Otimização de sistema e configuração de pavimentos de concreto pré-moldado via algoritmos evolucionários

• Ano de defesa: 2015
• Autor(a) principal: Sena, Rafael Wandson Rocha
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Engenharia de estruturas; Otimização; Concreto pré-moldado; Concreto protendido; Algoritmos genéticos; Pavimentos; Structural optimization; Conceptual design; Precast concrete floor; Prestressed concrete; Genetic algorithm
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/30826

Big is the search for cost reductions in civil construction. Production in assembly lines and large scale, modulation and waste reduction are features that highlight the use of the precast concrete (PC). Traditionally, the project is realized by trial and error process, where the project solution is determined directly by the experience of the designer engineer. The optimization is the searching for a best solution through a mathematic model, where a lot of combinations of possibilities of answers will be evaluated, and then there is a selection of the best combination or optimum response. The objective of this work is the formulation of a mathematical model for optimization of prestressed precast concrete floor, which optimizes the structural system (slab type), the layout of the components (placement of the components) and the components themselves (dimensions and details), based on normative constraints and architectural constraints. For the slabs are allowed two structural systems, alveolar and double tee, and for the beams is used the inverted tee type. The objective function is the total cost arising from manufacturing, transportation and assembly phases. Because the design variables are discrete, genetic algorithms are used due to its efficiency and simplicity in treating this class of problems. The parameters of the GA are calibrated with test examples and applications are made to literature examples. The Particle Swarm Optimization (Particle Swarm Optimization) is used as a means of comparison and validation of results. The sensibility of the solution is studied for the variation in the cost parameters of the objective function. Solutions for other market profiles can be easily implemented