(GRS-764) AnTeS-NeCom - Analysis of the Failure Propagation in the Network Communication of Digital I&C Systems with a Test System

Förderkennzeichen RS1590

The research and development project RS1590, funded by the Federal Ministry for Economic Affairs and Energy (BMWi) and later by the Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection (BMUV), was dedicated to investigating the impact of communication failures in the networks of digital instrumentation and control (I&C) systems in nuclear power plants. This project builds upon previous and, in part, concurrent initiatives in which the Gesellschaft für Anlagen- und Reaktorsicherheit (GRS) developed fundamental, model-based methods for analyzing the behavior of digital I&C in the event of failures (funding codes 3615R01343, 4718R01314, 4722R01215). A crucial component of this research work was the development of the Analysis and Testing System AnTeS, which includes both real and simulated I&C systems.

The methods applied within the framework of AnTeS include Failure Mode and Effects Analyses (FMEAs), automated impact analyses (an automation and extension of FMEA developed by GRS), Fault Tree Analyses (FTAs), and Monte Carlo simulations. These methods serve to identify and evaluate potential causes of failures and their impacts. 

Modern network technologies and topologies, used for both internal and external communication in I&C systems, also play a central role. The influence of these technologies on the reliability and safety of the systems was specifically examined in this project to address gaps in the existing methods and in the application of AnTeS.

The primary goal of the project was to develop an in-depth understanding of network communication within I&C systems. This included the development of methods for fault injection into network communication and the subsequent examination of the impacts of such failures on the reliability of various model systems. For these investigations, existing model systems were expanded, and new systems were designed and analyzed.

Through sensitivity analyses, the impact of different parameters on system reliability was evaluated. The project's findings indicate that digital I&C systems in nuclear power plants are highly robust against network failures, and that these failures only have a marginal impact on the overall reliability of the systems. These insights contribute significantly to the further development of the GRS methodology.