AP-TLB-IGWO: Adult-pup teaching–learning based interactive grey wolf optimizer for numerical optimization

Abstract

Bio-inspired optimization technique acts as the fundamental backbone of any AI assisted expert system and Grey wolf optimizer (GWO) is a noteworthy example of one such technique. GWO is new in nature but gaining its popularity due to its better exploration and avoidance of stagnation at local minima. However it needs a more intelligent and balanced exploration and exploitation management procedure. Also for the problems with nonzero solution in the search space GWO needs extra attention as it has a tendency of convergence at the origin of the coordinate system. In order to mitigate these challenges and for the better representation of the existing system, an adult-pup teaching–learning based interactive grey wolf optimization (AP-TLB-IGWO) algorithm is proposed in this paper. With a new structural framework, the proposed algorithm focuses on better generalization, search procedure and diversification. Here both the adult wolves and the pups act as the search agents. They both concurrently and independently explore the entire search space for optimal results. Energy driven hunting procedure is introduced in the update equation of GWO. Information sharing takes place between adult and pup wolves and prospective agents are identified. The performance of the proposed technique is tested on 23 well known classical benchmark functions, CEC2014 problem set, CEC2017 problem set and five real world optimization problems and compared with several established optimization techniques. The experimental results and related analysis proved that the proposed AP-TLB-IGWO is providing significantly promising results in comparison with existing techniques.