Computational modelling for advanced nuclear power plants

Engineers are building computer models that can predict exactly what happens inside a nuclear power plant during normal operation, a fault, or even a severe accident. These models matter because the safety of nuclear power depends on knowing how the reactor and its systems will behave under every possible condition. Current computational tools exist, but they need rigorous testing against real-world measurements before operators and regulators can trust them fully. Without validated models, engineers must rely more heavily on physical experiments, which are expensive, time-consuming, and cannot cover every scenario. This project will create a systematic framework for verifying and validating those computer models against well-defined benchmark cases. It will also develop advanced computational methods specifically for analysing behaviour during fault conditions and severe accidents. If successful, the research will give the nuclear industry predictive tools that are robust, efficient, and proven to be accurate. That could make it easier to design safer reactors, reduce the need for costly physical tests, and help regulators make faster, more confident decisions about plant safety. For the public, the direct benefit is a clearer guarantee that nuclear power stations pose minimal hazard—without requiring anyone to think about the complex calculations running behind the scenes.