Desenvolvimento de técnicas experimentais para análise dos parâmetros termofísicos presentes na equação da biotransferência de calor

• Ano de defesa: 2022
• Autor(a) principal: Oliveira, José Ricardo Ferreira
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Mecânica
• Palavras-chave em Português: Equação da Biotransferência de Calor; Bioheat Transfer Equation; Propriedades Termofísicas; Thermophysical Properties; Tecidos in-vitro; In-vitro tissues; Inferência Bayesiana; Bayesian Inference; Problemas Inversos; Inverse Problems; Quantificação de Incertezas; Uncertainty Quantification; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA::FENOMENOS DE TRANSPORTE::TRANSFERENCIA DE CALOR; Engenharia Mecânica; Meios de transferência de calor; Mecânica dos fluidos
• Link de acesso: https://repositorio.ufu.br/handle/123456789/36516; http://doi.org/10.14393/ufu.te.2022.578

This work aims to develop experimental techniques to evaluate the thermophysical properties of the Bioheat Transfer Equation. The study was performed on in-vitro tissues, called phantoms, which mimic human tissue. The phantoms are made of silicon and nano-sized magnetic particles. The experiments to estimate thermal diffusivity and thermal conductivity consist of partial heating of the phantoms on only one active surface. Heat flux and temperature are measured at two different points on the surface. The method was applied to two different thermal models using the same set of experimental data. The first model used the ratio between two measured surface temperatures to estimate the thermal diffusivity. The inverse problem was then solved with Bayesian Inference. The second model applied Bayesian Inference to the theoretical and experimental values of the temperatures to determine the maximum likelihood of the quadratic temperature error function, thus obtaining the thermal conductivity. Metabolic generation was analysed from uniform heating of the sample by electromagnetic induction. The time-dependent of the temperature variation was measured at two points on the surface and, through Bayesian Inference, the heat generation of the phantom was estimated. The influence of perfusion on thermophysical properties was evaluated on a phantom that has internal channels for water circulation. Temperature ratio and Bayesian Inference are applied to estimate the thermal diffusivity and thermal conductivity. It is found that perfusion has little influence on the volumetric heat capacity of the sample. Uncertainty quantification indicates that the estimated properties have a dispersion smaller than 18.0 %, with 95.45 % coverage probability and a coverage factor equal to 2.00. Finally, a study carried out in COMSOL® explores the practical application of the technique through analyses in simulated tissues with perfusion and metabolism.