A Validation Approach for the Multiscale Modeling of Additively Manufactured Lattice Structures

Abstract

A validation approach is proposed to ensure that the computational models corresponding to each scale level in a hierarchical multiscale modeling process can produce accurate simulations in the material characterizations at multiple levels. In the validation approach, the stochastic upscaling method is utilized for the homogenization of the geometry and material properties of material structure at each scale level of interest. The experimental data is utilized for the validation of the predictions made by the multiscale models with homogenized properties at each scale level. For the quantitative assessment of the model’s prediction accuracy, a validation metric that integrates the u-pooling method with the Kolmogorov-Smirnov test will be presented. Additively manufactured cellular structures are used as application examples to demonstrate the efficacy of the proposed multiscale validation approach. The results from the examples show that the proposed method effectively captures the predictive capability of the models with an unbiased decision criterion based on the test statistic at multiple scales.