A Neural Network Approach In A Backward Heat Conduction Problem
Fabio Tokio Mikki, Edison Issamoto, Jefferson I. da Luz, Pedro Paulo Balbi de Oliveira, Haroldo F. Campos-Velho, Jose Demisio Simoes da Silva · 2016
This paper describes the experiments conducted in determining the initial temperature distribution on a slab with adiabatic boundary conditions, from a transient temperature distribution, obtained at a given time. This is an ill-posed inverse problem, where the initial condition has to be estimated. Two different artificial neural networks have been applied to address the problem: backpropagation and radial basis functions (RBF). Both approaches use the whole temperature history mapping. In our simulations, RBF presented better solutions, faster training, but higher noise sensitiveness, as compared to backpropagation.