Caracterização experimental do escoamento intermitente líquido-gás em tubulações horizontais

• Ano de defesa: 2013
• Autor(a) principal: Vicencio, Fernando Enrique Castillo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Programa de Pós-Graduação em Engenharia Mecânica e de Materiais
• Palavras-chave em Português: Escoamento bifásico; Modelos matemáticos; Métodos de simulação; Engenharia Mecânica; Two-phase flow; Mathematical models; Simulation methods; Mechanical engineering
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/825

The goal of the present work is to broaden the knowledge on the dynamics of two-phase intermittent gas-liquid flows through the experimental characterization of their parameters. The experimental work was carried out at the LACIT-UTFPR labs and a two-phase flow apparatus consisting of a 25.8-mm ID, 9-m long transparent line was used. A pair of 12x12-nodes wire mesh sensors 3.75 cm apart was used to identify the existing phase at each node. A test grid comprehending a broad range of gas and liquid flow rates within the intermittent flow pattern region was chosen. The operational limitations of the experimental rig were taken into account on defining the boundaries of the aforementioned test grid. The temporal series of the data, as well as characteristic parameters of the intermittent flows such as frequency, void fraction, elongated bubble velocity and the lengths of the liquid slug and the elongated bubble were obtained after signal processing. The probability distribution of each parameter was found and approximated to a known statistical distribution such as the normal or the log-normal ones. Correlations to fit the averaged parameters and their standard deviations were proposed. The objective of such correlations is to predict the average values of the characteristic parameters of the intermittent flows so that such predicted values might be used as input data for the theoretical models developed. Finally, statistical distributions based on the empirical correlations and their normal and log-normal distributions were found. These distributions validated the aforementioned correlations, which can henceforth be used in future theoretical modelling in this field.