基于交互式递归分析的两相流流型识别方法
|
摘要
针对存在两相流流型识别速度慢且识别准确度不高的问题,提出了一种基于交互式递归分析无须成像的流型识别方法。利用邻域误差法对电容传感器测得的流型电容值进行分析得到最优嵌入维数,对几种典型流型建立交互式递归图的嵌入维数;再利用交互式递归对仿真流型及参考流型的电容值进行分析,得到一组交互式递归图;通过比较交互式递归图主对角线递归黑点的比例来确定流型的相似度。对典型油气两相流流型识别做了仿真实验,结果表明,该方法可准确识别两相流流型,且识别速度快。
Aiming at the problems that the two-phase flow regime recognition speed is slow and the recognition accuracy is low, a flow regime recognition method is proposed based on cross recursive analysis(CRA) without imaging. The false nearest neighbors are used to analyze flow regime capacitance values measured by capacitance sensor to obtain an optimal embedding dimension; and the embedding dimension of cross recursive plot is established for several typical flow patterns. The capacitance values of the simulation flow regimes and the reference flow regimes are analyzed using cross recursion to get a set of cross recursive plots. The similarity of two flow regimes is determined by comparing the ratio of black recursive dots on the main diagonal of cross recursive plot. The simulation experiments of typical oil-gas two-phase flow regime identification are conducted. The simulation result shows that the method can accurately identify different flow regimes of two-phase flow; and the recognition rate is fast.
Aiming at the problems that the two-phase flow regime recognition speed is slow and the recognition accuracy is low, a flow regime recognition method is proposed based on cross recursive analysis(CRA) without imaging. The false nearest neighbors are used to analyze flow regime capacitance values measured by capacitance sensor to obtain an optimal embedding dimension; and the embedding dimension of cross recursive plot is established for several typical flow patterns. The capacitance values of the simulation flow regimes and the reference flow regimes are analyzed using cross recursion to get a set of cross recursive plots. The similarity of two flow regimes is determined by comparing the ratio of black recursive dots on the main diagonal of cross recursive plot. The simulation experiments of typical oil-gas two-phase flow regime identification are conducted. The simulation result shows that the method can accurately identify different flow regimes of two-phase flow; and the recognition rate is fast.
引文
[1]李孝禄,王文越,张远辉,等.液压制动管路中气液两相流流型聚类分析识别[J].农业机械学报,2016,47(2):377-383.Li Xiaolu,Wang Wenyue,Zhang Yuanhui,et al.Identification of gas-liquid two-phase flow patterns in hydraulic braking pipeline based on cluster analysis[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(2):377-383.
[2]祁丽.管内气液两相流流型的智能识别[D].厦门:集美大学,2014:7-13.Qi Li.Intelligent Recognition of Flow Pattern of Gas-liquid Two Phase Flow in Horizontal Pipe[D].Xiamen:JiMei University,2014:7-13.
[3]何雪.面向管道多相流检测的电容层析成像研究[D].北京:北京工业大学,2014:23-30.He Xue.Research on Electrical Capacitance Tomography to Multiphase Pipeline Detection[D].Beijing:Beijing University of Technology,2014:23-30.
[4]宋蕾.基于神经网络的电容层析成像系统流型识别研究[D].哈尔滨:哈尔滨理工大学,2015:15-26.Song Lei.Study on Identification of Flow Regimes Based on Neural Networks for Electrical Capacitance Tomography System[D].Harbin:Harbin University of Science and Technology,2015:15-26.
[5]陈飞,周云龙,窦华容.基于图像小波变换的气液两相流流型识别[J].工程热物理学报,2007,22(5):752-756.Chen Fei,Zhou Yunlong,Dou Huarong.Identification of gas-liquid two-phase flow pattern based on image wavelet transform[J].Journal of Engineering Thermophysics,2007,22(5):752-756.
[6]周云龙,李洪伟,袁俊文.一种识别气液两相流流型的新方法[J].热能动力工程,2009,24(1):68-72.Zhou Yunlong,Li Hongwei,Yuan Junwen.A new method for identification of gas-liquid two-phase flow pattern[J].Thermal Power Engineering,2009,24(1):68-72.
[7]唐雪琴,王侃,徐宗昌,等.基于MAPSO算法的小波神经网络训练方法研究[J].系统仿真学报,2012,24(3):104-108.Tang Xueqin,Wang Kan,Xu Zongchang,et al.Research on WNN Training Algorithm Based on MAPSOAlgorithm[J].Journal of System Simulation,2012,24(3):104-108.
[8]周云龙,顾杨杨.基于独立分量分析和RBF神经网络的气液两相流流型识别[J].化工学报,2012,63(3):796-799.Zhou Yunlong,Gu Yangyang.Flow regime identification of gas/liquid two-phase flow based ICA and RBF neural network[J].CIESC Journal,2012,63(3):796-799.
[9]龙军.基于传感器数据融合的小通道气液两相流参数测量新方法研究[D].杭州:浙江大学,2013:31-44.Long Jun.Study on new measurement methods of gas-liquid two-phase flow in small channels based on sensor data fusion[D].Hangzhou:Zhejiang University,2013:31-44.
[10]闫润强,朱贻盛.基于信号递归度分析的语音端点检测方法[J].通信学报,2007,28(1):35-39.Yan Runqiang,Zhu Yisheng.Speech endpoint detection based on recurrence rate analysis[J].Journal on Communiacations,2007,28(1):35-39.
[11]WANG C H,ZHONG ZH P,E J Q.Flow Regime Recognition in Spouted Bed Based on Recurrence Plot Method[J].Powder Technology(S0032-5910),2012,219:20-28.
[12]董芳,金宁德,宗艳波,等.两相流流型动力学特征多尺度递归定量分析[J].物理学报,2008,57(10):6145-6154.Dong Fang,Jin Ningde,Zong Yanbo,et al.Multi-Scale Recurrence Quantification Analysis of the Dynamic Characteristics of Two Phase Flow Pattern[J].Acta Physica Sinina,2008,57(10):6145-6154.
[13]蒲晓川.基于递归分析方法的齿轮故障诊断[D].武汉:武汉科技大学,2014:16-28.Pu Xiaochuan.A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Master in Engineering[D].Wuhan:Wuhan University of Science and Technology,2014:16-28.
[14]AGUILA J J,MARIN I,ARIAS E,et al.High Performance Computing Applied to the False Neare st Neighbors Method:Box-assisted and kd-tree app roaches[J].Lecture Notes in Electrical Engineering(S1876-1110),2011,90(7):323-336.
[15]AMIRA Z,JAMAL CH,JEAN M G.Reducing Sojourn Points from Recurrence Plots to Improve Transition Detection:Application to Fetal Heart Rate Transitions[J].Computers in Biology and Medicine(S0010-4825),2015,63(1):251-260.
[16]MIQUEL F L,NARCIS G,FRANCESC X L.Recurrence Plots to Characterize Gas-Solid Fluidization Regimes[J].International Journal of Multiphase Flow(S0301-9322),2015,72(3):43-56.
[17]成卫青,唐旋.一种基于改进互信息和信息熵的文本特征选择方法[J].南京邮电大学学报(自然科学版),2013,33(5):63-68.Cheng Weiqing,Tang Xuan.A Text Feature Selection Method Using the Improved Mutual Information and Information Entropy[J].Journal of Nanjing University of Posts and Telecommunications(Natural Science),2013,33(5):63-68.
[18]徐大露,薛倩,马敏,等.基于小尺度ECT传感器的滑油在线监测研究[J].仪表技术与传感器,2015(11):34-37.Xu Dalu,Xue Qian,Ma Min,et al.Research on Online Lubrication Oil Monitoring Based on Small Scale ECTSensor[J].2015(11):34-37.
[2]祁丽.管内气液两相流流型的智能识别[D].厦门:集美大学,2014:7-13.Qi Li.Intelligent Recognition of Flow Pattern of Gas-liquid Two Phase Flow in Horizontal Pipe[D].Xiamen:JiMei University,2014:7-13.
[3]何雪.面向管道多相流检测的电容层析成像研究[D].北京:北京工业大学,2014:23-30.He Xue.Research on Electrical Capacitance Tomography to Multiphase Pipeline Detection[D].Beijing:Beijing University of Technology,2014:23-30.
[4]宋蕾.基于神经网络的电容层析成像系统流型识别研究[D].哈尔滨:哈尔滨理工大学,2015:15-26.Song Lei.Study on Identification of Flow Regimes Based on Neural Networks for Electrical Capacitance Tomography System[D].Harbin:Harbin University of Science and Technology,2015:15-26.
[5]陈飞,周云龙,窦华容.基于图像小波变换的气液两相流流型识别[J].工程热物理学报,2007,22(5):752-756.Chen Fei,Zhou Yunlong,Dou Huarong.Identification of gas-liquid two-phase flow pattern based on image wavelet transform[J].Journal of Engineering Thermophysics,2007,22(5):752-756.
[6]周云龙,李洪伟,袁俊文.一种识别气液两相流流型的新方法[J].热能动力工程,2009,24(1):68-72.Zhou Yunlong,Li Hongwei,Yuan Junwen.A new method for identification of gas-liquid two-phase flow pattern[J].Thermal Power Engineering,2009,24(1):68-72.
[7]唐雪琴,王侃,徐宗昌,等.基于MAPSO算法的小波神经网络训练方法研究[J].系统仿真学报,2012,24(3):104-108.Tang Xueqin,Wang Kan,Xu Zongchang,et al.Research on WNN Training Algorithm Based on MAPSOAlgorithm[J].Journal of System Simulation,2012,24(3):104-108.
[8]周云龙,顾杨杨.基于独立分量分析和RBF神经网络的气液两相流流型识别[J].化工学报,2012,63(3):796-799.Zhou Yunlong,Gu Yangyang.Flow regime identification of gas/liquid two-phase flow based ICA and RBF neural network[J].CIESC Journal,2012,63(3):796-799.
[9]龙军.基于传感器数据融合的小通道气液两相流参数测量新方法研究[D].杭州:浙江大学,2013:31-44.Long Jun.Study on new measurement methods of gas-liquid two-phase flow in small channels based on sensor data fusion[D].Hangzhou:Zhejiang University,2013:31-44.
[10]闫润强,朱贻盛.基于信号递归度分析的语音端点检测方法[J].通信学报,2007,28(1):35-39.Yan Runqiang,Zhu Yisheng.Speech endpoint detection based on recurrence rate analysis[J].Journal on Communiacations,2007,28(1):35-39.
[11]WANG C H,ZHONG ZH P,E J Q.Flow Regime Recognition in Spouted Bed Based on Recurrence Plot Method[J].Powder Technology(S0032-5910),2012,219:20-28.
[12]董芳,金宁德,宗艳波,等.两相流流型动力学特征多尺度递归定量分析[J].物理学报,2008,57(10):6145-6154.Dong Fang,Jin Ningde,Zong Yanbo,et al.Multi-Scale Recurrence Quantification Analysis of the Dynamic Characteristics of Two Phase Flow Pattern[J].Acta Physica Sinina,2008,57(10):6145-6154.
[13]蒲晓川.基于递归分析方法的齿轮故障诊断[D].武汉:武汉科技大学,2014:16-28.Pu Xiaochuan.A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Master in Engineering[D].Wuhan:Wuhan University of Science and Technology,2014:16-28.
[14]AGUILA J J,MARIN I,ARIAS E,et al.High Performance Computing Applied to the False Neare st Neighbors Method:Box-assisted and kd-tree app roaches[J].Lecture Notes in Electrical Engineering(S1876-1110),2011,90(7):323-336.
[15]AMIRA Z,JAMAL CH,JEAN M G.Reducing Sojourn Points from Recurrence Plots to Improve Transition Detection:Application to Fetal Heart Rate Transitions[J].Computers in Biology and Medicine(S0010-4825),2015,63(1):251-260.
[16]MIQUEL F L,NARCIS G,FRANCESC X L.Recurrence Plots to Characterize Gas-Solid Fluidization Regimes[J].International Journal of Multiphase Flow(S0301-9322),2015,72(3):43-56.
[17]成卫青,唐旋.一种基于改进互信息和信息熵的文本特征选择方法[J].南京邮电大学学报(自然科学版),2013,33(5):63-68.Cheng Weiqing,Tang Xuan.A Text Feature Selection Method Using the Improved Mutual Information and Information Entropy[J].Journal of Nanjing University of Posts and Telecommunications(Natural Science),2013,33(5):63-68.
[18]徐大露,薛倩,马敏,等.基于小尺度ECT传感器的滑油在线监测研究[J].仪表技术与传感器,2015(11):34-37.Xu Dalu,Xue Qian,Ma Min,et al.Research on Online Lubrication Oil Monitoring Based on Small Scale ECTSensor[J].2015(11):34-37.