Some Analyses of Energy Production from the NWTC Variable Speed Test Bed

Using the NREL 20-kilowatt, direct-drive, variable-speed wind turbine test bed, we used collected data to examine differences between constant-speed and variable-speed wind tubine operations. The assigned goal of the variable-speed control algorithm we used was to adjust continuously the turbine rotor speed so that its tip-speed ration (TSR) stays as close as possible to a precalculated optimumvalue. To examine the success of the variable-speed control, histograms of tip-speed ratio root-mean-square (rms) error for each data set were calculated and plotted. The resulting histograms of both constant and variable-speed control algorithms were compared. The results validated the expected conclusion that lower rms TSR errors are associated with higher measured wind turbine powercoefficients. A second comparison was made between variable-speed histograms and synthesized constant-speed histograms for the same wind. The decrease in rms TSR error of the variable-speed case was used to quantify the observed improvement in TSR tracking brought by the variable-speed control.