jjonkman
commented
1 year ago Dear @Wangzhenyu12,
Your approach sounds OK. I believe the issue is that you are outputting the blade tip-to-tower clearance from ElastoDyn rather than AeroDyn. TipClrnc1 is an ElastoDyn output whereas B1NβClrnc would be the equivalent output of AeroDyn, where β is a number in the range [1,9], corresponding to entry β in the BlOutNd list, which you'd want to set to the node at the tip (= NumBlNds). ElastoDyn output TipClrnc1 is calculated between the tip node in ElastoDyn and the yaw axis and AeroDyn output B1NβClrnc is calculated between the aerodynamic output node β and the tower axis, minus the local tower radius.
When you define the aerodynamic offsets in AeroDyn (BlCrvAC, BlSwpAC), these define offsets of the aerodynamic center from the pitch axis. When you use ElastoDyn to model the blades (as opposed to modeling the blades in BeamDyn), the undeflected blade is always straight and lying along the pitch axis (including tip node). Adding aerodynamic offsets in AeroDyn does not change the shape of the blade in ElastoDyn; rather, it only changes the calculation of the aerodynamic loads and how these loads get mapped to the structural model.
FYI: I would normally expect an aerodynamic imbalance to involve a change in aerodynamic twist, chord, or airfoil data between blades, not necessarily a change in aerodynamic centers.
Best regards,
Wangzhenyu12
Hello, everyone!
I want to study the effect of areodynamic imbalance on clearance. According to the OpenFAST Documentation Section 4.2.1: Aerodynamic imbalances are possible through the use of geometrical differences between each blade.
Then I modify BlCrvAC and BlCrvAng in AeroDyn_blade.dat to set initial bending for blade1 of a 5MW onshore wind turbine, the BlCrvAC of blade tip is about -10m. BlCrvAng are recalculated by the formula. BlCrvAng(n) = atan ( (BlCrvAC(n+1) - BlCrvAC(n-1) ) / ( BlSpn(n+1) - BlSpn(n-1) )*180/pi.
The AeroDyn_blade.dat of blade1 is shown below.
Then I replace ADBlFile(1) in AeroDyn15.dat and output TipClrnc1,TipClrnc2 and TipClrnc3. I do not use BeamDyn in the calculation. The raw clearance output data are like sine waves. I only care about the minimum clearance so I pick every lowest point of each cycle from the raw clearance data and draw the minimum clearance figures of every blade. The sampling frequency is same as the rotation frequency for every blade. I think the minimum clearance occurs at the moment when the blade passes the tower according to the definition of TipClrnc.
According to the description of BlCrvAC:upwind turbines have negative BlCrvAC for improved tower clearance. But this is not consistent with my calculation result. The clearance of blade1 has surprisingly become smaller than other two blades. And The clearance difference is very small. The clearance time-series are shown below under mean wind speed 3m/s with IECturbc B.
As far as I understand, the clearance of blade1 should be much longer than the other two blades because of the initial bending -10m. I also calculate the results under different mean wind speed such as 1m/s and 12m/s, The results are similar.
Please help me to solve this problem. Thanks and regards!