Optimal hybrid energy system for locomotive utilizing improved Locust Swarm optimizer

A novel methodology for optimum sizing of a hybrid energy (HE) system is presented in this paper to supply the driving force of a locomotive. The HE system includes a lithium-ion battery along with a polymer electrolyte membrane (PEM) fuel-cell. The idea behind this paper is to minimize the HE system's total cost under the PEM fuel-cell state of charge (SoC) constraint and capacity constraint of the battery. The minimization in this study is performed by an improved version of the Locust Swarm (LS) optimization algorithm (ILS). The algorithm uses oppositional learning and chaos mechanism to resolve the premature convergence and speed of the algorithm along with escaping from the local optimum point. The results of the final case study have been done for analyzing the locomotive speed demand, the average power demand, and the locomotive slant. A comparison of the outcomes of the suggested ILS algorithm with the standard LS algorithm and Particle swarm optimizer (PSO) from the literature and the results showed that in a maximum slope (2%), the total cost of the HE system for the suggested ILS algorithm, the basic LS algorithm, and the PSO algorithm are 3.8 x 10(6) $, 4.43x10(6) $, and 4.86 x 10(6) $, respectively which indicated that the achieved overall expense for the suggested ILS gives the best amount and the results are carried out to verify the superiority of the proposed method in solving a challenging real-world problem. (C) 2020 Elsevier Ltd. All rights reserved.

Automated summary

A novel methodology for determining the optimal size of a hybrid energy system is introduced in this paper, aiming to minimize the total cost of the system while meeting the constraints of fuel cells and battery capacity. The proposed improved algorithm addresses the issues of traditional algorithms through oppositional learning and chaos mechanism, and the analysis results show that the proposed algorithm outperforms in tackling real-world problems.