no code implementations • 14 Jun 2022 • Christian Perron, Darshan Sarojini, Dushhyanth Rajaram, Jason Corman, Dimitri Mavris
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.
2 code implementations • 8 Aug 2020 • Raphael Gautier, Piyush Pandita, Sayan Ghosh, Dimitri Mavris
The comparison shows that the proposed method improves the active subspace recovery and predictive accuracy, in both the deterministic and probabilistic sense, when only few model observations are available for training, at the cost of increased training time.
no code implementations • 29 Dec 2017 • Yao Zhang, Woong-Je Sung, Dimitri Mavris
The adaptability of the convolutional neural network (CNN) technique for aerodynamic meta-modeling tasks is probed in this work.