A hierarchical two-stage energy management for a home microgrid using model predictive and real-time controllers

Abstract

This paper presents a hierarchical two-layer home energy management system to reduce daily household energy costs and maximize photovoltaic self-consumption. The upper layer comprises a model predictive controller which optimizes household energy usage using a mixed-integer linear programming optimization; the lower layer comprises a rule-based real-time controller, to determine the optimal power settings of the home battery storage system. The optimization process also includes load shifting and battery degradation costs. The upper layer determines the operating schedule for shiftable domestic appliances and the profile for energy storage for the next 24 h. This profile is then passed to the lower energy management layer, which compensates for the effects of forecast uncertainties and sample time resolution. The effectiveness of the proposed home energy management system is demonstrated by comparing its performance with a single-layer management system. For the same battery size, using the hierarchical two-layer home energy management system can achieve annual household energy payment reduction of 27.8% and photovoltaic self-consumption of 91.1% compared to using a single layer home energy management system. The results show the capability of the hierarchical home energy management system to reduce household utility bills and maximize photovoltaic self-consumption. Experimental studies on a laboratory-based house emulation rig demonstrate the feasibility of the proposed home energy management system.