Lógica fuzzy para a automação da classificação de amostras de leite

Detalhes bibliográficos
Ano de defesa: 2020
Autor(a) principal: Jousiane Alves Martins
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de Minas Gerais
Brasil
Programa de Pós-Graduação em Produção Animal
UFMG
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://hdl.handle.net/1843/36413
Resumo: The demand for consumption of milk products has grown among the Brazilian population, which makes the monitoring of the quality of milk dairy products made by important both for industry and for the consumer and the producer. This study aimed to verify the efficiency of automation of the classification as regards the quality of the milk by means of Fuzzy logic. In the stage of fuzzyfication were considered as linguistic variables of the physicochemical characteristics of milk. We evaluated the levels of fat, protein, lactose, solids not fat (ESD), total solids (EST), titratable acidity, relative density at 15°C, crioscópico Index and alizarol Test, and created a linguistic variable for the output. The fuzzy systems have been developed using the R software, being used the methodology of Mandani Min in step of fuzzyfication and the centroid method in defuzzification, for which they were adopted the classifications (adulterated, inadequate and adequate) for each of the milk sample. In the simulation process used for the milk samples inappropriate verifies that the analyzes of fat, protein, lactose, solids not fat, total solids, density and acidity were within the pattern established by current legislation. While all samples (n = 25/100%) presented alteration on Alizarol test. All 25 samples for the analysis of adulterated milk were classified as adulterated by the fuzzy logic. This demonstrates that this methodology allowed the automation of this classification with efficiency. As the system modeled the appropriate milk samples from all showed appropriate values of protein, lactose, fat, solids not fat, total solids and density. All these samples were classified as adequate by the fuzzy logic. fuzzy logic is a powerful tool for the classification of the milk, and can be used advantageously by professionals of the area in order to reduce labor, human and financial resources. Furthermore, it should be emphasized that the relationship between the relevance and quality of the milk allows you to confirm the efficiency of the fuzzy system, which enables and enjoys the dairy sector.