Constructing a Statistical Model of Displacement Magnifying Structure Based on Modeling Validation

Abstract

Numerical simulation results for the optimum design are often different from the experimental results because of simulation errors resulting from the idealization and modeling errors. Less accurate simulation will make the optimization lower confidence. On the natural frequency test of displacement magnifying structure (DMS) which we have obtained by the optimum design, the experimental result had a large gap from the simulation results. In this study, the statistical model is constructed to quantify the gap that is considered as uncertainties. Full factorial numerical integration method that is one of the techniques of uncertainty propagation is adopted to construct the statistical model. The statistical model is used to quantify the effect of the uncertainty on the structural performance such as the reliability concerning the natural frequency of DMS. That differs from the deterministic model parameter estimation method such as finite element updating, that is difficult to quantify the effect of the uncertainties. The proposed method is adopted to construct the statistical model of DMS.