Dissertation entitled: “Bayesian Neural Networks for Calibration of Air Pollution Sensors and Fault Detection in Chemical Processes”.My research project aimed to study applications of bayesian neural network models for two study cases. The first case study focuses on the application of a Bayesian neural network model to solve a problem of calibration of sensors used to monitor air pollution in urban areas. The second case study focuses on the application of a Bayesian neural network model for autonomous detection of operating faults that occur in a real chemical process.

My master's degree research project was funded by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). The research was developed under the scope of the FAPESP project entitled “Application of Bayesian deep learning for Smart City monitoring and chemical industry processes operation” (grant number: 19/08280-9), which is associated with the FAPESP project entitled “Institutional development plan in the area of digital transformation: advanced manufacturing and smart and sustainable cities (PDIp)” (grant number: 17/50343-2) developed by the Instituto de Pesquisas Tecnológicas (IPT).

Contribution to the development of the the FAPESP project entitled “Institutional development plan in the area of digital transformation: advanced manufacturing and smart and sustainable cities (PDIp)” (grant number: 17/50343-2) developed by the Instituto de Pesquisas Tecnológicas (IPT).

Research internship developed in the Process Simulation and Control Laboratory (LSCP) at Escola Politécnica da USP. The research project focused on the development of CFD (Computational Fluid Dynamics) simulations of twelve case studies about fluid dynamics found in real chemical processes. The simulations were developed using the Ansys CFX software.

Research Internship developed in the Process Simulation and Control Laboratory at Escola Politécnica da USP. The research project focused on the development of a computational simulation of a control loop for a heating tank. The simulation was developed using the softwares MATLAB and Simulink. The P&ID diagram of the control loop was generated using SmartPlant P&ID software.

Research internship at the Functional Materials Engineering laboratory of the IMO-IMOMEC Institute at the University of Hasselt. The research project focused on the development of organometallic copper inks for printing conductive patterns using inkjet printers. The project aimed to cooperate in the development of new methods of manufacturing conductive patterns used in electronic devices such as IoT sensors, RFID tags, OLED electrodes, OPV electrodes, etc. Responsibilities: the study of the formation of ink droplets in the printer's inkjet nozzle, analysis of the quality of ink deposition on different substrates, and optimization of the heat treatment required for the patterns printed using the inks under development.