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

Predicting neutron diffusion eigenvalues with a query-based adaptive neural architecture.

M G Lysenko1, H I Wong, G I Maldonado

  • 1Vehicle CAE Integration Department of Ford Motor Co., Dearborn, MI 48121, USA.

IEEE Transactions on Neural Networks
|February 7, 2008
PubMed
Summary
This summary is machine-generated.

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

Resumption of the reproductive phase of Trypanosoma lewisi in recovered and near recovered rats following both reinoculation with the parasite and treatment with sodium salicylate.

The Journal of parasitology·2010
Same author

[Congenital coronary arteriovenous fistula: analysis of five cases].

Zhonghua Minguo xiao er ke yi xue hui za zhi [Journal]. Zhonghua Minguo xiao er ke yi xue hui·1989
Same author

The use of tissue culture-grown trophozoites of an avirulent strain of Toxoplasma gondii for the immunization of mice and guinea pigs.

The Journal of parasitology·1971
Same author

In Vivo and In Vitro Studies of Delayed-Type Hypersensitivity to Toxoplasma gondii in Guinea Pigs.

Infection and immunity·1971
Same author

EFFECT OF X-IRRADIATION ON TRYPANOSOMA LEWISI INFECTION IN THE RAT.

Experimental parasitology·1965
Same author

In vitro phagocytosis of Trypanosoma lewisi by rat exudative cells.

Experimental parasitology·1960
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces an adaptive artificial neural network (ANN) approach for rapid neutron diffusion eigenvalue prediction. The method enhances accuracy by optimizing training data and network architecture for nuclear reactor design.

Area of Science:

  • Nuclear Engineering
  • Computational Science
  • Artificial Intelligence

Background:

  • Nuclear reactor core design relies on solving the neutron diffusion equation, a complex process involving large-scale nonlinear partial differential equations (PDEs).
  • Traditional methods require iterative solutions, which can be computationally intensive and time-consuming.
  • Artificial Neural Networks (ANNs) offer a potential alternative for faster eigenvalue prediction.

Purpose of the Study:

  • To develop a query-based, adaptive retraining and restructuring approach for ANNs.
  • To improve the speed, accuracy, and generalization of ANN-based predictions for the fundamental mode eigenvalue of the neutron diffusion equation.
  • To enhance the performance of a multilayer perceptron (MLP) for nuclear reactor core design calculations.

Main Methods:

Related Experiment Videos

  • Implemented a query-based approach for adaptive selection of training and cross-validation data.
  • Focused on ANN architecture adjustments, including dynamic node architecture (DNA).
  • Utilized techniques such as nonrandom initial training set selection, adjoint function input weighting, and teacher-student queries for data generation.

Main Results:

  • Demonstrated significant improvements in accuracy and generalization for ANN-based eigenvalue predictions.
  • The adaptive methodology successfully upgraded a basic feedforward MLP.
  • The flexible approach can be integrated with various training algorithms.

Conclusions:

  • The developed query-based adaptive ANN approach provides a faster and more accurate method for predicting neutron diffusion eigenvalues.
  • This technique offers a promising alternative to traditional iterative PDE solutions in nuclear reactor design.
  • Further research can explore the application of this methodology to other complex scientific and engineering problems.