Desenvolvimento de uma metodologia para determinação e análise dos deslocamentos térmicos de máquinas-ferramenta usando o método dos elementos finitos e redes neurais artificiais

• Ano de defesa: 2013
• Autor(a) principal: Sousa, Romualdo Figueiredo de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraí­ba; BR; Engenharia Mecânica; Programa de Pós-Graduação em Engenharia Mecânica; UFPB
• Palavras-chave em Português: Máquinas-ferramentas; Deslocamentos térmicos; Redes neurais artificiais; Thermal deformation; Machine tool; Finite element method; Artificial Neural Network; ENGENHARIAS::ENGENHARIA MECANICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/5375

In the processes of manufacturing machine tools play an important role particularly in the manufacture of parts of complex geometries. Currently the requirements for high dimensional and geometric accuracy during the machining process require small dimensional tolerances. Much of the errors of a machine tool are those which are thermally induced from the heat sources internal and external factors acting on the machine, thus causing thermal deformations in its structure. In this paper, we present a methodology for determining and analyzing the thermal deformation of machine tools using finite element method (FEM) and artificial neural networks (ANN). After molding machine using FEM, and defining the location of the heat source were obtained the temperature gradient of the machine and the corresponding thermal deformation at predetermined periods. Results obtained with simulations using the software NX.7.5 and the measurement principle of the Ballbar system, showed that this methodology is an effective tool in determining the thermal deformation of the machine, correlating the temperature reading at strategic points with volumetric deformation at the tool tip. Allowing then the thermal analysis of the errors in the pair tool part. Additionally, these results were used to train an ANN. The parameters of "learning" network under conditions of transient contours, allowed the network training. After training and validation set, she will be able to make the prediction of thermal errors just stating the temperature values of specific points of each heat source. This methodology will contribute to the designs of machine tools with high accuracy and thermal stability, but also provide for compensation of thermally induced errors.