Desenvolvimento de um nariz eletrônico para monitoramento de gases poluentes em aterros sanitários

• Ano de defesa: 2023
• Autor(a) principal: Muniz, Juliene dos Santos Davi
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Engenharia Elétrica; Programa de Pós-Graduação em Engenharia Elétrica; UFPB
• Palavras-chave em Português: Aterros sanitários; Biogás; Sensores de gás; Nariz eletrônico; Rede neural artificial; Sanitary landfills; Biogas; Gas sensors; Electronic nose; Artificial neural network; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/30249

The biogas generated from the degradation of solid urban waste in sanitary landfills is composed of various gases, including gases that contribute to global warming and toxic gases harmful to the health of both people and animals. Monitoring the concentration of these gases in sanitary landfills and their vicinity is of great importance as it allows for the assessment of potential air contamination caused by biogas generation in landfills and the proposal of methods to minimize it. This work aims to develop an Electronic Nose capable of identifying the gases present in biogas and estimating their concentra- tions using an Artificial Neural Network (ANN). Several gas sensors were selected, and a printed circuit board was constructed with the sensors and their signal conditioning circuits. A signal acquisition system for the sensors was developed using LabVIEW soft- ware, which collects data read by the sensors and storages it in a spreadsheet used for data analysis and ANN training. After several data acquisition sessions from the sen- sors, the input and output data were organized into a single spreadsheet, which was used with a neural architecture developed to predict the concentrations of the target gases. The results obtained from the ANN training showed that it was possible to mea- sure the concentrations of gases with a root mean square error (RMSE) less than 1 for most of the target gases considered, confirming the accuracy of the developed system.