Since 2004, Quintessa has carried out research and development in support of EDF Energy (and formerly British Energy) for the continued safe operation of Advanced Gas Cooled Reactors (AGRs). A critical part of the AGR safety case is demonstrating the continued physical integrity of the graphite cores that are made up of a large number of bricks.
Quintessa provides support in a number of key areas associated with the evolution of the reactor cores as they age, and predictions of the evolution of graphite properties with time. Advanced statistical methods have been used to enable the safe period of reactor operation to be calculated on the basis of the history of brick defects observed to date, and to determine the optimum future inspection strategy. These methods are applied to each of the reactors when they are shut down for routine outages.
The CoreStats code is designed to complement process-based approaches by focussing on what information on future behaviour can be obtained from historical data. Several statistical models are postulated, and the available historical data are used to determine which of these models are consistent with the observations. It would be possible simply to use the 'best' model, but it is better to work with several models that are consistent with the data but which may give different forecasts of future core behaviour, enabling uncertainties to be quantified. The QuinStats software was developed in order to fit such statistical models and calculate predictions including variability and uncertainty.
Statistical modelling methods are also used to model the evolution of graphite properties including weight loss.; As the reactors age some of the graphite is oxidised, resulting in increased porosity and lower density. The models employed are based on physical understanding of the key processes taking place and have been shown to have good predictive power by comparing model predictions made before samples are taken from the reactor with measurements made at inspection outages.
Quintessa has developed a custom software tool, CHANSELA, which is used to visualise information on the reactor cores and to support decisions on which fuel channels should be inspected to support continued safe operation. CHANSELA is able to suggest the combination of channels that will provide the most useful information subject to user-specified constraints, and to compare this with user-defined selections. The software employs some novel concepts, including a genetic algorithm for optimising channel selections, and allows every stage of the process to be documented, providing a clear and transparent audit trail.
Other custom tools developed for EDF Energy include ASPECT, used to visualise and interpret information collected from eddy current scanning of the graphite core, and LoTAS, a web-based application used for managing and analysing data recorded as fuel is moved in the reactor.
Quintessa are using COMSOL Multiphysics ® to construct a diverse assessment of the reactor brick evolution, largely independent of its own internal modelling approach, which can be used to predict the evolution of the shape of the bricks and the rate at which keyway root cracking will occur. The approach combines statistical models of the main model inputs, a nonlinear coupled mechanical model of the graphite, and the Monte Carlo approach to determine the cracking rates.
Image courtesy of EDF Energy
