Title :
Regime recognition of two-phase pipe flow based on D-S evidence theory
Author :
Zhang, Fu-sheng ; Dong, Feng ; Tan, Chao ; Liu, Dong
Author_Institution :
Sch. of Electr. Eng. & Autom., Tianjin Univ., Tianjin
Abstract :
Flow regime recognition plays a fundamental role in two-phase flow measurement. Based on the characters information provided by dual-plane electrical resistance tomography (ERT) system, a method of flow regime recognition in horizontal pipe based on D-S evidence theory is brought forward. A particular way to acquire the basic probability assignment function in D-S evidence theory is introduced. Every single plane of the dual-plane ERT system has been regarded as an independent information resource. Through the fusion of features from both of the planes, less features are needed to be extracted, and better flow regime identifying result is achieved.
Keywords :
computational fluid dynamics; flow measurement; pipe flow; probability; tomography; two-phase flow; uncertainty handling; D-S evidence theory; dual-plane ERT system; dual-plane electrical resistance tomography system; flow regime recognition; probability assignment function; two-phase pipe flow measurement; Data mining; Electric resistance; Electric variables measurement; Electrical resistance measurement; Fluid flow; Fluid flow measurement; Machine learning; Plugs; Sensor fusion; Tomography; D-S evidence theory; Electrical resistance tomography; Flow regime recognition; Two-phase flow;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620386