modelica
MBS Multibody Instabilities
Issue: I am getting increasingly chaotic results when using the Multibody Library. In my case I have made custom linear functions to apply forces (damping, stiffness) onto bodies. The code is on my repo OceanEngineeringToolbox. I will list the issues that I am encoutering.
-
I measure kinematics with absolute sensor, when using the resolve in world frame I get reasonable results; however, when using the local resolve numerical instabilities arise and it crashes. Which doesn't really make any sense because the local and global quantities are very similar im magnitude and the same in frequency and phase being only a kinematic transformation to convert between the two.
-
When running single body simulations, results are expected and I can validate them with external software. However, when coupling two bodies through a prismatic constraint, the results are no longer representative of the true dynamics.
-
I have tested this with a simple mass and sinsuisoidal input to view the response. I notice that the response is not always harmonic, the amplitude of the troughs and crests are not the same magnitude, which I don't know how that woud be possible for a linear system subject to harmonic forcing.
I believe these are the base issues causing problems with my project and I can't seem to resolve them on my end. I can provide any files/results needed to help further debug. Thank you very much for any help,
Thomas
Please let us know which MSL version and which Modelica simulation tool you use.
As a start for easier analysis it would help if you could add a SSCCE (consisting only of MSL classes) here together with the expected trajectory and the observed behaviour.
I am using OMEdit v1.24.3 (64-bit) and v4.0.0 of the MSL.
I will try to develop a SSCCE @beutlich.