Quantification of interval field uncertainty in dynamic Finite Element models

In the context of integrating uncertainty and variability in Finite Element (FE) models, several advanced techniques for taking both inter-and intra-variability (variability within one part) into account have been introduced in literature. In the framework of non-probabilistic variability modelling, especially the theory of interval fields (IF) has proven to show promising results. Application of the IF concept to real life problems however requires identification of these parameters. Recently, the authors proposed a generic methodology to identify IF parameters based on measurement data. However, the methodology suffers greatly form the curse of dimensionality. This paper therefore introduces a methodology to reduce this dimensionality, while still maintaining the information that is necessary for the identification of the interval field. The proposed method is validated on a dynamic model of a cantilever beam, containing uncertainty. It is shown that an accurate identification is feasible following this technique at a reduced computational cost.