基于1D-CNN-AdaBoost及电阻层析成像的两相流流型辨识

doi:10.3969/j.issn.1000-1158.2022.12.16

摘要
图/表
参考文献(14)
相关文章 (15)

全文: PDF (2402 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要基于电阻层析成像(ERT)系统采集的垂直管气液两相流测量数据,将一维卷积神经网络(1D-CNN)与AdaBoost(Adaptive Boosting)相结合,构建1D-CNN-AdaBoost算法,进行了气液两相流流型辩识研究。该算法使用5个1D-CNN作为弱分类器,通过实验数据样本训练,结合AdaBoost形成最终的强分类器。将1D-CNN-AdaBoost算法与BP神经网络、支持向量机及决策树算法进行比较,结果表明,1D-CNN-AdaBoost算法辨识正确率高于其他算法,平均辨识正确率可达97%。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张立峰
	肖凯
	华回春

关键词 ：计量学, 流型辨识, 电阻层析成像, 卷积神经网络, 自适应提升

Abstract：Based on the vertical tube gas-liquid two-phase flow measurement data collected by the electrical resistance tomography (ERT) system, the one-dimensional convolutional neural network (1D-CNN) and AdaBoost (Adaptive Boosting) are combined to construct the 1D-CNN-AdaBoost algorithm, a study on the flow pattern identification of gas-liquid two-phase flow has been carried out. The algorithm uses five 1D-CNNs as weak classifiers, trained on experimental data samples, and combined with AdaBoost to form the final strong classifier. Comparing 1D-CNN-AdaBoost algorithm with BP neural network, support vector machine and decision tree algorithm, the results show that the recognition accuracy of 1D-CNN-AdaBoost algorithm is higher than other algorithms, and the average recognition accuracy can reach 97%.

Key words： metrology flow pattern identification electrical capacitance tomography convolutional neural network adaptive boosting

收稿日期: 2021-06-15 发布日期: 2022-12-28

PACS:

TB937

基金资助:国家自然科学基金(61973115)

作者简介: 张立峰(1979-),男,江西临川人,华北电力大学副教授,博士,主要从事多相流检测及电学层析成像技术方面的研究。Email:lifeng.zhang@ncepu.edu

引用本文:

张立峰,肖凯,华回春. 基于1D-CNN-AdaBoost及电阻层析成像的两相流流型辨识[J]. 计量学报, 2022, 43(12): 1622-1626.
ZHANG Li-feng,XIAO Kai,HUA Hui-chun. Identification of Two-phase Flow Pattern Based on 1D-CNN-AdaBoost and Electrical Resistance Tomography. Acta Metrologica Sinica, 2022, 43(12): 1622-1626.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2022.12.16 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2022/V43/I12/1622

［1］张立峰, 王化祥. 基于SVM及电容层析成像的两相流流型识别［J］. 仪器仪表学报, 2009, 30(4):812-816.
Zhang L F, Wang H X. Two-phase flow pattern identification based on SVM and electrical capacitance tomography［J］. Chinese Journal of Scientific Instrument, 2009, 30(4):812-816.
［2］陈露阳, 尹佳雯, 孙志强, 等. 基于EEMD-Hilbert谱的气液两相流钝体绕流流型识别［J］. 仪器仪表学报, 2017, 38(10):2536-2546.
Chen L Y, Yin J W, Sun Z Q, et al. Flow pattern identification of gas-liquid two-phase flow around bluff bodies based on EEMD-Hilbert spectrum ［J］. Chinese Journal of Scientific Instrument, 2017, 38(10):2536-2546.
［3］Zhan L, Lin Y L, Tu Q Y, et al. Investigation of gassolids flow hydrodynamics in a cold model of a dual fluidised bed gasifier using electrical capacitance tomography sensors［J］. Particuology, 2020, 51: 193-204.
［4］孙海亮, 李彦龙, 刘昌岭, 等. 电阻层析成像技术及其在水合物开采模拟实验中的应用［J］. 计量学报, 2019, 40(3):455-461.
Sun H L, Li Y L, Liu C L, et al. Electrical resistance tomography technology and its application in hydrate mining simulation experiment ［J］. Acta Metrologica sinica, 2019, 40(3):455-461.
［5］申超群, 卫怀玉, 陈德运. 基于神经网络的ERT系统流型辨识的研究［J］. 哈尔滨理工大学学报, 2008, 13(6):31-34.
Shen C Q, Wei H Y, Chen D Y. Research on Flow Pattern Identification of ERT System Based on Neural Network［J］. Journal of Harbin University of Science and Technology, 2008, 13(6):31-34.
［6］陈德运, 张华, 朱波, 等. 油水两相流电阻层析成像系统流型的辨识［J］. 电机与控制学报, 2007, 11(6):639-643.
Chen D Y, Zhang H, Zhu B, et al. Flow pattern iden-tification of oil-water two-phase flow electrical resistance tomography system ［J］. Journal of Electrical Engineering and Control, 2007, 11(6): 639-643.
［7］朱波. 基于小波分析和神经网络的ERT系统流型辨识研究［D］. 哈尔滨:哈尔滨理工大学, 2008.
［8］许燕斌, 王化祥, 崔自强. 基于独立成分分析的水平管气液两相流分相信息提取［J］. 化工学报, 2009, 60(12):3012-3018.
Xu Y B, Wang H X, Cui Z Q. Extraction of phase separation information of horizontal pipe gas-liquid two-phase flow based on independent component analysis［J］.CIESC Journal, 2009, 60(12):3012-3018.
［9］董健, 柳亦兵, 滕伟, 等. 基于BP_Adaboost算法的风电机组叶片结冰检测［J］. 可再生能源, 2021, 39(5): 632-636.
Dong J, Liu Y B, Teng W, et al. Wind turbine blade icing detection based on BP_Adaboost algorithm［J］. Renewable Energy Resources, 2021, 39 (5): 632-636.
［10］张立峰, 王智, 吴思橙. 基于卷积神经网络与门控循环单元的气液两相流流型识别方法［J］. 计量学报, 2022, 43 (10): 1306-1312.
Zhang L F, Wang Z, Wu Si C. Flow Pattern Recogn-ition of Gas-liquid Two-phase Flow Based on Convolu-tional Neural Network ［J］. Acta Metrologica sinica, 2022, 43(10): 1306-1312.
［11］张立峰, 朱炎峰. 基于MO-PLP-ELM及电容层析成像的两相流流型辨识［J］. 计量学报, 2021, 42(3):334-338.
Zhang L F, Zhu Y F. Flow pattern identification of two-phase flow based on MO-PLP-ELM and electrical capacitance tomography ［J］. Acta Metrologica sinica, 2021, 42(3):334-338.
［12］仝卫国, 朱赓宏, 顾浩. 基于层析成像的气液两相流相关流量测量方法［J］. 计量学报, 2020, 41(10):1245-1251.
Tong W G, Zhu G H, Gu H. A gas-liquid two-phase flow correlation flow measurement method based on tom-ography ［J］. Acta Metrologica Sinica, 2020, 41(10):1245-1251.
［13］许燕斌, 王化祥, 崔自强. 水平管气水两相流分相界面识别［J］. 天津大学学报, 2010, 43(8):743-748.
Xu Y B, Wang H X, Cui Z Q. Identification of phase separation interface of horizontal pipe gas-water two-phase flow［J］. Journal of Tianjin University, 2010, 43(8):743-748.
［14］仝卫国, 庞雪纯, 朱赓宏. 基于卷积神经网络的气液两相流流型识别方法［J］. 系统仿真学报, 2021, 33(4):883-891.
Tong W G, Pang X C, Zhu G H. Flow pattern identification method of gas-liquid two-phase flow based on convolutional neural network ［J］. Journal of System Simulation, 2021, 33(4):883-891.