Konstantinos Gryllias
KEYNOTE LECTURE
From Data to Decisions: Advancing Diagnostics through AI, Deep Learning, and Hybrid Intelligence
Konstantinos Gryllias
Department of Mechanical Engineering
KU Leuven, Belgium
ABSTRACT
Artificial intelligence is reshaping the way we understand, monitor, and maintain complex systems, such as wind turbines. This keynote will explore how deep learning, hybrid AI and Digital Twin approaches are transforming condition monitoring, from early fault detection and anomaly detection to prognostics and estimation of the remaining useful life.
Drawing on applications in industrial systems such as rotating machinery, the talk will highlight how data-driven models, when combined with physics-based knowledge and digital twins, can overcome key challenges such as data scarcity, domain shifts, and model interpretability. The keynote will showcase recent advances in transfer learning, anomaly detection, and simulation-driven AI, demonstrating how next-generation diagnostic systems can become more accurate, reliable, and explainable.
Ultimately, the lecture will provide a forward-looking perspective on how AI-enabled diagnostics can support safer, more efficient, and more sustainable engineering systems across different industrial domains, including wind turbines, industrial gearboxes, vehicle drivetrains, injection molding and 3d printing.
SPEAKER BIOGRAPHY
Konstantinos Gryllias received his Diploma and Ph.D. degrees in Mechanical Engineering from the National Technical University of Athens, Athens, Greece, in 2004 and 2010, respectively. He is currently Professor of Vibro-acoustics of machines and transportation systems at the Department of Mechanical Engineering, KU Leuven, Leuven, Belgium. In parallel he is the Manager of the University Core Lab Flanders Make@KU Leuven Motion Products, Belgium. Flanders Make is the strategic research centre for the manufacturing industry in Flanders, Belgium. He serves as associate editor of the Journal of Mechanical Systems and Signal Processing. His research interests include Condition Monitoring, Signal Processing, Artificial Inteligence, Prognostics, and Health Management of Mechanical and Mechatronic systems. He is chairing the ISMA course on Modal Analysis and is member of the organising commitee of ISMA International Conference on Noise and Vibration Engineering. Prof. Gryllias is actively involved in major AI initiatives including Leuven.AI and the Flanders AI Research Program, bridging academic research and industrial innovation. Additionally he is participating a number of national, international and industrial project of condition monitoring, including MSCA Patron, Satisfy.AI, HybridWind etc.
