An Efficient Hybrid Particle Swarm Optimization and Firefly Algorithm

Typically, particle swarm optimization (PSO) is known for its robust global exploration, but it often falls into the “local trap”. Compared to PSO, the firefly algorithm (FFA) might offer superior local exploitation, but it often lacks speed characteristics and struggles to expand beyond local searches. To harness the strengths of both algorithms, an efficient hybrid particle swarm optimization and firefly algorithm (HPSOFF) is proposed, in which particle swarm optimization with nonlinear inertia weight and attenuation factor (PSO-NIWAF) and firefly algorithm with dynamic population (FFA-DP) are jointly executed to avoid their limitations. Firstly, the mechanism of stochastic parameter mapping (SPM) is employed to obtain uniformly distributed particles, which can promote the optimization speed of the algorithm. Then, the strategy of nonlinear inertia weight and attenuation factor (NIWAF) is introduced into the PSO to bolster its global exploration. In the later stage of the algorithm, a dynamic population strategy is introduced into the FFA to adaptively construct a temporary population. By ranking the fitness values of all particles in the temporary population, superior solutions can be identified to ensure precise local exploitation. Finally, the proposed HPSOFF is adopted to test the CEC2005 benchmark functions as well as famous engineering applications. The experimental results demonstrate that the proposed HPSOFF has significant advantages among these state-of-the-art algorithms.