Institutional Repository of Vilnius University
Adaptation of the grey wolf optimizer for dynamic optimization problems /
Abstract
dc:descriptionThis work investigates swarm intelligence algorithms in dynamic optimization environments, where the global optimum changes over time. Three algorithms are evaluated: the Grey Wolf Optimizer (GWO), a multi-swarm Particle Swarm Optimization variant (MPSO), and a proposed Dynamic Grey Wolf Optimizer (DGWO). Experiments are conducted on four benchmark functions (Sphere, Rastrigin, Rosenbrock, and Griewank) under three dynamic shift mechanisms: linear, periodic, and teleport. Performance is assessed using offline error, tracking error, and additional metrics including fitness, distance to the moving optimum, and population diversity. Results show that standard GWO suffers from premature convergence and poor adaptability in dynamic settings. MPSO improves performance through population decomposition but remains sensitive to abrupt changes. The proposed DGWO manages to achieve comparable performance to the MPSO, demonstrating improved tracking ability, robustness, and diversity maintenance when compared to the regular GWO.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Mačeika, Mantvydas,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308337854&prefLang=en_US