Nuclear Techniques ›› 2017, Vol. 40 ›› Issue (7): 70603-070603.doi: 10.11889/j.0253-3219.2017.hjs.40.070603

• NUCLEAR ENERGY SCIENCE AND ENGINEERING • Previous Articles     Next Articles

Development of a GPU-accelerated thermal hydraulic code for pebble-bed fluoride-salt cooled high temperature reactor core

E Yanzhi1,2, ZOU Yang1, GUO Wei1, PENG Yu1,2, XU Hongjie1   

  1. 1. Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Jiading Campus, Shanghai 201800, China;
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2017-03-07 Revised:2017-04-20 Online:2017-07-10 Published:2017-07-10
  • About author:10.11889/j.0253-3219.2017.hjs.40.070603
  • Supported by:
    Supported by Strategic Pilot Science and Technology Project of Chinese Academy of Sciences (No.XDA02001002), the Frontier Science Key Programof Chinese Academy of Sciences (No.QYZDY-SSW-JSC016)

Abstract: Background: Pebble bed fluoride-salt cooled high temperature reactor (PB-FHR) is a kind of Generation IV reactor. Three-dimensional thermal hydraulic code of reactor core is expected to simulate cases with complex spatial effect for PB-FHR but takes very heavy time-consuming. Graphics processing unit (GPU) contains numerous computing cores, thus can be used to efficiently accelerate numerical calculation if applied properly. Purpose: This study aims to develop a GPU-accelerated thermal-hydraulic code (GATH) for PB-FHR core. Methods: Thermal non-equilibrium porous media theory is adopted to build the reactor core physical model. Efficient iterative algorithms are researched and implemented on GPU. A PB-FHR core model is built for thermal hydraulic analysis with GATH. Simulation results are compared with ANSYS CFX software to verify GATH code and the GPU acceleration performance is analyzed. Results: The results between GATH and CFX are in good agreement. The speedup ratio of GATH can reach 8.39 times. Conclusion: The physical model and calculation method adopted in GATH code are right. The GPU accelerating methods proposed in this paper can efficiently accelerate thermal hydraulic simulation.

Key words: PB-FHR, GPU-accelerated, Thermal hydraulic, Thermal non-equilibrium porous media

CLC Number: 

  • TL333