Inversion of structural dynamics simulations: State-of-the-art and orientations of the research

This publication offers an overview of the technology available for formulating inverse problems and correlating measured responses with simulations from finite element analysis. The application targeted is clearly structural dynamics although most of the techniques discussed here originate or find their counterparts in physics and other engineering fields. After reviewing the state-of-the-art practices in model updating where a mostly linear model is optimized to satisfy a series of modal criteria of correlation, an assessment of the advantages and limitations of this technology is offered. Current orientations of inverse problem solving are discussed, including the need to propagate variability through the simulation; the generation of fast running models; the adequate choice of data metrics for nonlinear dynamics; sampling strategies for the optimization; and hypothesis testing in the context of multivariate data analysis.