Authors:
Katharina Legler
;
Muhammad Sheheryar Jajja
and
Klaus Volbert
Affiliation:
Faculty of Computer Science and Mathematics, Ostbayerische Technische Hochschule (OTH) Regensburg, Germany
Keyword(s):
Digital Twins, Internet of Things, Machine Learning Models, Data Visualization.
Abstract:
The energy crisis, energy demand growth, and dependence on fossil fuels worldwide have made urgent action necessary for us to seek sustainability in energy production and use. Digital technologies, especially Digital Energy Twins, have immense potential to reduce energy consumption, thereby reducing environmental impacts, particularly in the building sector. This paper presents the development of a digital energy twin that supports sustainable energy consumption analysis and optimization. Our study begins with a comprehensive analysis of the energy consumption data, the weather data, and the building plans as a solid basis for the analysis. We identify key energy consumption trends and patterns across different timescales and device-specific details that could be optimized, such as base load consumption and device-specific inefficiencies. A key part of our work is forecasting energy consumption using time series models, such as the ARIMA model, which promises to be useful in identify
ing patterns for improving energy efficiency. Overall, our study provides valuable insights into energy optimization and could form the base for further advances in digital energy twins at OTH Regensburg, helping to contribute to its sustainable development goals and smart campus initiatives.
(More)