Begell House Inc.
Multiphase Science and Technology
MST
0276-1459
24
2
2012
A NEURAL NETWORK ALGORITHM FOR DENSITY MEASUREMENT OF MULTIPHASE FLOW
89-103
10.1615/MultScienTechn.v24.i2.10
Nabeel
Al-Rawahi
College of Engineering, Sultan Qaboos University
Mahmoud
Meribout
Electrical Engineering Department, Petroleum Institute, Abu Dhabi
Ahmed
Al-Naamany
College of Engineering, Sultan Qaboos University, Muscat, Oman
Ali
Al-Bimani
College of Engineering, Sultan Qaboos University, Muscat, Oman
Adel
Meribout
Electrical Engineering Department, Petroleum Institute, Abu Dhabi
neural network algorithm
density measurement
multiphase flow loop
Measurement of the density of multiphase flow is crucial in multiphase flow meters. While radioactive-based measuring methods are known to give good results, end users prefer nonradioactive devices for obvious reasons. This paper shows the capability of an online neural network algorithm, that uses the differential pressure along a vertical pipe and across a venturi meter, in predicting the average density of the multiphase flow. A multiphase flow loop was constructed to conduct the training and evaluating experiments needed for the algorithm. Experimental results performed on the multiphase flow loop demonstrate that the density measurement can be achieved with good accuracy for liquid and gas velocities for up to 4 and 25 m/s (at 30° C and 7 bars), respectively, while covering different complex flow regimes including annular flow, slug flow, and dispersed flow. The neural network algorithm that was developed for this purpose gave very good results in measuring the flow density.
A MULTISIZE MODEL FOR BOILING BUBBLY FLOWS
105-179
10.1615/MultScienTechn.v24.i2.20
Didier
Zaepffel
CEA Grenoble, DEN/DM2S/STMF/LMSF, 17; rue des Martyrs, 38054, Grenoble, Cedex 9, France
Christophe
Morel
Commissariat a l'Energie Atomique, DEN/DM2S/STMF/LMSF, 17; rue des Martyrs, 38054, Grenoble, Cedex 9, France
Daniel
Lhuillier
Laboratoire de Modélisation en Méecanique, Universitée Pierre et Marie Curie et CNRS, 8, rue du capitaine Scott, 75015 Paris, France
bubbles
multi-size
two-phase
moments method
coalescence
breakup
quadratic law
This paper concerns the modeling of boiling bubbly flow with a multisize approach. In industrial applications as well as in natural bubbly flows, bubbles generally exhibit a full spectrum of different sizes and shapes that influences their behavior. Two major methods have emerged in the literature to take into account the effects of these distributions. The first one is the moments method and the second one is the MUSIG (for MUlti SIze Group) method. In the moments method, a given mathematical law describes the bubble size distribution. In the MUSIG method, a discrete bubble size distribution is assumed and used in the numerical solution of the problem. In this paper a moments method is developed and applied to boiling bubbly flows in a vertical pipe. The results are compared to experimental data. The axial changes of the local void fraction and liquid temperature radial profiles are well reproduced by the model, despite the too-high values of the predicted interfacial area concentration (and consequently the too-low values of the bubble Sauter mean diameter). This discrepancy between a good prediction of the void fraction and an overestimated interfacial area concentration could be due to the nonsphericity of the bubbles.