Machine Learning for PAC1D and SESE

Abstract

This document outlines various machine learning approaches that were taken in an effort to surrogate numerical models Python Ablation Code 1-Dimension (PAC1D) and Scalable Effects Simulation Environment (SESE)[1], [2], with the ultimate objective of discovering the most efficient method for approximating SESE with sparse data utilization. The methods explored include; physics-informed neural networks, deep galerkin method, deep mixed residual methods, operator network, deep operator network, fourier neural operator, physics-informed fourier neural operator, and physics-informed kernal neural operator. Many of the methods showed strengths and weaknesses in their performance, with the physics-informed kernal neural operator showing the most potential for approximating SESE's behavior.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 03, 2023
Accession Number
AD1208932

Entities

People

  • Brett A. Bowman
  • Chad A. Oian
  • Jason A. Kurz
  • Matthew G. Seman
  • Taufiquar Khan

Tags

Communities of Interest

  • Autonomy
  • Human Systems

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Artificial Intelligence Software
  • Banach Space
  • Bayesian Networks
  • Boundaries
  • Computational Science
  • Deep Learning
  • Differential Equations
  • Governments
  • Human Performance
  • Information Systems
  • Inverse Problems
  • Kernel Functions
  • Machine Learning
  • Network Science
  • Neural Networks
  • Partial Differential Equations
  • Sequences
  • Simulators
  • Three Dimensional
  • United States
  • United States Government

Fields of Study

  • Computer science
  • Physics

Readers

  • Computational Modeling and Simulation
  • Linear Algebra
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks