Search in the site by keyword

reports - Deliverable

Development, testing, and analysis of the results of using residential energy management systems across varying consumption profiles in different contexts

reports - Deliverable

Development, testing, and analysis of the results of using residential energy management systems across varying consumption profiles in different contexts

The report outlines the development of an energy management system called SEM (Smart Energy Manager), which uses Machine Learning (ML) and Model Predictive Control (MPC) techniques to optimize the energy consumption of a building. The effectiveness of SEM was evaluated through tests in a specially equipped laboratory, and its economic and environmental benefits were assessed in various climatic contexts compared to a conventional control system.

This work is part of ongoing research to find efficient solutions for reducing primary energy consumption in residential “building-plant-user” systems, particularly in the operation of climate control systems coupled with renewable energy generation. To achieve this, a three-year project was conducted to identify and experimentally validate solutions for energy management in buildings. These solutions are designed to adapt to the changing conditions of the system through appropriate control techniques and optimize energy consumption while considering comfort, energy efficiency, climate conditions, and user behavior.

The work presented in this report relates to the third year of activity and focuses on the final development of the energy management system called SEM – Smart Energy Manager. The report describes how the management system software was structured, the simplified physical models implemented, and the tests conducted at RSE’s EffE (Energy Efficiency) Laboratory to verify the correct operation of the SEM system and quantify its economic and environmental potential compared to a conventional control system (chronothermostat). The experiments showed that SEM increased self-consumption in a home equipped with photovoltaic panels and a heat pump, depending on weather conditions and user comfort requirements, from 17-19% with a traditional chronothermostat to 31-35%. Additionally, the SEM system was transferred onto an embedded platform to create a demonstrative device, which functions as a modern intelligent chronothermostat with a touch display.

To estimate the potential benefits of the SEM system on energy consumption in Italian residential buildings, the MODENA (Energy Model for Dwellings) tool was developed. This tool uses a range of sector-specific statistical information, including data from the efficient appliances market. Evaluations conducted for three representative cities (Milan, Rome, Palermo) estimated that using SEM could lead to energy cost savings of 10% to 14% compared to a traditional chronothermostat.

A specific consumer opinion survey on residential comfort confirmed the need for greater awareness of efficiency and good energy-saving practices. The creation of the interactive system AIACE (Integrated Activities for Comfortable and Efficient Housing), accessible online, addresses this need by allowing users to virtually navigate a 3D building and view explanatory videos, providing detailed information on the technologies tested at the EffE Laboratory.

Projects

Comments