Geometrical and Dimensional Uncertainties Effects Quantification for Stress/Strain Field Characterization

Abstract

In the automotive industry a significant part of research is dedicated to design and produce better products. The unavoidable presence of uncertainties in manufacturing processes and structural parameters leads to the necessity to develop stochastic models able to correctly represent physical behaviours. This paper presents numerical and experimental investigations on the effects of the fitting process on the stress/strain map of assembled automotive rim-disk systems. Different strategies of geometrical decomposition are used to condense large volumes of data by retaining essential information for the whole design process including not only manufacturing pre-stresses and operational loads but also dimensional tolerances of mating parts by means of a probabilistic model. Efficient approaches are used to identify the relative importance of input factors on the output uncertainties. A real test case of a wheel assembly is presented and discussed.