Assessment of the accuracy of an ultrasound imaging system for the measurement of the thickness of pavements\' concrete slabs.

• Ano de defesa: 2019
• Autor(a) principal: Seiffert, Henrique Luiz de Lima
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Concrete pavements; Ensaios não destrutivos; Nondestructive evaluation; Pavimentos rígidos; Ultrasound imaging; Ultrassonografia
• Link de acesso: http://www.teses.usp.br/teses/disponiveis/3/3138/tde-17102019-145532/

The thickness of 150mm and 250mm thick slabs of plain concrete pavements was measured with an ultrasound imaging system and then with coring. The thickness measures determined with coring were taken as the gold standard. The accuracy of the thickness measures determined with the ultrasound imaging system were assessed in terms of absolute errors, relative errors, and the R-Squared. The significance of the changes in the accuracy measures with changes in the thickness range of the slabs and in the central frequency of the transmitted pulses was investigated. The thickness measures determined with the ultrasound imaging system for every central frequency were, on average, very inaccurate measures of the thickness of the concrete slabs. At the same time, the dispersion in the absolute and the relative errors for every central frequency was of such significance that the rectification of those measures with the mean errors would be uncertain. Furthermore, the absolute and the relative errors were, on average, smaller with increasing central frequency when the thickness range was disregarded. At the same time, the absolute errors of the thickness measures determined on 150mm slabs were, on average, significantly smaller, than the ones determined on the 250mm slabs. The absolute erros of the thickness measures determined on 150mm slabs were also not dependent on the central frequency. However, the absolute erros of the thickness measures determined on 250mm slabs were dependent on the central frequency, namely, they were, on average, smaller with increasing central frequency. The rectification of the measures on 150mm and 250mm slabs with their respective mean absolute erros would be uncertain too, because of the considerable degree of dispersion in those erros. The results for the relative errors was consistent with the above. The thickness of the 150mm and the 250mm thick slabs was measured by the ultrasound imaging system with different accuracy then. The implications of the changes in the central frequency was also different for the 150mm and the 250mm thick slabs.