A real-valued parallel clonal selection algorithm and its application to the design optimization of multi-layered frequency selective surfaces

A parallelized real-valued clonal selection algorithm (CLONALG) is successfully implemented in this paper utilizing message passing interface (MPI) to reduce the computational burden of a large clone pool. CLONALG is one of the many branches of Artificial Immune System (AIS) algorithms with unique inherent properties that make it a very efficient optimization techniques for multimodal problems such as the ones commonly encountered in computational electromagnetic design. As a demonstration of its effectiveness, a numerical study is carried out with known benchmark functions along with the optimization of multi-layered frequency selective surface (FSS) filters in the X-band. Our results show that the CLONALG can consistently outperform a standard GA implementation particularly in multi-modal optimization problems.