Seminar: Forecasting Bifurcations in Aeroelastic and Other Systems

Multiple stable dynamics can exist in systems that have thresholds and breaking/tipping points, such as aeroelastic and other systems. Predicting these breaking points or bifurcations before they happen is a crucial challenge for monitoring engineered systems and for understanding natural systems such as ecosystems. Unfortunately, such bifurcations can now be predicted only if an accurate model of the entire system is available. Recent model-less methods analyze the critical slowing down phenomenon which occurs in the dynamics as a critical transition is approached. However, these existing methods provide only qualitative information and cannot accurately identify when the breaking point will occur or how the system will behave after the critical transition. We show that it is possible to predict breaking points as well as the system behavior after those points without a model of the system. We use a novel technique to analyze the systems response to large perturbations well in advance of the critical transition. Such perturbations can be due to environmental events or intentional system interrogation. In this presentation we will show how to apply the new method to aeroelastic systems, vibration-based mass sensors, and ecosystems.

About the Speaker

Bogdan Epureanu is a Professor of Mechanical Engineering at the University of Michigan. He obtained is PhD from the Mechanical Engineering department at Duke in 1999. His research focuses on complex systems such as biological and epidemiological systems, aerospace and automotive structures, and turbomachinery. Examples include creating novel mechano-chemical dynamic models of nanoscale intracellular transport processes, developing the next generation of highly-sensitive diagnosis and monitoring techniques, discovering novel methods for forecasting tipping points in complex systems such as disease epidemics and ecological systems, developing innovative reduced order models of multi-physics systems such as Li-ion batteries and complex structures, creating advanced system identification and control methodologies for smart structures and fluid-structural systems. These blend novel methods and theory with fundamental experiments in linear and nonlinear dynamics from macro to nano-scale. Some of his honors include: the A. M. Strickland Prize, for the best paper published in the Proceedings of IMechE as a significant contribution to the field of manufacturing; the Young Innovator Award from Petro-Canada; the Pi Tau Sigma Gold Medal Award from ASME; the NSF CAREER Award; the Ferdinand P. Beer and E. Russell Johnston, Jr. Outstanding New Mechanics Educator Award from ASEE; the Junior Achievement Award from the American Academy of Mechanics; the Outstanding Achievement Award from the UofM Department of Mechanical Engineering; and the John Ulrich Education Excellence Award and the 1938E Award from the UofM College of Engineering.

For more information, please visit: http://www-personal.umich.edu/~epureanu/.

Hosted by Professor Kiran D'Souza