| 2 | 0 | 5 |
| 下载次数 | 被引频次 | 阅读次数 |
基于颗粒动力学理论和机器学习算法，本文建立了用于量化冰浆管道流动的两相流欧拉-欧拉模型，提出了以管道顶部冰颗粒堆积分数为分类指数的流型划分方法，构建了基于逻辑回归的冰浆输送管道流通滞阻预测模型。结果表明：两相流欧拉-欧拉数值模型能够详尽提供冰浆流动流场和浓度场等信息，实现了输送管道流通滞阻风险预测样本采集。提出的流型识别方法能够准确识别冰浆管道流动的流型，总识别准确率达94%。应用准确率、查准率、查全率和查准率与查全率调和平均值（F1-score）指标评价所建立的流通滞阻预测模型，在测试集上对于管道流通滞阻风险预测的准确率可达到92%。相比之下，在低滞阻风险时预测性能更优。
Abstract：Based on granular kinetic theory and machine learning algorithms, a two-phase Euler-Euler model is developed to quantify ice slurry pipe flow in this paper. A flow pattern classification method using the ice particle packing fraction at the pipe crown as the classification index is proposed. A predictive model based on logistic regression for flow blockage in ice slurry transportation pipelines is constructed. The results show that the Eulerian-Eulerian model could provide detailed information on the physical fields of ice slurry flow, which is used to obtain data samples under different operating conditions. The proposed method can accurately identify the flow pattern of ice slurry in pipes with an accuracy of 94%. The indicators of the accuracy, precision, recall, and the harmonic mean of precision and recall(F1-score) are applied to evaluate the prediction model of the blockage risk, and the accuracy reaches over 92%. Overall, the performance of the proposed method is better at low blockage risk of ice slurry flow in a horizontal pipe.
[1]闫金光,刘佳佳,刘晓华,等.加快发展“光储直柔”建筑的重要意义、挑战及政策建议[J].中国能源, 2022,44(8):33-38.
[2]吕笑冲,袁俊,霍天晴,等.基于全生命周期分析的热泵水果干燥机的R1234ze(E)型环保工质替代评价[J].制冷技术, 2023, 43(6):68-75.
[3]宋文吉,冯自平,肖睿.冰浆技术及其应用进展[J].新能源进展, 2019, 7(2):129-141.
[4]鲁威,刘刚,于静,等.动态冰浆系统在乳品行业的应用分析[J].制冷技术, 2020, 40(3):59-63.
[5]张雪,刘圣春,宋丽莹,等.不同管道内冰浆流动换热特性及冰堵分析[J].制冷技术, 2021, 41(3):55-61.
[6]黄逸宸,武卫东,黄华,等.应用于冰浆湿冷差压预冷系统的包装箱开孔结构实验研究[J].制冷技术, 2024,44(2):53-60.
[7] KAUFFELD M, WANG M J, GOLDSTEIN V, et al. Ice slurry applications[J]. International Journal of Refrigeration, 2010, 33(8):1-15.
[8] HIROCHI T, MAEDA Y, YAMADA S, et al. Flow patterns of ice/water slurry in horizontal pipes[J]. Journal of Fluids Engineering, 2004, 126(3):436-441.
[9] LI X, LI L, WANG W, et al. Machine learning techniques applied to identify the two-phase flow pattern in porous media based on signal analysis[J]. Applied Sciences, 2022,12(17):8575.
[10] MI Y, ISHII M, TSOUKALAS L H. Vertical wo-phase flow identification using advanced instrumentation and neural networks[J]. Nuclear Engineering and Design, 1998,184(2/3):409-420.
[11]计时鸣,胡科东,谭大鹏,等.基于小波包的固液两相流流型识别方法[J].湘潭大学自然科学学报, 2012,34(3):88-92.
[12] EKAMBARA K, SANDERS R S, NANDAKUMAR K, et al. Hydrodynamic simulation of horizontal slurry pipeline flow using ANSYS-CFX[J]. Industrial and Engineering Chemistry Research, 2009, 48(17):8159-8171.
[13] WANG J H, WANG S G, ZHANG T F, et al. Numerical investigation of ice slurry isothermal flow in various pipes[J]. International Journal of Refrigeration, 2013,36(1):70-80.
[14]周志华.机器学习[M].北京:清华大学出版社, 2016.
[15] BOYD S P, VANDENBERGHE L. Convex optimization[M]. Cambridge:Cambridge University Press, 2004.
[16] TIAN Q, HE G, WANG H, et al. Simulation on transportation safety of ice slurry in ice cooling system of buildings[J]. Energy and Buildings, 2014, 72:262-270.
[17] BARNES H A, HUTTON J F, WALTERS K. An introduction to rheology[M]. Amsterdam:Elsevier, 1989.
[18] KAUSHAL D R, TOMITA Y. Experimental investigation for near-wall lift of coarser particles in slurry pipeline using γ-ray densitometer[J]. Powder Technology, 2007,172(3):177-187.
[19] VUARNOZ D, SARI O, EGOLF P W, et al. Ultrasonic velocity profiler UVP-XW for ice-slurry flow characterization[C]//Proceedings of the 3rd International Symposium on Ultrasonic Doppler Method for Fluid Mechanics and Fluid Engineering. Viligen:Paul Scherrer Institute, 2002.
[20] KITANOVSKI A, VUARNOZ D, ATA-CAESAR D, et al.The fluid dynamics of ice slurry[J]. International Journal of Refrigeration, 2005, 28(1):37-50.
基本信息:
DOI：
中图分类号:TB657.2
引用信息:
[1]吕燕捷,邢启峰,王继红.水平管道内冰浆流通滞阻风险预测[J].制冷技术,2025,45(03):50-55.
基金信息:
建筑安全与环境国家重点实验室/国家建筑工程技术研究中心开放课题(No.BSBE2021-08); 国家自然科学基金(No.52378090,No.51508067)