DocumentCode
1901764
Title
Artificial Neural Networks for Inverse Heat Transfer Problems
Author
Cortés, Obed ; Urquiza, Gustavo ; Hernández, J.A. ; Cruz, Marco A.
Author_Institution
UAEM, Cuernavaca
fYear
2007
fDate
25-28 Sept. 2007
Firstpage
198
Lastpage
201
Abstract
We present a solution for an inverse heat transfer problem involving internal heat generation using artificial neural networks. The problem involves a heat conduction problem with internal heat source in cylindrical coordinates. The network is a feedforward with backpropagation algorithm. We compare the results with the Levenberg-Marquardt method and discuss advantages and disadvantages. The two methods recover very well the optimum parameters.
Keywords
heat conduction; inverse problems; mechanical engineering computing; neural nets; artificial neural networks; backpropagation algorithm; heat conduction problem; internal heat generation; internal heat source; inverse heat transfer problems; Artificial neural networks; Boundary conditions; Eigenvalues and eigenfunctions; Heat transfer; ISO standards; Inverse problems; Robot kinematics; Steady-state; Temperature measurement; Testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Electronics, Robotics and Automotive Mechanics Conference, 2007. CERMA 2007
Conference_Location
Morelos
Print_ISBN
978-0-7695-2974-5
Type
conf
DOI
10.1109/CERMA.2007.4367685
Filename
4367685
Link To Document