Metodologia numérica para projeto e previsão de desempenho de motor Stirling tipo Beta

• Ano de defesa: 2019
• Autor(a) principal: Bryan Castro Caetano
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: CFD; Projeto mecânico; Tipo Beta; Previsão de desempenho; Motor stirling; Motores Stirling; Projetos mecânicos; Engenharia mecânica; Desempenho
• Link de acesso: http://hdl.handle.net/1843/RAOA-BCLHCN

The design and performance prediction of a Stirling engine are little explored in the scientific literature, only a few works address how to design an engine. In addition, there are no models that satisfactorily simulate the behaviour of real Stirling engines, since performance prediction models estimate large deviations, a barrier to be broken in new engine designs. The purpose of this work is to contribute to the development of Stirling engines, proposing a methodology that allows designing, characterizing and predicting the performance of a Beta-type engine. Concepts available in the literature are used for the determination of geometric design parameters and the mechanical components of the engine are presented, which can operate with helium gas or atmospheric air with a working pressure of 700 kPa. The engine performance prediction is performed with the aid of a numerical simulation methodology, aiming at reducing the deviation in predicted power and facilitating extrapolation without the need for new experimental data. For this, strategies are combined to allow the calculation of the engines internal pressure at the beginning of the cycle (lower dead centre of the power piston) and the displacer piston faces temperature, to assist the transient model using three-dimensional computational fluid dynamics. The lower dead centre pressure is calculated with Schmidts first-order model. The calculated value serves as a pressure correction in the transient CFD model, reducing computational cost and increasing its accuracy. The temperature at the displacer piston faces is obtained with a steady-state simulation in CFD (without piston movement). The obtained temperature profile is implemented in a user defined function in the transient simulation using ANSYS Fluent software. Other relevant factor in this work is the analysis of the influence of the Discrete Ordinate radiation model, which increases the temperature at the displacers frontal face in 132.2 K, generating an improvement in the indicated power prediction with a reduction in the deviation from -14, 4% to -2.6%, when compared with experimental values. With the validated methodology, the designed engine is characterized and the temperature influence in the expansion chamber, the type and the fluid pressure are evaluated. In this work, different fluids, working pressures and temperatures were tested in the expansion chamber (673 K to 1073 K) for the same rotation of 600 rpm. the highest power point indicated (240.5 W) and higher thermal efficiency (19.8 %) were achieved by operating helium at a pressure of 700 kPa at a temperature of 1073 K and 873 K. With the proposed methodology, it is possible to extrapolate the simulation to engines with other dimensions, avoiding the necessity to build a prototype to know its behavior.