Release notes for October, 1998 SeaSoft release New version numbers for all software: CALMsim(1.65); Catsim(1.75); Discsim(3.95); Moorsim(3.95); SALMsim(3.95); Semisim(3.95); Shipsim(3.95); Slowsim(1.45); SPMsim(3.95); Statmoor(4.75) +++ Note: One of the changes in this release necessitated a minor data file modification affecting only Shipsim, Semisim and Discsim. This change will require user attention for older data files that have "velocity" or "displacement" point RAO requests. When this is the case, a detailed alert message is issued on the first execution of these data files. A list of major modifications, enhancements and bug fixes follows: ============================================================================= >>> Trim & Heel reporting conventions, bug fixes and consistency updates: ============================================================================= Net vessel loads reported in MEANOUT, LOWOUT and SNAPOUT when nonzero trim and/or heel was specified were confusing, inconsistent, and in some instances outright incorrect. The original "Vessel Coordinate" force and moment resolution reported in MEANOUT, LOWOUT and SNAPOUT used a coordinate system with Z "earth-vertical" as contrasted with deck-vertical. In the presence of vessel trim or heel, the force and moment reports thus produced had limited engineering utility since they were not truly "vessel fixed" and did not directly reflect the associated structural loads (which are, of course, vessel-fixed). The convention has been changed to produce directly the load components needed for vessel-fixed structural evaluations. Specifically, vessel-relative forces and moments reported in MEANOUT, LOWOUT and SNAPOUT are now given in a true vessel-fixed frame. In the *absence* of trim or heel, the new and old conventions are identical and should produce identical loads. In addition to the poorly documented and unhelpful choice of coordinate system for the vessel load reports, there was a bug in the evaluation of moments when trim and heel were specified which produced incorrect results (that were subsequently reported in the incorrect coordinate system...). This has been corrected. Finally, for consistency of display, the forces and moments in MEANOUT are now reported simply as mooring forces and moments; previously they were reported as a mix of mooring and environmental forces and moments ("environmental" values are equal and of opposite sign to "mooring" values). Mean environmental forces are still available in LOWOUT, however. Forces, moments and motions reported in the "Global" coordinate system have not changed, even in the presence of nonzero trim or heel. (The Global system comprises an earth-vertical z axis and x-axis oriented towards global zero.) We would emphasize that disentangling trim and heel is a somewhat treacherous exercise. Loads are most naturally computed in a global, earth-vertical system. In the presence of crossed environmental conditions, a turret-moored vessel requires calculations to be done in several different coordinate systems (including baseline, mean and instantaneous conditions resolved in turret, vessel, or global systems). These various results need to be properly transformed for the most suitable output system (anchor loads in a global system; turret bearing loads in instantaneous vessel system, fairlead loads in instantaneous turret system, etc.). A mean trim or heel requires a substantial number of three-dimensional rotational transformations in the analysis. So, use the trim/heel feature at your pleasure, but be watchful for problems and remember, physically it is unlikely that a realistic amount of trim or heel will produce profoundly different results from a zero trim/heel simulation. In any event, it would be prudent to always check any "final" trim/heel simulations with the same simulation using zero trim and heel. +++ A Note on user-supplied RAOs when requesting trim or heel: In the presence of nonzero trim or heel, the simulation must be made aware of whether the user-supplied RAOs are in a vessel-relative system (with heave perpendicular to the deck) or in an earth-relative (heave perpendicular to the water surface). This information is passed using a toggle flag on the trim and heel definition page of the editor. ============================================================================= >>> Enhancements to several wave and swell options; these primarily affect Shipsim, Semisim, Discsim: ============================================================================= 1. The number of points available for vessel displacement, velocity or acceleration evaluation has been increased from 4 to 49 (same as the mooring line limit in Moorsim). 2. The method for inputting coordinates of the points in item (1) above has been changed so that a *single* array of points now services all three types of motion output (i.e., displacement, velocity or acceleration). The old method required input of a *different* array of points for each of these three motion types. +++ Note: This change necessitated a data file modification that may require user attention for earlier data files that have velocity or displacement point requests. If so, a warning is issued on the first execution of such data files. This problem only affects Shipsim, Semisim, Discsim. 3. The number of irregular wave directions that can be processed in a single run is now 36. The number of RAOs that can be obtained in a single run is huge; this could be in excess of one thousand in the case of a dense angular spreading grid tied to a large number of irregular wave requests. 4. When multiple irregular wave directions are requested, you can now "lock" the swell direction to the irregular wave direction so that the swell component has a fixed angle relative to each requested irregular wave direction (and therefore a different global swell direction for each irregular wave direction). The old method, which is still available, used a single global swell direction for all irregular wave directions. 5. The azimuthal wave energy spreading feature has been reworked: You can now specify the angular sector inside of which wave energy is contained (it used to be fixed at 180 degrees). You can get, in effect, any wave energy distribution from highly directional (waves confined to a 10 degree azimuthal sector, say) to azimuthally symmetric (equal wave energy incident from 360 degrees, as in the eye of a hurricane). Also, the azimuthal grid for evaluation of the spread energy can now have up to 36 points (increased from 12 in previous releases). This enhancement applies to all simulations, including SPMsim, Moorsim, CALMsim, Towsim and SALMsim. ============================================================================= >>> Wave frequency line load updates & improvements: ============================================================================= - Improved the computation of longitudinal elastic wave speed, which is important in the wave-frequency load calculation for long lengths of chain-type materials. - Fixed an infrequent numeric overflow condition that could prevent meaningful output of wave-frequency line loads in certain situations. - Made several minor improvements in the "small amplitude" w.f. line dynamics algorithm - The wave-frequency line load algorithm has been improved to better handle the inertial contribution; previously the inertial contribution was ignored in favor of the (normally much greater) square-law drag contribution. This change should improve individual line and net vessel wave-frequency load estimates arising from buoyant risers, in particular. See item 11 on editor page 20 of Moorsim/SPMsim. The new option may also produce a noticeable change in minimum line loads for all types of lines (as well as the maximum loads for riser-type lines mentioned above). Line added-mass on-line information notes & recommendations have been changed. - DYNOUT received some logical and cosmetic improvements, particular in the spread sea case. - Improved error reporting for the "FOBLQ" errors associated with the recently implemented large-amplitude wave-frequency line load algorithm. - Reworked the [Wave Height]/[Water Depth] warning for large-amplitude wave-frequency line load algorithm. - Installed an option in Moorsim/SPMsim to force the simulation to compute *only* wave-frequency line load "RAOs" about the user-specified pretension value for each line. No equilibrium testing is performed and no low-frequency dynamics or vessel-wide quantities are estimated when this flag is set. The only output is static catenary-elastic interpolation tables and wave-frequency line load RAOs (in DYNOUT). All other output is turned off and data irrelevant to the load RAOs are ignored. This is a convenience feature to facilitate study of wave-frequency behavior for many different types of line in one simulation pass. Requires the "line extension" excitation option (as opposed to the "regular wave" excitation option) to be in effect. (See item 13 on editor page 20 after selecting the "line extension" option on page 16.) ============================================================================= >>> Wave spreading updates & improvements: ============================================================================= - Fixed a bug which prevented correct calculation of RMS line loads in Moorsim/SPMsim when the "spectral estimate" method was used for "characteristic" load evaluations *and* angular wave spreading was selected. 1. Previous versions of Moorsim/SPMsim had an error in the recently implemented "RMS spectral load" calculation when wave spreading is present; correcting this error generally increases the reported RMS loads (it about doubled them in the cases we looked at), but they are still not as large the RMS values of the corresponding *unspread* case. 2. The corrected calculation appears free of algorithmic errors. We tested this by forcing all vessel RAOs to be frequency-independent and the line load RAOs for each wave approach angle to be the same (but still frequency-dependent), thereby mimicking a linear system (to eliminate the complication of the true nonlinear transfer function between fairlead motions and line loads). When we did this, the spread and unspread calculations gave the same value in spite of the fact that one represents a sum over many different wave headings and frequencies with different weights for different angles and the other a calculation using a single wave direction. Therefore, when real vessel and line load RAOs are used rather than the frequency-independent ones used for testing, the differences between spread and unspread "RMS spectral load" values evidently reflect the underlying nonlinearity of the line load model bumping into the linear procedure used in the "RMS spectral load" evaluation. Generally speaking, it is not hard to see why the use of the "RMS spectral load" flag gives smaller RMS loads in spread versus unspread seas; this is because, all other things being equal, the linearization means that the effective line load RAOs for the spread case are smaller than for the unspread case (since they use effective wave heights/fairlead motions associated with only a fraction of the total wave energy and since the nonlinearity generally produces more-than-proportionate load increases for a given increase in fairlead motion). (Whew! Got that?) You can see this by increasing the number of angular wedges for the spread sea case with the "RMS spectral load" flag set; the RMS values grow ever smaller as the number of wedges increases. This is simply an artifact of the linearization. It means that if you are going to use the "RMS spectral load" option with spreading, you may want to use the smallest justifiable number of wedges (4 or 6, perhaps). We really don't know what is best here. Case-specific judgement is probably in order. The "RMS spectral load" procedure has been problematic from the outset. Still, it does serve to protect against the one condition it was designed for: cases in which the RMS line load response at the single frequency (associated with the peak of the fairlead tangential velocity response spectrum) used for the "original" RMS load calculation is anomalously small. If the RMS value is important, and you must run spread seas, you should run the simulation both ways to reduce the possibility of understating the RMS values. We are hard up against a fundamental limitation of the frequency-domain approach here... ============================================================================= >>> Shipsim roll damping updates ============================================================================= - Fixed a long-standing issue in Shipsim concerning the handling of bilge keels and their effects on quadratic roll damping. In previous versions of Shipsim, quadratic roll damping for vessels with bilge keels was computed, regardless of the keel details (keel length, width, etc.), as if the vessel had perfectly square bilges (i.e., without bilge keels and with zero bilge radius). The logic behind this procedure, while not hopelessly flawed, led to the unintuitive (and seemingly unsatisfactory) conclusion that roll damping for a vessel with bilge keels was independent of bilge keel particulars. The reason this conclusion is not as bad as might first appear is that to permit drydocking, bilge keels normally do not project beyond the planes formed by extension of the bottom and side vessel surfaces; that is, the outboard edge of an "optimal" bilge keel lies just inside the (imaginary) location of the corner of a square (zero-radius) bilge. This prevents damage to the bilge keels during drydocking. As a result, the large-scale potential flow around the bilges during rolling (and the associated square-law damping associated with eddy-like viscous disruptions of this flow) is qualitatively not too different for a vessel with bilge keels and one with a perfectly rectangular underwater cross-section. This is the underlying basis for the original Shipsim treatment. We have improved the quadratic roll damping model to more realistically accommodate bilge particulars; in most cases these changes should have a relatively small effect on the predicted roll performance. In addition, we have corrected an error which produced incorrect damping results in situations where zero bilge keel width and/or length was specified in conjunction with a nonzero bilge radius. To reproduce the bilge keel roll damping model implemented in earlier simulations, use a bilge radius of 0.0 combined with "no bilge keel" as input data. ============================================================================= >>> Miscellaneous bug fixes & updates: ============================================================================= - Fixed a bug that prevented use of the full 37 angular grid points and 51 frequencies for user-specified RAOs and wave drift force coefficients. - Installed additional "short-period traps" in the internal wave drift force calculation that will permit use of a shorter minimum period without creating numeric overflow problems. However, problems can still arise when values in the wave period array get too small. The "short period" problem threshold is not uniform, even across different machines using the same compiler, because the method of computing floating point quantities in modern machines actually depends upon CPU architecture. As a general rule, minimum periods of 3 seconds or greater (for typical applications) will avoid problems. The symptoms of a too-small period are usually readily apparent: the simulation will fail to execute to completion, or bizarre output values will appear (such as NAN's, INF's or ****'s). The susceptibility of a particular simulation and a particular vessel/water depth combination can easily be explored by simply running the simulation for increasingly small periods to see where it begins to fail. - Added an option for wind speed spectral density output in Slowsim. - Added an option for wind and current "group" spectra output in Slowsim. - Fixed several bugs and some confusing and inconsistent terminology in the output descriptions of SNAPOUT, RANOUT and DYNOUT. - The global vessel positions associated with "snapshot" loads given in LOWOUT and SNAPOUT are now output. It should be noted that there is no unique "n-sigma" snapshot position; rather the position given is associated with an "n-sigma" surge and equilibrium yaw (i.e., vessel at its equilibrium heading). - The sway-yaw mode damping in SPMsim in certain crossed environment cases has been improved; damping in aligned conditions should be unaffected by these modifications. - The vessel load output in SNAPOUT and DYNOUT suffered from an incorrect rotational transformation that could cause net vessel load errors to be reported in strongly crossed conditions. This has been fixed. - A number of output problems associated with spread sea conditions in DYNOUT have been fixed. - Mean mooring moments are now displayed in MEANOUT, about the user-specified moment datum. - Clarification of several output items' descriptions in Shipsim. - Added additional interpolation range warnings to Catsim. (Under some conditions Catsim could complete execution, using interpolation values in the problematic range between the first two interpolation table rows, without issuing a suitable warning. This has been fixed.) - Fixed a bug in the "mooring feedback" option for Moorsim/Discsim that prevented proper accounting of mooring system damping when the feedback option was in effect. - Installed traps to prevent "environment on, zero value" conditions from causing abnormal simulation termination (i.e., crashes). (Example: Wind flag "on" with zero wind speed specified) - Adjust tolerances for selected warning outputs in nonlinear wave-frequency load algorithm to eliminate occurrences of some unnecessary warnings. - Numerous small cosmetic and logical changes. ============================================================================= Ends release notes of October, 1998 =============================================================================