State estimation with limited sensors -- a deep learning based approach

DOI10.1016/J.JCP.2022.111081MaRDI QIDQ2135833zbMATH OpenOpenAlexFDO

Authors Yash Kumar, Pranav Bahl, Souvik Chakraborty

Publication date 9 May 2022

Published in Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2101.11513

zbMATH Keywords

autoencoderslimited sensorsLSTMstate estimation

Mathematics Subject Classification ID

Artificial intelligence (68Txx) Incompressible viscous fluids (76Dxx) Turbulence (76Fxx)

Recommendations

• Leveraging reduced-order models for state estimation using deep learning
• Shallow neural networks for fluid flow reconstruction with limited sensors
• Nonlinear reduced DNN models for state estimation
• Unsteady flow prediction from sparse measurements by compressed sensing reduced order modeling
• Flow state estimation in the presence of discretization errors

Cites work

• A heuristic approach to effective sensor placement for modeling of a cylinder wake
• A non-linear observer for unsteady three-dimensional flows
• Bayesian filtering and smoothing
• Conditional eddies in isotropic turbulence
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• Leveraging reduced-order models for state estimation using deep learning
• Linearly recurrent autoencoder networks for learning dynamics
• Machine learning. A probabilistic perspective
• Matching pursuits with time-frequency dictionaries
• Neural network modeling for near wall turbulent flow.
• Pattern recognition and machine learning.
• Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networks
• Reconstruction of unsteady viscous flows using data assimilation schemes
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit
• Stochastic estimation and flow sources associated with surface pressure events in a turbulent boundary layer
• Stochastic stability of the discrete-time extended Kalman filter
• The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions
• Unsteady flow sensing and estimation via the gappy proper orthogonal decomposition

Cited in

(3)

• Leveraging reduced-order models for state estimation using deep learning
• Shallow neural networks for fluid flow reconstruction with limited sensors
• Performance triggered adaptive model reduction for soil moisture estimation in precision irrigation

Describes a project that uses

Uses Software

• PRMLT
• PMTK
• OpenFOAM
• PyTorch
• Adam

This page was built for publication: State estimation with limited sensors -- a deep learning based approach

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2135833)