Akademik Veri Yönetim Sistemi
IEEE Access, cilt.10, ss.66770-66796, 2022 (SCI-Expanded, Scopus)
This paper presents an improved version of Linear Population Size Reduction Success History Based Adaptive Differential Evolution (LSHADE) algorithm for solving global optimization problems. Fitness Distance Balance (FDB) selection method was used to redesign the mutation operator in order to enhance the search performance of the LSHADE algorithm. In order to test and validate the performance of the proposed algorithm, a comprehensive experimental study was carried out. For this purpose, it was tested on the CEC14 and CEC17 benchmark problems, consisting of different problem types and dimensions. Results of the FDB-LSHADE was compared to the performance of 8 other up-to-date and highly preferred metaheuristic search (MHS) algorithms. According to Friedman test results, the proposed FDBLSHADE algorithm ranked first among the all competing algorithms. Moreover, the proposed algorithm was used to solve single- and multi-objective energy hub economic dispatch (EHED) problems, which were a non-convex, a nonlinear, and high dimensional problems. To analyze the results of the proposed algorithm obtained from experimental studies, two non-parametric statistical methods, which are Wilcoxon and Friedman tests, were used. The simulation results of the proposed algorithm were compared to the results of the 8 other MHS algorithms. The results demonstrated that the FDB-LSHADE was a superior performance compared to other MHS algorithms for solving both benchmark and EHED problems.