The negative mayfly optimization algorithm
Juan Zhao, Zheng-Ming Gao
Abstract
Abstract The global best or historical best positions were involved in updating positions of individuals in swarms for almost all of the swarm-based nature-inspired algorithms with exceptions for the averages. However, literal research to the particle swarm optimization (PSO) algorithm had proved that the all of the candidates would also leave the global best and the historical best candidates and such improvement would result in a better performance. Similarly, the negative mayfly optimization (MO) algorithm was proposed based on such conditions. Simulation experiments were carried out and verified that the negative MO algorithm could perform better than the original MO algorithm, especially in optimizing the multimodal benchmark functions.