Resilient & Secure Digital Distribution Grid

Domain #1
Sensitive infrastructure protection

Challenge 1.1. Development of cybersecurity solutions for sensitive infrastructure protection

Demonstrator

Demonstrator

Open Call #2 laureate

Description of the project

The electricity distribution grid is facing severe challenges due to the energy transition. A timely and accurate view of the current and future electrical behaviour of the electricity distribution grid is provided by the digital twin, which connects to heterogeneous IoT data sources. The resilience and security of the digital twin is essential, while challenged by the heterogeneous IoT connectivity and through zero-day attacks.

The RS2DG project will showcase the accurate and fast detection of faults and attacks on the digital twin and therefore will contribute to a smooth and secure energy transition. Major reliability and efficiency benefits will be demonstrated for the grid operator, for the operator of the digital twin which translate into the security and efficiency of the future electricity distribution grid.

Mid-term project update

The RS2DG project is assuring robustness of the digital twin of the electricity grid via novel methods based on machine learning.
The
project has identified two main usecases for anomaly and threat detection for the digital twin:
  • Missing data detection addressing the challenge of heterogeneous data collection for the digital twin with different update frequencies;
  • Identification of anomalies in values of energy measurements machine learning approaches will identify that unusual energy consumption data is provided by measurement devices; reasons for the latter can be wrong mountings of equipment (such as current transformers) or else malicious attacks on the measurement collection infrastructure.

The anomalies and attacks will be detected by a Resilience and Security component, which has been designed and implemented. A preliminary assessment of its integrated behavior has been obtained. The preliminary assessments demonstrate the potential of the solution to achieve a high true positive rate and to significantly reduce time to detection and effort for recovery from anomalous situations of the digital twin.

Final project update

The implementation of the energy transition poses strong challenges on the electricity distribution grid. New digital tools for identification and prediction of grid bottlenecks, for grid resilience and for grid extension planning are required. Due to the high dynamics over time and due to the high variability of consumption and generation behavior over different parts of the distribution grid, monitoring and planning of the grids must be able to use the true grid behavior and cannot rely on assumptions and standard load profiles any more. The Digital Twin of the electricity grid instead uses a combination of semi-static structural information about the grid topology together with electrical measurements from different grid locations and from different types of measurement devices and IT systems in order to automatically build up an model of the grid that accurately represents the time-series behavior of the true physical distribution infrastructure for planning. The trustworthiness of the result from the digital twin relies on the ability to detect anomalies in input measurement data fast.

The project RS2DG has integrated and demonstrated a software solution, called Security & Resilience (S&R) component, that was successfully demonstrated to detect cybersecurity threats as well as anomalies in the electrical measurements. Immediate alarms increase the resilience and the security of the digital twin of the electricity grid that is the basis for operations and planning of current and future the low and medium voltage grids.

Two project demonstrations have demonstrated the viability of the automated anomaly detection for input data, despite the challenging scenario of heterogeneous data sources for the Digital Twin. The assessment showed the benefit of the S&R component through three technical KPIs: accuracy of the detection was shown to be high and significant reductions of recovery times from data faults and attacks as well as significant reductions of operational efforts due to the novel S&R component were shown.

Workshop on security and resilience of digitalization in the distribution grid, Plattling, Germany – May 2024

Project consortium

GridData GmbH
GridData GmbH
🇩🇪
Germany
ResilTech S.R.L
ResilTech S.R.L
🇮🇹
Italy