A Close-to-Real-time Energy Management System for Smart Residential Buildings

This paper proposes a close to real-time energy management system for multi-family residential buildings. The aim of the model is to optimally manage the controllable energy sources in a rolling time window to reduce the total energy cost of the building, considering the time variable electricity prices. A mixed-integer linear programming model is developed to control the washing machines, dishwashers and the energy storage system in a smart building. The performance of the model is evaluated through simulating the operation of an energy management system during one day using the measurement data from a real building located within Chalmers University of Technology campus. The results show that the proposed model can help reduce the building's energy cost by 9% and reduce the peak demand 18.78% using the resource flexibilities in the building, without causing discomfort for the tenants.