Analysis Sequence

<< Click to Display Table of Contents >>

Navigation:  Flexcom > Theory > Applied Loading > Coupled Analysis > Floating Body >

Analysis Sequence

Previous pageNext page

The following sequence of steps would be recommended for analysing a complex multi-component model for a particular combination of waves, current, wind and other user-defined loading:

Step 1: In an initial static analysis, apply buoyancy and gravity loads, any forces critical to the stability of the system (such as a riser top tension), zero constant displacements (such as at riser and mooring line seabed connections), while restraining the vessel CoG node in all translational degrees of freedom except  the vertical.

Step 2: In a static restart from Step 1, release all restraints at the vessel CoG node.

Step 3: Complete any offset and/or decay test analyses being performed to verify the system response in a restart or series of restarts from Step 2.

Step 4: Apply all remaining static loads such as those due to wind and current, again in a restart from Step 2. If the dynamic response to a random wave is to be simulated in a subsequent step, include the wave spectrum definition in the input data for this step; Flexcom will then apply the static mean drift force in this step.

Step 5: Simulate the dynamic response of the system to the wave environment in a dynamic restart from Step 4 in either the time or the frequency domain.

Some key points in the rationale for performing analyses in the above sequence of steps are as follows:

Step 1 allows the build up of tension in the various lines of the model (mooring lines, risers, TLP tendons if present, etc.), while enabling the vessel to find its vertical equilibrium position. Building up tensions in this way is generally mandatory for the types of model Flexcom is routinely used to analyse, because so many components of such systems depend largely on tension to resist applied loads.

Step 2 then allows the vessel to find its horizontal equilibrium position. In some cases it may be possible to combine Steps 1 and 2 in a single static step. On the other hand, for some sensitive systems it may be necessary to perform the Step 2 analysis quasi-statically.

A quasi-static Flexcom analysis is one in which a static configuration is obtained from a highly damped dynamic analysis. In such analyses, high mass and stiffness damping coefficients are specified, and these have the effect of dissipating structural dynamics resulting from the release of the CoG restraints. The analysis continues until the structure settles into a static configuration, at which point all loads and displacements are constant.

The Step 3 analysis or analyses, if performed, are mainly verification or validation exercises. However they do have the advantage that they increase the engineer’s understanding of the behaviour of the overall system, and offer the opportunity to correct, refine and/or improve the Flexcom as appropriate, prior to the later static and dynamic stages.

The Step 4 analysis allows all of the remaining static loads to be added to the model.

Preceding a frequency domain dynamic analysis by some or all of Steps 1-4 is mandatory (a frequency domain analysis in Flexcom must restart from a converged static solution).

The advantage of the multi-step procedure for time domain dynamic analysis is that starting a dynamic run from a stable static configuration generally minimises the influence of initial transients associated with the application of the dynamic loads, with a consequent saving in CPU effort. In addition, there are of course no transients generated by the addition of static loads, since these are already present at the start of the dynamic phase.

Finally, using the sequence of steps allows for the possibility of restarting multiple dynamic runs from the same static configuration; you might want to look at a range of wave periods, amplitudes, directions and even wave specification types (regular or random). There is an obvious saving in restarting all of these from the same static configuration, rather than repeating all static stages in every dynamic run.