Wolverine Optimization Algorithm

Fuente: Zenodo
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Zhang, Jincheng
Format: Recurso digital
Veröffentlicht: Zenodo 2026
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866901801685483520
author Zhang, Jincheng
author_facet Zhang, Jincheng
contents <p><span>Swarm intelligence optimization algorithms have been widely studied in continuous optimization, combinatorial optimization, and engineering applications due to their lack of gradient information requirements, simple structure, and strong global search capabilities. However, most existing swarm intelligence algorithms suffer from severe strategy homogenization during the search process, reliance on manual parameter scheduling for switching between the exploration and exploitation phases, and susceptibility to getting trapped in local optima. To address these shortcomings, this paper proposes a novel swarm intelligence optimization method—the wolverine hyperevolutionary optimization algorithm with dual-awareness and structure destruction mechanisms. Based on the independent predation behavior of wolverines, this algorithm introduces a dual-awareness search mechanism, enabling individuals to adaptively switch between self-experience-driven and environmental information-driven approaches. Simultaneously, a destruction-reconstruction operator based on historical optimal structures is designed to effectively escape local optima. Furthermore, a nonlinear predation path model is proposed to enhance the algorithm's convergence stability in later stages. The algorithm achieves a natural transition from exploration to exploitation capabilities through the dynamic evolution of attack energy, thereby reducing dependence on external parameter scheduling strategies. Theoretical analysis shows that the algorithm has good interpretability in terms of search dynamics and energy allocation. Experimental results (omitted) demonstrate that this method exhibits strong global search capabilities and stability on multiple benchmark optimization problems</span>.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18366280
institution Zenodo
language
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Wolverine Optimization Algorithm
Zhang, Jincheng
<p><span>Swarm intelligence optimization algorithms have been widely studied in continuous optimization, combinatorial optimization, and engineering applications due to their lack of gradient information requirements, simple structure, and strong global search capabilities. However, most existing swarm intelligence algorithms suffer from severe strategy homogenization during the search process, reliance on manual parameter scheduling for switching between the exploration and exploitation phases, and susceptibility to getting trapped in local optima. To address these shortcomings, this paper proposes a novel swarm intelligence optimization method—the wolverine hyperevolutionary optimization algorithm with dual-awareness and structure destruction mechanisms. Based on the independent predation behavior of wolverines, this algorithm introduces a dual-awareness search mechanism, enabling individuals to adaptively switch between self-experience-driven and environmental information-driven approaches. Simultaneously, a destruction-reconstruction operator based on historical optimal structures is designed to effectively escape local optima. Furthermore, a nonlinear predation path model is proposed to enhance the algorithm's convergence stability in later stages. The algorithm achieves a natural transition from exploration to exploitation capabilities through the dynamic evolution of attack energy, thereby reducing dependence on external parameter scheduling strategies. Theoretical analysis shows that the algorithm has good interpretability in terms of search dynamics and energy allocation. Experimental results (omitted) demonstrate that this method exhibits strong global search capabilities and stability on multiple benchmark optimization problems</span>.</p>
title Wolverine Optimization Algorithm
url https://doi.org/10.5281/zenodo.18366280