Análise da influência de modelos dinâmicos de caminhar humano na resposta de aceleração em pisos mistos de aço e concreto

• Ano de defesa: 2021
• Autor(a) principal: Cunha, Rafael Nunes da
• Orientador(a): Argôlo, Higor Sérgio Dantas de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Programa de Pós-Graduação: Pós-Graduação em Engenharia Civil
• Palavras-chave em Português: Engenharia civil; Concreto; Aço; Piso misto; Modelos matemáticos
• Palavras-chave em Inglês: Mathematical models; Concrete structures; Excessive vibrations
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA CIVIL
• Link de acesso: https://ri.ufs.br/jspui/handle/riufs/14316

Modern architecture has required the use of structures without partitions, with the possibility of using lighter sealing elements that are easy to adapt to new layouts. In addition, the speed of construction is an important factor in choosing the structural conception, especially for spaces with commercial purposes. In this scenario, composite steel and concrete structures stand out. However, these aspects reduce the global stiffness of the structure, leading to a decrease in its fundamental frequency, which approach the most common frequencies of walking activities, resulting in problems related to excessive vibration. Therefore, in order to avoid excessive vibration problems, it is essential to verify the structure considering the most commons dynamic loads, either through simplified methods, or through refined numerical methods, such as the Finite Element Method. In the light of the foregoing, the present paper analysed different dynamic loading models that represent human walking activities, applying them in composite steel and concrete floors that were previously experimentally tested, obtained in the technical literature, which were modelled numerically in the software Ansys. For the execution of the work, 10 dynamic loading models and 4 floors were selected, 2 classified as low fundamental frequency and 2 as high. In total, 130 simulations were performed, obtaining 220 acceleration histories over time. Based on the historical data, it was determined which dynamic model best represent human walking for different situations. Issues such as measurement position (on the walk path or in nearby regions), integration pace in transient analyses, post-processing of acceleration results, removal of peaks considered as outliers, among others, were evaluated. All comparisons were made in terms of peak acceleration and rms acceleration, where the quality of each model was verified for these criteria. It was concluded that the biodynamic models have better quality than the other models in peak acceleration analysis. However, for analyses related to rms acceleration, the heel impact model and the single step model stood out. In addition, criteria were defined for the removal of outliers, based on the results obtained in this paper. Regarding the simplified methods, it was observed that the Brazilian standard method was not accurate in many simulations, with better results only for rigid structures. As for the American and European standards method, these provided better results on low fundamental frequency floors, as for the aforementioned frequency, however, for analyses related to acceleration, these methods led to values below those of reference.