NRG4-U: a novel home energy management system for a unique loadprofile

ABSTRACT

Home Energy Management Systems (HEMSs) are developed to facilitate residential energy management toward a more efficient way while enhancing the residents’ energy awareness. Within this paper, a novel non-intrusive, advisory and low-cost home eNeRGy management system FOR a Unique load profile (NRG4-U) is proposed to generate a target personalized load profile for each user to manage energy while preserving comfortable living conditions. The user is able to monitor all systems applications through a visualization dashboard. NRG4-U individual modules gradually generate a user-based profile including preferences, routines, and energy base load. Thermal and visual comfort are estimated through a fusion algorithm for better results. Occupancy and activity inference are estimated by auto-generating an energy base-load exploiting hidden Markov models without using any manual pre-processing task or ground truth. Personalized recommendations are sent to alert the users for energy saving actions while motivating them to adapt to more efficient energy usage. The main source of generating energy savings is the real time recommendations that inform the user about changing the operational states of appliances that may not be necessary at the given time, or may be functioning for excessive time. The suggested HEMS tool (NRG4-U) has been deployed in residential environments, while the experimental results demonstrate that it is a compelling tool for personalized load profile and energy awareness. Results demonstrate that using the NRG4-U HEMS can result in average monthly energy savings of 7%.

KEYWORDS:

• Comfortable living condition
• energy management
• energy recommendation energy awareness
• energy saving
• personal energy profile

Acknowledgments

This work is partially supported by the PRECEPT project funded by the EU H2020 under Grant Agreement No. 958284.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Asimina Dimara

Asimina Dimara is a research assistant in CERTH/ ITI. She received her diploma in Informatics from Hellenic Open University (HOU) on July 2017. She is now (2021) attending her MSc by research in “Intelligent Computer Systems” in the Aegean University. Her main research interests include energy awareness, energy management, data analysis & analytics, machine learning, artificial and computational intelligence, neural networks, and time- series prediction. She has authored or coauthored three (3) papers in international scientific peer review journals and more than seventeen (17) papers in international and national conferences. She has also been involved in three (3) research projects funded by the EC and the Greek secretariat of Research and Technology.

Vasileios Georgios Vasilopoulos

Vasileios Georgios Vasilopoulos is a research assistant in CERTH/ITI. He received his diploma in Eletric and Computer Engineering from Aristotle University of Thessaloniki (AUTH) on March 2019. His main research interests include data analytics, embedded systems development and programming, internet of things technologies and electro-acoustics. He has authored or coauthored three (3) papers in international and national conferences. He has been involved in three (3) research projects funded by the EC and the Greek secretariat of Research and Technology.

Stelios Krinidis

Stelios Krinidis is an Associate Professor at International Hellenic University - Management Science and Technology Department and a postdoctoral research associate in CERTH/ITI. He received the Diploma degree and the Ph.D. degree in Computer Science from the Computer Science Department of the Aristotle University of Thessaloniki (AUTh), Thessaloniki, Greece, in 1999 and 2004 respectively. He has also served as an adjunct lecturer at the Aristotle University of Thessaloniki, the Democritus University of Thrace, and at the Technological Institute of Kavala during the period 2005-2012. His main research interests include computational intelligence, computer vision, pattern recognition, artificial intelligence, machine learning, signal processing and analysis, 2D and 3D image processing and analysis, data and visual analytics, etc. He has authored thirty-nine (39) papers in international scientific peer review journals and more than seventy (70) papers in international and national conferences. He has also been involved in more than twenty (20) research projects funded by the EC and the Greek secretariat of Research and Technology.

Dimitrios Tzovaras

Dimitrios Tzovaras is a Senior Researcher (Researcher A’) and the Director of the Informatics and Telematics Institute (ITI) at the Centre for Research and Technology Hellas (CERTH). His main research interests include visual analytics, 3D object recognition, search and retrieval, behavioral biometrics, assistive technologies, information and knowledge management, multimodal interfaces, computer graphics and virtual reality. He has been working as a Researcher since September 1999 and he has been involved in more than 180 projects, funded by the EC and the Greek Ministry of Research and Technology. His involvement with those research areas has led to the coauthoring of over 200 articles in refereed journals, more than 570 papers in international conferences and more than 55 book chapters. He has served as a regular reviewer for a number of international journals and conferences. He is a member of IEEE and EURASIP.