Bee-foraging learning particle swarm optimization

Numerous particle swarm optimization (PSO) algorithms have been developed for solving numerical problems in recent years. However, most of existing PSO only one search phase. There is no strengthened phase the well-performed particles, and also re-initialization exhausted particles. These issues may still restrict performance complex problems. In this paper, inspired by bee-foraging mechanism artificial bee colony algorithm, a novel learning (BFL-PSO) algorithm proposed. Different from algorithms, proposed BFL-PSO has three different phases, namely employed learning, onlooker scout learning. The works like traditional one-phase-based PSO, while performs around those particles to exploit promising solutions, re-initializes introduce new diversity. comprehensively evaluated on CEC2014 benchmark functions, compared with state-of-the-art as well algorithms. experimental results show that achieves very competitive terms solution accuracy. addition, effectiveness newly introduced phases verified. • Bee-foraging integrated into optimization. developed. Three (employed, scout) are adopted BFL-PSO. exhibits