Date of Award
Pooya Moradian P.M. Zadeh
Ziad Z.K. Kobti
evolutionary, grey wolves, meta-heuristics, multi-population, optimization
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Various fields, such as engineering, physics, and economics, require optimization in the real world. Various meta-heuristic methods have gained popularity in recent decades to solve these optimization problems; evolutionary algorithms are one of the ways to solve these problems. This class of algorithms deal with a generation of candidate solutions that are evolved until a stopping criterion is achieved. Researchers are improving these algorithms' performance by introducing new ensemble strategies to tackle a variety of problems. This thesis focuses on creating a novel co-operative multi-population framework to solve single and bi-objective problems based on the hunting strategies and hierarchical structures of grey wolves. The structure of this framework allows to overcome several defects and improves the information flow and convergence of the search process. The framework is evaluated using IEEE's Congress of Evolution Congress benchmarks for single-objective real parameter optimization (2013) and unconstrained multi-objective optimization problems (2009). The performance is compared with the traditional grey wolf optimization algorithms and state-of-the-art for single and multi-objective optimization.
Verma, Nimish, "H-MPGWO: A Hierarchical Multi-Population Grey Wolf Optimization Framework" (2021). Electronic Theses and Dissertations. 8541.