A novel method for flow pattern identification in unstable operational conditions using gamma ray and radial basis function. (May 2017)

• Record Type: Journal Article
• Authors: Roshani, G.H.; Nazemi, E.; Roshani, M.M.
• Is Part Of: Applied radiation and isotopes. Volume 123(2017:May)
• Page Start: 60
• Page End: 68
• Publication Date: 2017-05
• Subjects: Radial basis function -- Neural network -- Flow pattern recognition -- Density independent -- Multi beam gamma ray; Radiology -- Periodicals; Radiation -- Industrial applications -- Periodicals; Nuclear chemistry -- Periodicals; Internet resource; Periodical; 660.298
• DOI: 10.1016/j.apradiso.2017.02.023
• Languages: English
• ISSNs: 0969-8043

Abstract

Abstract: Changes of fluid properties (especially density) strongly affect the performance of radiation-based multiphase flow meter and could cause error in recognizing the flow pattern and determining void fraction. In this work, we proposed a methodology based on combination of multi-beam gamma ray attenuation and dual modality densitometry techniques using RBF neural network in order to recognize the flow regime and determine the void fraction in gas-liquid two phase flows independent of the liquid phase changes. The proposed system is consisted of one 137 Cs source, two transmission detectors and one scattering detector. The registered counts in two transmission detectors were used as the inputs of one primary Radial Basis Function (RBF) neural network for recognizing the flow regime independent of liquid phase density. Then, after flow regime identification, three RBF neural networks were utilized for determining the void fraction independent of liquid phase density. Registered count in scattering detector and first transmission detector were used as the inputs of these three RBF neural networks. Using this simple methodology, all the flow patterns were correctly recognized and the void fraction was predicted independent of liquid phase density with mean relative error (MRE) of less than 3.28%. Highlights: Flow regime and void fraction were determined in two phase flows independent of the liquid phase density changes. An experimental structure was set up and the required data was obtained. 3 detectors and one gamma source were used in detection geometry. RBF networks were utilized for flow regime and void fraction determination.