Research Specifications

This paper proposes stochastic energy management for an electrical system consisting of a multi-microgrid and a distribution company (DISCO). The objective function aims at finding an optimal day-ahead energy schedule of dispatchable resources, energy storages, and demand response in the presence of intermittent renewable energy sources. Two different energy storage devices, i.e., battery and supercapacitor (SC), are modeled in the system to provide both high power and energy density. Wind speed, solar irradiation, and electrical demand uncertainties have been modeled using probability density functions of uncertain parameters, and the number of scenarios has been decreased using mixed-integer linear programming (MILP) based scenario reduction with a novel characteristic. In the proposed model, the DISCO, as a leader, attempts to maximize its profit while the microgrids (MGs), as followers, tend to minimize their operational costs. Karush-Kuhn-Tucker, strong duality, and Fortuny-Amat transformation are used to convert the non-linear bi-level problem into a linear single-level problem. The proposed model was tested on a hypothetical distribution system and the results showed the efficiency of the model in reducing operational costs of MGs by 7.5%, increasing the DISCO profit by 6.9%, and improving the reliability of the system by reducing the energy not supplied.