Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Aerodynamics and rotor design' taking place on Wednesday, 12 March 2014 at 09:00-10:30. The meet-the-authors will take place in the poster area.

Sebastian Flock RWTH Aachen, Germany
Co-authors:
Sebastian Flock (1) F P Ralf Schelenz (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

Sebastian Flock studied mechanical engineering at the RWTH Aachen specializing on aeronautical engineering. In 2009 he joined the Institute for Machine Elements and Machine Design as Research Engineer in the Drive Train Analysis - Group. He is currently working on Multi Body Simulations of Trailing Edge Flaps at the rotor of a wind turbine. The aim is to quantify the load reduction potential for the drive train. Mr. Flock is responsible for the Project “Aeroelastic Analysis of Large Wind Turbines”, which is funded by the German Federal Ministry of Environment.

Abstract

Active control of trailing-edge-flaps at the rotor of a large wind turbine

Introduction

Aerodynamic forces on large rotors impose dynamic loads on the drive train which may lead to damage or system failure. Therefore actively controlled devices are object to research, especially to what extent they can reduce dynamic loads and thus increase the availability of the turbine, for example [1-3].
The presentation will give an overview on the work within the project “Aeroelastic Analysis of Large Wind Turbines”, which is being financed by the German Federal Ministry of Environment (BMU). The focus is on integrating an individual trailing edge flap controller into a wind turbine model to quantify load reduction.


Approach

At the Institute for Machine Elements (IME) as well as the Chair for Wind Power Drives (CWD) at RWTH Aachen the generic offshore wind turbine “IME 6.0” has been developed in the last three years based on publicly available data as well as data provided by supplier companies [4-9], (see fig(1)). The rotor is a modification of the 5MW Reference Turbine Rotor definition given in [10].



The load simulations are based on a modified version of the AeroDyn – Code, which is made publicly available by the National Renewable Energy Laboratory (NREL) [11]. Changes to the code were made in adding the additional degrees of freedom for the flap angle.

With the help of the CFD-solver FLOWER [12], which was kindly made available by the German Centre for Aviation and Space Travel (DLR) and the Multi Block Grid Deformation-Tool MUGRIDO [13] provided by the Chair for Computational Analysis of Technical Systems (CATS), the 2D-aerodynamic pressure distribution of all possible combinations of angle of attack and flap deflection angle are simulated for the airfoils equipped with flaps.

The geometric dimensions and positions of the flaps are following the suggestions from [14] and [15]. In the model each blade is equipped with two flaps, which are controlled individually. The individual flap control was embedded into the turbine controller, which is a state of the art collective pitch control system [8].

Two different load cases are subject to simulation, for the proof of concept:
1.) A production load case at approximately 14 m/s wind with high turbulence (16%).
2.) An extreme coherent gust (Mexican Hat) without turbulence, following the load case definition given in [17].




Main body of abstract

Controller Concept

The focus of the presentation will be on the design of the turbine controller - especially the implementation of individual flap control. Starting from the turbine model it is explained, how the control task is handled in the simulation-chain. Figure 2 shows the interactions between the simulation model in SIMPACK, the controller in Simulink and the aerodynamic load application derived from AeroDyn in combination with the CFD-solver Flower.



This is followed by an explanation of the flap actuation in the simulation model. All effects, which are taken into account by the flap actuator, are shown in Figure 3. First the aerodynamic pressure distribution over the flap is numerically integrated to obtain the aerodynamic loads acting on the hinge. Furthermore the mass and the mass moment of inertia are considered. The flaps are actuated by two linear electric motors connected to a lever-gear. The inverse kinematic of this lever-gear has been modelled to take into account the current gear ratio, which is dependent on the operation point. The behaviour of the motors is represented by characteristic-curves, from data, which are provided by the manufacturer.



The flap controller shall only come into effect, as soon as two conditions are fulfilled. First, the generator must operate near its nominal operation point and second, the flaps are only to be activated for strong wind conditions to produce an optimal ratio of activity to load reduction over the lifetime of the turbine. A comparison is made between a mean wind speed at the Anemometer recorded and averaged by the main controller and the actual inflow wind speed at the flap position. Flaps are being fully deployed for a discrepancy of more than 20% [7]. The control of the requested flap angle is shown in figure 4.



Because all flaps have to be actuated individually, the absolute value of the inflow velocity at each flap is needed. Therefore this output value has been added to the AeroDyn-Code and is passed through the simulation model to the controller. This approach is equivalent to placing pitot-tubes at the leading edge of the blade, which has been suggested for example in [12] and was part of experiments in [13].
Figure 5 shows the resulting absolute velocities at all six flaps installed on the three bladed rotor for the load case “Extreme Gust”. Each three velocities for the outer flaps oscillate around a common mean value and are shifted in phase because of wind shear and tilting of the rotor. The common mean value for the inner flaps is lower because they are subject to a lower circumferential speed.



Results

To prove the flexibility of the controller design, two very different load cases are simulated. First a typical production load case defined by a strong, very turbulent wind at 14m/s and 16% turbulence, which has been pre-processed by TurbSim [18]. The second load case is an extreme gust following the definition given in [17]. Both are subject to a wind shear rate of 0.2. The nominal wind speeds at hub height is shown in figure 6.



Figures 7 and 8 show the load reduction for the root bending moment of one blade in frequency domain. Both figures show, that the controller produces significant reductions in the load, reducing the main excitation in the rotation-frequency of the rotor by 35% for the turbulent wind scenario and 24% for the extreme gust, respectively.








Conclusion

By the results, which are presented, it is shown, that active trailing edge flaps yield an enormous potential for load reduction on wind turbines. This corresponds very well with recent simulation results publicated for example in [19]. The simulations on the turbine model are the first ones known to the authors, to directly simulate all major effects of the flap actuation namely: The mass of the flap, inertia loads on the hinge, aerodynamic loads and the inverse kinematics of the gear to interconnect the flap with a linear electric motor and the characteristic curves of the motor. Everything is combined in a multibody simulation model including a detailed representation of the drivetrain. This way it is possible to seize the effects of load reduction on the turbine.
The time series and frequency analysis show a reduction in the main excitation frequency in root-bending moment of approximately 35%, which could be achieved for a production operation point with turbulent wind. The scenario of an extreme coherent gust in accordance to the IEC 61400 definition yielded an approximate reduction of 24% in the same respect.
This paper/presentation is a contribution to the control of wind turbines equipped with trailing edge flaps. The presented tool chain (SIMPACK, Simulink, FLOWER/AeroDyn) gives the opportunity to quantify the load reduction on the basis of a detailed multi body simulation model of a turbine including drive train dynamics and turbine controller.
The work shall assist turbine manufacturers in future design phases, to quantitatively assess the benefits of integrating trailing edge flaps to the rotor.



Learning objectives
Multi Body Simulation Model of the IME-6.0 Offshore Turbine, including detailed drivetrain representation. Controller design for trailing edge flaps on wind turbines, taking into account the actuation with linear electric motors, aerodynamics considering the local inflow velocities at the flaps as well as masses and inertia loads on the hinge moment. Load reductions realized for the IEC design load cases “extreme coherent gust” and strong turbulent wind.


References
[1] Yen Nakafuji, D., van Dam, C., Smith, R., and Collins, S., “Active Load Control for Airfoils using Microtabs,” J. Sol. Energy Eng., 123, pp. 282–289., 2001

[2] vanWingerden J-W, Hulskamp AW, Barlas T, Marrant B, van Kuik GAM, Molenaar D-P, Verhaegen M. On the proof of concept of a “smart” wind turbine rotor blade for load alleviation. Wind Energy 2008; 11: 265–280.

[3] Basualdo, S.: Load alleviation on wind turbine blades using variable airfoil geometry, WIND ENGINEERING VOLUME 29, NO. 2, 2005 PP 169–182, 2005

[4] Schelenz, R., Flock, S., Möller, D.: „Aeroelastische Rotoren in der Mehrkörpersimulation“ VDI-Tagung Schwingungen in Windenergieanlagen, Hannover, 2010

[5] Berroth, J., “ Design and Modelling of the structural components of an offshore
wind energy plant for multibody simulation,” Diploma Thesis, RWTH Aachen, 2011

[6] Vetten, D.: Gearbox Design for an Offshore Windturbine, B.-Sc. FH-Aachen, 2012

[7] Frenzel, F.: Active load alleviation by considering the aerodynamic flaps and actuators in the control of a wind turbine, B.-Sc. Thesis, RWTH Aachen, 2012

[8] Back, B.: Implementation of A Wind Turbine Controller And Conceptional Design Of Advanced Control Strategies, B.-Sc. Thesis, RWTH Aachen, 2012

[9] Kämper, T.: Modeling of a rotor blade and concept development for an actuator system of a trailing edge flap for active load alleviation on wind turbines, B.-Sc. Thesis, RWTH Aachen, 2012

[10] Jonkman, J., Butterfield, S., Musial, W., Scott, G.: Definition of a 5-MW Reference Wind Turbine for Offshore System Development. Technical Report NREL/TP-500-38060, National Renewable Energy Laboratory, 2009

[11] Moriarty, P.J.; Hansen, A.C.: AeroDyn Theory Manual, National Renewable Energy Laboratory, 2005 (NREL/TP-500-36881)

[12] Aumann, P., Bartelheimer, W., Bleecke, H., Eisfeld, J., Lieser, J., Heinrich, R., Kroll, N., Kuntz, M., Monsen, E., Raddatz, J., Reisch, U., Roll, B.: FLOWer Installation and USER Handbook Release 116. Tech. Rep. MEGAFLOW-1001, DLR (2000)

[13] Boucke, M.: Kopplungswerkzeuge für aeroelastische Simulationen, PhD, RWTH Aachen, 2003

[14] Troldborg, N., “Computational Study of the Risø-B1-18 Airfoil Equipped with Actively Controlled Trailing Edge Flaps”, Master Thesis, Technical University of Denmark ,Department of Mechanical Engineering, Fluid Mechanics, 2004

[15] Andersen PB: Advanced Load Alleviation for Wind Turbines using Adaptive Trailing Edge Flaps: Sensoring and Control, PhD, DTU, 2010

[16] Castaignet, D.: Model predictive control of trailing edge flaps on a wind turbine blade, PhD thesis, Roskilde TU Denmark, 2011

[17] International Electrotechnical Comission TC 88-MT1: IEC 61400-1 Ed.3: Wind Turbines – Part 1: Design Requirements”, Geneva 2005

[18] Kelley N.D., Jonkman J.B.: Overview of the TurbSim Stochastic Inflow Turbulence Simulator, NREL - Technical Report, NREL/TP-500-41137, 2007

[19] Barlas, T.K.;van der Veen, G.J.;van Kuik, G. A. M.: Model predictive control for wind turbines with distributed active flaps: incorporating inflow signals and actuator constraints, Wind Energ. 2012; 15:757–771, DOI: 10.1002/we.503