BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Memento EPFL// BEGIN:VEVENT SUMMARY:Inaugural Lecture - Prof. Olga Fink DTSTART:20230510T173000 DTEND:20230510T184500 DTSTAMP:20260916T034309Z UID:56e1d357ec2f9c4e526189f8195328309f0f9960ee152fbdc5eef420 CATEGORIES:Inaugural lectures - Honorary Lecture DESCRIPTION:Prof. Olga Fink\nDate: 10 May 2023\nTime: 17:30 - 18:45\nIntro ductions by the Dean\, lectures by Prof. Olga Fink and Prof. Stefana Paras cho. Followed by an Apero.\nPlace: CO2\nZoom link\n\nTitle:\nFrom physics to machine learning and back: Applications to intelligent maintenance and operation of complex systems\n\nAbstract\nThe amount of measured and colle cted condition monitoring data for complex infrastructure and industrial a ssets has been recently increasing significantly due to falling costs\, im proved technology\, and increased reliability of sensors and data transmis sion. However\, faults in safety critical systems are rare. The diversity of the fault types and operating conditions makes it often impossible to e xtract and learn the fault patterns of all the possible fault types affect ing a system. Consequently\, faulty conditions cannot be used to learn pat terns from. Particularly run to failure trajectories are rare. Even collec ting a representative dataset with all possible operating conditions can b e a challenging task since the systems experience a high variability of op erating conditions. Therefore\, training samples captured over limited tim e periods may not be representative for the entire operating profile. The collection of a representative dataset may delay the implementation of dat a-​driven fault detection\, diagnostics and prognostics systems. Moreove r\, some of the current limitations include limited scalability\, generali zation ability and interpretability of the developed models.\nThe talk wil l give insights into the currently ongoing research at the Intelligent Mai ntenance and Operations Systems Laboratory at EPFL\, focusing on two key a reas. Firstly\, the presentation will center around the fusion of physics- based and deep learning algorithms\, particularly in the context of fault diagnostics and prognostics. Secondly\, the presentation will delve into t he topic of domain adaptation and generalization and their impact on impro ving fault diagnostics and prognostics.\n\nAbout the speaker\nOlga Fink ha s been assistant professor at EPFL since March 2022\, heading the “Intel ligent Maintenance and Operations Systems” laboratory. Olga is also a re search affiliate at Massachusetts Institute of Technology. Before joining EPFL faculty\, Olga was assistant professor of intelligent maintenance sys tems at ETH Zurich from 2018 to 2022\, being awarded the prestigious profe ssorship grant of the Swiss National Science Foundation (SNSF). Between 20 14 and 2018 she was heading the research group “Smart Maintenance” at the Zurich University of Applied Sciences (ZHAW) where she was senior lect urer. Olga received her Ph.D. from ETH Zurich on the topic of “Failure and Degradation Prediction by Artificial Neural Networks: Applications to Railway Systems” and a diploma in industrial engineering from the Hambu rg University of Technology. She has gained valuable industry experience as a reliability engineer for railway rolling stock and as a reliability and maintenance expert for railway systems. Olga’s research focuses on Hybrid Algorithms Fusing Physics-Based Models and Deep Learning Algorithm s\, Hybrid Operational Digital Twins\, Transfer Learning\, Self-Supervised Learning\, Deep Reinforcement Learning and Multi-Agent Systems for Intell igent Maintenance and Operations of Infrastructure and Complex Assets.\n\n \n\n\n  LOCATION:CO2 https://plan.epfl.ch//?room==CO%201 https://epfl.zoom.us/j/68 989432342 STATUS:CONFIRMED END:VEVENT END:VCALENDAR