On-Line Identification of Horizontal Two-Phase Flow Regimes Through Gabor Transform and Neural Network Processing

The fundamental objective of this work is the construction of a specialist system capable of diagnosing different configurations of horizontal two-phase flow regimes. It is important to emphasize that the development of this know-how is capital to the efficient operation of facilities for manipulation and transportation of multiphase fluids, and represents today one of the most important challenges in petrochemical and thermonuclear industries. The working principle of the proposed system is based on the signals acquired by a rapid response pressure gradient sensor, and on its post processing through Gabor Transform and on a previously trained artificial neural network. The implementation is accomplished in way that the diagnosis operation is performed on-line, from acquisition of the signal to its post-processing. Experimental results were obtained on the experimental circuit at NETeF — Nu´cleo de Engenharia Te´rmica e Fluidos of USP — Universidade de Sa˜o Paulo, at Sa˜o Carlos, using a horizontal test section, with 12m length and 30mm internal diameter. Experiments were done with the following air-water flow regimes: stratified smooth, stratified wavy, intermittent, annular and bubbly. Results show that the percentage of correct flow regime diagnosis in steady state conditions is practically of 100%.