基于改进型SVM算法的气液两相流持液率计算模型Liquid Holdup Calculation Model of Gas-Liquid Two-Phase Flow Based on Improved SVM Algorithm
花靖,蒋秀,于超,谷成林,靳彦欣,逄铭玉
摘要(Abstract):
气液两相流持液率的精准计算对于流型识别、管道腐蚀评价与预测以及输气管道输送效率计算等方面的研究具有重要的意义。目前,国内外专家学者建立的持液率计算模型适应范围有限,缺乏一套适应于不同工况条件的持液率计算模型。本文首先统计分析了国内外不同专家学者共2 123组实验研究数据,构建了丰富的室内实验数据库;再选定支持向量机(SVM)对数据进行训练和回归预测,为了提高算法的可靠性,采用K-交叉验证(K-CV)、遗传算法(GA)以及粒子群优化算法(PSO)对SVM算法的惩罚因子c和核函数参数g寻优过程进行优化,分别建立了K-CV-SVM、GA-SVM、PSO-SVM模型,综合考虑计算精度和稳定性,最终优选了GA-SVM作为持液率计算模型的核心算法。将该模型与Eaton、BB、MB、BBE等经验关系式模型进行了计算精度对比分析,结果显示:建立的持液率计算模型在上倾管、水平管以及下倾管中都具有比经典模型更佳的计算精度。本文基于GA-SVM算法建立的持液率计算模型计算精度高、稳定性强、适应性广泛,可满足不同工况条件下持液率计算的需求。
关键词(KeyWords): 气液两相流;持液率;管道腐蚀评价与预测;计算模型;支持向量机;K-交叉验证;遗传算法;粒子群算法
基金项目(Foundation): 国家青年科学基金项目“油气混输管道水合物沉积与剥离脱落机理研究”(51904330)
作者(Author): 花靖,蒋秀,于超,谷成林,靳彦欣,逄铭玉
参考文献(References):
- [1] 吴浩江,蔡正敏,周芳德.应用神经网络法智能识别油气水多相流流型[J].西安石油学院学报(自然科学版),2000,15(2):11-13,37.WU Haojiang,CAI Zhengmin,ZHOU Fangde.Identification of the flow pattern of oil-gas-water multiphase flow by neural network [J].Journal of Xi’an Petroleum Institute (Natural Science Edition),2000,15(2):11-13,37.
- [2] 姜俊泽,张伟明.水平管道气液两相段塞流参数计算的精确模型[J].化工学报,2012,63(12):3826-3831.JIANG Junze,ZHANG Weiming.Accurate model for computing parameters of gas-liquid slug flow in horizontal pipe[J].CIESC Journal,2012,63(12):3826-3831.
- [3] 花靖,蒋秀,于超,等.低含液输气管道内积液特性研究现状及展望[J].安全、健康和环境,2021,21(2):10-16.HUA Jing,JIANG Xiu,YU Chao,et al.Current status and prospects of research on characteristics of liquid accumulation in gas pipelines with low liquid content[J].Safety Health & Environment,2021,21(2):10-16.
- [4] 管孝瑞,金有海,王建军,等.低含液输气管线内两相流动及其CO2腐蚀研究进展[J].化工机械,2017,44(3):245-251.GUAN Xiaorui,JIN Youhai,WANG Jianjun,et al.Progress in investigating gas-liquid flow and CO2 corrosion in gas transmission pipelines with low liquid loading[J].Chemical Engineering & Machinery,2017,44(3):245-251.
- [5] BEGGS H D,BRILL J P.A study of two-phase flow in inclined pipelines[J].J Pet Technol,1973,25:607-617.
- [6] HART N P,HAMERSMA P J,FOURTUIN J M H.Correlations predicting frictional pressure drop and liquid holdup during horizontal gas-liquid pipe flow with a small liquid holdup[J].International Journal of Multiphase Flow,1989,15(6):947-964.
- [7] TAITEL Y,DUKLER A E.A model for predicting flow regime transitions in horizontal and near horizontal gas-liquid flow[J].AIChE Journal,1976,22(1):47-54.
- [8] GROLMAN E,FORTUIN J M H.Gas-liquid flow in slightly inclined pipes[J].Chemical Engineering Science,1997,52(24):4461-4471.
- [9] CHEN X T,CAL X D,BRILL J P.Gas-liquid stratified-wavy flow in horizontal pipelines[J].Journal of Energy Resources Technology,1997,119(4):209-216.
- [10] BANAFI A,TALAIE M R.A new mechanistic model to predict gas-liquid interface shape of gas-liquid flow through pipes with low liquid loading[J].AIChE Journal,2015,61(3):1043-1053.
- [11] HAMERSMA P J,HART J.A pressure drop correlation for gas/liquid pipe flow with a small liquid holdup[J].Chemical Engineering Science,1987,42(5):1187-1196.
- [12] BANAFI A,TALAEI M R,GHAFOORI M J.A comprehensive comparison of the performance of several popular models to predict pressure drop in stratified gas-liquid flow with low liquid loading[J].Journal of Natural Gas Science & Engineering,2014,21(6):433-441.
- [13] 邵孟良,于颖敏.基于遗传算法的BP神经网络气液两相流持液率预测模型优化[J].西安石油大学学报(自然科学版),2019,34(6):44-49.SHAO Mengliang,YU Yingmin.Optimization of gas-liquid two-phase flow liquid hold-up prediction model with BP neural network based on genetic algorithm[J].Journal of Xi’an Shiyou University(Natural Science Edition),2019,34(6):44-49.
- [14] ROS N C.Simultaneous flow of gas and liquid as encountered in well tubing[J].SPE Journal of Petroleum Technology,1961,13(10):68-78.
- [15] 陈星杙,刘伟,袁宗明,等.基于ACE算法的水平管道持液率计算模型[J].中国海上油气,2018,30(2):180-186.CHEN Xingyi,LIU Wei,YUAN Zongming,et al.An ACE algorithm-based model to calculate liquid holdup in horizontal pipe[J].China Offshore Oil and Gas,2018,30(2):180-186.
- [16] BADIE S,HALE C P,LAWRENCE C J,et al.Pressure gradient and holdup in horizontal two-phase gas-liquid flows with low liquid loading[J].International Journal of Multiphase Flow,2000,26(9):1525-1543.
- [17] ABDUL-MAJEED G H.Liquid holdup in horizontal two-phase gas-liquid flow[J].Journal of Petroleum Science and Engineering,1996,15(2/4):271-280.
- [18] MINAMI K,BRILL J P.Liquid holdup in wet-gas pipelines[J].SPE Production Engineering,1987,2(1):36-44.
- [19] KOKAL S L,STANISLAV J F.An experimental study of two-phase flow in slightly inclined pipes—II.Liquid holdup and pressure drop[J].Chemical Engineering Science,1989,44(3):681-693.
- [20] MENG W,CHEN X T,KOUBA G E,et al.Experimental study of low-liquid-loading gas-liquid flow in near-horizontal pipes[J].SPE Production & Facilities,2001,16(4):240-249.
- [21] NGUYEN V T,SPEDDING P L.Holdup in two-phase,gas-liquid flow—II[J].Chemical Engineering Science,1977,32(9):1015-1021.
- [22] MUKHERJEE H.An experimental study of inclined two-phase flow[D].Tulsa:University of Tulsa,1979.
- [23] FAN Y.An investigation of low liquid loading gas liquid stratified flow in near-horizontal pipes[D].Tulsa:University of Tulsa,2005.
- [24] FAN Y,PEREYRA E,SARICA C,et al.Analysis of flow pattern transition from segregated to slug flow in upward inclined pipes[J].International Journal of Multiphase Flow,2019,115(1):19-39.
- [25] 许明阳,殷晨波,陈曦.基于改进型支持向量机的胶接强度预测模型[J].合肥工业大学学报(自然科学版),2021,44(2):164-169.XU Mingyang,YIN Chenbo,CHEN Xi.Prediction model of bonding strength based on improved support vector machine[J].Journal of Hefei University of Technology (Natural Science Edition),2021,44(2):164-169.
- [26] 薛浩然,张珂珩,李斌,等.基于布谷鸟算法和支持向量机的变压器故障诊断[J].电力系统保护与控制,2015,43(8):8-13.XUE Haoran,ZHANG Keheng,LI Bin,et al.Fault diagnosis of transformer based on the cuckoo search and support vector machine[J].Power System Protection and Control,2015,43(8):8-13.
- [27] 陈伟根,滕黎,刘军,等.基于遗传优化支持向量机的变压器绕组热点温度预测模型[J].电工技术学报,2014,29(1):44-51.CHEN Weigen,TENG Li,LIU Jun,et al.Transformer winding hot-spot temperature prediction model of support vector machine optimized by genetic algorithm[J].Transactions of China Electrotechnical Society,2014,29(1):44-51.
- [28] 李静,徐路路.基于机器学习算法的研究热点趋势预测模型对比与分析:BP神经网络、支持向量机与LSTM模型[J].现代情报,2019,39(4):23-33.LI Jing,XU Lulu.Comparison and analysis of research trend prediction models based on machine learning algorithm:BP neural network,support vector machine and LSTM model[J].Journal of Modern Information,2019,39(4):23-33.
- [29] 刘浩然,赵翠香,李轩,等.一种基于改进遗传算法的神经网络优化算法研究[J].仪器仪表学报,2016,37(7):1573-1580.LIU Haoran,ZHAO Cuixiang,LI Xuan,et al.Study on a neural network optimization algorithm based on improved genetic algorithm[J].Chinese Journal of Scientific Instrument,2016,37(7):1573-1580.
- [30] 高飞.MATLAB智能算法超级学习手册[M].北京:人民邮电出版社,2014:224-248.
- [31] 张俊,殷坤龙,王佳佳,等.基于时间序列与PSO-SVR耦合模型的白水河滑坡位移预测研究[J].岩石力学与工程学报,2015,34(2):382-391.ZHANG Jun,YIN Kunlong,WANG Jiajia,et al.Displacement prediction of baihuihe landslide based on time series and PSO-SVR model[J].Chinese Journal of Rock Mechanics and Engineering,2015,34(2):382-391.
- [32] 喻西崇,赵金洲,冯叔初.起伏多相流管路持液率计算方法研究[J].西南石油学院学报,2000,22(3):94-97.YU Xichong,ZHAO Jinzhou,FENG Shuchu.The prediction method of holdup in undulated multiphase pipelines[J].Journal of Southwest Petroleum Institute,2000,22(3):94-97.