Simulação computacional de secagem de frutos de café em leitofixo

• Ano de defesa: 2011
• Autor(a) principal: Santos, Emílio de Souza
• Orientador(a): Correa, Paulo César
• Banca de defesa: Ceballos, Maria Del Carmen Jarén , Coimbra, Jane Sélia dos Reis
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Engenharia Agrícola
• Departamento: Construções rurais e ambiência; Energia na agricultura; Mecanização agrícola; Processamento de produ
• País: BR
• Palavras-chave em Português: Café; Secagem; Armazenamento de produtos agrícolas
• Palavras-chave em Inglês: Coffee; Drying; Storage of agricultural products
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://locus.ufv.br/handle/123456789/3609

The aims of this work were to: model fixed-bed drying of coffee berries considering the product and air flow properties variation; validate this new model by comparing the simulation results with the experimental data; analyze the sensitivity of each model parameter; and compare the performance of the proposed model with a traditional model (Michigan Model). A thin-layer drying model was developed, based on mass and heat balance between the drying air and the product. Moreover, it was considered the product and air flow properties variation during the process. Thus a thick layer drying model was created by connecting four thin-layer models. The proposed model was implemented in Modelica language and simulated using OpenModelica® 1.6.0. The proposed model was validated by the analysis of model performance index and relative standard deviation between the experimental data and simulated results. The proposed model was compared with the Michigan model by the analysis of the statistical performance and relative bias of each model. Finally, in order to simplify the model structure and understands its dynamics, including the consideration that it is different of the Michigan model, it was done the differential and subjective model parameter and variable sensitivity analysis. The proposed model presented values of relative standard deviation lower than 10 % and values of performance index higher than 85 % for all simulated tests, indicating satisfactory representation of the studied phenomena. However, the Michigan model presented values of relative standard deviation higher than 10 % and/or values of performance index lower than 85 % for all simulated tests, indicating that the model does not satisfactorily represent of the studied phenomena. It was verified by the bias analysis that the proposed model slightly overestimates the moisture content values, unlike the Michigan model that significantly underestimates the values. After the model parameter and variable sensitivity analysis, it was concluded that: a) the proposed model is not sensitive to the product physical properties variation, excluding the water latent heat inside the product; b) probably the proposed model tends to overestimates the moisture content value due to the variation of the air properties and of the water latent heat inside the product, as well as the consideration of the constant temperature period during the process begin; c) probably the proposed model tends to underestimates the moisture content value due to the consideration of the product layer shrinkage and the air flow and pressure variation.