Open AccessIdentification of Dynamic Models by Using Metaheuristic Algorithms
Author(s) -
Mustafa Danacı,
Fehim Köylü,
Zaid Ali Al-Sumaidaee
Publication year - 2021
Publication title -
adi journal on recent innovation/adi journal on recent innovation (ajri)
Language(s) - English
Resource type - Journals
eISSN - 2685-9106
pISSN - 2686-0384
DOI - 10.34306/ajri.v3i1.492
Subject(s) - metaheuristic , algorithm , robustness (evolution) , computer science , ant colony optimization algorithms , swarm behaviour , parallel metaheuristic , meta optimization , identification (biology) , genetic algorithm , artificial bee colony algorithm , heuristic , noise (video) , simulated annealing , mathematical optimization , machine learning , artificial intelligence , mathematics , biochemistry , chemistry , botany , biology , image (mathematics) , gene
A modified versions of metaheuristic algorithms are presented to compare their performance in identifying the structural dynamic systems. Genetic algorithm, biogeography based optimization algorithm, ant colony optimization algorithm and artificial bee colony algorithm are heuristic algorithms that have robustness and ease of implementation with simple structure. Different algorithms were selected some from evolution algorithms and other from swarm algorithms to boost the equilibrium of global searches and local searches, to compare the performance and investigate the applicability of proposed algorithms to system identification; three cases are suggested under different conditions concerning data availability, different noise rate and previous familiarity of parameters. Simulation results show these proposed algorithms produce excellent parameter estimation, even with little measurements and a high noise rate.