Modeling a digital twin for the optimization of a self-supply energy system for residential use
The climate situation and the energy crisis have prompted a number of policies and strategies that foster the adoption of renewable energy sources. To tackle the intermittency and fluctuations associated with the operation of these sustainable energy sources, renewable hydrogen appears as an appeali...
Saved in:
Published in | 2024 IEEE International Systems Conference (SysCon) pp. 1 - 8 |
---|---|
Main Authors | , , , , , |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
15.04.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The climate situation and the energy crisis have prompted a number of policies and strategies that foster the adoption of renewable energy sources. To tackle the intermittency and fluctuations associated with the operation of these sustainable energy sources, renewable hydrogen appears as an appealing solution to decarbonize different economic sectors. In this sense, the design and implementation of a hybrid renewable energy-hydrogen system has led to the first electrically self-sufficient social housing in Spain, located in the town of Novales (Cantabria). On the other hand, the digitization of this type of self-sufficient systems would allow automatic adaptation to changing situations, increasing energy efficiency. In this context, we introduce the design and initial implementation phases of a digital twin architecture that, using machine learning and artificial intelligence techniques, facilitates the optimization of the performance of the physical system by interacting with its control components. This involves the use of telemetry solutions that allow the capture and storage of data from the physical system itself, as well as from the environment, such as instance meteorological data. We also discuss some initial results of the digital twin, which features models of the electrical components of the physical system, based on both their logical behavior and machine learning techniques. |
---|---|
ISSN: | 2472-9647 |
DOI: | 10.1109/SysCon61195.2024.10553483 |