An experimental hybrid control approach for wind turbine emulator.

Abstract

In this study, the experimental implementation of a Hybrid Controller (HC) for a Wind Turbine Emulator (WTE) is presented. Conventionally, the speed control of a WTE is based on two Proportional Integral (PI) controllers used for the outer (speed) loop and inner (torque) loop of a DC motor drive. However, based on the extensive literature survey, with the PI controllers alone, significant overshoot and substantial settling time in the speed response are unavoidable. The overshoot cannot be completely removed without sacrificing speediness. To eliminate the overshoot and concurrently improve the settling time, an HC, combination of Adaptive Neuro-Fuzzy Inference System (ANFIS) and PI controller for the outer and inner loops, respectively, is proposed, designed, implemented and analyzed. The actual wind turbine is emulated using a 2-quadrant DC motor drive based on DS1104 R&D controller board from dSPACE. These two control schemes, namely PI (both controllers are PI), and HC (PI controller for inner loop and ANFIS controller for outer loop) for the WTE are extensively simulated based on the small-scale wind turbine model and eventually implemented using a small-scale laboratory set-up. When implemented in the WTE, with a step-change in wind speed, the hybrid control scheme produces a better result. It is found out the Root Mean Square Error (RMSE) for the WTE rotor speed of the HC is 0.495, which is lower compared to PI control scheme RMSE of 0.7273. Finally, to verify the validity of the WTE based on HC, a random wind speed with wind turbine parameters extracted from the actual wind turbine is applied. The results indicated that the WTE perfectly emulated the wind turbine rotor speed with an RMSE of 0.5004.