Monte Carlo Investigation of the UK’s First EPR Nuclear Reactor Startup Core Using Serpent

• Main Author: Jinfeng Li
• Format: Article
• Language: English
• Published: MDPI AG 2020-10-01
• Series: Energies
• Subjects: computational neutronics; European pressurised reactor; Monte Carlo simulation; nuclear physics; nuclear reactor core modelling; nuclear energy
• Online Access: https://www.mdpi.com/1996-1073/13/19/5168

Computationally modelling a nuclear reactor startup core for a benchmark against the existing models is highly desirable for an independent assessment informing nuclear engineers and energy policymakers. For the first time, this work presents a startup core model of the UK’s first Evolutionary Pressurised Water Reactor (EPR) based on Monte Carlo simulations of particle collisions using Serpent 2, a state-of-the-art continuous-energy Monte Carlo reactor physics burnup code. Coupling between neutronics and thermal-hydraulic conditions with the fuel depletion is incorporated into the multi-dimensional branches, obtaining the thermal flux and fission reaction rate (power) distributions radially and axially from the three dimensional (3D) single assembly level to a 3D full core. Shannon entropy is quantified to characterise the convergence behaviour of the fission source distribution, with 3 billion neutron histories tracked by parallel computing. Source biasing is applied for the variance reduction. Benchmarking the proposed Monte Carlo 3D full-core model against the traditional deterministic transport computation suite used by the UK Office for Nuclear Regulation (ONR), a reasonably good agreement within statistics is demonstrated for the safety-related reactivity coefficients, which creates trust in the EPR safety report and informs the decision-making by energy regulatory bodies and global partners.