Back to the programme printer.gif Print

Wednesday, 12 March 2014
11:15 - 12:45 Advanced control concepts
Science & Research  

Room: Llevant
Session description

The application of advanced control can be utilised to improve turbine performance. The topics addressed in this session include wind turbine/farm control to provide frequency support including droop control, the collective control of a number of wind turbines through the use of a common bus bar and converter, and the stabilisation of floating wind turbines. The control design techniques used include model-predictive, nonlinear and robust control design.

Lead Session Chair:
William Leithead, University of Strathclyde, United Kingdom

Marta Barreras, Gamesa, Spain
Markus Marnett RWTH Aachen, Germany
Markus Marnett (1) F P Sören Wellenberg (1) Wolfgang Schröder (1)
(1) RWTH Aachen, Aachen, Germany

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Dr. Marnett has been working in the field of wind energy research for more than eight years. He studied aerospace engineering and business administration at RWTH Aachen University. Subsequently, he joined the Institute of Aerodynamics at RWTH Aachen University as a research scientist and has been involved in several numerical and experimental projects to investigate the fluid mechanics of vertical axis wind turbines. Moreover, he is a member of the Center for Wind Turbine Drives at RWTH Aachen University and responsible for the development of a real-time simulation environment describing the freestream wind field and resulting aeroelastic rotor behavior.


Systematic numerical design of optimal blade pitch control for vertical axis wind turbines


Especially the future potential of vertical axis wind turbines for the deep-water offshore installation has encouraged research and development efforts on that concept. To increase the turbine net capacity to ten megawatt and higher, the optimization of loads and efficiency using blade pitch control becomes essential. Therefore, a systematic and multi-objective numerical optimization procedure for blade pitch control of straight bladed vertical axis wind turbines was developed and applied to a test case. The resulting algorithm allows the detailed design of optimal blade pitch control for all relevant operating conditions satisfying various objectives.


The efficiency of vertical axis wind turbines can be improved using individual blade pitch control. Simultaneously, blade pitch control facilitates the reduction and harmonization of turbine loads. This is an exceptionally important topic for vertical axis wind turbines since they are typically exposed to intermittent loads. To achieve the desired turbine behavior, complex turbine pitch control depending on the operating conditions is necessary. Hence, a fast converging systematic numerical design tool to determine optimal blade pitch control was developed. It meets the requirements for multiple and competing objectives and allows handling multiple constraints at the same time.
This innovative design tool combines a multi-objective evolutionary algorithm, a performance prediction model, and a suitable parameterization of the blade pitch curve along the azimuthal angle of the turbine. The performance prediction model is based on a 2D free-vortex model describing the production, convection, and interaction of vortex systems. A vortex panel model combined with a boundary layer formulation applicable for attached flows represents the rotor blades. This defines a constraint to prevent local angles of attack from exceeding critical values within the optimization procedure. Moreover, it is necessary to limit the reduced frequencies occurring at the blades to a critical value to ensure the accuracy of the boundary layer model. Even if just a single-objective optimization criterion like the energy efficiency is selected, a multi-objective task arises from the constraints of the performance prediction model. For this reason, an optimization algorithm based on generalized differential evolution is selected and modified to tailor the algorithm more specifically to the optimization problem.
A simple convincing test case targeting a constant torque coefficient is presented in the framework of this paper and the performance of the design procedure is discussed.

Main body of abstract

Originally, blade pitch control was taken into consideration since vertical axis wind turbines at fixed pitch have been regarded inappropriate for stand-alone applications due to their lack of torque during start-up and at low tip speed ratios (see Kirke (1998)). Subsequently, a couple of research activities demonstrated the ability to improve the system efficiency applying blade pitch control (Cooper & Kenedy (2004), Cooper et al. (2006), Hwang et al. (2006), Pawsey (2002), Staelens et al. (2003) and Vandenberghe & Dick (1987)). A first numerical optimization tool was developed by Paraschivoiu et al. (2009) combining a double multiple streamtube model with a single-objective genetic algorithm. Their study focused on aerodynamic efficiency and could increase annual energy production compared to a fixed pitch design by almost 30%.
Discussing the future potential of vertical axis wind turbines for deep water offshore installations (see Paquette & Barone (2012)), additional objectives for blade pitch control become relevant. These are constant loading of both, tower and blades, a constant rotor torque, and the conflicting objective of an optimal reduced amount of required auxiliary energy. Hence, a systematic multi-objective design tool is important.
A multi-objective optimization using generalized differential evolution and an unsteady Reynolds-averaged Navier-Stokes solver as the performance prediction model satisfying the constraints of constant power and minimized dynamic loading of the turbine, was presented first in Marnett (2012). Unfortunately, this approach requires a high computational effort, which is inconvenient for industrial use. The design tool discussed in this study is clearly more efficient such that it meets the temporal engineering needs.
One major component is the aerodynamic performance prediction model. The framework to simulate the flow field of a vertical axis wind turbine is based on a 2D free-vortex model for incompressible irrotational flow. The blades are represented by a coupled viscous-inviscid interaction between a panel code and the integral boundary layer equations (see Riziotis & Voutsinas (2007) and Zanon et al. (2012)). The final flow field is computed as the sum of the undisturbed flow field and the induced velocities from the shed vortex filaments. However, to receive accurate results for the flow field around the blades, the local angle of attack and the occurring reduced frequencies have to be restricted from exceeding critical values in line with the optimization procedure. The applied performance prediction model is a perfect trade-off between physical accuracy and computational effort.
Another major component is the multi-objective optimization procedure based on the differential evolution method first introduced by Storn & Price (1995) as a heuristic approach for nonlinear and non-differentiable continuous space functions. Lampinen (2001a, b, 2002) first proposed an exceptionally simple and yet effective method for adapting differential evolution for the particular demands of multi-objective optimization. His formulation was used and tailored to ensure diversity preservation among the individuals of an optimization generation with the ideas of Deb et al. (2002). The resulting optimization algorithm ensures a very efficient and robust procedure for complex and noisy target functions. Moreover, it enables the procedure to meet the physical requirements of the applied performance prediction model.
Results of the combined design tool are discussed for a test case defined by the constraint of constant rotor torque using blade pitch control. The numerical results show a good match with the objective. Furthermore, the necessary computational effort is comparatively low. Note that the target value for rotor torque is chosen on a rather low level to simplify this test case against the background of the methodic requirements.
The paper discusses a systematic multi-objective numerical design tool for blade pitch control of vertical axis wind turbines, which facilitates optimal blade pitch control for all relevant kinds of operating conditions implicating various objectives. It is of special interest since the new performance prediction model is rather accurate in terms of an application within an optimization procedure and unlike existing design tools it allows for multi-objective aerodynamic performance optimization.
The applied procedures and a validation of the design tool will be discussed in detail in the final paper. Moreover, an in-depth analysis of the results, their impact for future work, and the necessity for code expansions will be given.


Blade pitch control is inevitable for the improvement of the aerodynamic efficiency and the reduction of dynamic loads for large vertical axis wind turbines. Due to the strongly coupled turbine flow field, the future design process is rather complex and has to meet a variety of constraints. Since the optimal pitch curve varies with the operating conditions of the turbine, the design has to be repeated for numerous design points. Hence, a systematic and accurate numerical design tool for blade pitch control of vertical axis wind turbines is needed.
Available tools are not able to meet the requirements of multi-objective design. The work conducted at the Chair of Aerodynamics at RWTH Aachen University resulted in a unique and innovative multi-objective design tool facilitating optimal blade pitch control for all relevant kinds of operating conditions. Its major innovation is the combination of different approaches and its application to blade pitch control. This multi-objective optimization procedure is derived from a generalized differential evolution and a performance prediction model based on a 2D free-vortex model with a coupled viscous-inviscid representation of the turbine blades. Furthermore, the constraints to ensure an accurate representation of the performance prediction model were added to the optimization process.
The test case containing the constraint of constant rotor torque shows convincing results and a comparatively low computational effort. Future work will focus on additional test cases, windfield modelling, further validation as well as speed-up, and necessary code expansions for detached flow conditions, dynamic stall effects, aeroelasticity, and a 3D flow field.

Learning objectives
A design procedure for optimal blade pitch control has been developed and preliminary tested. The required computational effort is comparatively low. Consequently, the proposed design procedure facilitates future application for the overall design process of vertical axis wind turbines featuring blade pitch control.

Cooper, P. & Kennedy, O. 2004 Development and analysis of a novel vertical axis wind turbine. Tech. Rep. School of Mechanical, Materials and Mechatronic Engineering, University of Wollongong, Australia.

Cooper, P., Kennedy, O. & Whitten, G. 2006 Aerodynamics of a novel active blade pitch vertical axis wind turbine. Tech. Rep. School of Mechanical, Materials and Mechatronic Engineering, University of Wollongong, Australia.

Deb, K., Pratap, A., Agarwal, S. & Meyarivan, T. 2002 A fast and elitist multiobjective genetic algorithm: NSGA-II. In IEEE Transactions on evolutionary computation, Vol. 6, No. 2, pp. 182-197, April 2002.

Drees, H. M. 1978 The cycloturbine and its potential for broad application. In 2nd Int. Symposium on Wind Energy Systems, pp. 81-88. October 3rd-6th.

Grylls, W., Dale, B. & Sarre, P. E. 1978 A theoretical and experimental investigation into the variable pitch vertical axis wind turbine. In 2nd Int. Symposium on Wind Energy Systems, pp. E9-101-E9-118. October 3rd-6th.

Hwang, I. S., Min, S. Y., Jeong, I. O., Lee, Y. H. & Kim, S. J. 2006 Efficiency improvement of a new vertical axis wind turbine by individual active control of blade motion. In Proc. SPIE 617311, pp. 617311-1-617311-8.

Kirke, B. K. 1998 Evaluation of self-starting vertical axis wind turbines for stand-alone applications. PhD thesis, School of Engineering, Griffith University.

Lampinen, J. 2001a DE’s selection rule for multiobjective optimization. Tech. Rep.. Lappeenranta University of Technology, Department of Information Technology, Laboratory of Information Processing.

Lampinen, J. 2001b Solving problems subject to multiple nonlinear constraints by the differential evolution. Tech. Rep.. Lappeenranta University of Technology, Department of Information Technology, Laboratory of Information Processing.

Lampinen, J. 2002 A constraint handling approach for the differential evolution algorithm. In Proceedings of the 2002 Congress on Evolutionary Computation (CEC 2002), pp. 1468–1473.

Marnett, M. 2012 Multiobjective Numerical Design of Vertical Axis Wind Turbine Components. PhD thesis, Chair of Fluid Mechanics and Institute of Aerodynamics, RWTH Aachen University.

Paquette, J. & Barone, M. 2012 Innovative Offshore Vertical-Axis Wind Turbine Rotor Project. Presented at the European Wind Energy Association Annual Event 2012, Copenhagen, Denmark.

Paraschivoiu, I., Trifu, O. & Saeed, F. 2009 H-darrieus wind turbine with blade pitch control. International Journal of Rotating Machinery 2009, 1-7.

Pawsey, N. C. K. 2002 Development and evaluation of passive variable-pitch vertical axis wind turbines. PhD thesis, University of New South Wales.

Riziotis, V. A. & Voutsinas, S. G. Dynamic stall modelling on airfoils based on strong viscous-inviscid interaction coupling. In International Journal for Numerical Methods in Fluids, 2008, Vol. 56, pp. 185-208.

Staelens, Y., Saeed, F. & Paraschivoiu, I. 2003 A straight-bladed variable-pitch VAWT concept for improved power generation. In ASME 2003 Wind Energy Symposium, pp. 146-154.

Storn, R. & Price, K. 1995 Differential evolution - a simple and efficient adaptive scheme for global optimization over continuous spaces. Tech. Rep.. International Computer Science Institute, Berkeley.

Vandenberghe, D. & Dick, E. 1986 A theoretical and experimental investigation into the straight bladed vertical axis wind turbine with second order harmonic pitch control. Wind Engineering 10 (3), 122-138.

Vandenberghe, D. & Dick, E. 1987 Optimum pitch control for vertical axis wind turbines. Wind Engineering 11 (5), 237-247.

Zanon, A., Giannattasio, P. & Ferreira, C. J. S. 2012 A vortex panel model for the simulation of the wake flow past a vertical axis wind turbine in dynamic stall. Wind Energy, 2012.