Improved Pelican optimization algorithm for solving load dispatch problems

Abstract

The Pelican Optimization Algorithm (POA) is a newly developed algorithm inspired by the hunting behavior of pelicans. Despite its fast convergence rate, it suffers from premature convergence, the imbalance between exploration and exploitation, and lack of population diversity. In this work, an improved POA is proposed to attenuate these shortcomings. IPOA benefits from three motion strategies and predefined knowledge-sharing factors that better describe the stochastic hunting behavior of pelicans, as well as a modified dimension-learning-based hunting (DHL) behavior to retain diversity. To test the effectiveness of these improvements, it was used to solve 23 benchmark functions, including unimodal, multimodal, and 6 composite functions (CEC 2017). To evaluate performance, it was applied to solve economic and combined economic emission load dispatch problems that play a critical role in real-world power system planning and operation while considering environmental impacts. This experiment includes 6, 10, 11, 40, 140, 160, and 320 generating units with nonconvex and non-smooth objective functions. The comparison is performed for benchmark functions and the optimal dispatch problems. The results confirm the competitive and almost superior performance of the IPOA in several cases, which proves the applicability and efficiency of the proposed approach in solving real-world problems.