Inverse estimation of various surface emissivity distributions using repulsive particle swarm optimization

摘要

The aim of the present study is to demonstrate further inverse solving capability of the repulsive particle swarm optimization (RPSO) method by estimating unknown surface radiative property of various distributions in a surface radiation problem. A distribution of the surface emissivity was selected as the unknown property to retrieve and was defined as arbitrary functional forms to simulate various profiles in an axisymmetric cylindrical enclosure with nonparticipating media condition. A total of four functional forms were examined with uniform, rectangular step, triangular ramp and polynomial curve profiles. Also, the number of unknowns was increased to verify if the RPSO method could still retrieve property values exactly when compared with a previous study. In conclusion, the present results proved that the RPSO method showed accurate and robust performance as the inverse solver due to a superior search capability for the inverse surface radiation problem involving numerous unknown parameters of various distributions. (C) 2018 Elsevier Ltd. All rights reserved.