Vodice The Imperial Park Hotel 26 - 28 March 2025
27 th International Conference on Heating, Cooling and Air-conditioning
ABSTRACT
PAPER
Vasilis CHRISTODOULIDES Frederick University Y. Frederickou 7, Pallouriotisa Nicosia 1036 Cyprus e-mail: st025542@stud.frederick.ac.cy
MP4
VIDEO
He is a student in mechanical engineering at the Frederick University in Nicosia (Cyprus).
Dr.-Ing. Paris A. FOKAIDES Frederick University, Nicosia, Cyprus Nicholas AFXENTIOU Frederick University, Nicosia, Cyprus
UTILIZING DIGITAL TWIN TECHNOLOGY FOR IMPROVED PREDICTIVE MODELLING OF ENERGY CONSUMPTION IN SMART BUILDINGS
The building sector is responsible for approxi- mately 30 - 34% of global energy consumption, highlighting its pivotal role in sustainability and energy efficiency. Digital Twin (DT) technology has emerged as a game-changer in smart build- ing applications, offering real-time monitoring, analysis, and optimization capabilities. A Digital Twin is a dynamic virtual model of a physical system, integrating sensor data, Building Information Models (BIM), and Internet of Things (IoT) technologies to simulate energy usage pat- terns. This enables the identification of inefficien- cies, optimization of energy consumption, and forecasting of future energy needs based on en- vironmental conditions, occupancy patterns, and equipment performance. This study develops a predictive energy consump- tion model utilizing DT technology for Frederick University's campus buildings. Energy data from 2021 to 2024, collected via smart meters and sensor networks, was analyzed using advanced predictive algorithms - XGBoost, SARIMAX, and
Prophet. Extensive feature engineering tech- niques, including time-based pattern recognition, rolling averages, and data normalization, were implemented to enhance model accuracy. The dataset was divided into training, validation, and testing subsets to ensure robust performance evaluation. A comprehensive literature review explores the opportunities and challenges of DT integration in building energy management. Key barriers in- clude interoperability constraints, cybersecurity concerns, and the seamless incorporation of re- newable energy systems. Despite these challeng- es, the potential of DTs to transform energy man- agement through real-time insights and predic- tive analytics is substantial. The findings of this research offer practical in- sights into optimizing energy consumption and advancing the deployment of DT technology in smart buildings, contributing to a more sustain- able and intelligent built environment.
22
2nd Thematic Section - FRAN BOŠNJAKOVIĆ DAY
Powered by FlippingBook