Hierarchical Energy Management System for Home Microgrids

Abstract

With the spread of distributed energy resources, sensing infrastructure, and automation facilities, modern homes are becoming “home microgrids.” This paper intends to support this trend and proposes a two-stage hierarchical energy management system for smart homes by considering both day-ahead and actual operation stages. In the day-ahead stage, an efficient scenario analysis approach is developed to account for residential photovoltaic solar power uncertainty. The approach performs solar power scenario generation and reduction based on the Wasserstein distance metric and K-medoids, respectively. This is then followed by the use of a stochastic day-ahead residential energy resource scheduling model. In the actual operation stage, a semi-scenario based rolling horizon optimization mechanism is proposed, based on which an actual operation model is established. Simulations are then conducted to validate the effectiveness of the proposed system.