Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2017-03-16T12:44:59Z No. of bitstreams: 1 Daniel Augusto Kestering_.pdf: 4709572 bytes, checksum: bd1166e3946f589fd86f700a714928c2 (MD5) / Made available in DSpace on 2017-03-16T12:44:59Z (GMT). No. of bitstreams: 1 Daniel Augusto Kestering_.pdf: 4709572 bytes, checksum: bd1166e3946f589fd86f700a714928c2 (MD5) Previous issue date: 2016-10-31 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / PROSUP - Programa de Suporte à Pós-Gradução de Instituições de Ensino Particulares / A investigação dos modelos de arrasto gás-sólido é fundamental para se obter bons resultados de fluidização utilizando dinâmica dos fluidos computacional. A tecnologia de fluidização é muito utilizada para conversão térmica de combustíveis sólidos e tem como principal vantagem a boa mistura entre gás e sólido. O presente trabalho utiliza dois softwares para simulação de leitos fluidizados, MFIX e Ansys Fluent, para comparar os modelos de arrasto de Syamlal e O`Brien (1987) e Di Felice (1994). A abordagem utilizada para modelagem do problema é o modelo de dois fluidos (Two Fluid Model, TFM), juntamente com a teoria cinética para escoamento laminar (Kinetic Theory for Granular Flow, KTGF). Um método para ajuste do modelo de DF (DI FELICE, 1994), baseado no trabalho de Esmaili e Mahinpey (2011), é sugerido, assim como o modelo de SO (SYAMLAL; O`BRIEN,1987) é ajustado utilizando dados em condição de mínima fluidização. Foram conduzidos experimentos para obtenção de velocidade e fração de vazios em condição de mínima fluidização a fim de ajustar ambos os modelos. As partículas utilizadas nos experimentos foram esferas de vidro de 1,21 mm, 0,8 mm e areia de fundição de 0,29 mm. O método proposto representa de forma adequada os dados obtidos em mínima fluidização das três partículas. Com os modelos de arrasto ajustados, simulações numéricas em regime de fluidização foram conduzidas em domínio bidimensional e tridimensional. Os resultados obtidos nestas simulações apresentam boa concordância com resultados experimentais em queda de pressão do leito e borbulhamento. Concomitantemente, um código para obtenção de modelo de arrasto utilizando o algoritmo EMMS/Bubbling foi desenvolvido e simulações numéricas bidimensionais foram conduzidas, para teste e validação. Os resultados do código mostram que o modelo segue a mesma tendência de Shi, Wang e Li (2011), que desenvolveram o modelo EMMS/Bubbling. / The investigation of gas-solid drag models is a key to obtain good results of fluidization by using computational fluid dynamic tools. The fluidization technology is used for solid fuel thermal conversion and its main advantage is the high gas-solid mixture. The present effort uses two software for fluidized beds simulation, MFIX and Ansys Fluent, in order to compare the drag models of Syamlal and O`Brien (1987) and Di Felice (1994). Two Fluid Model is the approach used to model together with Kinetic Theory for Granular flow. A method to adjust DF drag model (DI FELICE, 1994), based on Esmaili and Mahinpey (2011), is suggested, as well as SO drag model (SYAMLAL; O’BRIEN, 1987) is adjusted using data obtained from minimum fluidization condition. Experiments were realized to obtain velocity and void fraction at minimum fluidization condition in order to adjust both models. Glass beads with diameter of 1,21 mm and 0,8 mm and sand with diameter of 0,29 mm were used on experiments. The purposed method fits the data obtained on minimum fluidization condition of the three particles, in accordance with experimental data. With the models adjusted, numerical simulation were conducted using drag models for two- and three-dimensional domain. The results of this simulations agrees with experimental data of pressure drop and bubble formation. Simultaneously, a code to obtain a drag model using EMM/Bubbling algorithm was developed and numerical simulation were conducted. Results of EMMS show that the model have the same tendency of results of Shi, Wand and Li (2011), who developed EMMS/Bubbling model.

This research focuses on developing a comprehensive framework for designing and modeling experiments in the presence of multiple sources of competing prior knowledge. In particular, methodology is proposed for process optimization in high-cost, low-resource experimental settings where the underlying response function can be highly non-linear. In the first part of this research, an initial experimental design criteria is proposed for optimization problems by combining multiple, potentially competing, sources of prior information--engineering models, expert opinion, and data from past experimentation on similar, non-identical systems. New methodology is provided for incorporating and combining conjectured models and data into both the initial modeling and design stages. The second part of this research focuses on the development of a batch sequential design procedure for optimizing high-cost, low-resource experiments with complicated response surfaces. The success in the proposed approach lies in melding a flexible, sequential design algorithm with a powerful local modeling approach. Batch experiments are designed sequentially to adapt to balance space-filling properties and the search for the optimal operating condition. Local model calibration and averaging techniques are introduced to easily allow incorporation of statistical models and engineering knowledge, even if such knowledge pertains to only subregions of the complete design space. The overall process iterates between adapting designs, adapting models, and updating engineering knowledge over time. Applications to nanomanufacturing are provided throughout.