Sistema inteligente para controle da climatização de aviários para produção de frangos de corte

• Ano de defesa: 2019
• Autor(a) principal: Silva, Cosme Teixeira da
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Lavras; Programa de Pós-Graduação em Engenharia de Sistemas e Automação; UFLA; brasil; Departamento de Engenharia
• Palavras-chave em Português: Avicultura de corte; Frangos de corte - Conforto térmico; Internet das coisas; Agroindústria 4.0; Poultry Industry; Broilers - Thermal comfort; Internet of things; Agribusiness 4.0; Sistemas de Computação
• Link de acesso: http://repositorio.ufla.br/jspui/handle/1/42146

In this study, the control and monitoring process for a broiler house was improved to mitigate or eliminate thermal stresses in the birds’ environment. An architecture based on the Internet of Things (IoT) and agribusiness 4.0 was used to implement a fuzzy controller embedded in an ESP8266 microcontroller. A multiplatform web application (consisting of open-source applications and free software, such as Node-RED, a tool based on data streams) communicated the conditions of the aviary climate system to the fuzzy controller. Data were collected by a DS18B20 temperature sensor and a DHT22 temperature and humidity sensor. A Raspberry Pi 3 served as a low-cost central computer system to store these data and the rating provided by the microcontroller-embedded fuzzy controller. This information was transmitted to the central web server via message queue telemetry transport (MQTT). Data persistence was achieved using a MySQL database installed on the Raspberry Pi 3. The system performed decision-making to control the aviary thermal environment of broilers ranging from one to 49 days old. The input variables of the system were the black globe-humidity index (BGHI) and the bird age. Defuzzification by the center of gravity method produced environmental ratings as output variables that were used to control the aviary thermal environment in an automatic and smart manner. The inference was performed using the Mamdani method, and the accuracy of the results was determined by comparison with an environmental classification performed in MATLAB. Using the low-cost intelligent prototype produced a 100% accuracy (based on the MATLAB results) for the thermal environment score for 200 scenarios simulated with different thermal environments. Thus, the intelligent prototype can be used to automatically supervise evaluation variables and equipment control in aviary thermal environments. This alternative system can be used as an agribusiness 4.0 application to mitigate thermal stress conditions in aviaries and, consequently, reduce productivity losses.