When reconstructing the stress state from a known strain field, the number of frames used significantly affects the computation time of implicit IRMA algorithm. Therefore, down-sampling is necessary to obtain results within a reasonable time-scale. In contrast, the NICE algorithm does not require this, as its computation time is almost independent of the number of frames while maintaining accuracy. In the thesis, We demonstrate this by using different sample loading conditions. In addition to this comparison, we also show that equal plastic strains during different loading paths cause differences in the final stresses despite strains being identical.
|
