Process System

Process systems engineering research currently focuses on developing advanced predictive models for improved design, operation, and control of industrial processes. This involves integrating physics-based models with data-driven approaches, such as Bayesian optimization and hybrid machine learning architectures incorporating Bayesian neural networks, to create more accurate and robust representations of complex systems. Key advancements include the use of knowledge graphs and large language models to automate problem-solving and improve decision-making, as well as the development of physics-aware proxy models for efficient and generalizable predictions. These improvements promise significant advancements in process optimization, safety, and sustainability across various industries.