Jason Laks

Control of wind turbines: Past, present, and future

We review the objectives and techniques used in the control of horizontal axis wind turbines at the individual turbine level, where controls are applied to the turbine blade pitch and generator. The turbine system is modeled as a flexible structure operating in the presence of turbulent wind disturbances. Some overview of the various stages of turbine operation and control strategies used to maximize energy capture in below rated wind speeds is given, but emphasis is on control to alleviate loads when the turbine is operating at maximum power. After reviewing basic turbine control objectives, we provide an overview of the common basic linear control approaches and then describe more advanced control architectures and why they may provide significant advantages.

A tutorial of wind turbine control for supporting grid frequency through active power control

As wind energy becomes a larger portion of the worlds energy portfolio and wind turbines become larger and more expensive, wind turbine control systems play an ever more prominent role in the design and deployment of wind turbines. The goals of traditional wind turbine control systems are maximizing energy production while protecting the wind turbine components. As more wind generation is installed there is an increasing interest in wind turbines actively controlling their power output in order to meet power setpoints and to participate in frequency regulation for the utility grid. This capability will be beneficial for grid operators, as it seems possible that wind turbines can be more effective at providing some of these services than traditional power plants. Furthermore, establishing an ancillary market for such regulation can be beneficial for wind plant owner/operators and manufacturers that provide such services. In this tutorial paper we provide an overview of basic wind turbine control systems and highlight recent industry trends and research in wind turbine control systems for grid integration and frequency stability.

Model Predictive Control Using Preview Measurements From LIDAR y

Light detection and ranging (LIDAR) systems are able to measure conditions at a distance in front of wind turbines and are therefore suited to providing preview information of wind disturbances before they impact the turbine blades. In this study, a time-varying model predictive controller is developed that uses preview measurements of wind speeds approaching the turbine. Performance of the controller is evaluated using ideal, undistorted measurements at positions that rotate with the turbine blade and measurements obtained at the same locations, but including distortion characteristic of LIDAR systems. Using these measurements, the model predictive controller is simulated in turbulent wind conditions and its performance is compared against previously designed, linear-time-invariant H1 preview controllers and industry standard controllers. Surprisingly, even though the LIDAR distortions produce signicant measurement error, controller performance is found to surpass that obtained using individual-pitch feedback-only controllers without preview. In previous studies, errors introduced articially, but of the same order of magnitude, were shown to degrade the performance of preview control so that it is worse than using feedback only. In this study, we also incorporate a simple error model to compensate the eect of LIDAR induced error, but nd that it does not improve performance.

Combined Feed-forward/Feedback Control of Wind Turbines to Reduce Blade Flap Bending Moments y

In above rated conditions, wind turbines are often subjected to undesirable high structural loading. We investigate two feedback control techniques in combination with a feedforward control method for reducing blade ap bending loads in above rated wind conditions. The feedback controls studied include both disturbance accommodating/tracking and integral augmented/repetitive types that incorporate models of persistent disturbances at DC (step changes in wind) and at the once per revolution frequency. Each method is combined with a feed-forward method utilizing the wind speed as measured at the blade tips and also based on the blade tip average wind speed. Performance is assessed by simulating the combined feed-forward/feedback systems on a three bladed turbine model with the National Renewable Energy Lab’s FAST wind turbine code. It is found that feedforward of blade tip average wind speed measurements can provide signicant reduction of blade root loads and improved speed regulation in time varying wind that is uniform across the rotor plane. However, the improvements are not as great in non-uniform and turbulent conditions in which case the blade tip average wind speed measurements provide incomplete information for conditions that can be unique at each blade. We use an extension of the linearized turbine model to design feed-forward compensation that uses individual measurements of the wind at each blade tip. Results suggest that using blade local measurements provides substantially greater reduction in blade loads, but assessing the full potential of using this more detailed information requires more accurate modeling of the way perturbations local to each blade couple into the turbine.

A spectral model for evaluating the effect of wind evolution on wind turbine preview control

As wind turbines become larger and more flexible, the potential benefits of load mitigating control systems become more important to reduce fatigue and extend component life. In the last five years, there has been significant research activity exploring the effectiveness of preview control techniques that may be feasible using advanced wind measurement technologies like LIDAR (light detection and ranging). However, most control development tools use Taylors frozen turbulence hypothesis. The end result is that preview measurements made up-stream from the rotor can be obtained with unrealistic accuracy, because the same wind velocities eventually arrive at the turbine. In this study, we extend the spectral methods commonly used to generate turbulent wind fields for controls simulation, but in a way that emulates wind evolution. This changes preview measurements made upwind from the rotor, in such a way that the differences- between the preview measurements and speeds arriving at the turbine- increase with distance from the rotor. We then evaluate the degradation in load mitigation performance of a controller that uses preview measurements obtained at various distances in front of the rotor.

Comparison of wind turbine operating transitions through the use of iterative learning control

In below-rated wind conditions, a wind turbine operates to maximize the amount of available power harvested from the wind and is said to be operating in region 2. In above-rated wind conditions, where regulation is the main objective to prevent over power and speed faults and to mitigate loads, the turbine is said to be in region 3. There is no standard method for operation at the boundary of the two regions and transitions between them can be problematic. In this study, we use iterative learning control to determine the control actuation necessary to accurately track idealized candidate trajectories during the transition between regions 2 and 3. The amount of control actuation required to track a transition trajectory and the ability to do so with minimal collateral loading determines which trajectory is most amenable for a given turbine. Trajectories are also graded by the average power produced during transition since they take the turbine off of the optimal power point.

Optimal Control of Wind Energy Systems: Towards a Global Approach (Munteanu, I. et al.; 2008) [Bookshelf]

This book is organized into eight chapters and three appendices. The first three chapters introduce wind energy and discuss modeling of both the wind and wind turbines, while the next three chapters discuss basic and advanced controllers for wind turbines. The last two chapters outline experimental systems for validating control algorithms and give some general conclusions. The appendices provide the wind turbine parameters considered in particular case studies, give an overview of the main ideas behind three advanced control methodologies, and present some photos and diagrams of experimental evaluation testbeds. This book provides a reasonably good overview of wind energy systems for control engineers, but being a monograph, it is not comprehensive enough to be the sole source of information for control researchers wanting to move into the wind energy area.

Multi-Blade Coordinate and direct techniques for asymptotic disturbance rejection in wind turbines

Multi-Blade Coordinates (MBC) is a technique used to transform rotating degrees of freedom into a reference frame that is fixed. In application to wind turbines, it has a number of advantages, a primary one being that its application tends to make dynamics that change with rotor position, appear fairly time invariant. Another advantage is that with MBC, controls that mitigate cyclic loads can be implemented using straight forward integral control that operates over a wide range of rotor speeds. However, using the standard MBC approach, the control system will not provide asymptotically perfect rejection of cyclic loads unless the rotor is symmetric with respect to each blade. On the other hand, at constant rotor speeds, a non-MBC approach can provide asymptotic rejection of cyclic loads even when the rotor is not symmetric. In this study, we investigate to what extent asymptotically perfect load mitigation is lost for MBC based controllers, in conditions that include rotor asymmetry and time varying wind speeds. It is found that although the standard MBC approach fails to achieve asymptotic rejection of cyclic loads, its performance is comparable to higher-order non-MBC approaches in turbulent wind conditions.