Modelos matemáticos para a determinação da cinética de degradação in vitro de coprodutos agroindustriais

• Ano de defesa: 2017
• Autor(a) principal: Bezerra, Rayane Pinho
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Mato Grosso; Brasil; Instituto de Ciências Agrárias e Ambientais (ICAA) – Sinop; UFMT CUS - Sinop; Programa de Pós-Graduação em Zootecnia
• Palavras-chave em Português: Arroz; DDGS; Gases; Maracujá; Não lineares; CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA; Non-linear; Passion fruit; Rice
• Link de acesso: http://ri.ufmt.br/handle/1/4287

The objective of this study was to evaluate and identify among the non-linear mathematical models (Exponential, Exponential Bicompartimental, Gompertz, Logistic and Logistic Bicomparti-mental,) the one that presents better quality of adjustment to cumulative gas production curves in vitro the ethanol co-products DDGS of sorgum (DDGSS) and maize (DDGSM); Rice co-products: rice bran (QA), whole rice bran (FI), de-sengordurado rice bran (FD) and rice husk (CA); (MPG), passion fruit sifting (MPF), passion fruit seed (SM) and passion fruit peel (CM). In order to determine the total gas production, the automatic in vitro gas production technique was used, the ANKOM® RF - Gas production system (GAPS). The adjustment of the non-linear regression models was performed using the iterative GaussNewton method, inserted in the procedure PROC NLIN of the SAS. The criteria adopted to assess the quality of the adjustment were: joint nullity (P> 0.05), Akaike's ponderation criterion (Wi), RQMR and number of iterations to reach convergence (N.I.). Considering the evaluated parameters, the model that best fit the in vitro gas production profile of DDGSM, FI and FD was the exponential; Of DDGSS Gompertz; For CM and QA was logistic. For MPF, MPG and SM the best fit of the bicompartmental exponential model and for the bicompartmental logistic CA. Of the five models evaluated for the in vitro gas production adjustment, all of them fit the degradation profiles of some of the foods tested. Comparison between models for data adjustment is necessary since there is not a single model that fits the cumulative gas production profiles of all evaluated foods.