Physical knowledge improves prediction of EM Fields

• URL: http://arxiv.org/abs/2503.11703v1
• Date: Wed, 12 Mar 2025 10:22:57 GMT
• Title: Physical knowledge improves prediction of EM Fields
• Authors: Andrzej Dulny, Farzad Jabbarigargari, Andreas Hotho, Laura Maria Schreiber, Maxim Terekhov, Anna Krause,
• Abstract summary: We propose a 3D U-Net model to predict the spatial distribution of electromagnetic fields inside a radio-frequency (RF) coil with a subject present.<n>We introduce a physics-augmented variant, U-Net Phys, which incorporates Gauss's law of magnetism into the loss function using finite differences.
• Score: 2.772393030165581
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose a 3D U-Net model to predict the spatial distribution of electromagnetic fields inside a radio-frequency (RF) coil with a subject present, using the phase, amplitude, and position of the coils, along with the density, permittivity, and conductivity of the surrounding medium as inputs. To improve accuracy, we introduce a physics-augmented variant, U-Net Phys, which incorporates Gauss's law of magnetism into the loss function using finite differences. We train our models on electromagnetic field simulations from CST Studio Suite for an eight-channel dipole array RF coil at 7T MRI. Experimental results show that U-Net Phys significantly outperforms the standard U-Net, particularly in predicting fields within the subject, demonstrating the advantage of integrating physical constraints into deep learning-based field prediction.

Related papers

• RadioDiff-$k^2$: Helmholtz Equation Informed Generative Diffusion Model for Multi-Path Aware Radio Map Construction [69.96295462931168]
  We propose a physics-informed generative learning approach, termed RadioDiff-$bmk2$, for accurate and efficient multipath-aware radio map (RM) construction. We establish a direct correspondence between EM singularities, which correspond to the critical spatial features influencing wireless propagation, and regions defined by negative wave numbers in the Helmholtz equation.
  arXiv  Detail & Related papers  (2025-04-22T06:28:13Z)
• Time-inversion of spatiotemporal beam dynamics using uncertainty-aware latent evolution reversal [46.348283638884425]
  This paper introduces a reverse Latent Evolution Model (rLEM) designed for temporal phase of forward beam dynamics. In this two-step self-supervised deep learning framework, we utilize a Conditional Autoencoder (CVAE) to project 6D space projections of a charged particle beam into a lower-dimensional latent distribution. We then autoregressively learn the inverse temporal dynamics in the latent space using a Long Short-Term Memory (LSTM) network.
  arXiv  Detail & Related papers  (2024-08-14T23:09:01Z)
• EM-GANSim: Real-time and Accurate EM Simulation Using Conditional GANs for 3D Indoor Scenes [55.2480439325792]
  We present a novel machine-learning (ML) approach (EM-GANSim) for real-time electromagnetic (EM) propagation. In practice, it can compute the signal strength in a few milliseconds on any location in 3D indoor environments.
  arXiv  Detail & Related papers  (2024-05-27T17:19:02Z)
• Global-local Fourier Neural Operator for Accelerating Coronal Magnetic Field Model [17.256941005824576]
  We propose a global-local Fourier Neural Operator (GL-FNO) that contains two branches of FNO. The performance of the GLFNO is compared with state-of-the-art deep learning methods, including FNO, U-NO, U-FNO, Vision Transformer, CNN-RNN, and CNN-LSTM. The results demonstrate that GL-FNO not only accelerates the MHD simulation (a few seconds for prediction, more than times 20,000 speed up) but also provides reliable prediction capabilities, thus greatly contributing to the understanding of space weather dynamics.
  arXiv  Detail & Related papers  (2024-05-21T13:04:53Z)
• Physics-informed generative neural networks for RF propagation prediction with application to indoor body perception [2.679900758407988]
  The paper discusses a Variational Auto-Encoder (VAE) model which is trained to reproduce the effects of human motions on the EM field. Proposed physics-informed generative neural network models are verified against both classical diffraction-based EM tools and full-wave EM body simulations.
  arXiv  Detail & Related papers  (2024-05-03T14:35:02Z)
• Gaussian Plane-Wave Neural Operator for Electron Density Estimation [11.850515912491657]
  We introduce the Gaussian plane-wave neural operator (GPWNO), which operates in the infinite-dimensional functional space. In particular, both high- and low-frequency components of the density can be effectively represented due to the complementary nature of the two bases. Experiments on QM9, MD, and material project datasets demonstrate GPWNO's superior performance over ten baselines.
  arXiv  Detail & Related papers  (2024-02-05T06:05:30Z)
• Pre-training via Denoising for Molecular Property Prediction [53.409242538744444]
  We describe a pre-training technique that utilizes large datasets of 3D molecular structures at equilibrium. Inspired by recent advances in noise regularization, our pre-training objective is based on denoising.
  arXiv  Detail & Related papers  (2022-05-31T22:28:34Z)
• Magnetic Field Prediction Using Generative Adversarial Networks [0.0]
  We predict magnetic field values at a random point in space by using a generative adversarial network (GAN) structure. The deep learning (DL) architecture consists of two neural networks: a generator, which predicts missing field values of a given magnetic field, and a critic, which is trained to calculate the statistical distance between real and generated magnetic field distributions. Our trained generator has learned to predict the missing field values with a median reconstruction test error of 5.14%, when a single coherent region of field points is missing, and 5.86%, when only a few point measurements in space are available.
  arXiv  Detail & Related papers  (2022-03-14T12:31:54Z)
• WaveY-Net: Physics-augmented deep learning for high-speed electromagnetic simulation and optimization [0.0]
  We introduce WaveY-Net, a hybrid data- and physics-augmented convolutional neural network that can predict electromagnetic field distributions with ultra fast speeds. We show that the high speed simulator can be directly and effectively used in the local and global freeform optimization of metagratings.
  arXiv  Detail & Related papers  (2022-03-02T17:11:33Z)
• Predicting the Thermal Sunyaev-Zel'dovich Field using Modular and Equivariant Set-Based Neural Networks [1.7709249262395885]
  We train neural networks on the IllustrisTNG-300 cosmological simulation to predict the continuous electron pressure field in galaxy clusters. We employ a rotational DeepSets architecture to operate directly on the set dark matter particles.
  arXiv  Detail & Related papers  (2022-02-28T19:00:07Z)
• Simultaneous boundary shape estimation and velocity field de-noising in Magnetic Resonance Velocimetry using Physics-informed Neural Networks [70.7321040534471]
  Magnetic resonance velocimetry (MRV) is a non-invasive technique widely used in medicine and engineering to measure the velocity field of a fluid. Previous studies have required the shape of the boundary (for example, a blood vessel) to be known a priori. We present a physics-informed neural network that instead uses the noisy MRV data alone to infer the most likely boundary shape and de-noised velocity field.
  arXiv  Detail & Related papers  (2021-07-16T12:56:09Z)
• Spin Entanglement and Magnetic Competition via Long-range Interactions in Spinor Quantum Optical Lattices [62.997667081978825]
  We study the effects of cavity mediated long range magnetic interactions and optical lattices in ultracold matter. We find that global interactions modify the underlying magnetic character of the system while introducing competition scenarios. These allow new alternatives toward the design of robust mechanisms for quantum information purposes.
  arXiv  Detail & Related papers  (2020-11-16T08:03:44Z)
• Ferromagnetic Gyroscopes for Tests of Fundamental Physics [49.853792068336034]
  A ferromagnetic gyroscope (FG) is a ferromagnet whose angular momentum is dominated by electron spin polarization and that will precess under the action of an external torque. We model and analyze FG dynamics and sensitivity, focusing on practical schemes for experimental realization.
  arXiv  Detail & Related papers  (2020-10-17T07:13:50Z)
• OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Features [42.96944345045462]
  textscOrbNet is shown to outperform existing methods in terms of learning efficiency and transferability. For applications to datasets of drug-like molecules, textscOrbNet predicts energies within chemical accuracy of DFT at a computational cost that is thousand-fold or more reduced.
  arXiv  Detail & Related papers  (2020-07-15T22:38:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.