Dynamic search fireworks algorithm with chaos

Abstract

As a relatively new algorithm for swarm intelligence, fireworks algorithm imitates the explosion process of fireworks. A different amplitude in dynamic search fireworks algorithm is presented for an improvement of enhanced fireworks algorithm. This paper integrates chaos with the dynamic search fireworks algorithm so as to further improve the performance and achieve global optimization. Three different variants of dynamic search fireworks algorithm with chaos are introduced and 10 chaotic maps are used to tune either the amplification coefficient $C_a$ or the reduction coefficient $C_r$ . Twelve benchmark functions are verified in use of the dynamic search fireworks algorithm with chaos (dynamic search fireworks algorithm). The dynamic search fireworks algorithm significantly outperformed the Fireworks Algorithm, enhanced fireworks algorithm, and dynamic search fireworks algorithm based on solution accuracy. The highest performance was seen when dynamic search fireworks algorithm was used with a Gauss/mouse map to tune Ca. Additionally, the dynamic search fireworks algorithm was compared with the firefly algorithm, harmony search, bat algorithm, and standard particle swarm optimization (SPSO2011). Study results indicated that the dynamic search fireworks algorithm has the highest accuracy solution among the five algorithms.