Multi-Layer Model Predictive Optimization of Energy Efficient Building Microgrids

Abstract

This paper introduces a multi-layer model predictive optimization (mLMPO) framework for energy management of building microgrids with Internet of Things (IoT)-enabled dispatchable loads and Distributed Energy Resources (DERs). The goal is to achieve high energy efficiency and demand response capability, while satisfying occupants’ comfort. Due to the diversity of on-site resources and complexity of occupancy modeling, traditional building management systems (BMS) cannot always optimize energy efficiency and maintain occupant comfort simultaneously. This paper will address this gap and develop a new framework for implementing mLMPO in building microgrids. The data from a large academic building in California is used for simulation studies. The results of this paper can provide a road map for co-optimization of energy efficient and occupants comfort in IoT-enabled smart buildings and microgrids.

Keywords

• Building microgrid
• energy efficiency
• IoT-enabled dispatchable loads
• multi-layer model predictive optimization
• sensitivity analysis