Safe Reinforcement Learning for Chemical Process Control
Safe reinforcement learning (RL) in chemical process control is an emerging field that addresses the complexities and inherent risks of controlling chemical reactions and systems in real-time. This area of research focuses on designing AI-driven algorithms that can predict and mitigate potential model mismatches, ensuring operations' safety and stability.
José Torraca
Safe Reinforcement Learning for Process Control
José Torraca is a visiting PhD student at Imperial College London, specializing in safe reinforcement learning for chemical processes. His research is focused on developing algorithms that enhance safety and efficiency, ensuring robust process control in dynamic and uncertain environments. He is also a PhD researcher in a partnership between the Laboratory of Software Development (LADES), PEQ/COPPE/UFRJ, and Petrobras. His work includes implementing process monitoring and control algorithms, focusing on virtual flow metering, and controlling slug flow in oil and gas applications.