Aplicação de análise wavelet para detecção de instabilidades fluidodinâmicas em um processo de umedecimento de partículas em leito de jorro
Date
2022-02-02
Authors
Machado, Vinicius Seferian Scheffer [UNIFESP]
relationships.isAdvisorOf
Silva, Carlos Alexandre Moreira da [UNIFESP]
item.page.type-of
Trabalho de conclusão de curso de graduação
Journal Title
Journal ISSN
Volume Title
Abstract
O leito de jorro é um equipamento utilizado para operações unitárias de secagem, recobrimento e granulação de partículas, nas indústrias química, farmacêutica e de alimentos. Este equipamento possui a vantagem de promover um movimento dinâmico cíclico do sistema particulado, que permite altas taxas de transferência de calor e massa, sendo o regime de jorro estável, uma condição operacional desejável. Nesse contexto, é de extrema importância o monitoramento dos regimes fluidodinâmicos que ocorrem durante as operações em um leito de jorro, a fim de detectar instabilidades fluidodinâmicas. As medidas de sinais de flutuação de pressão são ferramentas confiáveis para detecção de instabilidades fluidodinâmicas em operações de leito úmido. Portanto, o objetivo deste trabalho foi empregar a transformada wavelet em sinais de flutuações de pressão, para a identificação das instabilidades fluidodinâmicas em ensaios de umedecimento de partículas esféricas de vidro de 2,6 e 3,1 mm, usando massas de leito de 1,0 e 1,5 kg e vazões de atomização de 13, 16 e 19 mL/min, além da aplicação de controle estatístico para delimitar uma faixa de controle para as variáveis de análise na região de jorro estável. Os sinais de pressão utilizados foram obtidos por Butzge (2012) e foram processados de forma off-line por transformada wavelet, uma técnica alternativa e promissora, que tem recebido destaque nas últimas décadas em estudos de identificação de regimes em leitos fluidizados. Em síntese, verificou-se a robustez e reprodutibilidade das variáveis de análise da transformada wavelet na identificação de instabilidades fluidodinâmicas nos ensaios processados. Com base na evolução das variáveis analisadas nos 12 ensaios, concluiu-se que o controle do sistema tem de ser realizado nos princípios de instabilidades em dois grupos de análise, de acordo com a carga do leito, quando o conteúdo energético da mesoescala atinge o Limite Inferior de Controle (LIC) médio (59,336% para 1,0 kg e 47,971% para 1,5 kg) ou quando a entropia atinge o Limite Superior de Controle (LSC) médio (1,76 bits para 1,0 kg e 2,026 bits para 1,5 kg)
The spouted bed is used for drying, coating, and particle granulation operations in chemical, pharmaceutical, and food industries. This equipment has the advantage of promoting a cyclic dynamic movement of the particulate system, which allows high rates of heat and mass transfer, with stable spouting being a desirable operating condition. It is essential to monitor the fluid dynamic regimes that occur during operations in a spouted bed to detect fluid dynamic instabilities in this context. The pressure fluctuation signal measurements are reliable tools for detecting fluid dynamics instabilities in wat spouted bed operations. Therefore, the objective of this work is to employ the wavelet transform on signals of pressure fluctuations to identify fluid dynamic instabilities in wetting tests. Glass beads particles of 2.6 and 3.1 mm were used. The wetting experiments were conducted with 1.0 e 1.5 kg of solids and three water atomization flow rates (13, 16, and 19 mL/min). The pressure drop signals used were obtained by Butzge (2012) and were processed off-line using wavelet transform, an alternative and promising technique, which has been highlighted in recent decades in studies to identify regimes in fluidized beds. In summary, it was verified the robustness and reproducibility of the wavelet transform variables in the identification of fluid dynamic instabilities in the processed essays. Based on the evolution of the variables analyzed in the 12 essays, the control of the system must be carried out on the instabilities in two groups of analysis, based on the solids mass, in principles of when the energy content of the mesoscale reaches the average lower control limit (59,336% for 1,0 kg and 47,971% for 1,5 kg) or the entropy reaches the average upper control limit (1,76 bits for 1,0 kg and 2,026 bits for 1,5 kg).
The spouted bed is used for drying, coating, and particle granulation operations in chemical, pharmaceutical, and food industries. This equipment has the advantage of promoting a cyclic dynamic movement of the particulate system, which allows high rates of heat and mass transfer, with stable spouting being a desirable operating condition. It is essential to monitor the fluid dynamic regimes that occur during operations in a spouted bed to detect fluid dynamic instabilities in this context. The pressure fluctuation signal measurements are reliable tools for detecting fluid dynamics instabilities in wat spouted bed operations. Therefore, the objective of this work is to employ the wavelet transform on signals of pressure fluctuations to identify fluid dynamic instabilities in wetting tests. Glass beads particles of 2.6 and 3.1 mm were used. The wetting experiments were conducted with 1.0 e 1.5 kg of solids and three water atomization flow rates (13, 16, and 19 mL/min). The pressure drop signals used were obtained by Butzge (2012) and were processed off-line using wavelet transform, an alternative and promising technique, which has been highlighted in recent decades in studies to identify regimes in fluidized beds. In summary, it was verified the robustness and reproducibility of the wavelet transform variables in the identification of fluid dynamic instabilities in the processed essays. Based on the evolution of the variables analyzed in the 12 essays, the control of the system must be carried out on the instabilities in two groups of analysis, based on the solids mass, in principles of when the energy content of the mesoscale reaches the average lower control limit (59,336% for 1,0 kg and 47,971% for 1,5 kg) or the entropy reaches the average upper control limit (1,76 bits for 1,0 kg and 2,026 bits for 1,5 kg).