Generalized Additive Models to Optimize the Hydrophobicity Process of Kaolinite

The wide industrial use of kaolinite requires that the extraction processes be modeled to determine the appropriate conditions of the benefit. Although classic linear regression models have been used, these have not been appropriate due to the non-compliance with normal distribution for the response variable. The data analyzed in this study correspond to a kaolinite extraction process by surface physicochemistry carried out in La Unión, Antioquia, Colombia. The response variable was the zeta potential and the explanatory variables were type of collecting solution, concentration, and pH. In this article, the recovery of kaolinite is modeled through generalized additive models, which can choose the statistical distribution and model all the parameters based on explanatory variables. Five distributions were selected for the response variable according to the Akaike information criterion ($AIC$). The model with generalized distribution Beta 2 was the model that presented the best performance according to the metrics used and it was found that the best-operating conditions obtained are the type of oleic acid collector, the concentration of 10 units, and pH 6

• Fecha de publicación: 2023
• Institución: Universidad Pedagógica y Tecnológica de Colombia
• Repositorio: RiUPTC: Repositorio Institucional UPTC
• Idioma: spa
• OAI Identifier: oai:repositorio.uptc.edu.co:001/15360
• Acceso en línea: https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/14424; https://repositorio.uptc.edu.co/handle/001/15360
• Palabra clave: Caolinita; hidrofobicidad; modelos aditivos; modelos de regresión; potencial zeta; additive models, hydrophobicity, kaolinite, regression models, zeta potential
• License: http://purl.org/coar/access_right/c_abf2