Stochastic multi-layer optimization for cooperative multi-microgrid systems with hydrogen storage and demand response

Due to the growing demand for hydrogen and advancements in converting and utilizing various energy sources, hydrogen has garnered significant attention. This research presents a stochastic day-ahead scheduling method for a cooperative Multi-Microgrid (MMG). Additionally, this paper introduces a new application of the Social Network Search optimizer (SNS) for solving the energy management (EM) problem. Furthermore, a stochastic sizing for Hydrogen Storage is proposed. The proposed energy management system (EMS) consists of three layers aimed at optimizing economic, environmental, and technical objectives simultaneously. First, the economic layer is simulated to minimize operating and transaction costs. The simulation results of this layer based on SNS algorithm are compared with other algorithms such as African vultures optimization algorithm (AVOA), Northern goshawk optimization (NGO), Student psychology based optimization algorithm (SPBO), Artificial Rabbits Optimization (ARO), and Equilibrium optimizer (EO). The results demonstrate the superior performance of the SNS algorithm in solving the EM problem. Next, the SNS algorithm is employed to determine the optimal sizing of Hydrogen Energy Storage System (HESS) integrated into the EMS of the MMG. Additionally, in the first layer, the operating costs are fairly allocated among the cooperative MMG based on the Shapely value. The results indicate that the cooperation among the MMG systems leads to an 11.13% reduction in operating costs. The developed Multi-layer multi-objective problem (MLMO) adopts a hybrid weighted-sum Ε-lexicography approach to eliminate the need for normalizing different objectives. In the second and third layers, efforts are made to minimize greenhouse gas (GHG) emissions and maximize peak load reduction (PLR) to enhance the reliability of the MMG. The stochastic nature of renewable energy sources (RESs), load demand, and energy prices are addressed using Monte Carlo simulation (MCS) and backward reduction technique (BRT). The simulation results for a MMG test system demonstrate a 4.3% reduction in GHG emissions in the optimization layer and a 4.45% enhancement in PLR in the respective layer. © 2024 Hydrogen Energy Publications LLC