Estudo estatístico de produção de concretos com adições minerais

• Autor(a) principal: Furquim, Paulo Ricardo de Vargas
• Data de Publicação: 2006
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• dARK ID: ark:/26339/001300000rtpq
• Texto Completo: http://repositorio.ufsm.br/handle/1/7909
• topic: Civil engineering; Civil construction; Materials of construction; Concrete; Engenharia civil; Construção civil; Materiais de construção; Concreto; CNPQ::ENGENHARIAS::ENGENHARIA CIVIL

The GEPECON, Studies and Researches Concrete Group, has developed a researches collection about the concrete with mineral additions over the last 10 years. Starting from these papers we have decided to gather some projects data to be statistically analyzed. The present paper had the objective of studying the influence of the independent variables (water/binder relation, compression strength, type and content of mineral addition) on the dependent variables of microstructure and durability of the concrete through the micro-statistic tools. The variance analysis and the sim pie and multiple lineal regression were applied 50 that it was possible to determine which was the most significant independent variable for these concrete properties. Data were analyzed in five projects and later they were gathered by variables in order to accomplish the statistical analysis. In the general model, through the multiple lineal regression, the independent variables which has been exerted more influence on the dependent variables were the addition content and the compression strength, while for durability they were the content and the addition type. However, in the single models with the simple lineal regression, the most highlighted independent variables on the content of combined water and remainder calcium hydroxide were the content and the addition type; in the poresymmetry they were the compression strength and the water/binder relation, while for the dependent variables of the durability the water/binder relation and the compression strength were, respectively, the most significant ones. For the addition type, the most influent mixtures were the binary mixtures with CCA and CV, the ternary ones CCA-CV on the microstructure and CV-E on the durability. The most significant addition content was 25% for the microstructure and 70% for the durability. As it was already expected, starting from these analyzes has been emphasized the results of the sim pie lineal regression that showed a larger dispersion and lower significance in relation to the ones of the multiple lineal regression.