Can MOSES Questions

Q: Can MOSES do stability analysis of a multi-body system connected by connectors ?
REV 7.05

A: This depends on what you mean by "stability analysis". If you mean finding the restoring force for a given displacement, then you can find it with a procedure similar to that outlined in the sample C_RARM.

Q: Can MOSES simulate a two body, frequency domain lift in the Frequency Domain?
REV 7.03

A: Of course MOSES gives good results for what you have modeled. In reality, there are several ways that your model will break down.

  • The frequency domain cannot take into account the nonlinearities of the connection such as a line going slack,
  • You cannot look at the nonlinearities of the lifted body impacting the structure from which it was lifted,
  • You cannot look at the slam when the lifted structure enters the water,
  • You cannot look at the nonlinearities when the lifted body is set down, and finally
  • There is probably no damping on the lifted body when it is in the air. This will cause unreasonably large motions of the lifted body in the neighborhood of the natural period of the "swinging mode". You should add some damping by either using one of the damping options on the connector definition or with #DRAG.
To fix all of the first four points, you can do the problem in the time domain.

Q: Can MOSES perform quasi-static, one-damage, and transient mooring analyses?
REV 7.00

A: In a word, yes!

Quasi-Static mooring analysis is what you get when you find equilibrium and then either compute RAOs or SRESPONSE. The mooring line tensions are the ones which should be compared to the allowable.

One-Damage is not so well defined. You can place the vessel in its extreme position as predicted by the quasi-static analysis, deactivate the most heavily loaded line and perform a time domain analysis. You may or may not have an environment applied.

Transient mooring analysis is simply a time domain simulation for a moored vessel.

Click Here for a sample which does the quasi-static and transient analysis. Also Click Here for more discussion on the subject.

Q: Can MOSES compute the maxima as outlined in API RP2SK?
REV 7.00

A: In a word, yes!

In the frequency domain:

Use the CULL command to split the results into low and high frequency parts and then use STATISTICS to compute the statistics of each part and combine according to RP2SK; i.e. Ur = Max + Sig Where Max is the maximum response of the larger of the two and Sig is the significant response of the smaller.

In the time domain:

You use add_col s_high -filter period 1 5 30 1e5 add_col s_low -combine 5 1 23 -1 to split a signal into high and low frequency values and then compute the statistics for each part and combine.

Click Here for an example.

Notice that we can write the RP2SK recipe as:

    Ur = 3.72 Rl [ 1 + 2/3.27  Rs/Rl ]
where Rl is the RMS of the larger and Rs is the RMS of the smaller. Alternatively, the maximum of the whole response is
    Ut = 3.72 sqrt[ Ml + Ms]
    Ut = 3.72 Rl sqrt[ 1  + Ms/Ml ]
and Ml and Ms are the squares of Rl and Rs respectively. The ratio
    Ur/Ut = [1 + .5376 sqrt(Ms/Ml)] / sqrt[ 1 + Ms/Ml]
The following is a plot of this ratio as a function of the ratio of the zeroth moments of the low and high frequency parts.

As you can see, the maximum of this function is around 1.13; i.e. the maximum error you will get by using the MOSES computed maximum in lieu of the RP2SK method is 1.13.

Thus if you set the multiplier for the "maximum" to 3.72 * 1.13 = 4.2 you can use the maximum computed by MOSES in the frequency domain instead of what is done in the example. This would probably be a reasonable practice in the time domain also.

Q: Can MOSES compute the natural periods of a vessel?
REV 7.00

A: Not directly. In general, the equation of motion here are integro differential and therefore classical natural frequencies do not exist. As an alternative you can find the period where you get a peak response. If you have a system with classical natural periods then a place of maximum response is normally the natural period if the knowledge of the period is of interest. (i.e. you have an under damped system)

You need to be careful here, however. If the natural period is outside the wave energy range you will need to use the command SRESPONSE and use wind as an exciting force.

Q: Can MOSES analyze the "springing effect" on a ship hull girder in waves?
REV 6.02

A: Yes, you need to model the structural attributes of the ship hull, and use Generalized Degrees of Freedom.

Q: Can MOSES consider the Green Water Effect?
REV 6.02

A: This depends on exactly what you mean by "Green Water Effect". If you include a #plate or use pgen -cs_current 1 1 1 in your model, and compute vessel motions in the time domain, you will get drag and slam loading on the deck. You will also want to use &DEFAULT -WAVE_RUN YES.

Q: Can MOSES be used to compute the air gap for a structure?
REV 6.02

A: Yes, MOSES can compute the air gap for any structure. You need to compute time domain motions for the vessel, and then post process the results using the POINTS command. You will need to define the point on the vessel where you want to measure the air gap, and use VLIST to show the available variables. The wave clearance is the variable you want, and you will probably want to obtain statistics of the wave clearance over the time history, to illustrate the maximum and minimum values. You can also compute the air gap in the frequency domain, using ST_CLEARANCE.

Q: Can MOSES compute wave induced shear force and wave induced bending moment on a ship in the frequency domain?
REV 5.07

A: Yes, but perhaps not in the way you think. Other programs that deal with only monohull ship structures may provide this in their frequency domain post processors. Since a hull girder is statically determinant, this is relatively straight forward. However, for other vessel types, such as semisubmersibles or TLPs, this is not so easy. Since MOSES is a very general program, the way to include frequency domain wave loads in longitudinal strength is perform a proper structural analysis. The hull girder can be modeled as a single beam along the centerline, and the structural post processor provides the loads in the beam. Start with the sample JT_FREQ. (For details, click here)

Q: Can MOSES evaluate relative velocity between jacket structure and wave during tow analysis?
REV 7.06

A: Yes and No. In the time domain, the forces on Morrisons' elements are evaluated based on the relative velocity. You can get a time trace of the force on a member or a load group during a time domain simulation. Or you can use the POINTS command to find the releative velocity between a point on the jacket and a point set at global Z = 0. We do not, however, have an ability to "print out" the relative velocity between a point and the wave.

Q: Can MOSES simulate a problem at model test tank scale?
REV 5.03

A: No. There are several constants and factors embedded in the software that come from real size problems.