Submitted by DANIEL FERNANDO DA SILVA null (danielprofessorengemec@gmail.com) on 2018-01-04T13:41:19Z No. of bitstreams: 1 Dissertação - Daniel Fernando - versão final.pdf: 4270080 bytes, checksum: a1076d940ede22de07f7319dd405f059 (MD5) / Approved for entry into archive by Cristina Alexandra de Godoy null (cristina@adm.feis.unesp.br) on 2018-01-04T17:07:06Z (GMT) No. of bitstreams: 1 silva_df_me_ilha.pdf: 4270080 bytes, checksum: a1076d940ede22de07f7319dd405f059 (MD5) / Made available in DSpace on 2018-01-04T17:07:06Z (GMT). No. of bitstreams: 1 silva_df_me_ilha.pdf: 4270080 bytes, checksum: a1076d940ede22de07f7319dd405f059 (MD5) Previous issue date: 2017-07-07 / O presente trabalho estudou a influência do aporte térmico do processo de soldagem GTAW automatizado nas propriedades mecânicas do aço inoxidável austenítico AISI 316L. Esse aço é amplamente utilizado na indústria alimentícia e de petróleo e gás, por ser muito resistente a corrosão. Utilizou-se chapas de 1,0 mm de espessura, soldadas por junta de topo sem chanfro. Os parâmetros de soldagem foram selecionados e ajustados através de pré-testes, pois existem poucas fontes de pesquisa sobre a soldagem de chapas finas de aço inoxidável. Para analisar a influência da variação do aporte térmico, variou-se a velocidade de soldagem na faixa de 375 a 500 mm/min, com incremento de 25 mm/min. O processo de soldagem foi autógeno. Foram realizadas análises macrográficas para avaliar a largura e profundidade do cordão de solda. Foram realizados também ensaios mecânicos de microdureza Vickers e ensaios de tração para avaliar as propriedades mecânicas do cordão de solda formado. Os valores encontrados nestes ensaios foram comparados com o aporte térmico de soldagem, para avaliar a influência do mesmo nas propriedades mecânicas da junta de solda. Os resultados obtidos com os ensaios descritos, mostraram que o controle da energia de soldagem, ou seja, do aporte térmico é essencial na soldagem de chapas finas de aço inoxidável austenítico AISI 316L, gerando juntas de solda livres de descontinuidades e com propriedades mecânicas próximas ao do metal puro. / The present research studied the influence of the thermal input of the automated GTAW welding process on the mechanical properties of AISI 316L austenitic stainless steel. This steel is widely used in the food and oil and gas industry because it is very resistant to corrosion. Panels of 1.0 mm thickness were used, welded by a non-chamfered top joint. The welding parameters were selected and adjusted through pre-tests, as there are few sources of research on the welding of stainless steel thin sheets. In order to analyze the influence of the variation of the thermal input, the welding speed was varied in the range of 375 to 500 mm / min, with an increase of 25 mm / min. The welding process was autogenous. Macrographic analyzes were performed to evaluate the width and depth of the weld bead. Mechanical tests of Vickers microhardness and tensile tests were also performed to evaluate the mechanical properties of the weld bead formed. The values found in these tests were compared with the thermal input of welding, to evaluate the influence of the same on the mechanical properties of the weld joint. The results obtained with the described tests showed that the control of the welding energy, that is, of the thermal input is essential in the welding of thin sheets of AISI 316L austenitic stainless steel, generating solder joints free of discontinuities and with mechanical properties close to the of the pure metal.

Orientador: Vicente Afonso Ventrella / Resumo: O presente trabalho estudou a influência do aporte térmico do processo de soldagem GTAW automatizado nas propriedades mecânicas do aço inoxidável austenítico AISI 316L. Esse aço é amplamente utilizado na indústria alimentícia e de petróleo e gás, por ser muito resistente a corrosão. Utilizou-se chapas de 1,0 mm de espessura, soldadas por junta de topo sem chanfro. Os parâmetros de soldagem foram selecionados e ajustados através de pré-testes, pois existem poucas fontes de pesquisa sobre a soldagem de chapas finas de aço inoxidável. Para analisar a influência da variação do aporte térmico, variou-se a velocidade de soldagem na faixa de 375 a 500 mm/min, com incremento de 25 mm/min. O processo de soldagem foi autógeno. Foram realizadas análises macrográficas para avaliar a largura e profundidade do cordão de solda. Foram realizados também ensaios mecânicos de microdureza Vickers e ensaios de tração para avaliar as propriedades mecânicas do cordão de solda formado. Os valores encontrados nestes ensaios foram comparados com o aporte térmico de soldagem, para avaliar a influência do mesmo nas propriedades mecânicas da junta de solda. Os resultados obtidos com os ensaios descritos, mostraram que o controle da energia de soldagem, ou seja, do aporte térmico é essencial na soldagem de chapas finas de aço inoxidável austenítico AISI 316L, gerando juntas de solda livres de descontinuidades e com propriedades mecânicas próximas ao do metal puro. / Abstract: The present research studied the influence of the thermal input of the automated GTAW welding process on the mechanical properties of AISI 316L austenitic stainless steel. This steel is widely used in the food and oil and gas industry because it is very resistant to corrosion. Panels of 1.0 mm thickness were used, welded by a non-chamfered top joint. The welding parameters were selected and adjusted through pre-tests, as there are few sources of research on the welding of stainless steel thin sheets. In order to analyze the influence of the variation of the thermal input, the welding speed was varied in the range of 375 to 500 mm / min, with an increase of 25 mm / min. The welding process was autogenous. Macrographic analyzes were performed to evaluate the width and depth of the weld bead. Mechanical tests of Vickers microhardness and tensile tests were also performed to evaluate the mechanical properties of the weld bead formed. The values found in these tests were compared with the thermal input of welding, to evaluate the influence of the same on the mechanical properties of the weld joint. The results obtained with the described tests showed that the control of the welding energy, that is, of the thermal input is essential in the welding of thin sheets of AISI 316L austenitic stainless steel, generating solder joints free of discontinuities and with mechanical properties close to the of the pure metal. / Mestre

Weld joint penetration determination is the key factor in welding process control area. Not only has it directly affected the weld joint mechanical properties, like fatigue for example. It also requires much of human intelligence, which either complex modeling or rich of welding experience. Therefore, weld penetration status identification has become the obstacle for intelligent welding system. In this dissertation, an innovative method has been proposed to detect the weld joint penetration status using machine-learning algorithms. A GTAW welding system is firstly built. Project a dot-structured laser pattern onto the weld pool surface during welding process, the reflected laser pattern is captured which contains all the information about the penetration status. An experienced welder is able to determine weld penetration status just based on the reflected laser pattern. However, it is difficult to characterize the images to extract key information that used to determine penetration status. To overcome the challenges in finding right features and accurately processing images to extract key features using conventional machine vision algorithms, we propose using convolutional neural network (CNN) to automatically extract key features and determine penetration status. Data-label pairs are needed to train a CNN. Therefore, an image acquiring system is designed to collect reflected laser pattern and the image of work-piece backside. Data augmentation is performed to enlarge the training data size, which resulting in 270,000 training data, 45,000 validation data and 45,000 test data. A six-layer convolutional neural network (CNN) has been designed and trained using a revised mini-batch gradient descent optimizer. Final test accuracy is 90.7% and using a voting mechanism based on three consequent images further improve the prediction accuracy.

No description available.