Normally a rotor on bearings does not require pre-stresses modal analysis, since neither the rotor or the bearing stiffness does not change with loading (in this case transverse loading). In order to use pre-stressed modal analysis, you need to have loading which has an effect of changing the overall stiffness of the system. For instance, if you are interested on the modal analysis of a blade on a rotating hub, the rotational speed causes a radial force on the blade which extends it. This extension increases the overall stiffness of the blade. Therefore, results of standart modal analysis corresponds to zero rotational speed. At any other rotational speed, you need do pre-stressed modal analysis to determine the actual mode shapes and natural frequencies corresponding to that speed of rotation.

In rotordynamic analysis, gyroscopic effects are included in the analysis as a result, the modal analysis is done depending on the speed of rotation, since gyroscopic effects depend on the rotational speed. Moreover, rotordynamic analysis may include the stress stiffening (as in the blade example, pre-stressed modal analysis) or spin softening effects depending on the options selected.

If you think that, stress stiffening and/or spin softening effects are important you can turn on these option the rotordynamic analysis. This depends on the rotordyanmic system. If you are not sure what to do, you can always perform two different analysis with and without these effects included and compare your results to see if these effects make a significant change.

Thanks Ender Cigeroglu ! I understand what you are saying and thank you for putting it succinctly. I have tried doing both and I see that there isn't much stress stiffening in rotors like shafts and hence the rotordynamic analysis results are as expected and match with tests. However, when I have a rotor that contains non-axisymmetric features e.g. blades, ANSYS, by default, does not take into account the gyroscopic effects (and prestress effects) on those features. Having read the documentation further I believe I ought to do a prestress analysis to get the prestress effects of the rotating reference frame analysis to include the gyroscopic effects and Workbench needs to be tweaked to include relevant APDL commands. It would be good to know if someone does this and what their experiences are. I think now my question is more specific to the tool i.e. ANSYS Workbench.

Bladed disks are much stiffer compared to rotor and therefore, you can study the vibration of the bladed disk by only considering that part. This means you will specify a fixed BC on the disk-shaft contact surface. Define the rotational speed and do the prestressed modal analysis by including gyroscopic effects. This type of analyses can be done like this in ANSYS.