The wind speed has a huge impact on the dynamic response of wind turbine. Because of this, many control algorithms use a measure of the wind speed to increase performance, e.g. by gain scheduling and feed forward. Unfortunately, no accurate measurement of the effective wind speed is online available from direct measurements, which means that it must be estimated in order to make such control methods applicable in practice. In this paper a new method is presented for the estimation of the effective wind speed. First, the rotor speed and aerodynamic torque are estimated by a combined state and input observer. These two variables combined with the measured pitch angle is then used to calculate the effective wind speed by an inversion of a static aerodynamic model.
SUMMARYThis paper considers the design of linear parameter varying (LPV) controllers for wind turbines in order to obtain a multivariable control law that covers the entire nominal operating trajectory.The paper first presents a controller structure for selecting a proper operating trajectory as a function of estimated wind speed. The dynamic control law is based on LPV controller synthesis with general parameter dependency by gridding the parameter space.The controller construction can, for medium-to large-scale systems, be difficult from a numerical point of view, because the involved matrix operations tend to be ill-conditioned. The paper proposes a controller construction algorithm together with various remedies for improving the numerical conditioning the algorithm.The proposed algorithm is applied to the design of a LPV controller for wind turbines, and a comparison is made with a controller designed using classical techniques to conclude that an improvement in performance is obtained for the entire operating envelope.
www.imm.dtu.dk Also a special thank to M.Sc.Eng. Bo J. Pedersen for numerous fruitful discussions on fatigue loads as well as general aspects of the analysis of dynamic systems. Finally, a very special thank to my supervisor, Assoc. Prof. Niels Kjølstad Poulsen, IMM, DTU, for providing qualified and dedicated guidance on any matter during the project period.
Abstract-Large scale wind turbines are lightly damped mechanical structures driven by wind that is constantly fluctuating. In this paper, we address the design of a model-based receding horizon control scheme to reduce the structural loads in the transmission system and the tower, as well as provide constant (or at least smooth) power generation. Our controller incorporates two optimization problems: one to predict or estimate mean wind speed, given LIDAR data, and the other to carry out receding horizon control to choose the control inputs. The method is verified against an existing wind turbine control system, and shows reductions in both extreme loads and power fluctuations by 80% and 90% respectively, when compared to a conventional controller.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.