Abstract:A slow decaying Kolmogorov n-width of the solution manifold of a parametric partial differential equation precludes the realization of efficient linear projection-based reduced-order models. This is due to the high dimensionality of the reduced space needed to approximate with sufficient accuracy the solution manifold. To solve this problem, neural networks, in the form of different architectures, have been employed to build accurate nonlinear regressions of the solution manifolds. However, the majority of the implementations are non-intrusive black-box surrogate models, and only a part of them perform dimension reduction from the number of degrees of freedom of the discretized parametric models to a latent dimension. We present a new intrusive and explicable methodology for reduced-order modelling that employs neural networks for solution manifold approximation but that does not discard the physical and numerical models underneath in the predictive/online stage. We will focus on autoencoders used to compress further the dimensionality of linear approximants of solution manifolds, achieving in the end a nonlinear dimension reduction. After having obtained an accurate nonlinear approximant, we seek for the solutions on the latent manifold with the residual-based nonlinear least-squares Petrov-Galerkin method, opportunely hyper-reduced in order to be independent from the number of degrees of freedom. New adaptive hyper-reduction strategies are developed along with the employment of local nonlinear approximants. We test our methodology on two nonlinear time-dependent parametric benchmarks involving a supersonic flow past a NACA airfoil with changing Mach number and an incompressible turbulent flow around the Ahmed body with changing slant angle.

Shape-morphing reduced-order models for nonlinear Schrodinger equations

Simulation of nonlinear thermomechanical waves with an empirical low dimensional model

Fast and scalable computation of shape-morphing nonlinear solutions with application to evolutional neural networks

Modeling Wave Packet Dynamics and Exploring Applications: A Comprehensive Guide to the Nonlinear Schrödinger Equation

Phase Space Models for Stochastic Nonlinear Parabolic Waves: Wave Spread and Singularity

Structure-preserving Reduced Order Modeling of non-traditional Shallow Water Equation

Fast Numerical Approximation of Linear, Second-Order Hyperbolic Problems Using Model Order Reduction and the Laplace Transform

Dynamics of rogue waves on a multi-soliton background in a vector nonlinear Schrodinger equation

Contraction of variational principle and optical soliton solutions for two models of nonlinear Schrödinger equation with polynomial law nonlinearity

Explicable hyper-reduced order models on nonlinearly approximated solution manifolds of compressible and incompressible Navier-Stokes equations

Learning physics-based reduced-order models from data using nonlinear manifolds

Investigation of Analytical Soliton Solutions to the Non-Linear Klein–Gordon Model Using Efficient Techniques

An effective semilocal model for wave turbulence in 2D nonlinear optics

Linear High-Order Energy-Preserving Schemes for the Nonlinear Schrödinger Equation with Wave Operator Using the Scalar Auxiliary Variable Approach

Projection-Based Reduced Order Model for Simulations of Nonlinear Flows with Multiple Moving Objects

Preserving Lagrangian structure in data-driven reduced-order modeling of large-scale dynamical systems

The solutions of classical and nonlocal nonlinear Schrödinger equations with nonzero backgrounds: Bilinearisation and reduction approach

Solitons, rogue waves and interaction behaviors of a third-order nonlinear Schrödinger equation

Dynamics of Rogue Waves on a Multisoliton Background in a Vector Nonlinear Schrödinger Equation

Novel wave solutions to a generalized third-order nonlinear Schrödinger’s equation

Reduced Order Model Based Nonlinear Waveform Inversion for the 1D Helmholtz Equation