http://repositorio.unb.br/handle/10482/34427| File | Description | Size | Format | |
|---|---|---|---|---|
| 2018_VanessadeMouraSantana.pdf | 2,49 MB | Adobe PDF | View/Open |
| Title: | Avaliação de danos estruturais em vigas considerando a influência da temperatura via redes neurais artificiais |
| Other Titles: | Evaluation of structural damage in beams considering the influence of temperature using artificial neural networks |
| Authors: | Santana, Vanessa de Moura |
| Orientador(es):: | Carvalho, Graciela Nora Doz de |
| Assunto:: | Dano - estruturas Vigas Dano - avaliação Redes neurais artificiais Análise térmica Estruturas metálicas |
| Issue Date: | 25-Apr-2019 |
| Data de defesa:: | 30-Aug-2018 |
| Citation: | SANTANA, Vanessa de Moura. Avaliação de danos estruturais em vigas considerando a influência da temperatura via redes neurais artificiais. 2018. xix, 97 f., il. Dissertação (Mestrado em Estruturas e Construção Civil)—Universidade de Brasília, Brasília, 2018. |
| Abstract: | Over the years the concern with the structural integrity of the constructions has been increasing. Among the many methods for evaluating structural integrity, those based on the observation of dynamic properties stand out, since frequencies and modes depend fundamentally on structural rigidity. The methods that evaluate this change in rigidity, demand a continuous monitoring of the structure, and the search for suitable methodologies for the treatment of the vast set of collected data becomes necessary. In this sense, Artificial Neural Networks (ANNs) constitute an interesting tool for the treatment of these data. However, environmental effects, such as temperature, humidity and wind, limit the successful application of these methods. Among the mentioned environmental effects the temperature stands out, being able to camouflage the influence caused by small damages. Based on this discussion, this work presents a study on the influence that the temperature variation generates on the dynamic properties and, consequently, on the damage evaluation method looking for an alternative evaluation that overcomes these problems. The data used for the construction of this study were obtained through ANSYS finite element software, with which numerical models of beam were created in intact configuration, at different temperatures, and with different damage scenarios. These data were used in different ANNs with the backpropagation learning algorithm to construct a tool capable of locating damage, since little is known about the use of RNAs for damage assessment taking into account the effects caused by changes in temperature. Finally, with the purpose of analyzing the performance of ANNs created in the face of real problems, the ANNs were tested with experimental data. |
| metadata.dc.description.unidade: | Faculdade de Tecnologia (FT) Departamento de Engenharia Civil e Ambiental (FT ENC) |
| Description: | Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Civil e Ambiental, 2018. |
| metadata.dc.description.ppg: | Programa de Pós-Graduação em Estruturas e Construção Civil |
| Licença:: | A concessão da licença deste item refere-se ao termo de autorização impresso assinado pelo autor com as seguintes condições: Na qualidade de titular dos direitos de autor da publicação, autorizo a Universidade de Brasília e o IBICT a disponibilizar por meio dos sites www.bce.unb.br, www.ibict.br, http://hercules.vtls.com/cgi-bin/ndltd/chameleon?lng=pt&skin=ndltd sem ressarcimento dos direitos autorais, de acordo com a Lei nº 9610/98, o texto integral da obra disponibilizada, conforme permissões assinaladas, para fins de leitura, impressão e/ou download, a título de divulgação da produção científica brasileira, a partir desta data. |
| Agência financiadora: | Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) e Fundação de Apoio à Pesquisa do Distrito Federal (FAPDF). |
| Appears in Collections: | Teses, dissertações e produtos pós-doutorado |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.