Estudo experimental e análise sobre os mecanismos de transferência de calor no resfriamento de chapas metálicas por jato impingente de água

• Ano de defesa: 2024
• Autor(a) principal: Helio Enio de Oliveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA MECÂNICA; Programa de Pós-Graduação em Engenharia Mecanica; UFMG
• Palavras-chave em Português: Laminação a quente; Resfriamento; Jato impingente; Coeficiente de transferência de calor; Problema inverso da condução de calor; Método dos elementos finitos; Engenharia mecânica; Calor - Transmissão; Calor - Condução; Laminação (Metalurgia); Energia; Sustentabilidade
• Link de acesso: http://hdl.handle.net/1843/81771

The cooling technique using circular impinging laminar jets is one of the most widely used in steel mills, especially in continuous hot rolling lines for high-quality steel plates. This method produces cooling rates of around 85°C/s, with heat fluxes of around 10 MW/m². To evaluate the phenomena involved, an experimental apparatus was adopted containing a heating furnace, a sample positioning support, a water tank and a U-tube type distributor tube. Stainless steel sheets of series 304 with controlled surface roughness were instrumented with type K thermocouples and heated in a controlled furnace, and were subsequently cooled by a subcooled circular jet of impinging water on their heated surface, varying the initial cooling temperature from 150°C to 1100°C, with water flow measurement. Temperature transients were collected with the aid of a datalogger following heating and cooling cycles. A mathematical model based on the finite element method was used, based on a Python 3.11 code. By solving the inverse heat conduction problem, the surface temperatures, heat fluxes and heat transfer coefficients were calculated for the 750ºC, 900ºC and 1000ºC tests. The regularization parameter by the “Future Steps” method and optimization based on the BOBYQA algorithm were compared to the results obtained experimentally. The impacts of the variation of the initial temperatures of the samples, behavior of the heat flux curves and heat transfer coefficient according to the time variation and in relation to the numerically calculated surface temperature were evaluated. The 220-node mesh used demonstrated an excellent compromise between computational effort and results. The temperatures presented variations of less than 0.22ºC. The study concluded that the heat transfer coefficient is highly influenced by the initial temperature of the samples and physical characteristics of the jet. The maximum value for the heat transfer coefficient and heat flux was reached within the nucleate boiling and transition regions, respectively. This behavior was observed in all three samples. The maximum heat flux reached a value of 1.77 MW/m² for a temperature of 1000ºC. It decreased with decreasing temperature. Cooling rates of 43 ºC/s were observed for a temperature of 1000ºC in the stagnation region.