Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Finite-element neural networks for solving differential equations.

Pradeep Ramuhalli1, Lalita Udpa, Satish S Udpa

  • 1Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI 48824, USA. rpradeep@egr.msu.edu

IEEE Transactions on Neural Networks
|December 14, 2005
PubMed
Summary

This study introduces a finite-element neural network (FENN) for solving partial differential equations (PDEs). The FENN offers a computationally efficient alternative to traditional methods for both forward and inverse problems in engineering.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Efficient data acquisition and reconstruction for air-coupled ultrasonic robotic NDE.

Scientific reports·2024
Same author

Gradient Index Metasurface Lens for Microwave Imaging.

Sensors (Basel, Switzerland)·2022
Same author

Machine learning for ultrasonic nondestructive examination of welding defects: A systematic review.

Ultrasonics·2022
Same author

Ultra-Wideband Microwave Imaging System for Root Phenotyping.

Sensors (Basel, Switzerland)·2022
Same author

Bioelectromagnetic Platform for Cell, Tissue, and In Vivo Stimulation.

Biosensors·2021
Same author

Negative Index Metamaterial Lens for Subwavelength Microwave Detection.

Sensors (Basel, Switzerland)·2021

Area of Science:

  • Computational Engineering
  • Numerical Analysis
  • Electromagnetics

Background:

  • Partial differential equations (PDEs) are crucial in engineering.
  • Traditional numerical methods like finite-element (FEM) and finite-difference methods face computational challenges with complex geometries.

Purpose of the Study:

  • To propose a novel finite-element neural network (FENN) for efficient PDE solutions.
  • To evaluate FENN's performance in solving electromagnetic forward and inverse problems.

Main Methods:

  • Embedding a finite-element model within a neural network architecture.
  • Applying the FENN to electromagnetic forward and inverse problems.
  • Utilizing an iterative approach with FENN for inverse problem-solving.

Related Experiment Videos

Main Results:

  • FENN demonstrates comparable accuracy to conventional FEM for forward problems.
  • FENN effectively solves inverse problems when provided with measured signals.
  • The parallel nature of FENN facilitates hardware and software implementation.

Conclusions:

  • FENN presents a fast and accurate method for solving PDEs.
  • FENN offers a promising approach for complex engineering simulations and inverse problem analysis.
  • The FENN architecture is suitable for parallel computing environments.