It is important to realize that the wind turbine performance is achieved to a large extent during normal operation. The cost, on the other hand, are generated during all operational modes (including power generation cases with the occurrence of fault, parking at storm, and idling). The energy capture of a variable speed wind turbine depends for the greater part on its ability to successfully operate at the peak of the Cp(X, 0)-curve in partial load. After all, maximum aerodynamic efficiency is achieved at the optimum tip-speed ratio A = Xopt at which the power coefficient Cp has its maximum value: Cp,max. This implies that the electromechanical torque Tem must be adapted in proportion to the wind velocity Vw squared (or rotor speed squared since Vw = urR/A) to maintain the optimum tip-speed ratio. In steady-state, the electromechanical torque balances the aerodynamical torque reduced by the amount of any mechanical torque loss in the drive-train. This fact is used in the commonly used "omega-squared" control law which is given by:
opt where T^ is the set-point for the electromechanical torque, p is the air density, R is the rotor radius, Cp,max is the power coefficient at the optimum tip-speed ratio Aopt, the rotor rotational speed, and Tioss the mechanical torque loss (which may itself be a function of the rotational speed and torque). Although this relationship is valid only in steady-state, it is also used dynamically to control the electromechanical torque demand of variable speed turbines as a function of the measured rotor (or generator speed since, in steady-state, = itr with itr the transmission ratio) speed. The measured rotational speed is often filtered using a first order low pass filter to avoid rapid electromechanical torque set-point changes. This way of controlling the electromechanical torque below rated wind speed may work satisfactory on relatively rigid turbines located on sites with uniform wind fields. More flexible variable speed turbines exposed to more turbulent wind conditions, however, will require a dynamic control law taking also the phase information into account.
Fingersh and Carlin  applied the "omega-squared" control law to the NREL Variable Speed Test Bed turbine and showed that this resulted in an overall energy capture loss of over 5 % when compared with the idealized variable speed operation. Modification of the law indicates a preliminary energy capture improvement of between 0.3% and 3%. It can be concluded that the losses associated with variable speed turbines running off the optimal point can be reduced by applying more advanced control algorithms to track the optimal tip-speed ratio in partial load.
Was this article helpful?