Home // SIMUL 2025, The Seventeenth International Conference on Advances in System Modeling and Simulation // View article

Algorithm for Predicting Radioactivity of Decommissioning Nuclear Power Plant

Authors:
Changyeon Yoon

Keywords: EPRI algorithm, SOR, Gauss-Jordan Elimination

Abstract:
During the decommissioning of nuclear power plants, it is essential to accurately predict worker radiation exposure, which is a key requirement for regulatory approval and ensuring radiation safety. However, access to high-radiation areas is often restricted, making it difficult to obtain direct measurements and leading to significant limitations in dose assessment. To address this issue, we developed a new algorithm based on the EPRI algorithm that accurately estimates source radioactivity using only minimal input information, including source locations, measured dose rates, and shielding conditions. The proposed algorithm consists of eight computational steps: it converts measured dose rates into energy-dependent flux, calculates gamma emission rates, estimates source radioactivity, and iteratively recalculates dose rates to minimize uncertainty. The estimation process employs both Successive Over-Relaxation (SOR) and Gauss-Jordan elimination methods. Initial guesses are refined through repeated iterations until convergence is achieved. When the location of the radiation source is known, the algorithm achieves higher accuracy and faster convergence by minimizing the residuals between calculated and measured dose rates. To verify the algorithm, we modeled a virtual work area containing Cs-137 and Co-60 point sources under varying shielding and measurement conditions. In these simulations, the SOR-based method yielded radioactivity estimates within a 10% error margin, while the Gauss-Jordan method demonstrated even higher accuracy with errors below 5%. These results demonstrate that the proposed algorithm provides reliable dose predictions even with limited input data and can be immediately applied to decommissioning sites as a practical and effective tool for radiological assessment.

Pages: 25 to 27

Copyright: Copyright (c) IARIA, 2025

Publication date: September 28, 2025

Published in: conference

ISSN: 2308-4537

ISBN: 978-1-68558-300-2

Location: Lisbon, Portugal

Dates: from September 28, 2025 to October 2, 2025