Dynamic Surrogate-Assisted Multiobjective Enhanced Beluga Whale Optimization Algorithm Based on K -Means++ and Its Applications to Conical Quadrifilar Helical Antenna

Le résumé fourni par la source

This letter proposes a dynamic surrogate-assisted enhanced multi-objective beluga whale optimization (DSA-EMOBWO) algorithm based onK-means++ clustering, which is applied to the optimization of a stepped conical quadrifilar helical antenna (CQHA). The proposed algorithm leveragesK-means++ clustering to construct localized dynamic surrogate-assisted model, reducing the computational cost of antenna optimization and providing improved localized surrogate accuracy. Furthermore, an enhanced multi-objective beluga whale optimization algorithm is employed to optimize the design variables of the stepped CQHA, solving the multi-performance requirements of the stepped CQHA. Three performance requirements of the stepped CQHA are studied in this work, including operational bandwidth,S-parameters, and axial ratio performance. The experimental example demonstrates that compared with conventional optimization methods, the proposed algorithm significantly improves the efficiency of multi-objective optimization and obtains the stepped CQHA with superior performance at a lower computational cost.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.