WIT Press


A Comparison Of Traditional And Non-classical Methods For Solving Inverse Heat Conduction Problem

Price

Free (open access)

Volume

12

Pages

10

Published

1996

Size

961 kb

Paper DOI

10.2495/HT960431

Copyright

WIT Press

Author(s)

J. Krejsa, L. Slama, J. Horsky, M. Raudensky & B. Patikova

Abstract

The paper deals with the inverse heat conduction (IHC) problem. The usage of artificial intelligence (AI) methods is presented. The artificial neural networks (ANN) and the genetic algorithm (GA) are presented as members of the AI methods group. The above mentioned methods are compared with the classic Beck's method using real data, measured on a metal plate cooled by a twin fluid nozzle. The main criteria for comparing the methods are the numerical precision of results, computational requirements and stability. Strengths and weaknesses of each approach are discussed. 1. Introduction Methods based on AI became widely developed in the last years. Mainly neural networks and genetic algorithms has

Keywords