On the (Im-)Possibility of Controlling a Stirring Process

Speaker:  Prof. Steffi Knorn

19 Marzo 2026 | 10:30
DEIB, Sala Seminari "N. Schiavoni" (Ed. 20A)

Contatti:  Prof. Lorenzo Fagiano

Sommario

On March 19^th, 2026, at 10:30 am the seminar on "On the (Im-)Possibility of Controlling a Stirring Process" will take place in DEIB Seminar Room "Nicola Schiavoni (Building 20A).

This talk addresses the challenges of developing feedback control systems for multiphase stirring processes, using a continuous stirred-tank reactor (CSTR) with settler as a case study. Despite the wealth of theoretical models in process engineering literature, the practical implementation of control for such systems remains largely unexplored, with most industrial operations running open-loop.
The presentation examines three modeling approaches: the physically motivated but practically unidentifiable Particle Balance Equation (PBE) model, simplified moment models, and data-driven models. Through experimental identification, the talk demonstrates the severe limitations of complex first-principles models for control purposes and highlights the practical value of simpler black-box approaches combined with extended Kalman filtering and model predictive control.
Key challenges are discussed throughout, including measurement technology based on image-based soft sensors for droplet size distribution, observability and identifiability limitations, and the significant gap between theoretical control concepts and real-world applicability. The talk concludes with a critical reflection on the state of control engineering in process industries, emphasizing that the transition to renewable raw materials and resource-efficient operation will require pragmatic control solutions that bridge the gap between academic literature and industrial reality.

At 11:30 a.m., the joint presentation of the ENHANCE guided course in Control and Automation Engineering will follow (Prof. Steffi Knorn (TUB), Damiano Varagnolo (NTNU), Lorenzo Fagiano (PoliMi).

Biografia

Steffi Knorn is a professor of control engineering at Technische Universität Berlin. Her research focuses on modelling, system identification and feedback control of complex dynamical systems, with applications ranging from process and manufacturing systems to energy and transportation networks. She studied systems engineering and technical cybernetics at Otto von Guericke University Magdeburg between 2003 and 2008 and received her PhD from the Hamilton Institute at the National University of Ireland, Maynooth, in 2013 followed by postdoctoral positions in Australia and Sweden. Before joining TU Berlin in 2021, she held academic positions at Uppsala University and Otto von Guericke University Magdeburg.