Physics-Informed Deep Neural Networks for Transient Electromagnetic Analysis (2020-08-04T00:00:00.000000Z)

TL;DR

A recurrent neural network based model, which learns representations of long-term spatial-temporal dependencies in the sequence of its input data, and an encoder-recurrent-decoder architecture, which is trained with finite difference time domain simulations of plane wave scattering from distributed, perfect electric conducting objects.

Abstract

In this paper, we propose a deep neural network based model to predict the time evolution of field values in transient electrodynamics. The key component of our model is a recurrent neural network, which learns representations of long-term spatial-temporal dependencies in the sequence of its input data. We develop an encoder-recurrent-decoder architecture, which is trained with finite difference time domain simulations of plane wave scattering from distributed, perfect electric conducting objects. We demonstrate that, the trained network can emulate a transient electrodynamics problem with more than 17 times speed-up in simulation time compared to traditional finite difference time domain solvers.

Authors

Oameed Noakoasteen

1 papers

Shu Wang

1 papers

Z. Peng

1 papers

TL;DR

Abstract

Authors

References43 items

Artificial Neural Networks and Machine Learning techniques applied to Ground Penetrating Radar: A review

A Machine-Learning-Based Epistemic Modeling Framework for Textile Antenna Design

Gaussian Processes for Direction-of-Arrival Estimation With Random Arrays

Wave physics as an analog recurrent neural network

Deep-Learning Schemes for Full-Wave Nonlinear Inverse Scattering Problems

Compressed Convolutional LSTM: An Efficient Deep Learning framework to Model High Fidelity 3D Turbulence

Direction-of-Arrival Estimation Based on Deep Neural Networks With Robustness to Array Imperfections

DeepNIS: Deep Neural Network for Nonlinear Electromagnetic Inverse Scattering

Inversion of Rough Surface Parameters From SAR Images Using Simulation-Trained Convolutional Neural Networks

Deep Learning for Super-Resolution Channel Estimation and DOA Estimation Based Massive MIMO System

Deep Learning Image Reconstruction Simulation for Electromagnetic Tomography

Deep D-Bar: Real-Time Electrical Impedance Tomography Imaging With Deep Neural Networks

Lat-Net: Compressing Lattice Boltzmann Flow Simulations using Deep Neural Networks

Single-Snapshot DoA Estimation in Array Antennas With Mutual Coupling Through a Multiscaling BCS Strategy

Generating Videos with Scene Dynamics

Convolutional Neural Networks for Steady Flow Approximation

Accelerating Eulerian Fluid Simulation With Convolutional Networks

Target Classification Using the Deep Convolutional Networks for SAR Images

A paradigm for data-driven predictive modeling using field inversion and machine learning

Deep Residual Learning for Image Recognition

The Finite-Difference Time-Domain Method for Electromagnetics with MATLAB® Simulations

Unsupervised Learning of Visual Structure using Predictive Generative Networks

Deep multi-scale video prediction beyond mean square error

Data-driven fluid simulations using regression forests

Discovering governing equations from data by sparse identification of nonlinear dynamical systems

Automated adaptive inference of phenomenological dynamical models

Recurrent Network Models for Human Dynamics

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

Equation-free mechanistic ecosystem forecasting using empirical dynamic modeling

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Very Deep Convolutional Networks for Large-Scale Image Recognition

Efficient Inference of Parsimonious Phenomenological Models of Cellular Dynamics Using S-Systems and Alternating Regression

Detecting Causality in Complex Ecosystems

Automated refinement and inference of analytical models for metabolic networks

Deconvolutional networks

Equation-free multiscale computation: algorithms and applications.

Equation-Free, Coarse-Grained Multiscale Computation: Enabling Mocroscopic Simulators to Perform System-Level Analysis

Preconditioning the Matrix Exponential Operator with Applications

Computational Electrodynamics the Finite-Difference Time-Domain Method

Applying Deep Learning Approach to the Far-Field Subwavelength Imaging Based on Near-Field Resonant Metalens at Microwave Frequencies

A machine learningbased fast-forward solver for ground penetrating radar with application to full-waveform inversion

TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems

Equations of Motion from a Data Series

Field of Study

Journal Information

Name

Page

Volume

Venue Information

Name

Type

URL

Alternate Names