Open AccessDeterministic‐like solution to the non‐convex economic dispatch problem
Author(s) -
ElSayed Wael T.,
ElSaadany Ehab F.,
Zeineldin Hatem H.,
AlDurra Ahmed,
ElMoursi Mohamed S.
Publication year - 2021
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/gtd2.12031
Subject(s) - economic dispatch , mathematical optimization , artificial bee colony algorithm , benchmark (surveying) , computer science , spinning , local search (optimization) , core (optical fiber) , algorithm , power (physics) , electric power system , mathematics , engineering , mechanical engineering , physics , telecommunications , geodesy , quantum mechanics , geography
This paper proposes a novel stochastic solution method for the static nonconvex power economic dispatch problem. All practical features such as valve point effects, ramp rate limits, prohibited operating zones, multiple fuel options, spinning reserve constraints, and transmission losses are considered. To develop the proposed algorithm, two modifications that significantly enhance the exploitation capability of the artificial bee colony algorithm are introduced. The first modification concerns onlooker and employed bees to focus their search around the best solution found so far. The second modification optimizes the escaping behavior from local minimums using scout bees. The proposed modifications form a multi‐level concentric search around the best solution found. Finally, combining the modified artificial bee colony algorithm with Levy flight cycles, which improve the escaping capability from local minimums, leads to the proposed algorithm. The algorithm parameters have been tuned to provide a deterministic‐like solution with ten benchmark problems. The improvement added by each proposed modification to the artificial bee colony algorithm has been confirmed using the Wilcoxon rank‐sum test. The obtained results by the proposed algorithm are superior compared to those reported in the literature. Moreover, estimated probabilities of more than 99.9% to obtain the global optimal solution are achieved.