Simplified Finite Element model of the Solution Space for Frontal Crash Analysis for Uncertainty Propagation

Passenger safety is a key aspect during the development of a new vehicle. How- ever, designing for safety is a very complicated task that results in high costs. The extensive use of finite element (FE) simulations helps accelerate develop- ment and contain the overall costs. Nevertheless, recent developments in data science can further help obtain the desired information at a fraction of the cur- rent costs. In this thesis, I propose a method to test a component in the early stage of development. This method considers the effects of development uncertainties to study how they affect the behavior of a component. For this purpose, I propose a simplified FE model that couples 1D elements with the full mesh of a component to be studied. Surrogate models are then used to predict the behavior of the other components. The solution space method is used in combination with the surrogate models to characterize the performance of the 1D element under the development uncertainties. To propagate the uncertainty, I finally run multiple samples. This confirms the capability of the method I propose to capture how a component is affected by the variations of the design parameters of the other components. This information provides an indication of the robustness of the design of the studied component. My method offers a tool for engineers to quickly test early-stage designs of a component, identifying which could work in the full model. This strategy reduces the cost of the overall development.