Research and verification on calculation method of multi-ring fuel resonance

[Background] Accurate resonance calculation for multi-ring fuel elements remains a significant challenge in reactor physics due to the complex spatial self-shielding effects and strong mutual interference between resonant nuclides. Traditional resonance methods, such as the subgroup method, often struggle to achieve a balance between computational efficiency and accuracy when dealing with such configurations. This is particularly critical for high-fidelity analysis of advanced reactors and experimental facilities like the high-flux engineering test reactor, where precise characterization of resonance phenomena is essential for predicting core performance and safety parameters. [Purpose] This study aims to address the limitations of existing resonance calculation methods for multi-ring fuel systems by developing a novel global-local coupling framework. The primary objectives are to enhance the accuracy of effective self-shielding cross-section computation for resonant nuclides, improve computational efficiency, and validate the method’s applicability for both assembly-level and full-core simulations. [Methods] A multi-ring fuel resonance calculation method based on the global-local coupling method (MRFRCM) was proposed specifically for multi-ring fuel analysis. In this approach, when handling global spatial effects, the entire multi-ring fuel is treated as an integrated black body. This process simplifies the multi-ring fuel problem into an equivalent rod-type fuel problem for calculating the global Dancoff correction factor. Subsequently, an equivalent one-dimensional local problem is established through a conservation-based search of the Dancoff correction factor. Finally, the problem is reverted to a one-dimensional multi-ring fuel configuration, where the ultra-fine group method is employed to obtain precise self-shielding cross-sections. The method was implemented and tested on multi-ring fuel assembly problems to evaluate its precision and efficiency. Furthermore, it was applied to two-dimensional and three-dimensional full-core models of a high-flux engineering test reactor to assess its performance in practical scenarios. [Results] The proposed method demonstrated superior accuracy and computational efficiency compared to the traditional subgroup method. In assembly-level calculations, the global-local approach reduced errors in effective cross-section estimation while maintaining competitive computation times. For full-core simulations, the results showed good agreement with reference solutions. The method also exhibited robust performance in handling complex geometries and heterogeneous material configurations. [Conclusions] The MRFRCM provides an effective solution for high-accuracy resonance modeling in multi-ring fuel systems. It significantly outperforms the traditional subgroup method in both precision and efficiency, making it suitable for large-scale reactor physics applications. The successful application to 2D and 3D full-core analyses confirms its practicality and reliability for simulating high-fidelity reactor core behavior. Future work will focus on extending the method to broader energy ranges and more complex reactor types.