/032010, 2024. a, b
Milani, S., Leoni, J., Cacciola, S., Croce, A., and Tanelli, M.: A machine-learning-based approach for active monitoring of blade pitch misalignment in wind turbines, Wind Energ. Sci., 10, 497–510, https://doi.org/10.5194/wes-10-497-2025, 2025. a, b
Qin, S., Tao, J., and Zhao, Z.: Deep learning-based fault diagnosis of wind turbine pitch systems using SCADA data, Energy AI, 14, 100259, https://doi.org/10.1016/j.egyai.2023.100259, 2023. a
Rotea, M. A.: Analysis of multivariable extremum seeking algorithms, Automatica, 36, 687–698, https://doi.org/10.1016/S0005-1098(99)00177-6, 2000. a
Stetco, A., Dinmohammadi, F., Zhao, X., Robu, V., Flynn, D., Barnes, M., and Keane, J.: 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
Tang, B., Song, T., and Deng, L.: Fault detection of wind turbine pitch system based on data-driven methods, Front. Energy Res., 9, 750983, https://doi.org/10.3389/fenrg.2021.750983, 2021. a
Tautz-Weinert, J. and Watson, S. J.: Using SCADA data for wind turbine condition monitoring – A review, IET Renew. Power Gener., 11, 382–394, https://doi.org/10.1049/iet-rpg.2016.0248, 2017. a
Yang, W., Court, R., and Jiang, J.: Wind turbine condition monitoring by the approach of SCADA data analysis, Renew. Energy, 126, 1–15, https://doi.org/10.1016/j.renene.2018.03.013, 2018. a
Zaher, A., McArthur, S. D. J., Infield, D. G., and Patel, Y.: Online wind turbine fault detection through automated SCADA data analysis, Wind Energy, 12, 574–593, https://doi.org/10.1002/we.319, 2009. a
Short summary
This work introduces a novel method to detect and quantify uniform pitch misalignment in wind turbines. This fault, where all blades are equally misaligned, is hard to detect because it causes no immediate imbalance but reduces efficiency over time. By combining physics-based features with machine learning techniques, our approach reliably identifies and quantifies this fault under various wind regime conditions, improving turbine maintenance and energy production.
This work introduces a novel method to detect and quantify uniform pitch misalignment in wind...
Altmetrics
Final-revised paper
Preprint