Geometric optimization of radiative enclosures using PSO algorithm

A Shape optimization of a two-dimensional radiative enclosure with diffuse-gray surfaces is studied by the particle swarm algorithm. The goal of the design problem is to achieve an optimized geometry to satisfy the desired temperature and heat flux distributions over a part of radiative enclosures, namely the design surface. The radiative transfer equation is solved by the discrete transfer method. In order to optimize the geometry of enclosure, the positions of the control points of the B-spline curves which represent the geometry of the enclosure are optimized. Some numerical examples are presented to demonstrate the effects of mesh refinements and shape parameters on the resultant geometry and the convergence rate.

► The particle swarm algorithm is used for optimization of radiative enclosures.
► The effects of angular mesh refinement, and shape parameters are investigated.
► The performance of the PSO is compared with the mGA.
► Using the PSO, the fitness function is improved for the same number of iterations.