Análise de desempenho de controladores em uma unidade experimental de tanques em série

Abstract

The appropriate selection of control strategies is essential to ensure the stability, performance, and efficiency of industrial processes. Tank systems configured in series are widely used in sectors such as chemical, petrochemical, and water treatment, and require precise control techniques due to their dynamic characteristics and inter-stage coupling. In this context, this project proposes a performance analysis of conventional Proportional-Integral-Derivative Control and Model Predictive Control applied to the level control of the second tank in an experimental two-tank-in-series setup. The entire system was developed from scratch, including the physical structure, electronic design with printed circuit board, and a graphical interface for real-time monitoring. Additionally, the embedded implementations and all analyses were conducted on open-source platforms, with a particular focus on the programming languages Python and C++. The study aims to understand the advantages, limitations, and practical applications of each control method, exploring open-source tools to support decision-making in real-world industrial automation projects.