Enhancing fireworks algorithms for dynamic optimization problems

Abstract

Dynamic optimization problems have been captivating the interest of the researchers, since most real world problems in different domains have various characteristics of dynamism. Different evolutionary and swarm intelligence techniques are proposed to solve dynamic optimization problems. Fireworks Algorithm (FWA) is a recently proposed swarm intelligence algorithm for global optimization of complex functions. This paper proposes two extensions on the FWA, called the EFWA-D1 and the EFWA-D2 algorithms, in order to adapt on dynamic optimization problems. We validate performance of the EFWA-D1 and the EFWA-D2 with the Moving Peaks Benchmark (MPB), a well-known synthetic dynamic optimization problem that generates and updates a multidimensional landscape consisting of several peaks. Experimental evaluation on various instances of MPB validates the applicability of our extensions on the FWA for a dynamic optimization problem. © 2016 IEEE.