Optimal sizing of grid connected multi-microgrid system using grey wolf optimization

Abstract

Renewable distributed energy resources (DERs) offer a promising and environmentally sustainable solution for providing energy. Nowadays, there has been significant attention in wind, solar Photovoltaic (PV), and hydrogen-based fuel cell (FC) systems due to their ability to provide cost-effective energy to replace conventional generations. However, the intermittent nature of renewable energy sources presents challenges and operational issues for fully renewable energy systems. To address this, integrating energy storage systems and effectively managing uncertainties related to both load and generation resources are crucial for mitigating such challenges. This paper proposes a hybrid grid-connected PV-wind-FC generation-based Multi-microgrid (MMG) system integrated with a Battery Energy Storage System (BESS) to meet the entire load demand of the adopted MMG-based IEEE 14-bus system. The aim is to ensure cost-effectiveness and enable energy trading with the main grid by optimizing system configurations. The study incorporates stochastic analysis to handle uncertainties related to load, meteorological data, and energy prices to optimize the configuration of DERs and BESS in the MMGs. A Grey Wolf Optimization (GWO) algorithm is employed to determine the optimal sizing of the proposed grid-connected MMG. The proposed algorithm has reduced the NPC from $431.796 million to $428.832 million and LCOE to 0.267$/kWh when load and generation data uncertainty and dynamic energy price has considered. The robustness of the proposed approach is evaluated by comparing results with those obtained using Particle Swarm Optimization (PSO) and JAYA algorithms. The GWO method demonstrates superior performance, resulting in lower total Net Present Cost (NPC), lower system capacity, and a lower Levelized Cost of Energy (LCOE) compared to its counterparts. Moreover, the GWO algorithm exhibits the fastest convergence, indicating its accuracy and robustness compared to PSO and JAYA algorithms.