Inteligência artificial aplicada ao controle de biorreatores

Abstract

This work analyzes the application of Artificial Intelligence (AI) in process control in bioreactors, with emphasis on fermentation vats used in the sugarcane-based ethanol industry. The research was conducted through a literature review and comparative analysis of studies developed at the University of São Paulo (USP) and the State University of Campinas (UNICAMP), which apply artificial neural networks integrated with Model Predictive Control (MPC) strategies. Additionally, a bibliometric analysis was performed using the Cytoscape software, enabling the mapping of research evolution and the identification of emerging trends. The results indicate that the use of AI in this type of control provides significant gains in stability, robustness, and efficiency, particularly in multivariable and nonlinear systems, such as biotechnological processes. The incorporation of techniques such as soft sensors, digital twins, and hybrid models expands predictive capability and operational flexibility, representing an advance toward Industry 4.0. The study also shows that the adoption of these technologies in ethanol plants is an interesting alternative to improve energy efficiency and productivity. However, the research also indicates that there are still challenges related to data availability, model interpretability, and the integration of new systems with existing industrial infrastructure.