Dear Roger,
Firstly, thank you for your efforts to solve my problem. It works well after changing the values mentioned above.
Moreover, i would like another question:
for this numerical instability problem, why did you choose increase the platform Yaw inertia and adjust the Rayleigh Damp parameters to deal with this problem? In other words, why do you choose these two ways to eliminate the instability?
Best regards
Yiqi
Dear @Han.Yiqi,
I am happy to hear that the OpenFAST model is running.
When I was testing your model, I observed that the instability was related to the yaw degree of freedom. In the yaw direction, there is the NacelleYInertia and the YawSpr at the upper side (tower top) and the PlatformYInertia and the monopile at the lower side. I guess there is a torsional mode (that it doesn’t have to be necessarily the 1st torsional mode) or a coupling between modes that makes the system numerically unstable. By increasing the PlatformYInertia, this issue seemed to disappear. To better understand this behavior, it would be necessary to perform an eigenanalysis of the system.
By looking at the files, I also observed that the monopile (defined in SubDyn) did not have any damping applied. There was only a damping applied to the internal modes, but no internal modes were retained (Guyan approach was adopted). Since the system was experiencing numerical instabilities, having a damping (and specially a stiffness proportional damping that increases linearly with the frequency) would help to damp out any possible high frequency that challenged the solver.
I hope that helps!
Roger