Does shear make velocity a function of height?

[danrhouck]

Typically, to calculate the "actual" velocity from the output induced velocities in the FieldData, it would be induced velocity*U_inf+Uinf in order to denormalize it and then add the freestream to the induced velocity. When shear is turned on, does the added velocity need to be a function of height to match the shear profile?

Related to the use of shear, is it possible that the use of shear causes the wake to drift upward? I struggle to think about this in terms of a free vortex method, but I'm thinking the shear is perhaps effectively superimposed as I suggest above, which would cause the top of the wake to advect faster than the bottom. In full fidelity, I think this would, in turn, increase the vertical mixing and, perhaps, drive the wake upward. Just spit balling on that. Basic question is how realistic is the use of shear in these simulations?

[whophil]

The induced velocity is only that due to the vortex and wall system, so you will need to add the height dependent profile to it.

CACTUS uses a "one-way" coupling to model the shear, i.e. shear is just a superposition to the inflow. This is known to cause a non-physical lifting of the wake system. Some discussion on this topic (not specific to CACTUS) can be found in Section 4.2 of the reference below:

• E Branlard et al 2015 J. Phys.: Conf. Ser. 625 012019. Aeroelastic large eddy simulations using vortex methods: unfrozen turbulent and sheared inflow
  https://iopscience.iop.org/article/10.1088/1742-6596/625/1/012019/pdf

I wrote some CACTUS-specific observations a never-released SAND report, which I will send you privately.

[xflow-ian]

I would also be interested in seeing this report if possible. I am interested in implementing addition IEC 61400 design load cases into the free stream input and it would be good to understand the consequences to the results in CACTUS from previous work.

Thanks,
Ian Brownstein, PhD
Co-founder, XFlow Energy Company
xflowenergy.com