WSEAS Transactions on Systems and Control
Print ISSN: 1991-8763, E-ISSN: 2224-2856
Volume 15, 2020
An Adaptive Differential Evolution Algorithm with Restart for Solving Continuous Optimization Problems
Authors: ,
Abstract: A new adaptive differential evolution algorithm with restart (ADE-R) is proposed as a general-purpose method for solving continuous optimization problems. Its design aims at simplicity of use, efficiency and robustness. ADE-R simulates a population evolution of real vectors using vector mixing operations with an adaptive parameter control based on the switching of two selected intervals of values for each scaling factor and crossover rate of the basic differential evolution algorithm. It also incorporates a restart technique to supply new contents to the population to prevent premature convergence and stagnation. The method is tested on several benchmark functions covering various types of functions and compared with some well-known and state-of-art methods. The experimental results show that ADE-R is effective and outperforms the compared methods.
Search Articles
Keywords: Continuous optimization, optimization method, adaptive differential evolution algorithm, adaptive parameter control, restart technique
Pages: 254-269
DOI: 10.37394/23203.2020.15.27