Forward Monte Carlo Method for the Solution of Time-Dependent Heat Conduction Equation.

Abstract

A forward Monte Carlo method has been developed to solve the time-dependent heat conduction or diffusion equation. The general method has been implemented so as to cover a variety of boundary conditions. The method is based on a floating volume random walk, similar to the one in an adjoint method. A novel problem has been posed, and successfully solved, on the treatment of boundary conditions. The solution required the introduction of a biasing function, leading to a biased random walk. The first step of the random walk is particular to the forward method. The succeeding steps can be considered self-adjoint, as they are identical in the forward and adjoint case. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1977
Accession Number
ADA048268

Entities

People

  • Eugene S. Troubetzkoy

Organizations

  • Mathematical Applications Group

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Algorithms
  • Boundaries
  • Coordinate Systems
  • Diffusion Coefficient
  • Discrete Distribution
  • Distribution Functions
  • Efficiency
  • Equations
  • Inequalities
  • Integral Equations
  • Monte Carlo Method
  • Probability
  • Probability Distribution Functions
  • Probability Distributions
  • Random Variables
  • Random Walk
  • Sampling

Readers

  • Computational Modeling and Simulation
  • Fluid Dynamics.
  • Linear Algebra