Inputs of MoorDyn_Step function

[soydanahmet]

Hi,

I am currently in the process of coupling Mordyn v2.2.1 with my code, and I am seeking clarification regarding the inputs of the MoorDyn_Step function, specifically in terms of their definitions and units. Your assistance in clarifying this matter would be greatly appreciated.

The MoorDyn_Step function has the following input parameters:

MoorDyn_Step(MoorDyn system, const double x, const double xd, double f, double t, double *dt)

Based on my understanding, the input vectors x (Position vector) and xd (Velocity vector) should be structured as follows:

x : {CoM_X, CoM_Y, CoM_Z, phi, theta, psi}

CoM_X: Coordinate the center of mass in the x-direction (meter) CoM_Y: Coordinate the center of mass in the y-direction (meter) CoM_Z: Coordinate the center of mass in the z-direction (meter)

phi: roll angle (rad) theta: pitch angle (rad) psi: yaw angle (rad)

xd : {V_X, V_Y, V_Z, omega_I(0), omega_I(1), omega_I(2)}

V_X: Velocity of the center of mass in the x-direction (m/s) V_Y: Velocity of the center of mass in the y-direction (m/s) V_Z: Velocity of the center of mass in the z-direction (m/s)

omega_I(0): Angular velocity around the x-direction (rad/s) omega_I(1): Angular velocity around the y-direction (rad/s) omega_I(2): Angular velocity around the z-direction (rad/s)

Thanks in advance.

[RyanDavies19]

Hi @soydanahmet, if you are working with a single coupled body then yes that is correct. The units for rotations are radians for the inputs into the step and init functions.

If you are working with other coupled objects then these vectors will change according to the object type. You can find more information in the coupling section of the documentation.

[soydanahmet]

Thank you so much for your prompt response. Now my code is working. I have one more (and last :) question about the moorDyn time step.

I am currently fine-tuning its integration with my computational framework. However, I might encounter a situation where the time step size in my code, which is controlled under the CFL condition, occasionally results in very small time step sizes. In contrast, the MoorDyn time step, as initially defined in the input file, may remain relatively large. So I'm curious whether it is possible to adjust the MoorDyn time step dynamically during the simulation.

[sanguinariojoe]

If you ask MoorDyn to integrate a time step, dt, which is smaller than the inner/model time step, dtM, then dtM is automagically corrected as dtM = dt

Even more, dt will be always honoured. If you ask to integrate a dt which is not divisible by dtM, then Moordyn will do as many steps with dtM as possible followed by a last smaller one.

On Tue, 3 Oct 2023, 17:25 Ahmet Soydan, @.***> wrote:

Thank you so much for your prompt response. Now my code is working. I have one more (and last :) question about the moorDyn time step.

I am currently fine-tuning its integration with my computational framework. However, I might encounter a situation where the time step size in my code, which is controlled under the CFL condition, occasionally results in very small time step sizes. In contrast, the MoorDyn time step, as initially defined in the input file, may remain relatively large. So I'm curious whether it is possible to adjust the MoorDyn time step dynamically during the simulation.

— Reply to this email directly, view it on GitHub https://github.com/FloatingArrayDesign/MoorDyn/issues/144#issuecomment-1745210385, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAMXKKF3JURXXHPQQ4G4DIDX5QU7DAVCNFSM6AAAAAA5PWH36OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONBVGIYTAMZYGU . You are receiving this because you are subscribed to this thread.Message ID: @.***>