A discussion on damage identification in parametrically excited rods.

Detalhes bibliográficos
Ano de defesa: 2021
Autor(a) principal: Costa, Vinicius Vaguetti da
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: https://www.teses.usp.br/teses/disponiveis/3/3144/tde-12082022-080832/
Resumo: Offshore risers are crucial structures employed in deepwater exploitation. They are responsible for connecting the floating unit to the sub-sea system, being used on drilling operations and for conveying oil and gas from the bottom of the sea to the surface. These pipes, however, are constantly subjected to several dynamic forces and, not rarely, fail. These forces have a complex nature and are difficult to model in order to predict risers failure. An alternative approach to the problem of preventing failure in this structures is the employment of Structural Healthy Monitoring (SHM) techniques. SHM makes use sensors distributed along the structure and monitors changes in the measurements to describe the structure condition. The present work studies the application of a particular SHM methodology in order to identify damages on vibrating rod models, being the dynamics of risers one of the motivations. The methodology employs Wavelet Transform and Second-Order Blind Identification (SOBI) aiming at recovering the most energetic mode shapes of undamaged and damaged rod models, herein studied using Finite Element Method (FEM). Subsequently, a damage index (DI) is applied to compare the mode shapes of the undamaged and damaged structures in order to estimate damage spot. The methodology is applied to a 500 m long vertical rod model subjected to parametric excitation. Vibration data is acquired in equidistant nodes simulating sensors distributed along the structure and a white noise signal is added to the vibrational signals. Rod models with damages in different spots and with different severities are generated and simulated. The rods are parametrically excited in the 10th, 15th and 20th modes separately and the results for these excitation conditions are compared. The results show that the technique is successful on estimating the damage spots on the rod models, depending on the damage position, damage severity, excited mode and distance between sensors.