A problem about Fluent reference value
 

I'm solving a 3D flow around a circular cylinder case with air dielectric and LES model. However, the drag coefficient of cylinder wall doesn't match the value announced on papers.

Reference value is set as follows:

length = diameter of cylinder
area = diameter * height of cylinder; for 2D case, diameter of cylinder
velocity = velocity of inlet
viscosity = viscosity of air

Are they right or wrong?

Thanks in advance.

There is not necessarily a right or wrong when it comes to reference values.
The ones you have are an obvious choice, but someone else might find the surface of the cylinder more appropriate than the projected surface for the area reference value...

Since you are comparing your results to those in some papers, you simply take the same reference values as reported in these papers.
If they give coefficient values without stating the reference values used to obtain them, I would start searching for other sources to compare your results.