Researchers from IT4Innovations National Supercomputing Center, part of VSB – Technical University of Ostrava, have, together with partners, developed a system that enables more precise analysis of the risk of outages in electrical and communication networks in the Czech Republic. Using open-source software and Barbora and Karolina supercomputers, they can quickly simulate long-term network operation, identify critical points, and plan maintenance to reduce the risk of large-scale failures.
Interconnected energy and communication networks form the backbone of modern society. Their failure can lead not only to power and internet outages, but also to financial losses and threats to public safety. Existing approaches have relied on commercial tools, which limit transparency and further development. The new methodology developed by Czech researchers offers an open, reproducible, and scalable solution that enables detailed simulation of both individual components and the entire infrastructure in real time.
The methodology uses open-source R packages. The FaultTree package is used to clearly visualise the network structure and provide an initial estimate of failure probabilities. A key role is then played by the innovative use of the Ftaproxim package, which tracks the status of individual components – including ageing, failures, and repairs – and systematically evaluates all possible outage scenarios.
The researchers validated the method's performance and accuracy on the Barbora supercomputer, where they tested various computational scenarios and fine-tuned the algorithm's parameters. The results were compared with those obtained using the solution implemented in the commercial MATLAB software. “Ftaproxim uses so-called proxel-based simulation, which tracks all possible changes in the system state in discrete time steps. Each component is described by two probabilistic models: one representing the time to failure and the other the repair duration. This allows us to identify which parts of the infrastructure have the greatest impact on system reliability and to focus preventive measures on them,” explains Michal Běloch from IT4Innovations.
Simulations of long-term network operation have shown that, with appropriate configuration of computational parameters, it is possible to significantly reduce computation time while maintaining accuracy comparable to commercial tools.“By selecting the right combination of parameters, we can greatly speed up simulations while preserving high result accuracy. Calculations that would normally take several days have been reduced to just a few hours or even minutes, with only minimal deviation from commercial tools. The combination of open-source software, supercomputing power, and modelling of interconnected energy and communication networks therefore represents an important step forward in the reliability analysis of critical infrastructure,” explains Pavel Praks from IT4Innovations. The model enables faster testing of different failure scenarios and more efficient operational planning of critical infrastructure.
The research team, consisting of authors from IT4Innovations, the Faculty of Electrical Engineering and Computer Science at VSB – Technical University of Ostrava, and Brno University of Technology, has also developed computational code optimised for the Karolina and Barbora supercomputers. This code enables larger-scale simulations and systematic testing across a wide range of scenarios.
Researchers from the above-mentioned institutions have also developed a system to optimise preventive maintenance for interconnected energy and communication networks. Unlike the conventional approach, which reacts only after a failure occurs, this method can recommend which components should be serviced and when to minimise outages while simultaneously reducing costs. “Using advanced mathematical models and time-dependent simulations, we take into account equipment ageing, repair times, and the probability of failure of individual components. This makes it possible not only to monitor risks, but also to actively plan network operation and maintenance interventions so that the infrastructure remains reliable even during long-term operation affected by failures,” adds Daniel Krpelík from IT4Innovations.
This approach enables testing different failure and maintenance scenarios, identifying optimal strategies, and balancing reliability and cost efficiency, which is essential for modern interconnected networks, where the failure of a single component can affect the operation of the entire infrastructure.
Research articles
Evaluating the unavailability of interconnected power and communication networks with open-source tools on a petascale cluster
https://doi.org/10.1177/01445987251377791
Multi-objective optimization of smart grid operations via preventive maintenance scheduling using time-dependent unavailability
https://doi.org/10.1016/j.ress.2025.111567
The research was carried out with the support by the European Union in the Increasing the resilience of power grids in the context of decarbonisation, decentralisation, and sustainable socioeconomic development project (CZ.02.01.01/00/23_021/0008759) under the OP JAC, and and the Czech Ministry of the Interior under the OPSEC Programme within the Research of a Holistic Model of Interconnected Critical Power and Communication Infrastructure project, registration number VK01030109.
