This presentation was made at the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

The successful design of a vehicle and its optimisation with regards to collision and impact can be a time-consuming iterative process, with costs in time and resources proportional to the optimisation level targeted and the number of iterations performed. Replacing iterative testing by numerical simulation allows to achieve a significantly higher number of iterations within a shorter time frame and for a reduced cost; given that reliable models and input data are used. Unfortunately, model construction, and specifically the establishment of input parameters, can make the numerical simulation solution ill-suited for a vehicle development process squeezed in short time frames due to industry standards and reality. This paper presents a methodology with successful applications where the implementation of the numerical simulation process has been accelerated based directly on experimental results for the enhancement of a snowmobile design submitted to frontal impacts. The models initially showed good agreement with heritage test results; however, the use of experimental results to set parameters that would otherwise suffer from large uncertainties allowed to enhance the precision of the predictions. Moreover, that approach shows to be necessary due to short schedule brought by a competitive design environment. The numerical model has been first used to explore the coherence between the scenario toward which the design enhancement is directed and the test protocol, meant to reproduce the scenario within an experimental set-up. The numerical model has then been used to represent design iterations of the test protocol to remain in-line with a strong test heritage and to allow experimental testing to be used to enhance the model and its predictions. Slight tuning of material properties has been made to compensate for discrepancies between material mechanical test conditions and vehicle test conditions. Finally, global action of all dissipative mechanisms has been lumped into a single mass weighted damping term. That approach, based on a single iteration, has been shown to enhance the agreement of successive iterations of numerical simulation and physical testing. Moreover, the lumped parameter value correlated on a first design showed to be applicable to other vehicles. That approach is the same that has been used for decades in vibration analysis of metallic structures for space vehicle launch environment or in vibration analysis of buildings for strength assessment within a seismic environment.