Machine learning-based surrogate modeling approaches for fixed-wing store separation

Development of Pod-Based Reduced Order Models Applied to Shallow Water Equations Using Augmented Riemann Solvers

“…Other works on the combination of SWE and ROMs are focused on reducing the computational complexity by using DEIM with implicit numerical schemes [43], parametric sensitivity analysis [52,53] or meshless radial basis functions technique [47]. It is also worth mentioning interesting recent works investigating the combination of neural networks with POD methods applied to Navier-Stokes equations [54,55] or even deep-learning techniques [56] and Physics-informed neural networks [57]. However, in this work the time averaging approach in combination with the proper interval decomposition (PID, see [58]), which has been reported as a useful tool [28,33], is used.…”