Makers of MOSES - The Final Answer in Marine Simulation
What Is New In Rev 5.01

General Changes:

Over the past several years, we have been asked to add numerous new features and new computational algorithms into our software. As a result, it was beginning to look as if it were patched together rather than being a true whole. Also, we had been requested to implement some new algorithms which simply did not fit into the existing structure. The result of these two pressures is this, the first release of the next generation of our software. As a result, quite a few things have changed. We have tried to keep compatibility with existing command and data files when possible, but there are a few things which simply must be changed. In keeping with the major changes to the program, the manual has been altered dramatically. Please take the time to read the manual from cover to cover. It will give you a better appreciation of how things now fit together than tackling the changes on an ad hoc basis.

While we have added many new capabilities with this release, it is just the beginning. Our current primary goal is incorporate the capability of "structural dynamics" during a simulation. Thus, in the near future, you will be able to consider the effect of flexibility during a launch or tow, not only at the stress analysis stage. Also, you will be able to consider the modes of vibration during any fatigue or stress analysis. Another objective is to again improve the computation of viscous damping.

What Answers Will Change:

Since the entire foundation of the program has been changed, expect the results to change. Most of the change is directly related to the way in which viscous damping is treated, but not entirely. Time domain simulations will change because the time variation of the drift force is now simulated better, and because of better numerical algorithms in computing the inverse Fourier transform of the damping. Some change is also attributable to the new computation of the hydrostatic force. While in some cases, the differences may be substantial, the new and old results normally yield the same engineering results. We have given a great deal of consideration to these changes, and feel that the new results are superior to the old ones.

Hydrostatic Models:

The changes are more apparent from a philosophical point of view, than from actually running the program itself. The basic operation of the program is much the same except that many command names have been changed to accommodate the new features. Fundamentally, however, things are quite different. First, hydrostatics are no longer computed using stations and offsets. Now, everything (except strip theory) is based on a panel (like a diffraction) model of the body surface. This results in better accuracy for the hydrostatic computations, and it allows us to do things we could not before. For example, previously, the forces on a "hull" in waves during a time domain simulation were not computed correctly. You can now replace the simplifying assumptions with a true computation of the pressure over the current wetted surface.

Since we no longer have to cope with stations and offsets, we can "rationalize" the way in which loads are mapped to the structure for a stress analysis. In the past, this mapping performed manually. Instead, we now have the concept of a point. All locations are described in terms of points, and nodes are simply points which have structural elements attached to them. Loads on panels are applied to points defining the panels, and then mapped to the nearest node, automatically.

Frequency Response:

Another major change is the way of viewing frequency response. In the past, we computed "response operators". These were simply responses to a wave of unit height and specified period and direction. This created some difficulty in how to account for nonlinear effects such as viscous damping, and it made it impossible to consider response at frequencies where there was not wave energy. Now, instead of "response operators",we consider "frequency response". While you can still get response operators, you can now get frequency response of the system to combined wind, wave, and nonlinear wave force spectra. This allows one to investigate the slow drift behavior of a moored system without resorting to the time domain. As a result of the new frequency response computation, the wave drift force treatment has been changed. In the past, only the mean drift on a stationary vessel was considered. Now, one can ask that the last set of RAOS be used to correct the drift force. Also, "wave drift" damping can now be added. As a result of this generalization, the FRQPOST menu had to be changed. Almost all of the commands in this menu have simply changed names, and now you can compute relative motion, velocity raos, and acceleration raos.

Defaults and Parameters:

Perhaps the most difficult change is that &DEFAULT and &PARMS are no longer valid commands. Three new commands replace them. &DIMEN changes dimensions, &MODEL_DEF defines defaults which are used only during the definition or editing of a model, and &PARAMETER handles most of the things which old &PARMS did as well as the things which used to be on &DEFAULT that are not applicable to modeling. The best approach here is to find the &PARMS which alter dimensions, change these to &DIMEN and then change all occurrences of &PARMS to &PARAMETER. Next, change all occurrences of &DEFAULT to &MODEL_DEF. This gets most things except for the "strange" options of &DEFAULT. The program will quickly tell you of these with errors.

For our unix users, we have provided a script, g4to5, which changes existing data files into new ones. It does the things outlined above as well as most of the other changes needed. This script is used as g4to5 root, where root is the root name to be converted. It not only converts the files, but saves the old ones with a prefix of o_, and builds files compare-root.x which show the changes made to the files.

As mentioned above, the way hydrostatic properties are computed is completely different. As a result, the commands defining hulls, stations, and offsets are missing from the documentation, but they are still honored. In this release, the same set of panels is used for hydrostatic and diffraction computations. Thus, if you have a mesh defined in an existing data file, you must remove the existing hull description that used to define the hydrostatics for the mesh. If you do not, you will get duplicate contributions from the hull and the mesh. To facilitate defining meshes, a new menu, &SURFACE, has been added. In essence, this menu allows one to build simple blocks and operate on them to create a complex structure. Another function of this menu is to convert old mesh descriptions into new ones. To do this, you first create a file containing only the mesh description. Also, you may have to change the order of the file so that the vertices are defined before the panels. Now, suppose that this file is "mesh". Then to convert the mesh, do the following: 

 
     moses new
     &surface
       &insert mesh
       rename @
       emit name -part
       emit name -3ddif
       &fini
Here, name is the name you wish to give to the mesh. The result of these commands is that a converted mesh will be in the file new.ppout. Again for unix users, this is automatically accomplished with g4to5.

An implication of the creation of the point concept discussed above is that now interest points are associated with existing points and need not be defined explicitly. For this reason, the command, POINT, in the &INTEREST MENU was removed. You should now define the point previously and "associate" it with various interest types. The association is accomplished with a new command IP_ASSOCIATE.

&DESCRIBE:

The way in which &DESCRIBE works has been changed and extended. It now has a form of: 
 
   &DESCRIBE TYPE NAME, -OPTIONS
where TYPE is something like body, part, etc. Now, however, the options can be used to alter attributes of the thing named, NAME. By doing this, we have eliminated the commands: BODIES, BO_ACTIVE, PA_ACTIV, CMP_ATTR. The old form of &DESCRIBE still works, but will be retired in the future.

Hydrodynamic Database:

The entire hydrodynamic database structure has been changed. In the past, forces, added mass, and damping matrices were stored. Now, pressures are stored. This results in rather dramatic savings of disk space for large problems. In fact, one problem went from 1000 megabytes for the root.hyd file to 20 with the new storage scheme. The data exported and "imported" reflect these changes. We now have two types of files that can be exported. H_DBASE is similar to the file we had before, but one can no longer use this type of database to perform a structural analysis. It can still be used however, for simulation purposes. The new type H_PRESSURE can be used for everything and is the preferred storage mechanism. In the past, the hydrodynamic database also contained the mean drift force and the viscous damping. These are now separate exportable items. In fact, we now do not "import" things but instead we define them in the &DATA MENU. This allows for more flexibility in the types of force data which can be read in.

We have made it easier to alter the mean drift force for a body. In particular, you can now define your own drift force table in &DATA and instruct the program to use your table with &DESCRIBE BODY -M_DRIFT name. As a result, the -WDRIFT option on &ENV has been removed.

In the past, there was some confusion about how Morison forces were treated with respect to the importation of a hydrodynamic database. As of now, no Morison contributions are included. They are recomputed whenever they are needed. Also, we now consider the viscous portion of the Morison force due to wave velocity.

A final implication of abandoning stations and offsets is that we can no longer universally do a good job of computing viscous roll damping via the method of Tanaka and Kato. For single hull vessels, the results remain much as before. For multi-hulled vessels, however, the results are rubbish, since we cannot really establish where eddies are shed. To resolve this problem, we have provided the user with a mechanism for defining exactly how to compute these coefficients. If you really do not want to expend the effort to uses these options, then you should "turn off" the viscous roll damping for any multi-hulled vessel. To turn if off, issue the command: 

 
   &DESCRIBE BODY NAME -TANAKA NONE
or
   &DESCRIBE BODY NAME -FM_TANAKA 0. -TM_TANAKA 0.
where NAME is the name of the multi-hulled vessel.

Miscellaneous Changes:

The manner in which launch ways are handled during a nonlinear stress analysis has been changed. We now consider the length and stiffness of the tilt beam.

One can now input a model completely in MEDIT without doing an inmodel. To make this possible, a new command TRANS_CON was created which replaces the old commands: JLLEGS, JLLEGP, LWAYS, LWAYP, and LOCJ. Thus, these commands will also be retired in the future.

As a result of these modeling changes the names available for &STATUS, &SUMMARY, and &PLTMOD have been changed. Most of these changes simply reflect the new nomenclature. &PLTMOD, however, has some substantial changes. Now, the VESSEL plots show only the exterior compartments. To look at the interior compartmentation, one uses COMPARTMENT. Also, the JOINS_SUM command has new types. COORDINATES produces the coordinates of nodes only while POINTS produces results for only those points which are not nodes.

Summary Of Command Changes:



Old Command                   Replacement






&PARMS                        &DIMEN  and &PARAMETER


&DEFAULT                      &MODEL_DEF and  &PARAMETER


&INSTATE -CURRENT              


&DATA -SHAPES                 &DATA SHAPES                    


&DATA -SOIL                   &DATA SOIL                      


&DATA -GRID                   &DATA ENV ; GRID                


&DATA -F_SPECT                &DATA ENV ; F_SPECT             


&DATA -P_SPECT                &DATA ENV ; P_SPECT             


&DATA -PROFILE                &DATA ENV ; PROFILE             


&DATA -TEMPRES                &DATA ENV ; TEMPRES             


&DATA -MGROWTH                &DATA ENV ; MGROWTH             


POINT                         IP_ASSOCIATE


IRR_EXFO                      ST_EXFOR


TRANSLAT                      FR_POINT


CFORCE                        FR_FCARG


RAO_CFOR                      FR_CFORC


IRREG -MOTION YES             ST_POINT


IRREG -MOTION NO              ST_FCARG


IRR_CFOR                      ST_CFORC


IRR_RFOR                      ST_RFORC


IRR_RSTR                      ST_RSTRE


SPECTRUM -MOTION YES          SP_POINT


SPECTRUM -MOTION NO           SP_FCARG


TIME -MOTION YES              TS_POINT


TIME -MOTION NO               TS_FCARG


MESH                          &DESCRIBE PIECE


&ENV -WDRIFT                  &DESCRIBE BODY -G_DRIFT


SECT                          &DESCRIBE BODY -SECT


LOCATION                      &DESCRIBE BODY -LOCATION


TDOM  -FACT_CONVEL            &DESCRIBE BODY -FACT_CONVEL     


TDOM  -NOCONV                 &DESCRIBE BODY -FACT_CONVEL  0. 


TDOM  -NO_DIRECT              &DESCRIBE BODY -NO_DIRECT      


TDOM  -USE_ZERO                                               


PRCOMP  -LGEN                  


PRCOMP or RAO -VDFACT         &DESCRIBE BODY -FM_TANAKA -FM_MORIS 


PRCOMP or RAO -VDFACT         &PARAMETER -FM_ROD    


PRCOMP or RAO -CD_TUBE        &PARAMETER -F_CD_TUBE

Obsolete Commands:



Obsolete                      Replacement



 



JLLEGS,JLLEGP,LWAYS,LWAYP,LOCJTRANS_CON


HULL                          PGEN


STAT                          PLANE


OSET,CIRC,RECT                PLANE, -RECT, -CIRC, -CART 


PART_MOVE                     &DESCRIBE PART -MOVE


&DESCRIBE -BODY               &DESCRIBE BODY


&DESCRIBE -PART               &DESCRIBE PART


&DESCRIBE -PROCESS            &DESCRIBE PROCESS


&DESCRIBE -LOAD_GROUP         &DESCRIBE LOAD_GROUP


&DESCRIBE -COMPARTMENT        &DESCRIBE COMPARTMENT


CMP_ATTR                      &DESCRIBE COMPARTMENT


&COMPARM -MINIMUM             &DESCRIBE COMPARTMENT -MINIMUM


&COMPARM -CONTENTS            &DESCRIBE COMPARTMENT -CONTENTS


&COMPARM -INT_PRES            &DESCRIBE COMPARTMENT -INT_PRES


&COMPARM -RESTORE             &INSTATE -PREVIONS             


BODIES                        &DESCRIBE SYSTEM


PA_ACTIVE                     &DESCRIBE ACTIVITY


BO_ACTIVE                     &DESCRIBE ACTIVITY


&NODE(  )                     &POINT( )         


&OFFSET( )                    &POINT(OFFSET )         


&EULERANG( )                  &POINT(EULERANG  )         


&D_NODE( )                    &POINT(D_NODE     )         


&HOOKLOC(  )                  &POINT(HOOKLOC    )         


&N_2NODE(  )                  &POINT(N_2NODE    )         


&EL_ENDS(  )                  &ELEMENT(EL_ENDS  )