基于Pybind11的铅冷快堆核热耦合方法

Neutronics/Thermal-hydraulics Coupling Method for Lead-cooled Fast Reactor Using Pybind11

  • 摘要: 为满足铅冷快堆堆芯核热耦合计算的需求,本文基于Pybind11对蒙特卡罗程序OpenMC进行封装,利用其轻量高效的C++与Python无缝交互能力,实现跨语言的API调用与内存数据传递。在此基础上,将OpenMC与pin级全堆芯子通道程序KMC-FBc以及点堆动力学模型PK以内耦合方式集成,开发了核热耦合程序OpenMC/KMC-FBc/PK。该框架在保持蒙特卡罗与子通道计算精度的同时,引入点堆动力学模型用于高效求解堆芯瞬态功率变化。为了验证所提出的方法,进行了稳态和瞬态耦合计算,结果表明所开发的耦合计算程序有效可靠,能够准确描述堆芯功率分布和局部热工特性。本文开发的程序可为铅冷快堆的性能评估和安全分析提供可靠工具。

     

    Abstract: The lead-cooled fast reactor (LFR), as one of the generation Ⅳ nuclear systems, attracts considerable attention due to its excellent thermophysical and neutronic characteristics. With increasing demands for core safety and economic performance, accurate neutronics/thermal-hydraulics coupling analysis becomes increasingly important for evaluating reactor behavior. The Monte Carlo method, characterized by its use of continuous-energy cross sections and flexible geometric modeling, enables detailed and reliable neutron transport simulations. Meanwhile, in the field of thermal-hydraulic analysis, subchannel codes achieve an optimal balance between computational precision and efficiency, making them suitable for pin-level core simulations. This study aims to develop a high-resolution internal coupling framework through API integration and in-memory data exchange techniques, which enables the deep exploration of complex multi-physics feedback mechanisms in LFRs under both steady-state and transient conditions. The primary objective is to satisfy the requirements for detailed neutronics/thermal-hydraulics coupling calculations in LFRs, thereby providing a reliable analysis tool for core design and safety assessment. The OpenMC/KMC-FBc/PK internal coupling system was established in this study by integrating the Monte Carlo neutron transport code OpenMC, the pin-level full-core subchannel program KMC-FBc, and a point kinetics (PK) model. To overcome the efficiency bottlenecks of cross-language communication, the Pybind11 library was employed to encapsulate the Monte Carlo module, enabling the C++ main control program to invoke application programming interfaces and perform direct in-memory data exchange. In this framework, the PK model was implemented to solve for transient power evolution, while the Monte Carlo module provided the detailed three-dimensional spatial power distribution, and the subchannel program was utilized to execute the thermal-hydraulic response during transients. Consequently, this approach maintained the high spatial resolution of both the Monte Carlo and subchannel codes while ensuring computational efficiency for transient solutions. The system was verified through code-to-code comparisons using a small-scale LFR model designed by Shanghai Jiao Tong University and the SNCLFR-100 reactor model developed by the University of Science and Technology of China. Furthermore, the system was applied to the NCLFR-Oil reactor for comprehensive steady-state and unprotected loss of heat sink (ULOHS) analysis. The results indicate that the spatial power distributions and key thermal-hydraulic parameters calculated by OpenMC/KMC-FBc/PK show excellent agreement with benchmark codes, verifying its accuracy and reliability in fine-grained coupled analysis for LFRs. Steady-state and transient analyses conducted on the NCLFR-Oil core demonstrate that while the coupling effects in LFRs have a less significant impact on power distribution compared to pressurized water reactors, they markedly influence key thermal-hydraulic parameters. Under ULOHS conditions, the peak temperatures calculated by OpenMC/KMC-FBc/PK are higher than those from the system-level code ATHLET, providing a more accurate reflection of local hot-spot characteristics. Overall, the developed neutronics/thermal-hydraulics coupling program accurately describes the interactions between neutron transport and thermal-hydraulics. It proves suitable for both steady-state and transient LFR analysis, serving as a reliable tool for the performance evaluation and safety assessment of LFRs.

     

/

返回文章
返回