Metaheuristic optimization techniques to design solar-fuel cell-battery energy system for locomotives

Abstract

This work proposes a design methodology of a Solar-Fuel Cell-Battery Energy System (SFCBES), which is proficient in supplying the energy requirements of a passenger train and with least cost, decide ratings of fuel cell pack and battery. The prospective work considers Indian Railways passenger locomotive with its real-time driving cycles. A design optimization method is articulated to solve cost minimization objective function, under the constraints such as fuel cell capacity, battery SOC limits, and average power demand of the locomotive. Detailed modelling of the hybrid power sources and locomotive is done and estimated for two practical driving cycles of the locomotive. Various cases including two intelligent energy management methods, three gradients, a hybrid power system with and without solar panels are considered. In this work, six metaheuristic algorithms namely Artificial Bee Colony (ABC), Differential Evolution (DE), Firefly Algorithm (FA), Particle Swarm Optimization (PSO), Invasive Weed Optimization (IWO), and Teaching Learning Based Optimization (TLBO) are employed in the solution of cost minimization problem. The results declared that, among various algorithms considered, TLBO gave a better performance with minimum cost and earlier convergence among the cases considered.