Data-driven and AI-based modelling and control

Overview

Here, contributions related to data-driven and learning-based modelling and control are presented, focusing on complex systems such as solar desalination facilities and microalgae-based bioprocesses, where accurate first-principles modelling is often unfeasible due to nonlinear dynamics, time-varying behaviour, and strong coupling with environmental conditions. The research aims to integrate advanced control strategies —including nonlinear predictive, hierarchical, and reinforcement learning-based control— with data-driven and hybrid (physics-informed) modelling approaches to achieve robust and efficient operation. Special attention is devoted to model-free learning frameworks that enable autonomous adaptation and optimization directly from process data, improving performance, stability, and resource efficiency. These contributions strengthen the link between classical control theory and artificial intelligence, supporting the sustainable development of energy–water–food nexus technologies. Moreover, the use of AI-based solutions for microalgae classification and contamination detection are explored.