Flexcom 2025.1.2 |
 
|
Flexcom 2025.1.2 |
|
The following faults are present in Flexcom 2025.1.2. More detailed information on each fault is provided in the following sections.
No |
Issue |
Severity |
1 |
Hoop stress is incorrectly computed where the internal diameter of an element is zero |
Minor |
2 |
Slug display may not always function if a model contains several overlapping slugs |
Minor |
3 |
Variable time step selected by Flexcom lower than expected |
Minor |
4 |
LifeTime Mode 2 (cycle counting) not running for keyxm & keyxi files |
Minor |
5 |
LifeTime Mode 1 (fatigue analysis) presents the default stress histogram for a non-critical hotspot location |
Minor |
6 |
LifeTime Mode 1 (fatigue analysis) presents stress histograms at one angular location on the local cross-section only |
Minor |
7 |
LifeTime plot numbers not sequential in terms of formatting |
Minor |
8 |
Aerodynamic forces on wind turbine blade applied at structural centre rather than aerodynamic centre |
Major |
•Related Topics: Postprocessing
•Description: Hoop stress is incorrectly computed where the internal diameter of an element is zero. This is unlikely to occur in practice, unless you are modelling a solid bar for example.
•Workaround: You could specify a small, but non-zero value for internal diameter under *PROPERTIES. Or you could compute hoop stress manually using the internal and external pressure outputs which are readily available from Flexcom.
•Related Topics: Visualising Slug Flow
•Description: Slug display may be unreliable if a model contains several overlapping slugs. You may encounter the warning message "Problems occurred during slug display. Slugs will not be displayed after Time X", beyond which point no slugs are displayed in the structural animation. Note that the slugs are still accommodated correctly in the numerical model, so it is purely a display issue. You may use *PRINT->OUTPUT=SLUG FLOW to verify that each element has the correct internal contents.
•Workaround: Use *PRINT->OUTPUT=SLUG FLOW to verify that each element has the correct internal contents.
•Related Topics: Variable Time Stepping
•Description: The automatic time stepping algorithm fails to increase the solution time step in a restart analysis if the simulation time is lower than the end time of the preceding simulation. In this case, the program incorrectly believes that the current simulation is within an active ramp period and therefore prevents any time stepping increase, even if the solution is repeatedly achieving numerical convergence on the first iteration. This behaviour dates back to early versions of Flexcom where the start and end times of consecutive restart simulations had to correspond exactly. Nowadays all simulations tend to start from time t=0s, regardless of any preceding stages.
•Workaround: There are several possible workarounds.
1.Increase the start time of the restart analysis to be equal to the end time of the preceding stage.
2.Ignore the issue and wait for the simulation to complete normally.
3.Increase the minimum time step size, forcing the simulation to proceed more quickly.
4.Insert a token restart analysis into the chain, with a start and end time of 0s and 0.1s respectively. This effectively resets the end time of the preceding simulation to a very low value.
•Related Topics: LifeTime Cycle Counting (Mode 2)
•Description: LifeTime cycle counting does not function for keyword files which have in-build metric (keyxm) or imperial (keyxi) units. The LifeTime analysis terminates prematurely with a spurious error message, "The extension of this keyword file (.keyxm) must match that of the keyword file(s) that generated the seastate files (.keyx)". This occurs even if the Flexcom seastate simulations have the same keyword file extension as the LifeTime input file.
•Workaround: Rename all the input files with a .keyx file extension and re-run the simulations. Contact Wood if you require any assistance converting keyxm/keyxi files to keyx format.
•Related Topics: LifeTime Fatigue Analysis (Mode 1)
•Description: When you run a fatigue analysis, the main output file (*.out) presents you with a stress histogram for the hot spot with the lowest fatigue life (and the location on the cross-section at which this minimum occurs). But occasionally LifeTime can present the stress histogram at a hotspot which does not correspond to the critical location. This fault relates to histogram output only and does not affect the main fatigue computations.
•Workaround: Manually request a stress histogram at the critical location via *HISTOGRAM DATA.
•Related Topics: Stress Histograms
•Description: LifeTime allows you to request stress histograms at any location of interest, by specifying a hotspot and an angular location on the cross-section via an integer input (1 = 0°, 2 = 45°, 3 = 90° etc.). However it is not possible to request a histogram for more than one angular location at any given hotspot, as LifeTime incorrectly issues an error message regarding request duplication.
•Workaround: Run a series of separate fatigue analyses, requesting a stress histogram at a different location each time under *HISTOGRAM DATA.
•Related Topics: LifeTime Fatigue Analysis (Mode 1)
•Description: All plots generated by LifeTime are numbered consecutively but the number of digits is not sufficient. This causes a minor inconvenience in the sense that the plots do not appear in alphabetical order in the File View or Windows Explorer. For example, the plots are named FileName.S1.mplt, ..., FileName.S9.mplt, FileName.S10.mplt. They should be named FileName.S01.mplt, ..., FileName.S09.mplt, FileName.S10.mplt.
•Workaround: Rename the plot files manually if you would like them to appear in alphabetical order.
•Related Topics: Wind Turbine Modelling
•Description: The application of aerodynamic forces and moments in the flexible blade model are not taking into account the spatial offset between the structural centre and the aerodynamic centre on the blade cross section. A similar limitation applies to the application of mass terms at the centre of gravity, although this is likely to be less significant. So, although the blade structural properties are correctly represented by the relevant stiffness terms, the twisting moment induced by the spatial offset is neglected. This could lead to some inaccuracy in deformational blade twist induced by the wind loads, particularly near rated wind speed where the forces are largest. Notwithstanding this limitation, the structural twist due the physical shape of the blade, and the rigid body twist imparted by the control system, are both correct. As these terms account for the majority of the overall twist angle, the local angle of attack and the corresponding aerodynamic forces should be quite close to their theoretical values.
•Workaround: There is no workaround, so you are advised to upgrade to Flexcom 2025.1.3 or later.