Articles | Volume 5, issue 3
Wind Energ. Sci., 5, 885–896, 2020

Special issue: Wind Energy Science Conference 2019

Wind Energ. Sci., 5, 885–896, 2020

Research article 13 Jul 2020

Research article | 13 Jul 2020

Real-time optimization of wind farms using modifier adaptation and machine learning

Leif Erik Andersson and Lars Imsland

Related subject area

Control and system identification
Model-based design of a wave-feedforward control strategy in floating wind turbines
Alessandro Fontanella, Mees Al, Jan-Willem van Wingerden, and Marco Belloli
Wind Energ. Sci., 6, 885–901,,, 2021
Short summary
Active flap control with the trailing edge flap hinge moment as a sensor: using it to estimate local blade inflow conditions and to reduce extreme blade loads and deflections
Sebastian Perez-Becker, David Marten, and Christian Oliver Paschereit
Wind Energ. Sci., 6, 791–814,,, 2021
Short summary
Wind inflow observation from load harmonics: initial steps towards a field validation
Marta Bertelè, Carlo L. Bottasso, and Johannes Schreiber
Wind Energ. Sci., 6, 759–775,,, 2021
Short summary
Control-oriented model for secondary effects of wake steering
Jennifer King, Paul Fleming, Ryan King, Luis A. Martínez-Tossas, Christopher J. Bay, Rafael Mudafort, and Eric Simley
Wind Energ. Sci., 6, 701–714,,, 2021
Short summary
Condition monitoring of roller bearings using acoustic emission
Daniel Cornel, Francisco Gutiérrez Guzmán, Georg Jacobs, and Stephan Neumann
Wind Energ. Sci., 6, 367–376,,, 2021
Short summary

Cited articles

Andersson, J. A. E., Gillis, J., Horn, G., Rawlings, J. B., and Diehl, M.: CasADi – A software framework for nonlinear optimization and optimal control, Math. Program. Comput., 11, 1–36,, 2019. a
Andersson, L. E., Bradford, E. C., and Imsland, L.: Distributed learning and wind farm optimization with Gaussian processes, in: American Control Conference (ACC), online conference, accepted, 2020a. a, b
Andersson, L. E., Bradford, E. C., and Imsland, L.: Gaussian processes modifier adaptation with uncertain inputs using distributed learning and optimization on a wind farm, in: IFAC World congress 2020, 11–17 July 2020, Berlin, Germany, accepted, 2020b. a
Andersson, L. E., Doekemeijer, B., van der Hoek, D., van Wingerden, J.-W., and Imsland, L.: Adaptation of Engineering Wake Models using Gaussian Process Regression and High-Fidelity Simulation Data, arXiv preprint: arXiv:2003.13323, 2020c. a, b
Annoni, J., Seiler, P., Johnson, K., Fleming, P., and Gebraad, P.: Evaluating wake models for wind farm control, in: 2014 American Control Conference, 4–6 June 2014, Portland, Oregon, USA, 2517–2523,, 2014. a
Short summary
The article describes a hybrid modeling approach to optimize the energy capture of wind farms. Hybrid modeling combines mechanistic and data-driven models. The data-driven part is used to correct inaccuracies of the mechanistic model. The hybrid approach allows for adjustment of the mechanistic model beyond simple parameter estimation. It is, therefore, an attractive approach in wind farm control. The approach is illustrated in several numerical case studies.