Modelling of Weld and Adhesive Connections in Crashworthiness Applications with LS-DYNA
The increasing demands with regard to the predictive capabilities and the exactness of crash simulations require more and more investigations into numerical models in order to capture the physical behaviour reliably. Steps towards this goal are the usage of finer meshes which allow for a better geometrical representation and more sophisticated material models which allow better prediction of failure scenarios. Another important playground towards improved crash models is the area of connection modelling. Validation in this area is usually closely related to very detailed models which cannot be easily translated into a crash environment due to time step restrictions. Therefore, representative substitute models have to be developed and foremost validated. The aspect that failure of the connections has to be considered as well adds another dimension to the complexity of the task. The present paper highlights the conflict between predictive capability, capture of physical reality and manageable numerical handling. Another aspect of the paper is the attempt to raise the awareness of the topics verification and validation of numerical models in general. This concept is illustrated using latest developments for modelling of spotwelds and adhesive bonding in LS-DYNA.
https://www.dynamore.de/de/download/papers/sonstiges/nafems-joints/view
https://www.dynamore.de/@@site-logo/DYNAmore_Logo_Ansys.svg
Modelling of Weld and Adhesive Connections in Crashworthiness Applications with LS-DYNA
The increasing demands with regard to the predictive capabilities and the exactness of crash simulations require more and more investigations into numerical models in order to capture the physical behaviour reliably. Steps towards this goal are the usage of finer meshes which allow for a better geometrical representation and more sophisticated material models which allow better prediction of failure scenarios. Another important playground towards improved crash models is the area of connection modelling. Validation in this area is usually closely related to very detailed models which cannot be easily translated into a crash environment due to time step restrictions. Therefore, representative substitute models have to be developed and foremost validated. The aspect that failure of the connections has to be considered as well adds another dimension to the complexity of the task. The present paper highlights the conflict between predictive capability, capture of physical reality and manageable numerical handling. Another aspect of the paper is the attempt to raise the awareness of the topics verification and validation of numerical models in general. This concept is illustrated using latest developments for modelling of spotwelds and adhesive bonding in LS-DYNA.