Stochastic gradient descent for optimization for nuclear systems

Williams, Austin; Walton, Noah; Maryanski, Austin; Bogetic, Sandra; Hines, Wes; Sobes, Vladimir

doi:10.1038/s41598-023-32112-7

Stochastic gradient descent for optimization for nuclear systems

Journal Article · Thu May 25 00:00:00 EDT 2023 · Scientific Reports

DOI:https://doi.org/10.1038/s41598-023-32112-7· OSTI ID:2417878

Williams, Austin ^[1]; Walton, Noah ^[2]; Maryanski, Austin ^[2]; Bogetic, Sandra ^[2]; Hines, Wes ^[2]; Sobes, Vladimir ^[2]

Univ. of Tennessee, Knoxville, TN (United States); University of Tennessee, Knoxville, TN (United States)
Univ. of Tennessee, Knoxville, TN (United States)

The use of gradient descent methods for optimizing k-eigenvalue nuclear systems has been shown to be useful in the past, but the use of k-eigenvalue gradients have proved computationally challenging due to their stochastic nature. ADAM is a gradient descent method that accounts for gradients with a stochastic nature. This analysis uses challenge problems constructed to verify if ADAM is a suitable tool to optimize k-eigenvalue nuclear systems. ADAM is able to successfully optimize nuclear systems using the gradients of k-eigenvalue problems despite their stochastic nature and uncertainty. Furthermore, it is clearly demonstrated that low-compute time, high-variance estimates of the gradient lead to better performance in the optimization challenge problems tested here.

Research Organization:: Univ. of California, Oakland, CA (United States)

Sponsoring Organization:: USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: NA0000979; NA0003180

OSTI ID:: 2417878

Journal Information:: Scientific Reports, Journal Name: Scientific Reports Journal Issue: 1 Vol. 13; ISSN 2045-2322

Publisher:: Nature Publishing GroupCopyright Statement

Country of Publication:: United States

Language:: English

References (12)

On the convergence of the Newton/log-barrier method Wright, Stephen J. Mathematical Programming, Vol. 90, Issue 1 https://doi.org/10.1007/PL00011421	journal	March 2001
Thermo-mechanical coupled analysis based design of ventilated brake disc using genetic algorithm and particle swarm optimization Belhocine, Ali; Shinde, Dinesh; Patil, Rahul JMST Advances, Vol. 3, Issue 3 https://doi.org/10.1007/s42791-021-00040-0	journal	June 2021
Application of a self-organizing fuzzy logic controller to nuclear steam generator level control Park, Gee Yong; Seong, Poong Hyun Nuclear Engineering and Design, Vol. 167, Issue 3 https://doi.org/10.1016/S0029-5493(96)01303-9	journal	April 1997
A modeling method for aero-engine by combining stochastic gradient descent with support vector regression Ren, Li-Hua; Ye, Zhi-Feng; Zhao, Yong-Ping Aerospace Science and Technology, Vol. 99 https://doi.org/10.1016/j.ast.2020.105775	journal	April 2020
Nuclear reactor vessel water level prediction during severe accidents using deep neural networks Koo, Young Do; An, Ye Ji; Kim, Chang-Hwoi Nuclear Engineering and Technology, Vol. 51, Issue 3 https://doi.org/10.1016/j.net.2018.12.019	journal	June 2019
Gradient descent-particle swarm optimization based deep neural network predictive control of pressurized water reactor power Ejigu, Derjew Ayele; Liu, Xiaojing Progress in Nuclear Energy, Vol. 145 https://doi.org/10.1016/j.pnucene.2021.104108	journal	March 2022
Consistency regularization-based deep polynomial chaos neural network method for reliability analysis Zheng, Xiaohu; Yao, Wen; Zhang, Yunyang Reliability Engineering & System Safety, Vol. 227 https://doi.org/10.1016/j.ress.2022.108732	journal	November 2022
Gradient-Informed Design Optimization of Select Nuclear Systems Pevey, John; Hiscox, Briana; Williams, Austin Nuclear Science and Engineering https://doi.org/10.1080/00295639.2021.1987133	journal	November 2021
The method of steepest descent for non-linear minimization problems Curry, Haskell B. Quarterly of Applied Mathematics, Vol. 2, Issue 3 https://doi.org/10.1090/qam/10667	journal	January 1944
Constrained consensus via logarithmic barrier functions Lee, Unsik; Mesbahi, Mehran IEEE Conference on Decision and Control and European Control Conference https://doi.org/10.1109/CDC.2011.6161496	conference	December 2011
Estimation of IMU and MARG orientation using a gradient descent algorithm Madgwick, S. O. H.; Harrison, A. J. L.; Vaidyanathan, R. 2011 IEEE International Conference on Rehabilitation Robotics https://doi.org/10.1109/ICORR.2011.5975346	conference	June 2011
Perturbation Theory Eigenvalue Sensitivity Analysis with Monte Carlo Techniques Rearden, B. T. Nuclear Science and Engineering, Vol. 146, Issue 3 https://doi.org/10.13182/NSE03-03	journal	March 2004

Similar Records

A Stochastic Gradient Descent Approach for Stochastic Optimal Control

Journal Article · Sat Aug 01 00:00:00 EDT 2020 · East Asian Journal on Applied Mathematics · OSTI ID:1814258

Stochastic gradient descent algorithm for stochastic optimization in solving analytic continuation problems

Journal Article · Sat Feb 29 23:00:00 EST 2020 · Foundations of Data Science · OSTI ID:1632071

Convergence of Hyperbolic Neural Networks Under Riemannian Stochastic Gradient Descent

Journal Article · Thu Oct 05 00:00:00 EDT 2023 · Communications on Applied Mathematics and Computation · OSTI ID:2007675

Related Subjects

73 NUCLEAR PHYSICS AND RADIATION PHYSICS
Science & Technology - Other Topics

Stochastic gradient descent for optimization for nuclear systems

Citation Formats

References (12)

Similar Records

Related Subjects