Multi-Objective Grey Wolf Optimizer Based on Improved Head Wolf Selection Strategy

The grey wolf optimizer (GWO) is an emerging swarm intelligence optimization algorithm that has been successfully applied in many research fields. The improved multi-objective grey wolf optimizer (IMOGWO) is proposed to address the problems of unreasonable head wolf selection and slow convergence speed in multi-objective grey wolf optimizer (MOGWO). Firstly, in response to the problem of MOGWO using roulette wheel method to select the head wolf from non dominated solutions, which can easily lead to blind search, a head wolf selection strategy based on reference vector is designed to make the solution set distribution of IMOGWO more uniform. Secondly, aiming to the problem of slow convergence speed and easy falling into local optima caused by the existing gray wolf position update strategy, the Levy flight operator is introduced into position update strategy to improve convergence speed and make IMOGWO easier to jump out of local optima. Finally, the effectiveness of the IMOGWO was verified through the UF series of problems.