Manifold Alignment-Based Multi-Fidelity Reduced-Order Modeling Applied to Structural Analysis
This addresses the challenge of integrating multi-fidelity data in structural analysis for engineers, though it appears incremental as it applies an existing method to new scenarios.
The paper tackles the problem of combining inconsistent field outputs from high- and low-fidelity structural simulations by using manifold alignment to project them onto a common subspace, resulting in a multi-fidelity reduced-order model that achieves higher predictive accuracy at lower computational cost compared to single-fidelity models.
This work presents the application of a recently developed parametric, non-intrusive, and multi-fidelity reduced-order modeling method on high-dimensional displacement and stress fields arising from the structural analysis of geometries that differ in the size of discretization and structural topology.The proposed approach leverages manifold alignment to fuse inconsistent field outputs from high- and low-fidelity simulations by individually projecting their solution onto a common subspace. The effectiveness of the method is demonstrated on two multi-fidelity scenarios involving the structural analysis of a benchmark wing geometry. Results show that outputs from structural simulations using incompatible grids, or related yet different topologies, are easily combined into a single predictive model, thus eliminating the need for additional pre-processing of the data. The new multi-fidelity reduced-order model achieves a relatively higher predictive accuracy at a lower computational cost when compared to a single-fidelity model.