A two stages Deep Learning Architecture for Model Reduction of Parametric Time-Dependent Problems
This work addresses computational efficiency for parametric time-dependent problems in fields like fluid dynamics, though it appears incremental as it builds on existing model reduction techniques with a novel architectural twist.
The authors tackled the challenge of generalizing solutions for parametric time-dependent systems with limited computational resources by proposing a two-stage deep learning framework, achieving a 97% reduction in computational time for incompressible Navier-Stokes equations compared to numerical resolution.
Parametric time-dependent systems are of a crucial importance in modeling real phenomena, often characterized by non-linear behaviors too. Those solutions are typically difficult to generalize in a sufficiently wide parameter space while counting on limited computational resources available. As such, we present a general two-stages deep learning framework able to perform that generalization with low computational effort in time. It consists in a separated training of two pipe-lined predictive models. At first, a certain number of independent neural networks are trained with data-sets taken from different subsets of the parameter space. Successively, a second predictive model is specialized to properly combine the first-stage guesses and compute the right predictions. Promising results are obtained applying the framework to incompressible Navier-Stokes equations in a cavity (Rayleigh-Bernard cavity), obtaining a 97% reduction in the computational time comparing with its numerical resolution for a new value of the Grashof number.