基于集成学习和信息融合的段塞流分相流量测量

doi:10.3969/j.issn.1000-1565.2024.05.011

摘要/Abstract

摘要： 段塞流是气液两相流中典型流型,准确测量其分相流量有利于实时监控生产过程,优化工艺控制,确保系统在安全、经济的工况下运行.本文在改进长喉文丘里管的基础上,设计了一种集近红外(NIR)、声发射(AE)技术于一体的水平气液流量智能多传感系统.利用AE传感器和NIR传感器检测气液两相的流动噪声信息和截面信息,采用经验模态分解法(EMD)提取气体体积分数的特征变量.通过集成学习算法进行特征级融合,融合后的段塞流体积含气率预测模型平均绝对百分比误差(MAPE)为4.11%,92.45%的预测结果偏差在±10%以内.在Collins模型的基础上,提出了基于梯度提升决策树(GBDT)的段塞流质量流量预测模型,其MAPE值为0.96%,全部预测结果的偏差在±20%以内.本研究为气液两相流段塞流参数混合不分离测量提供了一种新方法,为气液两相流动机理研究奠定了基础.

关键词: 气液两相流, 数据融合, 段塞流, 多传感器, 集成学习算法

Abstract: Plug flow is a typical flow pattern in gas-liquid two-phase flow, and accurate measurement of plug flow is conducive to real-time monitoring and optimizing process of production process, ensuring safe and economical operation of the system. Based on the improved Venturi tube with long throat diameter, an intelligent multi-sensor system for horizontal gas-liquid flow is designed, which integrates near infrared(NIR)and acoustic emission(AE)technology. AE sensor and NIR sensor were used to detect the gas-liquid phase interaction and disturbance information, and empirical mode decomposition(EMD)was used to extract the characteristic variables of gas volume fraction. The integrated learning algorithm was used- DOI:10.3969/j.issn.1000-1565.2024.05.011基于集成学习和信息融合的段塞流分相流量测量温佳祺^1,2,杨叙宁^1,2,李金硕¹,丁振君^1,3,董芳^1,3(1.河北大学质量技术监督学院,河北保定 071002;2.零碳能源建筑与计量技术教育部工程研究中心,河北保定 071002;3.计量仪器与系统国家地方联合工程研究中心,河北保定 071002)摘要:段塞流是气液两相流中典型流型,准确测量其分相流量有利于实时监控生产过程,优化工艺控制,确保系统在安全、经济的工况下运行.本文在改进长喉文丘里管的基础上,设计了一种集近红外(NIR)、声发射(AE)技术于一体的水平气液流量智能多传感系统.利用AE传感器和NIR传感器检测气液两相的流动噪声信息和截面信息,采用经验模态分解法(EMD)提取气体体积分数的特征变量.通过集成学习算法进行特征级融合,融合后的段塞流体积含气率预测模型平均绝对百分比误差(MAPE)为4.11%,92.45%的预测结果偏差在±10%以内.在Collins模型的基础上,提出了基于梯度提升决策树(GBDT)的段塞流质量流量预测模型,其MAPE值为0.96%,全部预测结果的偏差在±20%以内.本研究为气液两相流段塞流参数混合不分离测量提供了一种新方法,为气液两相流动机理研究奠定了基础.关键词:气液两相流;数据融合;段塞流;多传感器;集成学习算法中图分类号:TH814 文献标志码:A 文章编号:1000-1565(2024)05-0541-10Phase separation flow measurement of plug flow based on ensemble learning and information fusionWEN Jiaqi^1,2,YANG Xuning^1,2,LI Jinshuo¹, DING Zhenjun^1,3, DONG Fang^1,3(1. School of Quality and Technical Supervision, Hebei University, Baoding 071002, China; 2. Engineering Research Center of Zero-carbon Energy Bulidings and Measurement Techniques,Ministry of Education,Baoding 071002,China;3. National and Local Joint Engineering Research Center for Measuring Instruments and Systems, Baoding 071002, China)Abstract: Plug flow is a typical flow pattern in gas-liquid two-phase flow, and accurate measurement of plug flow is conducive to real-time monitoring and optimizing process of production process, ensuring safe and economical operation of the system. Based on the improved Venturi tube with long throat diameter, an intelligent multi-sensor system for horizontal gas-liquid flow is designed, which integrates near infrared(NIR)and acoustic emission(AE)technology. AE sensor and NIR sensor were used to detect the gas-liquid phase interaction and disturbance information, and empirical mode decomposition(EMD)was used to extract the characteristic variables of gas volume fraction. The integrated learning algorithm was used- 收稿日期:2024-03-06;修回日期:2024-05-02 基金项目:国家自然科学基金资助项目(62173122);河北省自然科学基金资助项目(F2022201034) 第一作者:温佳祺(2000—),女,河北大学在读硕士研究生,主要从事多相流理论及参数检测技术研究.E-mail:benben93657@163.com 通信作者:董芳(1980—),女,河北大学讲师,主要从事多相流参数检测及信息处理技术研究.E-mail:dongfang1023@163.com第5期温佳祺等:基于集成学习和信息融合的段塞流分相流量测量河北大学学报(自然科学版) 第44卷for feature-level fusion. The mean absolute percentage error(MAPE)of the fused plug flow volume gas content prediction model was 4.11%, and the deviation of 92.45% of the predicted results was within ±10%. On the basis of Collins model, a mass flow prediction model of plug flow based on gradient lifting decision tree(GBDT)is proposed. The MAPE value of GBDT is 0.96%, and the deviation of all prediction results is within ±20%. This study provides a new method for measuring the parameters of gas-liquid two-phase plug flow, which provides a research basis for sensing mechanism and measurement of multiphase flow.

Key words: gas-liquid two-phase flow, data fusion, plug flow, multi-sensor, ensemble learning algorithm

中图分类号:

TH814

温佳祺,杨叙宁,李金硕,丁振君,董芳. 基于集成学习和信息融合的段塞流分相流量测量[J]. 河北大学学报(自然科学版), 2024, 44(5): 541-550.

WEN Jiaqi,YANG Xuning,LI Jinshuo, DING Zhenjun, DONG Fang. Phase separation flow measurement of plug flow based on ensemble learning and information fusion[J]. Journal of Hebei University(Natural Science Edition), 2024, 44(5): 541-550.

参考文献

[1] YANG Y, WANG D, NIU P, et al. Measurement of vertical gas-liquid two-phase flow by electromagnetic flowmeter and image processing based on the phase-isolation[J]. Experimental Thermal and Fluid Science, 2019, 101: 87-100. DOI: 10.1016/j.expthermflusci.2018.10.010.
[2] 张立峰,王智.基于多域特征提取的气液两相流流型识别[J].计量学报, 2023, 44(10): 1509-1516. DOI: 10.3969/j.issn.1000-1158.2023.10.05.
[3] 方立德,田梦园,王配配,等.基于物质吸光度的气液两相流相含率测量模型[J].光电子·激光,2021, 32(4): 409-418. DOI: 10.3969/j.issn.1000-1158.2019.06.17.
[4] 金卫栋,戴明岩.气液两相流相含率确定方法的研究综述[J].山东化工, 2023, 52(5): 78-80. DOI: 10.19319/j.cnki.issn.1008-021x.2023.05.026.
[5] 安雅丽,徐志鹏,刘铁军,等.双锥流量计气水两相流流量测量实验研究[J].实验流体力学, 2016, 30(5): 67-73.DOI: 10.11729/syltlx20160044.
[6] 董芳,方立德,李小亭.功率谱熵在垂直于水平流向的气液两相流压差信号中的应用[J].河北大学学报(自然科学版), 2014, 34(5): 541-546. DOI: 10.3969/j.issn.1000-1565.2014.05.017.
[7] 方立德,赵敏慧,杨英昆,等.气液两相流动噪声特性及相含率测量模型[J].哈尔滨工程大学学报, 2019, 40(8): 1530-1536. DOI: 10.7666/d.D531037.
[8] 李小亭,卢庆华,方立德,等.基于声发射技术和小波变换的气-液两相流动噪声特性研究[J].电子测量与仪器学报, 2012, 26(12): 1031-1036. DOI: 10.3724/SP.J.1187.2012.01031.
[9] 刘月圆,梁玉娇,郑盟,等.新型均速管涡街流量传感器的测量特性研究[J].电子测量与仪器学报, 2022, 36(12): 10-18. DOI: 10.13382/j.jemi.B2205890.
[10] 赵宁,王超,孙宏军,等.一种测量环状流液膜参数的传感器研究[J].中南大学学报(自然科学版), 2018, 49(4): 995-1002. DOI: 10.11817/j.issn.1672-7207.2018.04.030.
[11] VIEIRA E R, PARSI M, MCLAURY S B, et al. Experimental characterization of vertical downward two-phase annular flows using Wire-Mesh Sensor[J]. Chem Eng Sci, 2015, 134: 324-339. DOI: 10.1016/j.ces.2015.05.013.
[12] XU Y, ZUO Y J, YUAN C, et al. Research on the measurement method of gas-water two-phase flow based on dual-sensor system[J]. Flow Meas Instrum, 2023, 93: 102393. DOI: 10.1016/j.flowmeasinst.2023.102393.
[13] LIU S, WANG K, LIU L, et al. Characterization of liquid film distribution and sub-flow regimes of annular flow in helically coiled tubes using ring island array sensor[J]. Int J Multiphas Flow, 2023, 160: 104357. DOI: 10.1016/j.ijmultiphaseflow.2022.104357.
[14] WEI Z, LI M, LIAO G, et al. Gas slug parameter measurement based on double embedded U-tube electromagnetic wave phase sensor in gas-liquid slug flow[J]. Flow Meas Instrum, 2024(96): 102530. DOI: 10.1016/j.flowmeasinst.2024.102530.
[15] QU X, GUO Q, ZHANG Y, et al. A new vector-based signal processing method of four-sensor probe for measuring local gas-liquid two-phase flow parameters together with Its assessment against one bubbly flow[J]. Appl Sci-basel, 2020, 10(16):54-63. DOI: 10.3390/app10165463.
[16] 赵晓云,宋红伟,王明星,等.基于人工智能算法的水平井气水两相生产测井流型识别方法研究[J].当代化工研究, 2023,(8): 173-175. DOI: 10.20087/j.cnki.1672-8114.2023.08.058.
[17] WANG L, LIU J, YAN Y, et al. Gas-liquid two-phase flow measurement using Coriolis flowmeters incorporating artificial neural network, support vector machine, and genetic programming algorithms[J]. IEEE Trans Instrum Meas, 2016, 66(5): 852-868. DOI: 10.1109/I2MTC.2016.7520458.
[18] WANG H, ZHANG M, YANG Y. Machine learning for multiphase flowrate estimation with time series sensing data[J]. MEAS, 2020, 10: 100025. DOI: 10.1016/j.measen.2020.100025.
[19] ALI N, VIGGIANO B, TUTKUN M, et al. Data-driven machine learning for accurate prediction and statistical quantification of two phase flow regimes[J]. J Petrol Sci Eng, 2021, 202: 108488. DOI: 10.1016/J.PETROL.2021.108488.
[20] MAMONOV V N, SERCOV A F. Experimental determination of the volume concentration of the gas phase in gas-liquid flow[J]. J Appl Mech Tech Ph+, 2021, 62(1): 57-62. DOI: 10.1134/S0021894421010077.
[21] COLLINS D B, GACESA M. Measurement of steam quality in two-phase upflow with venturimeters and orifice plates[J]. J. Fluids Eng, 1971, 93(1): 11-20. DOI: 10.1115/1.3425169.
[22] 陈雨桐.集成学习算法之随机森林与梯度提升决策树的分析比较[J].电脑知识与技术, 2021, 17(15): 32-34. DOI: 10.14004/j.cnki.ckt.2021.1441.
[23] 赵俊琴,王慧,王彤.基于LASSO的高维数据线性回归模型统计推断方法比较[J].中国卫生统计, 2017, 34(2): 250-252. DOI: 10.7666/d.D649822. (