Leveraging the Potential of Novel Data in Power Line Communication of Electricity Grids

Christoph Balada,Max Bondorf,Sheraz Ahmed,Andreas Dengela,Markus Zdrallek

2023-09-08

Abstract:Electricity grids have become an essential part of daily life, even if they are often not noticed in everyday life. We usually only become particularly aware of this dependence by the time the electricity grid is no longer available. However, significant changes, such as the transition to renewable energy (photovoltaic, wind turbines, etc.) and an increasing number of energy consumers with complex load profiles (electric vehicles, home battery systems, etc.), pose new challenges for the electricity grid. To address these challenges, we propose two first-of-its-kind datasets based on measurements in a broadband powerline communications (PLC) infrastructure. Both datasets FiN-1 and FiN-2, were collected during real practical use in a part of the German low-voltage grid that supplies around 4.4 million people and show more than 13 billion datapoints collected by more than 5100 sensors. In addition, we present different use cases in asset management, grid state visualization, forecasting, predictive maintenance, and novelty detection to highlight the benefits of these types of data. For these applications, we particularly highlight the use of novel machine learning architectures to extract rich information from real-world data that cannot be captured using traditional approaches. By publishing the first large-scale real-world dataset, we aim to shed light on the previously largely unrecognized potential of PLC data and emphasize machine-learning-based research in low-voltage distribution networks by presenting a variety of different use cases.

Machine Learning,Artificial Intelligence,Systems and Control

What problem does this paper attempt to address?

The paper attempts to address a series of complex challenges encountered during the transition of power systems to smart grids. Specifically, with the widespread adoption of renewable energy sources (such as photovoltaic power generation, wind power generation, etc.) and the emergence of an increasing number of energy consumers with complex load characteristics (such as electric vehicles, home battery systems, etc.), traditional methods are proving inadequate in coping with these changes. To bridge this gap, the paper proposes using machine learning (ML) to address these issues through data-driven approaches.

The main contributions of the paper include:
1. **Release of the FiN-2 dataset**: This is the first large-scale real-world broadband power line communication (PLC) dataset, containing over 13 billion data points, collected by more than 5100 sensors, covering a part of the low-voltage grid in Germany, and serving over 4.4 million people. This dataset aims to address the limitations of the FiN-1 dataset in terms of spatial coverage and the number of sensors.
2. **Demonstrating the advantages of data-driven methods**: Through the FiN-1 and FiN-2 datasets, the paper demonstrates the advantages of data-driven methods in various application areas such as asset management, grid state visualization, forecasting, predictive maintenance, and anomaly detection.
3. **Exploring novel machine learning architectures**: The paper particularly emphasizes the ability of novel machine learning architectures to extract rich information from real-world data, which cannot be obtained through traditional methods.

In summary, the paper aims to advance smart grid research, particularly in the application of low-voltage distribution networks, by releasing large-scale real-world datasets and demonstrating the application of data-driven methods.