BMDA: applying biogeography-based optimization algorithm and Mexican hat wavelet to improve dragonfly algorithm
- PDF / 1,694,392 Bytes
- 26 Pages / 595.276 x 790.866 pts Page_size
- 99 Downloads / 157 Views
(0123456789().,-volV)(0123456789(). ,- volV)
FOUNDATIONS
BMDA: applying biogeography-based optimization algorithm and Mexican hat wavelet to improve dragonfly algorithm Mohammad Reza Shirani1,2 • Faramarz Safi-Esfahani1,2 Published online: 1 October 2020 Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract One of the methods for solving optimization problems is applying metaheuristic algorithms that find near to optimal solutions. Dragonfly algorithm is one of the metaheuristic algorithms which search problem space by the inspiration of hunting and emigration behavior of dragonflies in nature. However, it suffers from the premature convergence of the population to an undesirable point in the detection ability (global search). In this research, an improved dragonfly algorithm called BMDA (applying Biogeography-based algorithm, Mexican hat wavelet, and Dragonfly algorithm) is presented to resolve the premature convergence in high workloads by creating a mutation phase based on the combination of the biogeography-based optimization (BBO) migration process and the Mexican hat wavelet transform in dragonfly algorithm (DA). The algorithm was evaluated for the mean error in comparison with standard dragonfly algorithm (DA), Memorybased Hybrid Dragonfly Algorithm (MHDA), chaotic dragonfly algorithm version 9 (CDA9), Adaptive_DA algorithm, bat algorithm (BAT), particle swarm optimization algorithm (PSO), raven roosting optimization (RRO) and whale optimization algorithm (WOA) using the CEC2017 benchmark functions. The implementation results of the proposed BMDA algorithm applying different benchmark functions outweighed the DA-based algorithm, MHDA algorithm, CDA9 algorithm, Adaptive_DA algorithm, BAT algorithm, PSO algorithm, RRO, and WOA algorithms in terms of mean error. Keywords Dragonfly algorithm Biogeography-based optimization algorithm Mexican hat wavelet
1 Introduction So far, several metaheuristic algorithms have been introduced to resolve the optimization problems in an iterative process during a limited time. The baseline particle swarm optimization (PSO) algorithm (Kennedy and Eberhart 1995) with simple implementation and excellent performance in global searches is not robust in local searches. It also suffers from the premature convergence problem that
Communicated by A. Di Nola. & Faramarz Safi-Esfahani [email protected] Mohammad Reza Shirani [email protected] 1
Faculty of Computer Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran
2
Big Data Research Center, Najafabad Branch, Islamic Azad University, Najafabad, Iran
means it is easily trapped in local optimums in high-dimensional optimization problems (Jena 2015). The recent dragonfly metaheuristic algorithm (DA) (Mirjalili 2016) inspires by the collective hunting behavior (as static congestion) and migration of dragonflies in nature (as dynamic congestion). Both static and dynamic congestions of the dragonflies indicate the exploration and exploitation capabilities of the algorithm. Despite many benefits,
Data Loading...
