Validation Study for AeroDyn and YawDyn Using Phase III Combined Experiment Data

This paper reports results of a study to determine the accuracy of two computer programs, YawDyn and AeroDyn for ADAMS, when predicting selected parameters on the Phase III Combined Experiment Rotor at the National Wind Technology Center in Golden, Colorado. Copmparisons are made between predicted and measured yaw motions (in free yaw), yaw moments (in fixed yaw), blade root flap movements, andaerodynamic lift coefficients and angle of attack at three blade stations. A comparison of blade root fatigue lifetime based upon predicted and measured loads is also shown to provide a simple measure of the overall accuracy of the simulations. Predictions are made using both measured winds and simulated turbulence as inputs to the models. Results include a demonstration of the ability of themodels to predict upwind motion of the (normally) downwind rotor. More typical downwind, free-yaw motion is also predicted quite well. The match between predicted and measured blade load rainflow cycle counts is generally good, with the largest discrepancies appearing at the low-cycle, high-amplitude end of the load spectrum where there is the greatest amount of scatter in both the measured andpredicted cycle counts. Aerodynamic lift and angle of attack are also predicted quite closely at the 47% and 63% stations. At the 95% station there is a clear failure of the models to properly account for three-dimensional flow effects near the blade tip. The measured lift curve slope is lower and the angle-of-attack range is narrower than predicted.