Nonintrusive Reduced Order Modelling of Convective Boussinesq Flows

In this paper, we formulate three nonintrusive methods and systematically explore their performance in terms of the ability to reconstruct quantities interest predictive capabilities. The include deterministic dynamic mode decomposition (DMD), randomized DMD nonlinear proper orthogonal (NLPOD). We apply these a convection dominated fluid flow problem governed by Boussinesq equations. analyze reconstruction results primarily at two different times for considering noise levels synthetically added into data snapshots. Overall, our indicate that, with selection number retained modes neural network architectures, all approaches make predictions that are good agreement full order model solution. However, find NLPOD approach seems more robust higher compared both approaches.