Dear Mustafa,
Just a few comments:
As discussed in the following topic on our forum, WT_Perf is no longer supported by NREL, but has been effectively replaced by the standalone driver for AeroDyn v15: WT_Perf being removed from distribution - #4 by Kisorthman.Vimalakan.
As far as I can tell, your results are quite close to those of WT_Perf and because you are using a different high-induction correction; perhaps that explains some of the differences you are seeing. AeroDyn v15–and it sounds like your code–replace the subiterations for axial and tangential induction with a single iteration covering both; perhaps this explains some of the differences in the windmill state. Regardless, though, I would not be too considered with your BEM implementation. You could compare your results to those of AeroDyn v15 if you wanted another set of results.
WT_Perf, like the standalone driver for AeroDyn v15, does not include structural weight/inertia terms.
I can’t comment on specific releases of WT_Perf; you’d have to look at the source-code changes between releases to answer your related questions.
There should not be imaginary numbers resulting from the Glauert/Buhl correction, as the correction is not applied for cases beyond the bounds of the equation’s validity.
Best regards,