Abstract
Data‐driven and Hybrid Modelling for Reactor Engineering
Developing advanced digital technologies to operate chemical and biochemical reaction processes is one of the primary research themes highlighted by the 4th Industrial Revolution. However, selecting and combining effective modelling techniques (e.g. kinetic modelling, machine learning, process analytical technology, and soft-sensing) to enable smarter reactor design and process optimisation remains a challenging topic. In our work, we develop different hybrid and surrogate modelling techniques that integrate machine learning into physical models to resolve otherwise intractable problems. This presentation will illustrate the use of these techniques in the field of (bio)chemical reaction engineering for automatic model structure identification, reaction optimisation, knowledge transfer and large-scale reactor design. An outlook of how to exploit these techniques for multi-scale and multi-physics reaction systems visualisation and knowledge generation will also be provided.
Biography
Process Systems Engineering and Machine Learning, Centre for Process Integration, Department of Chemical Engineering and Analytical Science, University of Manchester
Dr. Dongda Zhang is a Lecturer at the Department of Chemical Engineering and Analytical Science, the University of Manchester, and an Honorary Research Fellow at the Centre for Process Systems Engineering, Imperial College London. He currently leads research in the field of machine learning at the Centre for Process Integration, the University of Manchester, and is a member of the BBSRC Pool of Experts in Industrial Biotechnology. He holds BSc degree (2011) from Tianjin University and MSc (Distinction) degree (2013) from Imperial College London. He completed his PhD research within 2 years at the University of Cambridge, and graduated in 2016 after the university special approval for Thesis Early Submission. Prior to his appointment at Manchester, he was a postdoctoral research associate at Imperial and was a recipient of a Leverhulme Early Career Fellowship. His research focuses on developing advanced machine learning and hybrid modelling technologies to accelerate chemical and biochemical process development, optimisation, monitoring, scale-up and visualisation, with primary applications in reaction engineering and bioprocess systems engineering.