Modified Grey Wolf Optimization Hybridized with Teaching and Learning Mechanism for Solving Optimization Problems

Abstract

The Grey Wolf Optimization (GWO) algorithm, inspired by grey wolf social behaviors, has shown excellent performance in various optimization problems. However, it faces limitations in handling dynamic optimization problems. To address this, we propose an enhanced version, Merged Teaching and Learning Grey Wolf Optimization (MTLGWO). MTLGWO introduces a two-phase teaching and learning strategy, improving global exploration and local exploitation capabilities. The core improvements include using Latin Hypercube Sampling for better population initialization and adopting a group teaching mechanism to simulate diverse teaching strategies. Through comprehensive performance testing on CEC2017 basic test functions, MTLGWO demonstrates superior performance in terms of convergence accuracy, stability, and convergence speed. Compared with other classical heuristic optimization algorithms, MTLGWO proves its potential and reliability as an efficient tool for solving optimization problems. These results highlight MTLGWO's potential as an efficient tool for practical optimization problems.