Dear Ramya,
Here are my answers to your questions:
Q1. The IEA Wind 15-MW reference wind turbine has a direct drive (GBRatio = 1), with a shaft that is short and a stiffness that is high. So, there is no reason to enable the drivetrain-torsion DOF.
Q2. My guess is enabling the drivetrain DOF results in a very high natural frequency that would require a very small time step to otherwise avoid a numerical instability in the solution.
Q3. As with any floating wind turbine simulation, to avoid long start-up transients, I would set the the initial conditions of the model appropriate for the mean wind speed being simulated. Proper initial conditions should be set for the rotor speed and blade-pitch angles, as well as the platform surge and platform pitch DOFs based on their expected mean values for the mean wind speed simulated.
Q4. Again, there is no need to run this model with the drivetrain-torsion DOF enabled. I’m not sure what is going on with the simulation with the drivetrain-torsion DOF disabled, but the solution appears to be suffering a numerical or physical instability. Similar problems have been discussed many times on this forum. Did you change the system masses, stiffness, or time step?
Best regards,