Swarm intelligence based State-of-Charge optimization for charging Plug-in Hybrid Electric Vehicles

Abstract

Transportation electrification has undergone major changes since the last decade. Success of the smart grid with renewable energy integration solely depends upon the large-scale penetration of Plug-in Hybrid Electric Vehicles (PHEVs) for a sustainable and carbon-free transportation sector. One of the key performance indicators in the hybrid electric vehicle is the State-of-Charge (SoC), which needs to be optimized for the betterment of charging infrastructure using stochastic computational methods. In this paper, a newly emerged accelerated particle swarm optimization (APSO) technique was applied and compared with standard Particle swarm optimization (PSO), considering charging time and battery capacity. Simulation results obtained for maximizing the highly non-linear objective function indicate that APSO achieves some improvement in terms of best fitness and computation time.

Intelligent sizing of a series hybrid electric power-train system based on Chaos-enhanced accelerated particle swarm optimization

“…PSO also has the capability of dealing with integer variables [14,21] and is widely used in intelligent sizing and multi-objective optimization in the automotive industry [21][22][23][24][25][26]. In order to accelerate PSO's convergence property, accelerated particle swarm optimization (APSO) is proposed, and evidences have showed that optimizing HEV with standard APSO outperforms the one with PSO [27,28]. Nevertheless, in real engineering practice, similar to most metaheuristic methods, APSO algorithm sometimes forces the agents to fall into local optima instead of global optima.…”