化工学报 ›› 2021, Vol. 72 ›› Issue (9): 4649-4657.DOI: 10.11949/0438-1157.20210156

• 流体力学与传递现象 • 上一篇    下一篇

基于GA-BP神经网络的超临界CO2传热特性预测研究

颜建国1(),郑书闽1,郭鹏程1(),张博2,毛振凯2   

  1. 1.西安理工大学西北旱区生态水利国家重点实验室,陕西 西安 710048
    2.中国电建集团西北勘测设计研究院有限公司,陕西 西安 710065
  • 收稿日期:2021-01-25 修回日期:2021-07-08 出版日期:2021-09-05 发布日期:2021-09-05
  • 通讯作者: 郭鹏程
  • 作者简介:颜建国(1987—),男,博士,副教授,jgyan@xaut.edu.cn
  • 基金资助:
    国家自然科学基金项目(51839010);陕西省教育厅科研计划项目(21JY029);陕西高校青年科技创新团队项目(2020-29);清洁能源与生态水利工程研究中心项目(QNZX-2019-05)

Prediction of heat transfer characteristics for supercritical CO2 based on GA-BP neural network

Jianguo YAN1(),Shumin ZHENG1,Pengcheng GUO1(),Bo ZHANG2,Zhenkai MAO2   

  1. 1.State Key Laboratory of Eco-hydraulics in Northwest Arid Region, Xi’an University of Technology, Xi’an 710048, Shaanxi, China
    2.Power China Northwest Engineering Corporation Limited, Xi’an 710065, Shaanxi, China
  • Received:2021-01-25 Revised:2021-07-08 Online:2021-09-05 Published:2021-09-05
  • Contact: Pengcheng GUO

摘要:

超临界二氧化碳(S-CO2)动力循环在能源利用领域中拥有广阔的应用前景,其中超临界CO2的传热特性对其能量转换效率至关重要。开展了超临界CO2在水平小圆管内对流传热实验研究,并通过建立遗传算法优化的BP神经网络模型(GA-BP),对其在不同工况下的传热特性进行预测分析。实验参数范围:系统压力7.5~9.5 MPa,质量流速1100~2100 kg/(m2?s),热通量120~560 kW/m2。实验结果表明,超临界CO2传热系数随流体温度的升高先增大后减小,在拟临界温度附近达到最大值。GA-BP神经网络模型能有效地预测超临界CO2的传热系数,预测数据的决定系数R2为0.99662,超过95%的数据误差位于±10%范围内,平均误差为3.55%,为超临界流体传热预测提供新的思路。

关键词: 超临界二氧化碳, 对流, 传热, GA-BP神经网络, 传热预测

Abstract:

Supercritical carbon dioxide (S-CO2) power cycle has a broad application prospect in the field of energy utilization, in which the heat transfer of supercritical CO2 plays a significant role in its energy conversion efficiency. Therefore, an experiment has been conducted to determine the convective heat transfer characteristics of supercritical CO2 flowing in a mini horizontal circular tube, and a BP neural network optimized by genetic algorithm has been established to predict the heat transfer coefficient of supercritical CO2 under different conditions. The experimental parameters are as follows: system pressure 7.5—9.5 MPa, mass flux 1100—2100 kg/(m2?s), heat flux 120—560 kW/m2. The experimental results show that the heat transfer coefficient of supercritical CO2 increases first and then decreases with increasing fluid temperature, and reaches maximum near the pseudo-critical temperature. The model of the GA-BP neural network can effectively predict the heat transfer coefficient of supercritical CO2, the determinate coefficient of predicted data R2 is 0.99662, and more than 95% of the data are within the error range of ±10%, the average error is 3.55%. GA-BP neural network model provides a novel idea for heat transfer prediction for supercritical fluids.

Key words: supercritical carbon dioxide, convection, heat transfer, GA-BP neural network, prediction of heat transfer

中图分类号: