Optimal Energy Management for Networked Microgrids via Distributed Neurodynamic

Abstract

Optimal energy management of a power system enables the economic operation of grids. However, traditional centralized methodologies face limitations when dealing with large-scale power systems. To overcome these challenges, an optimal energy management model is developed for networked microgrids (NMGs), and a distributed neurodynamic optimization algorithm is proposed for the optimization model. Before facilitating power exchange between neighboring microgrids (MGs), a strict priority scheme for power distribution is outlined. Distributed power generation takes precedence in meeting load requirements, followed by charging and discharging operations. Furthermore, to ensure the stability of the overall grid, power transmission is established between neighboring MGs. To solve the optimization problem, a distributed recurrent neural network is developed, and it is demonstrated to converge towards an optimal solution. An illustrative example involving 3-MG NMGs is elaborated to demonstrate the validity and effectiveness of the proposed algorithm.