Suggestions to improve result
 

Hello,
i'm here to ask for suggestions to improve the results of the following simulation.

I'm trying to do a validation, using the experimental data found in:
"Near wall characterization of the flow over a two-dimensional steep smooth hill", J. B. R. Loureiro Æ F. T. Pinho Æ A. P. Silva Freire

the inlet velocity profile, and the inlet turbulence profile are shown in the first attachment.

The simulation setup is the following:
Models: Standard k-epsilon with Enanced wall function.
Boundary:
inlet: velocity-inlet, with user definied function for velocity and turbulent kinetic energy.
outlet: pressure-outlet
top: simmetry
ground: wall

In the second attachment you can see the yplus for the ground surface, hence i've used the Enhanced wall treatment option.
In the last attachment you see the velocity profiles (normalized in respect to the hill height). The simulation is able to "capture" the "trends" of the experimental data, but there are some significative variations.

What can i do to improve the results?
Thank you very much.