ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs

Abstract This paper presents a novel approach, ReactorNet, a machine learning framework leveraging thermal neutron flux imaging to enable real-time monitoring of pressurized water reactors (PWRs). By integrating EfficientNetB0 with a hybrid classification-regression architecture, the model accuratel...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmed Omar, Mohamed K. Elhadad, Moamen G. El-Samrah, Tarek F. Nagla, Tamer Mekkawy
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-13794-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849736207364259840
author Ahmed Omar
Mohamed K. Elhadad
Moamen G. El-Samrah
Tarek F. Nagla
Tamer Mekkawy
author_facet Ahmed Omar
Mohamed K. Elhadad
Moamen G. El-Samrah
Tarek F. Nagla
Tamer Mekkawy
author_sort Ahmed Omar
collection DOAJ
description Abstract This paper presents a novel approach, ReactorNet, a machine learning framework leveraging thermal neutron flux imaging to enable real-time monitoring of pressurized water reactors (PWRs). By integrating EfficientNetB0 with a hybrid classification-regression architecture, the model accurately identifies control rod positions and operational parameters through thermal neutron flux patterns detected by ex-core sensors. Principal Component Analysis (PCA) and Clustering Analysis decode radial flux variations linked to rod movements, while simulations of a 2772-MW(th) PWR using TRITON FORTRAN validate the framework. This framework outperforms Vision Transformers and ResNet50, achieving superior multi-class accuracy (97.5%) and reduced the mean absolute error (MAE) of regression. Test-Time Augmentation and cross-validation mitigate data limitations, ensuring robustness. This work bridges AI and nuclear engineering, demonstrating EfficientNetB0’s potential for precise, real-time reactor monitoring, enhancing operational safety and efficiency.
format Article
id doaj-art-f97dba9143b54f919f0bf2202a0f3a2e
institution DOAJ
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-f97dba9143b54f919f0bf2202a0f3a2e2025-08-20T03:07:20ZengNature PortfolioScientific Reports2045-23222025-08-0115111910.1038/s41598-025-13794-7ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRsAhmed Omar0Mohamed K. Elhadad1Moamen G. El-Samrah2Tarek F. Nagla3Tamer Mekkawy4Nuclear Engineering Department, Military Technical CollegeDepartment of Computer Engineering and AI, Military Technical CollegeNuclear Engineering Department, Military Technical CollegeNuclear Engineering Department, Faculty of Engineering, Alexandria UniversityAvionics Department, Military Technical CollegeAbstract This paper presents a novel approach, ReactorNet, a machine learning framework leveraging thermal neutron flux imaging to enable real-time monitoring of pressurized water reactors (PWRs). By integrating EfficientNetB0 with a hybrid classification-regression architecture, the model accurately identifies control rod positions and operational parameters through thermal neutron flux patterns detected by ex-core sensors. Principal Component Analysis (PCA) and Clustering Analysis decode radial flux variations linked to rod movements, while simulations of a 2772-MW(th) PWR using TRITON FORTRAN validate the framework. This framework outperforms Vision Transformers and ResNet50, achieving superior multi-class accuracy (97.5%) and reduced the mean absolute error (MAE) of regression. Test-Time Augmentation and cross-validation mitigate data limitations, ensuring robustness. This work bridges AI and nuclear engineering, demonstrating EfficientNetB0’s potential for precise, real-time reactor monitoring, enhancing operational safety and efficiency.https://doi.org/10.1038/s41598-025-13794-7Pressurized water reactor (PWR)Nuclear reactor safetyControl rod predictionEfficientNetB0Vision transformer (ViT)AI-driven reactor operations
spellingShingle Ahmed Omar
Mohamed K. Elhadad
Moamen G. El-Samrah
Tarek F. Nagla
Tamer Mekkawy
ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
Scientific Reports
Pressurized water reactor (PWR)
Nuclear reactor safety
Control rod prediction
EfficientNetB0
Vision transformer (ViT)
AI-driven reactor operations
title ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
title_full ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
title_fullStr ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
title_full_unstemmed ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
title_short ReactorNet based on machine learning framework to identify control rod position for real time monitoring in PWRs
title_sort reactornet based on machine learning framework to identify control rod position for real time monitoring in pwrs
topic Pressurized water reactor (PWR)
Nuclear reactor safety
Control rod prediction
EfficientNetB0
Vision transformer (ViT)
AI-driven reactor operations
url https://doi.org/10.1038/s41598-025-13794-7
work_keys_str_mv AT ahmedomar reactornetbasedonmachinelearningframeworktoidentifycontrolrodpositionforrealtimemonitoringinpwrs
AT mohamedkelhadad reactornetbasedonmachinelearningframeworktoidentifycontrolrodpositionforrealtimemonitoringinpwrs
AT moamengelsamrah reactornetbasedonmachinelearningframeworktoidentifycontrolrodpositionforrealtimemonitoringinpwrs
AT tarekfnagla reactornetbasedonmachinelearningframeworktoidentifycontrolrodpositionforrealtimemonitoringinpwrs
AT tamermekkawy reactornetbasedonmachinelearningframeworktoidentifycontrolrodpositionforrealtimemonitoringinpwrs