Dimensionamento de descontinuidades em peças metálicas por aproximação de Born da amplitude de espalhamento

• Ano de defesa: 2018
• Autor(a) principal: Omi, Guilherme Massaharu
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Elétrica e Informática Industrial; UTFPR
• Palavras-chave em Português: Ultrassom; Testes não-destrutivos; Espalhamento (Física); Metais - Fratura; Testes de ultrassom; Métodos de simulação; Ultrasonics; Non-destructive testing; Scattering (Physics); Metals - Fracture; Ultrasonic testing; Simulation methods; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::ELETRONICA INDUSTRIAL, SISTEMAS E CONTROLES ELETRONICOS::AUTOMACAO ELETRONICA DE PROCESSOS ELETRICOS E INDUSTRIAIS; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::MEDIDAS ELETRICAS, MAGNETICAS E ELETRONICAS INSTRUMENTACAO::INSTRUMENTACAO ELETRONICA; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/3781

This work presents a sizing method for inner metal bodies flaws, through ultrasonic A-scan signals generated by a phased array probe. The sizing technique makes use of the scattering matrix, from which the flaw pulse-echo response can be extracted. A mathematical model based on Born approximation is developed to analyze and extract valuable information about the scattering matrix related to the flaw. From the scattering matrix, the flaws pulse-echo response is obtained, which can be used to equate expressions for the flaw size and orientation. These expressions are not useful for all situations. Thus a new approach is developed for flaw sizing in the present work. This new approach uses the scattering matrices from experimental data as well as from the mathematical model by Born approximation. These scattering matrices are applied on a cost function to assess the minimum cost for a range of flaw length and frequencies. In this work, the sizing method is tested and validated on simulated data, providing good results for small flaws. However, the algorithm's performance decreases as the flaw size increases.