Pattern Recognition–Based Technique for Control Rod Position Identification in Pressurized Water Reactors

Elsamahy, Mohamed; Nagla, Tarek F.; Abdel‐Rahman, Mohamed A.E.

doi:10.1080/00295450.2020.1792742

Cited by 6 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When it is desired to identify groupings of related items that can be regarded as belonging to a "category," cluster analysis algorithms are used in unsupervised pattern recognition [8]. Additionally, clustering can be used with variables to identify collections of related variables, typically with the goal of choosing the right variables for calibration methods.…”

Section: Cluster Analysis (Dendrogram)mentioning

confidence: 99%

Multivariate Image Analysis for Core Monitoring in PWRs

Mohsen,

Nagla,

Elsamahy

et al. 2023

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

In pressurized water reactor (PWR) it is crucial for the operator to monitor the reactor parameters at the same time, such as temperature, pressure, boron concentration, control rod position, coolant density, etc., in order to make proper decision. However, the huge size of data reading from the different instrumentations, in addition to the limited human ability to visually detect, interpret, assess add a lot of uncertainty to the operator qualitative and quantitative analysis of the reactor performance. Therefore, this paper proposes the utilization of radial thermal flux maps (Neutron Images) technique, positioned on the reactor core as a sensitive monitoring technique for all changes of the reactor parameters as a result of the position of the control rods changing. Hence, the features contained in these neutron images are extracted (Multivariate image analysis and regression) via Principal Component Analysis (PCA), and Cluster Analysis (Dendrogram). To determine the effectiveness of the suggested technique in determining the location of the control-rods, several simulations are run. The 3D TRITON FORTRAN-code was utilized to simulate the radial thermal neutron flux of the Westinghouse 2775-MWth PWR benchmark at 100% thermal power generation. The SIMCA software programme is used to develop, test, and generalise the PCA model. Additionally, clustering analysis (CA) is carried out using the statistics software programme Minitab in order to demonstrate the effectiveness of the suggested method.

show abstract

Section: Cluster Analysis (Dendrogram)mentioning

confidence: 99%