Active-learning-driven surrogate modeling for efficient simulation of parametric nonlinear systems
From MaRDI portal
Abstract: When repeated evaluations for varying parameter configurations of a high-fidelity physical model are required, surrogate modeling techniques based on model order reduction are desired. In absence of the governing equations describing the dynamics, we need to construct the parametric reduced-order surrogate model in a non-intrusive fashion. In this setting, the usual residual-based error estimate for optimal parameter sampling associated with the reduced basis method is not directly available. Our work provides a non-intrusive optimality criterion to efficiently populate the parameter snapshots, thereby, enabling us to effectively construct a parametric surrogate model. We consider separate parameter-specific proper orthogonal decomposition (POD) subspaces and propose an active-learning-driven surrogate model using kernel-based shallow neural networks, abbreviated as ActLearn-POD-KSNN surrogate model. To demonstrate the validity of our proposed ideas, we present numerical experiments using two physical models, namely Burgers' equation and shallow water equations. Both the models have mixed -- convective and diffusive -- effects within their respective parameter domains, with each of them dominating in certain regions. The proposed ActLearn-POD-KSNN surrogate model efficiently predicts the solution at new parameter locations, even for a setting with multiple interacting shock profiles.
Recommendations
- Non intrusive reduced order modeling of parametrized PDEs by kernel POD and neural networks
- Non-intrusive double-greedy parametric model reduction by interpolation of frequency-domain rational surrogates
- A New Certified Hierarchical and Adaptive RB-ML-ROM Surrogate Model for Parametrized PDEs
- Nonintrusive Reduced-Order Models for Parametric Partial Differential Equations via Data-Driven Operator Inference
- Projection-based model reduction: formulations for physics-based machine learning
Cites work
- A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs
- A greedy non-intrusive reduced order model for shallow water equations
- A multistep technique with implicit difference schemes for calculating two- or three-dimensional cavity flows
- A non-intrusive multifidelity method for the reduced order modeling of nonlinear problems
- A parameterized non-intrusive reduced order model and error analysis for general time-dependent nonlinear partial differential equations and its applications
- An adaptive sampling approach for the reduced basis method
- Automated solution of differential equations by the finite element method. The FEniCS book
- Data-driven reduced order modeling for time-dependent problems
- Dynamic mode decomposition of numerical and experimental data
- Error estimation of the parametric non-intrusive reduced order model using machine learning
- LAPACK Users' Guide
- Linearly recurrent autoencoder networks for learning dynamics
- Machine-learning error models for approximate solutions to parameterized systems of nonlinear equations
- Model reduction and neural networks for parametric PDEs
- Model reduction for transport-dominated problems via online adaptive bases and adaptive sampling
- Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders
- Moment approximations and model cascades for shallow flow
- Non-intrusive reduced order modeling of nonlinear problems using neural networks
- Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem
- Online adaptive model reduction for nonlinear systems via low-rank updates
- Parametric model order reduction using pyMOR
- Physics-informed machine learning for reduced-order modeling of nonlinear problems
- PyMOR -- generic algorithms and interfaces for model order reduction
- Rank-Sparsity Incoherence for Matrix Decomposition
- Reduced order modeling for nonlinear structural analysis using Gaussian process regression
- Robust principal component analysis?
- Spatial interpolation of high-frequency monitoring data
- Spectral analysis of nonlinear flows
- The shifted proper orthogonal decomposition: a mode decomposition for multiple transport phenomena
Cited in
(9)- Directional filter combined with active learning for rare failure events
- Physics-based active learning for design space exploration and surrogate construction for multiparametric optimization
- The novel graph transformer-based surrogate model for learning physical systems
- Maximum a posteriori estimation for linear structural dynamics models using Bayesian optimization with rational polynomial chaos expansions
- An adaptive learning strategy for surrogate modeling of high-dimensional functions - application to unsteady hypersonic flows in chemical nonequilibrium
- Model reduction for advection dominated hyperbolic problems in an ALE framework: offline, online phases and error estimator
- Modeling partially observed nonlinear dynamical systems and efficient data assimilation via discrete-time conditional Gaussian Koopman network
- Active learning informed proper orthogonal decomposition for reduced order modelling of heat transfer in porous medium
- Online multi-fidelity data aggregation via hierarchical neural network
This page was built for publication: Active-learning-driven surrogate modeling for efficient simulation of parametric nonlinear systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6185211)
