Accelerated grey wolf optimization for global optimization problems

Grey Wolf Optimizer (GWO) belongs to the family of swarm intelligence and population-based algorithm which mimics the social behaviors of gray wolf packs. Due to its ease and viability, various researchers from different domains applied GWO to solve their research problems. However, GWO ought to have solid yet adjusted strengthening and broadening procedure to upgrade its execution. This paper introduces a novel SI algorithm, namely Accelerated gray Wolf Optimization (AGWO) which incorporates the enhanced hierarchy into GWO technique. Firstly, we introduce a mathematical model to strengthen the local and global search process. Then, we propose a diversity measure to eradicate the local confinement and to keep perfect balance between intensification and diversification process. Further, parameter tuned strategy is incorporated to speed up the convergence rate. The proposed methodology is experimented using MATLAB and tested with twenty-three mathematical benchmark problems (such as unimodal, multimodal and fixed dimensional multimodal functions). The search capability of the AGWO is compared with the generic GWO and various state-of-art swarm intelligence algorithms. Empirical results expose that AGWO provides better performance as per the standard deviation, best value and convergence curve compared to other algorithms. The sensitivity study of the proposed AGWO algorithm provides better results for 19 benchmark functions out of 23 benchmark functions in terms of minimal computation time and better convergence rate.