留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

球床式高温堆气固两相耦合半解析函数研究

赵蓬 王郑阳 吴浩 牛风雷 刘洋

赵蓬, 王郑阳, 吴浩, 等. 球床式高温堆气固两相耦合半解析函数研究[J]. 强激光与粒子束. doi: 10.11884/HPLPB202638.250238
引用本文: 赵蓬, 王郑阳, 吴浩, 等. 球床式高温堆气固两相耦合半解析函数研究[J]. 强激光与粒子束. doi: 10.11884/HPLPB202638.250238
Zhao Peng, Wang Zhengyang, Wu hao, et al. Semi-resolved function research on gas-solid two-phase coupling of high-temperature pebble beds[J]. High Power Laser and Particle Beams. doi: 10.11884/HPLPB202638.250238
Citation: Zhao Peng, Wang Zhengyang, Wu hao, et al. Semi-resolved function research on gas-solid two-phase coupling of high-temperature pebble beds[J]. High Power Laser and Particle Beams. doi: 10.11884/HPLPB202638.250238

球床式高温堆气固两相耦合半解析函数研究

doi: 10.11884/HPLPB202638.250238
基金项目: 国家自然科学基金项目(12105101); 中央高校基本科研业务费专项项目(2025MS066)
详细信息
    作者简介:

    赵 蓬,zpxq0407@163.com

    通讯作者:

    吴 浩,wuhao1938@hotmail.com

  • 中图分类号: TL331

Semi-resolved function research on gas-solid two-phase coupling of high-temperature pebble beds

  • 摘要: 为精确模拟高温球床堆内数万计燃料颗粒的气固两相耦合传热过程,并克服传统CFD-DEM方法因网格粗大导致的精度不足及全解析方法计算成本过高的问题,提出了一种适用于精细流体网格的半解析函数模型。该模型通过引入高斯核函数,对颗粒周围物理属性进行平滑与加权平均,从而实现在亚网格尺度下对颗粒所受流体作用力的精确计算。沃罗单元体分析表明,无量纲扩散时间的最优取值为0.6。超过此值会导致核函数影响域过度扩展,致使空间分布过度平滑而难以捕捉球床局部特征。在HTR-10球床堆的耦合传热仿真中,采用该模型计算得到的温度场分布与经验模型高度吻合。结果表明,本模型能够准确捕获颗粒间的相间作用力,为高温气冷堆热工流体仿真提供了一个兼具精度与效率的解决方案。
  • 图  1  不同网格尺寸计算孔隙率方案

    Figure  1.  Porosity schemes for different cell sizes

    图  2  沃罗单元体球床结构

    Figure  2.  Voronoi cell pebble beds structure

    图  3  配位数分布

    Figure  3.  Distribution of coordination number

    图  4  堆积密度分布

    Figure  4.  Stacking density distribution

    图  5  堆积密度变化曲线

    Figure  5.  Stacking density curve

    图  6  两种方式堆积密度分析

    Figure  6.  Compared stacking density by two methods

    图  7  均方误差曲线

    Figure  7.  MSE curve

    图  8  颗粒沉降速度随时间变化曲线

    Figure  8.  Particle settling velocity versus time curve

    图  9  雷诺数与阿基米德数关联曲线

    Figure  9.  Reynolds number and Archimedes number curve

    图  10  球床流化结果分析

    Figure  10.  Analysis of pebble bed fluidization results

    图  11  HTR-10球床温度分析

    Figure  11.  HTR-10 pebble bed temperature analysis

  • [1] 王连杰, 孙伟, 夏榜样, 等. 球床先进高温堆堆芯设计研究[J]. 核动力工程, 2018, 39(s2): 87-91

    Wang Lianjie, Sun Wei, Xia Bangyang, et al. Research on core design of pebble bed advanced high temperature reactor[J]. Nuclear Power Engineering, 2018, 39(s2): 87-91
    [2] 李睿, 任成, 杨星团, 等. 基于单元体理论的球床堆积结构与传热算法研究[J]. 核动力工程, 2016, 37(3): 39-42

    Li Rui, Ren Cheng, Yang Xingtuan, et al. Analysis of structures and heat transfer for packed beds[J]. Nuclear Power Engineering, 2016, 37(3): 39-42
    [3] 丁旺. 基于CFD-DEM的颗粒流体两相耦合模型的研究[D]. 上海: 上海交通大学, 2021

    Ding Wang. Research on the particle-fluid two-phase coupling model based on CFD-DEM[D]. Shanghai: Shanghai Jiao Tong University, 2021
    [4] 苏东升. 基于CFD-DEM耦合模拟方法的水流泥沙运动研究[D]. 天津: 天津大学, 2015

    Su Dongsheng. Investigation of fluid-sediment particle motion based on CFD-DEM coupling simulation method[D]. Tianjin: Tianjin University, 2015
    [5] 姚鹏, 王远, 冒刘鹏, 等. CFD-DEM模型最优网格尺寸的理论推证[J]. 泥沙研究, 2023, 48(5): 1-7,34

    Yao Peng, Wang Yuan, Mao Liupeng, et al. Theoretical validation on the optimal mesh size of CFD-DEM model[J]. Journal of Sediment Research, 2023, 48(5): 1-7,34
    [6] Xie Zhouzun, Wang Shuai, Shen Yansong. A novel hybrid CFD-DEM method for high-fidelity multi-resolution modelling of cross-scale particulate flow[J]. Chemical Engineering Journal, 2023, 455: 140731. doi: 10.1016/j.cej.2022.140731
    [7] Wang Zekun, Liu Moubin. Semi-resolved CFD-DEM for thermal particulate flows with applications to fluidized beds[J]. International Journal of Heat and Mass Transfer, 2020, 159: 120150. doi: 10.1016/j.ijheatmasstransfer.2020.120150
    [8] 杨星团, 刘志勇, 胡文平, 等. HTR-10堆芯球流运动的唯象学DEM模拟[J]. 原子能科学技术, 2013, 47(12): 2231-2237

    Yang Xingtuan, Liu Zhiyong, Hu Wenping, et al. DEM simulation of pebble flow in HTR-10 core by phenomenological method[J]. Atomic Energy Science and Technology, 2013, 47(12): 2231-2237
    [9] 徐泳, 孙其诚, 张凌, 等. 颗粒离散元法研究进展[J]. 力学进展, 2003, 33(2): 251-260

    Xu Yong, Sun Qicheng, Zhang Ling, et al. Advances in discrete element methods for particulate materials[J]. Advances in Mechanics, 2003, 33(2): 251-260
    [10] Wang Zekun, Teng Yujun, Liu Moubin. A semi-resolved CFD-DEM approach for particulate flows with kernel based approximation and Hilbert curve based searching strategy[J]. Journal of Computational Physics, 2019, 384: 151-169. doi: 10.1016/j.jcp.2019.01.017
    [11] Pirker S, Kahrimanovic D, Goniva C. Improving the applicability of discrete phase simulations by smoothening their exchange fields[J]. Applied Mathematical Modelling, 2011, 35(5): 2479-2488. doi: 10.1016/j.apm.2010.11.066
    [12] Rycroft C H. Multiscale modeling in granular flow[D]. Cambridge: Massachusetts Institute of Technology, 2007.
    [13] 严安, 孙喜明, 董玉杰. 球床式高温堆堆芯气固两相流耦合模拟研究[J]. 哈尔滨工程大学学报, 2022, 43(12): 1743-1749

    Yan An, Sun Ximing, Dong Yujie. Simulation study of gas-solid two-phase flow coupling of pebble-bed cores in a high-temperature reactor[J]. Journal of Harbin Engineering University, 2022, 43(12): 1743-1749
    [14] 蔡国庆, 刁显锋, 杨芮, 等. 基于CFD-DEM的流-固耦合数值建模方法研究进展[J]. 哈尔滨工业大学学报, 2024, 56(1): 17-32

    Cai Guoqing, Diao Xianfeng, Yang Rui, et al. Research progress of fluid-solid coupling model based on CFD-DEM coupling[J]. Journal of Harbin Institute of Technology, 2024, 56(1): 17-32
    [15] Suikkanen H, Ritvanen J, Jalali P, et al. Discrete element modelling of pebble packing in pebble bed reactors[J]. Nuclear Engineering and Design, 2014, 273: 24-32. doi: 10.1016/j.nucengdes.2014.02.022
    [16] 王晨洋. 颗粒堆积孔隙结构的统计性质研究[D]. 上海: 华东师范大学, 2024

    Wang Chenyang. A statistical study of the pore structures of granular packings[D]. Shanghai: East China Normal University, 2024
    [17] Kloss C, Goniva C, Hager A, et al. Models, algorithms and validation for opensource DEM and CFD–DEM[J]. Progress in Computational Fluid Dynamics, 2012, 12(2/3): 140-152. doi: 10.1504/PCFD.2012.047457
    [18] Di Felice R. The sedimentation velocity of dilute suspensions of nearly monosized spheres[J]. International Journal of Multiphase Flow, 1999, 25(4): 559-574. doi: 10.1016/S0301-9322(98)00084-6
    [19] Müller C R, Scott S A, Holland D J, et al. Validation of a discrete element model using magnetic resonance measurements[J]. Particuology, 2009, 7(4): 297-306. doi: 10.1016/j.partic.2009.04.002
    [20] Zhou Yandaizi, Wang Tielin, Zhu J. Investigation on minimum fluidization velocity in a modified Geldart’s diagram[J]. Chemical Engineering Journal, 2023, 453: 139984. doi: 10.1016/j.cej.2022.139984
    [21] Wu Hao, Zhao Houjian, Hao Zulong, et al. A non-linear transform approach for conduction-radiation heat transfer in the extended thermal discrete element method[J]. International Journal of Heat and Mass Transfer, 2021, 176: 121432. doi: 10.1016/j.ijheatmasstransfer.2021.121432
    [22] Chen Fubing, Dong Yujie, Zheng Yanhua, et al. Benchmark calculation for the steady-state temperature distribution of the HTR-10 under full-power operation[J]. Journal of Nuclear Science and Technology, 2009, 46(6): 572-580. doi: 10.1080/18811248.2007.9711564
  • 加载中
图(11)
计量
  • 文章访问数:  18
  • HTML全文浏览量:  11
  • PDF下载量:  1
  • 被引次数: 0
出版历程
  • 收稿日期:  2025-07-25
  • 修回日期:  2025-11-06
  • 录用日期:  2025-10-28
  • 网络出版日期:  2025-11-15

目录

    /

    返回文章
    返回