For general interest, I have posted my dissertation on swept wind turbine blade dynamic analysis (1.7 MB):
flight.engr.ucdavis.edu/~smlarwo … arwood.pdf
It discusses modifications to FAST for swept blades, including an additional degree of freedom that is used for the first torsion mode.
I have added a dissertation page that has errata and source files for codes mentioned in dissertation.
flight.engr.ucdavis.edu/~smlarwo … _page.html
Hi Scott,
I have read part of your work on the modelling of swept wind turbine blades; I found the thesis really interesting. I have a question for you, with regard to the aerodynamic modelling of such curved blades.
In your thesis you use the classical approach used for swept wing, i.e. projecting the flow perpendicularly to the LE. I found this approach also in other work of the past (as in “S. Liebst, WT gust load alleviation utilizing curved blades” for example).
Nevertheless, in this way you reduce the speed flow component that is useful for the power generation at the blade elements. Therefore, by consideirng all the other parameters (blade geometry, wind speed, angular speed etc.) as fixed, you obtain less power yielded for the WT with curved blades with respect to the one with straight blades. I do not know whether this is realistic or not and I would like if you could explain why you used this modelling.
Resuming, my question is:
if you curve a blade by translating the sections perpendicularly to the radial direction (without rotating the section), why you cannot consider that the flow is two-dimensional on sections perpendicular to radial direction rather than on sections perpendicular to the blade LE?
From my physical understanding it would seam that the flow could be approximated two-dimensional on sections perpendicular to the radial direction, even thought the blade is curved. On the other hand, in literature the other approach is used; so, probably I’am missing something and I would like to understand what it is.
Regards
Marco
Hi Marco,
Thank you for the question. I have never thought of translating the sections perpendicular to the radial direction; however, the flow would still have some angle relative to the leading edge because of the omega X position vector component. Also there might be difficulties in constructing the blade.
Yes you lose some efficiency in sweep, but you make up for it for a given turbine by increasing the rotor diameter. There is some debate on how the aerodynamics should be modeled; but we found at UC Davis through two separate verification efforts (CFD and panel vortex method) that considering the flow perpendicular to the leading edge for the aerodynamic loads was acceptable.
Hi Scott,
thanks for the reply.
I see your points, but I don’t understand what you wrote about the omega X position vector. If you add the wind speed (approximated as constant on the rotor plane and perpedicular to it) and the rotational speed, which is perpendicular to the radial direction of the rotor (not to the blade axis or to the point position vector), you should obtain a 2D flow in a plane perpendicular to such radial direction. Therefore, if the airfoil section is kept whitin this plane, you can consider the whole flow as the 2D flow on the section.
Sorry if I ask again, but this point is not clear to me.
Marco
I have attached a drawing showing how I visualized your proposed method- I call it “shearing” the sections. From what I understand, there is some flow crosswise to the leading edge due to the alignment of the omega x r vector; therefore the flow is not purely 2-D. If I am mistaken, please let me know.
You could design a swept curve such that the flow is 2-D; however, this might not be optimal for the structural benefits of sweep.
Hi Scott,
thanks a lot for the reply. Yes you are right 100%. Now I see what you meant, the drawing is very clear. So, we can conclude that however you design the swept blade, a lost of power is always unavoidable.
Clearly, this is true consideiring blades of the same radius.
Marco