Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2018-04-02T12:49:22Z
No. of bitstreams: 1
GiorgioAndreBritoOliveira_DISSERT.pdf: 3458330 bytes, checksum: 434d48526b1ff4a3e12b0d1b440b6198 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-04-04T14:49:22Z (GMT) No. of bitstreams: 1
GiorgioAndreBritoOliveira_DISSERT.pdf: 3458330 bytes, checksum: 434d48526b1ff4a3e12b0d1b440b6198 (MD5) / Made available in DSpace on 2018-04-04T14:49:23Z (GMT). No. of bitstreams: 1
GiorgioAndreBritoOliveira_DISSERT.pdf: 3458330 bytes, checksum: 434d48526b1ff4a3e12b0d1b440b6198 (MD5)
Previous issue date: 2018-02-09 / Os materiais comp?sitos s?o um novo destaque no avan?o tecnol?gico, impondo novas pesquisas relacionadas ao assunto devido a sua crescente demanda nas mais diversas ?reas. Dentre essas pesquisas surgem as que tem como objetivo facilitar as aplica??es desses materiais, atrav?s de uma r?pida apura??o das suas propriedades mec?nicas sem a necessidade de procedimentos experimentais, sendo essa fator primacial na prepara??o de projetos. Assim surgiram os modelos micromec?nicos, que ganharam destaque devido a sua praticidade, como exemplo das equa??es da Regra das Misturas e das equa??es de Halpin-Tsai. Recentemente, novos modelos computacionais vem combinando modelos micromec?nicos e aperfei?oando-os para se ter a m?xima acur?cia, como por exemplo as redes neurais artificiais (RNAs). Com base nisso, este trabalho visa a cria??o de arquiteturas de RNAs capazes de modelar o m?dulo de cisalhamento (G12) e a tens?o ?ltima de tra??o longitudinal (Xt) de comp?sitos unidirecionais. Com as RNAs treinadas e testadas, essas v?o servir como ferramentas computacionais, semelhante a fun??es, em que fornecendo as entradas teremos uma sa?da desejada. Para isso, fez-se necess?rio uma coleta de dados da literatura, que foram divididos em um conjunto de treino e um conjunto de teste, para realiza??o da valida??o cruzada. Se desenvolveram sete tipos de arquiteturas diferentes, tr?s para o G12 e quatro para o Xt, na qual essas possuem entre duas, tr?s e quatro entradas. Dentre esses modelos tr?s deles s?o considerados modelos mistos, que combina valores da sa?da da RNA com os valores obtidos vindos de modelos micromec?nicos, como o modelo de Halpin-Tsai. Ap?s o treinamento das RNAs, foi realizada uma an?lise comparativa dos valores vindos da RNA e dos valores experimentais, e ainda an?lises quantitativas e qualitativas com base no modelo de compara??o (modelo de Halpin-Tsai e modelo da Regra das Misturas), apresentando maiores valores de coeficiente de correla??o e menores valores de erro quadr?tico m?dio. / The composite materials are a new highlight in the technological advancement, consequently leading to the development of new researches due to its growing demand in the most diverse areas. Among these researches, arise those that have the objective to facilitate the application of these materials, through a fast estimation of its mechanical properties, without the need for experimental procedures, with this being the main factor in the projects preparation. Thus the micromechanical models appeared, which gained importance due to its practicality, such as the Mix Rule and the Halpin-Tsai equations. Recently, new computational models are combining micromechanical models and perfecting them to obtain maximum accuracy, as for instance in the Artifical Neural Networks application. Therefore, this work aims to create an Artificial Neural Network (ANN) architecture capable of modeling the shear modulus and ultimate longitudinal stress of unidirectional composites. When the ANN?s are trained and tested, they will serve as computational tools, similar to functions, where an input is supplied to obtain a desired output. To achieve this goal, it was necessary a collection of data in literature, which were divided in a training group and a testing group, with the cross validation between them being performed. Seven different types of architectures were developed, three for the G12 and four for the Xt, each of these with two, three and four inputs. Among these models, three of them are considered mixed models, which combines values from the output of the ANN with values obtained from the micromechanical models, such as the Halpin-Tsai. After the ANN training, a comparative analysis was performed between the values from the ANN and the experimental values, with quantitative and qualitative analysis being performed with the Halpin-Tsai model as a base for comparison, presenting higher values for the correlation coefficient and smaller values for the root mean square error.
| Identifer | oai:union.ndltd.org:IBICT/oai:repositorio.ufrn.br:123456789/24978 |
| Date | 09 February 2018 |
| Creators | Oliveira, Giorgio Andr? Brito |
| Contributors | 02306444498, Belisio, Adriano Silva, 56574681472, Costa J?nior, Jo?o Carlos Arantes, 42364973368, Bessa, Wallace Moreira, 05362438751, Freire J?nior, Raimundo Carlos Silv?rio |
| Publisher | PROGRAMA DE P?S-GRADUA??O EM ENGENHARIA MEC?NICA, UFRN, Brasil |
| Source Sets | IBICT Brazilian ETDs |
| Language | Portuguese |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/masterThesis |
| Source | reponame:Repositório Institucional da UFRN, instname:Universidade Federal do Rio Grande do Norte, instacron:UFRN |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.009 seconds