Gas-Water Two-Phase Flow Regime Recognition with Data and Feature Fusion from a Dual-Plane ERT System

Author

Tan, Chao ; Dong, Feng

Author_Institution

Tianjin Key Lab. of Process Meas. & Control, Tianjin Univ., Tianjin, China

Volume

2

fYear

2009

fDate

14-16 Aug. 2009

Firstpage

218

Lastpage

222

Abstract

Gas-water two-phase flow is of significance in industrial process and scientific research field. Its flow regime determines the flow parameters and the method of flow measurement. Precise identification of flow regime has been a popular subject for a long time. In this work, a series experiments on gas-water two-phase flow were conducted in a 50 mm diameter horizontal pipe, the flow parameters were measured with a dual-plane electrical resistance tomography (ERT) system. A method of feature extraction from the measurement data is presented and the flow regime was recognized by using a support vector machine (SVM) method. Additionally, the data fusion methods are selected and compared to discuss the method of improving the flow regime recognition performance.

Keywords

electric resistance; feature extraction; flow measurement; mechanical engineering computing; pipe flow; sensor fusion; support vector machines; tomography; two-phase flow; ERT system; SVM method; data fusion method; dual-plane electrical resistance tomography system; feature extraction method; feature fusion methid; flow measurement method; flow parameters; gas-water two-phase flow regime recognition; horizontal pipe; industrial process; size 50 mm; support vector machine method; Conductivity; Electric resistance; Electric variables measurement; Electrical resistance measurement; Feature extraction; Fluid flow measurement; Gas industry; Plugs; Tomography; Voltage; data fusion; electrical resistance tomography; feature extraction; feature fusion; flow regime;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2009. ICNC '09. Fifth International Conference on

Conference_Location

Tianjin

Print_ISBN

978-0-7695-3736-8

Type

conf

DOI

10.1109/ICNC.2009.293

Filename

5365515