Estimation of drill-string torsional vibration severity using field data and machine learning

• Ano de defesa: 2019
• Autor(a) principal: Vaio, Matheus Vera Di
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia Mecânica; UFRJ
• Palavras-chave em Português: Perfuração; Dados de campo; Vibração torcional; Aprendizado de máquina; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA
• Link de acesso: http://hdl.handle.net/11422/13774

This work aims to develop a method for real-time estimation of the drill string torsional vibration. This estimation during the drilling operation gives important information to the operator so that he can control the drilling parameters assertively. For that, it is made a presentation of the wells, and the possessed data. An adaptation of the PCA is proposed to make the preprocessing of the data that feeds a proposed deep neural network. Finally, the method is tested through four distinct cases, each one with its singular characteristics, with or not domain extrapolation. The proposed preprocessing tool and the use of the raw data have its results compared and evaluated. The conclusion provides a resume and some discussions of the results, its limitations, and its characteristics.