Traditional inverse modelling frameworks have been used to identify material parameters for full scale finite element simulations such as a vehicular impact, however the integrity of such methods has not often come under scrutiny. Discrepancies may arise between experimental results and finite element simulations, however most research has concluded that the weakness of a material model causes the discrepancy of results. In this paper, the authors have identified a number of other factors that could also potentially cause discrepancies between experimental and numerical data sets, and have presented a novel inverse modelling framework that will resolve some of the alternative issues that cause the disagreements. To improve the inverse modelling procedure further, it would be ideal to model the exact geometry of a specimen (that is tested experimentally) in order to adequately represent the geometrical imperfections that may alter the results in a FE simulation. The model reconstruction process is based upon Digital Close Range Photogrammetry (DCRP) principles to create an accurate CAD representation of an experimental specimen. In turn, the CAD model of the specimen can be meshed for FE analysis thus taking account of any significant geometrical variations. An error analysis presented in this paper demonstrates that the accuracy of an FE model constructed with the DCRP method is an improvement over using an average model geometry.
Materials by design for sustainable solutions in the energy, medical, transport, and engineering sectors
