Transient analysis
In: Computation concepts-Non-linear FE solution
The non-existence of a reliable tool allowing conversion of elastic FE-calculation to close to real elastic-plastic values similarly to the Neuber concept but under multiaxial loading, forced PragTic to turn to another way, if the low-cycle fatigue should be evaluated.
The concept of transient analysis allows a work with the local load states history at each examined point on the FE-model. This local load history is not retrieved by superposition of different load modes (channels) in one load regime. The actual external loads act directly at the FE-model and are reflected by the non-linear FE-solution to local points. This means that one local load history cannot be superposed with any other one, since the relation between external force and the actual local loads is not linear (elastic) any more.
PragTic thus only reads the listed local load states. Still, there are different ways of getting the local load histories into PragTic. Currently, PragTic supports two of them:
* Reading of complete local load histories at individual points/nodes/elements. This option is currently checked with output files prepared in ABAQUS (see this link how to prepare it). The import uses the File Recognition dialogue. There are nevertheless problems with formatting, that ABAQUS produces. Since it writes all the data for the whole output set to one line for each time instant, the successful import is very dependent on complying of format structure. ABAQUS nevertheless sometimes misses one blank space. This error can collapse the import - above all if results at more nodes/elements are read at once (say e.g. more than 20 nodes, but test it by yourself).
* Reading of separate result files with stress and strain tensors for each time instant and assembling the transient analysis in the Load Regimes window.
Note: I recommend the latter option as more clear and less error-prone. For load histories with more time instants it nevertheless means a quite lengthy and tedious import phase. I therefore expect to prepare later a new version, where the load regime would be assembled from files and not from imported data_vectors as it is today. The data_vectors would be imported after the load regime definition would be completed.
Another option is to prepare the import of the complete non-linear FE-analysis results in some recognized transfer format (e.g. ABAQUS *.fil). This nevertheless includes quite complicated and lengthy programming so unless I find some sponsor for it I do not start to prepare it.
More:
effect of localized plasticity
© PragTic, 2007
This help file has been generated by the freeware version of HelpNDoc