Search

Gas circulator (GC) units are an important rotating asset used in the advanced gas-cooled reactor design, facilitating the flow of CO$_2$ gas through the reactor core. The ongoing maintenance and examination of these machines are important for operators in order to maintain safe and economic generation. GCs experience a dynamic duty cycle with periods of nonsteady state behavior at regular refueling intervals, posing a unique analysis problem for reliability engineers. In line with the increased data volumes and sophistication of available technologies, the investigation of predictive and prognostic measurements has become a central interest in rotating asset condition monitoring. However, many of the state-of-the-art approaches finding success deal with the extrapolation of stationary time series feeds, with little to no consideration of more complex but expected events in the data. In this paper, we demonstrate a novel modeling approach for examining refueling behaviors in GCs, with a focus on estimating their health state from vibration data. A machine learning model was constructed using the operational history of a unit experiencing an eventual inspection-based failure. This new approach to examining GC condition is shown to correspond well with explicit remaining useful life measurements of the case study, improving on the existing rudimentary extrapolation methods often employed in rotating machinery health monitoring. [ABSTRACT FROM PUBLISHER]

This paper presents a self-tuning framework for the diagnosis of routine alarms in steam turbine generators utilizing a combination of inductive machine learning and knowledge-based heuristics. The techniques provide a novel basis for initializing and updating time series feature extraction parameters used in the automated decision support of vibration events due to operational transients. The data-driven nature of the algorithms allows for machine-specific characteristics of individual turbines to be learned and reasoned about. The paper provides a case study illustrating the routine alarm paradigm, and the applicability of systems using self-tuning techniques. The approaches discussed throughout are presented to provide useful diagnosis tools for the reliability and maintenance analysis of steam turbine generators. [ABSTRACT FROM PUBLISHER]

• Dimensional changes in fuel channels can potentially show signs of defects. • Understanding defects in fuel channels helps support the safety case. • Estimations of dimensional changes can be produced using monitoring data. • Generalization and improved performance was shown using data from multiple reactors. One of the major life-limiting factors of an Advanced Gas-cooled Reactor (AGR) nuclear power station is the graphite core as it cannot be repaired or replaced and therefore detailed information about the health of the core is vital for continued safe operation. The graphite bricks that comprise the core experience gradual degradation during operation as a result of irradiation. Routine physical inspection of the graphite core fuel channels is performed by specialist inspection equipment during outages every 12 months to 3 years. It has also been shown to be advantageous to supplement this periodic inspection information with analysis of operational data which can provide additional insights into the core health. One such approach is through the use of online monitoring data called the Fuel Grab Load Trace (FGLT). An FGLT is a measure of the perceived load of the fuel assembly with contributions from aerodynamic forces and frictional forces, which is related to bore diameter. This paper describes enhancements to existing analysis of FGLT data which, to date, has focussed solely on using data from a single reactor at a time to build bore estimation models, by considering data from multiple reactors to produce a generalised model of bore estimation. This paper initially describes the process of producing a bore estimation from an FGLT by isolating the contribution that relates to the fuel channel bore and then discusses the limitations with the existing bore estimation model. Improvements are then proposed for the bore estimation model and a detailed assessment is undertaken to understand the effect of each of these proposed improvements. In addition, the effect of introducing non-linear regression models to further enhance the bore estimation is explored. The existing model is trained on data from one reactor in the UK and therefore the results produced from it are only applicable to this reactor. However, out of the remaining 13 nuclear reactors currently in operation, 3 also have a similar construction to the reactor the model is trained on, and these should all produce similar FGLT data. Therefore, a generalised model is proposed that produces bore estimations for four AGRs stations reactors, compared with one previously. It is shown that this approach offers an improved overall bore estimation model. [ABSTRACT FROM AUTHOR]