Home > Research > Publications & Outputs > Prediction of Fuel Debris Location in Fukushima...

Research

Links

Text available via DOI:

View graph of relations

Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning

Research output: Contribution to Journal/Magazine › Conference article › peer-review

Published

Standard

Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning. / Alrawash, Saed; Hale, Matthew; Lennox, Barry et al.
In: EPJ Web of Conferences, Vol. 302, 15.10.2024.

Research output: Contribution to Journal/Magazine › Conference article › peer-review

Harvard

Alrawash, S, Hale, M, Lennox, B, Joyce, M, West, A, Watanabe, M, Zhang, Z & Aspinall, M 2024, 'Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning', EPJ Web of Conferences, vol. 302. https://doi.org/10.1051/epjconf/202430217004

APA

Alrawash, S., Hale, M., Lennox, B., Joyce, M., West, A., Watanabe, M., Zhang, Z., & Aspinall, M. (2024). Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning. EPJ Web of Conferences, 302. https://doi.org/10.1051/epjconf/202430217004

Vancouver

Alrawash S, Hale M, Lennox B, Joyce M, West A, Watanabe M et al. Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning. EPJ Web of Conferences. 2024 Oct 15;302. doi: 10.1051/epjconf/202430217004

Author

Alrawash, Saed ; Hale, Matthew ; Lennox, Barry et al. / Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning. In: EPJ Web of Conferences. 2024 ; Vol. 302.

Bibtex

@article{fe4cb13b7fb94cc5ba9c1e4ae771fb3c,
title = "Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning",
abstract = "Accurate fuel debris location is crucial part of the decommissioning of the Fukushima Nuclear Power plants. Conventional methods face challenges due to extreme radiation and complex structure of the materials involved. In this study, we propose a novel approach utilising neutron detection and machine learning to estimate fuel material location. Geant4 simulations and pythonTM scripts have been used to generate a comprehensive dataset to train a machine learning model using MATLAB{\textquoteright}s regression learner. A Gaussian Process Regression model was chosen for training and prediction. The results show excellent prediction performance to estimate the corium thickness effectively and to locate the nuclear fuel material with a mean square error (MSE) of 0.01. By combining the machine learning with nuclear simulation codes, this promises to enhance the nuclear decommissioning efforts to retrieve nuclear fuel debris. ",
author = "Saed Alrawash and Matthew Hale and Barry Lennox and Malcolm Joyce and Andrew West and Minoru Watanabe and Zhongming Zhang and Michael Aspinall",
year = "2024",
month = oct,
day = "15",
doi = "10.1051/epjconf/202430217004",
language = "English",
volume = "302",
journal = "EPJ Web of Conferences",
issn = "2100-014X",
publisher = "EDP Sciences",
note = "SNA + MC 2024 : Joint International Conference on Supercomputing in Nuclear Applications + Monte Carlo ; Conference date: 20-10-2024 Through 24-10-2024",
url = "https://www.sfen.org/evenement/sna-mc-2024/",

}

RIS

TY - JOUR

T1 - Prediction of Fuel Debris Location in Fukushima Nuclear Power Plant using Machine Learning

AU - Alrawash, Saed

AU - Hale, Matthew

AU - Lennox, Barry

AU - Joyce, Malcolm

AU - West, Andrew

AU - Watanabe, Minoru

AU - Zhang, Zhongming

AU - Aspinall, Michael

PY - 2024/10/15

Y1 - 2024/10/15

N2 - Accurate fuel debris location is crucial part of the decommissioning of the Fukushima Nuclear Power plants. Conventional methods face challenges due to extreme radiation and complex structure of the materials involved. In this study, we propose a novel approach utilising neutron detection and machine learning to estimate fuel material location. Geant4 simulations and pythonTM scripts have been used to generate a comprehensive dataset to train a machine learning model using MATLAB’s regression learner. A Gaussian Process Regression model was chosen for training and prediction. The results show excellent prediction performance to estimate the corium thickness effectively and to locate the nuclear fuel material with a mean square error (MSE) of 0.01. By combining the machine learning with nuclear simulation codes, this promises to enhance the nuclear decommissioning efforts to retrieve nuclear fuel debris.

AB - Accurate fuel debris location is crucial part of the decommissioning of the Fukushima Nuclear Power plants. Conventional methods face challenges due to extreme radiation and complex structure of the materials involved. In this study, we propose a novel approach utilising neutron detection and machine learning to estimate fuel material location. Geant4 simulations and pythonTM scripts have been used to generate a comprehensive dataset to train a machine learning model using MATLAB’s regression learner. A Gaussian Process Regression model was chosen for training and prediction. The results show excellent prediction performance to estimate the corium thickness effectively and to locate the nuclear fuel material with a mean square error (MSE) of 0.01. By combining the machine learning with nuclear simulation codes, this promises to enhance the nuclear decommissioning efforts to retrieve nuclear fuel debris.

U2 - 10.1051/epjconf/202430217004

DO - 10.1051/epjconf/202430217004

M3 - Conference article

VL - 302

JO - EPJ Web of Conferences

JF - EPJ Web of Conferences

SN - 2100-014X

T2 - SNA + MC 2024

Y2 - 20 October 2024 through 24 October 2024

ER -