Smart Power/Energy Management and Optimization in Microgrids

Abstract

Smart energy management of community-based microgrids (CMGs) can improve the performance of distribution networks (DNs) from many facets including flexibility, resiliency, and reliability. However, due to the continuous increase in energy demands and electricity prices, as well as the rising need for a transition toward flexible, decentralized, and decarbonized power systems, developing decent methods for sizing community energy storage (CES) systems is essential for smart and efficient energy management of CMGs. As a result, in this chapter, a machine learning–based approach for CES system sizing is proposed, in which the size of the CES system is determined via forecasting the load consumption and rooftop photovoltaic (PV) generation in the community while circumventing the need for acquiring household energy meter data. Moreover, by taking advantage of the forecasted data, the system economics and the cost-minimization analysis are performed using Hybrid Optimization Model for Multiple Energy Resources (HOMER) software. To evaluate the performance and applicability of the proposed method, a real energy community in Konstanz (a city in Germany) is selected as a case study. The results show that two batteries, each with 14.4 kWh capacity, are adequate to store the extra energy generated by rooftop PVs and to deliver the energy to the community when needed, leading to a 60.2% augmentation in PV penetration in the community. Therefore, proper sizing of CES systems via the proposed machine learning–based method may contribute to creating more economical, green, and self-sufficient energy communities, causing a considerable reduction in air pollution emissions and investment costs linked to DNs' reinforcement and expansion.