Exploiting Low Dimensional Structures for Control, Design, and Optimization

19 February 2016
12:00 pm
Exploiting Low Dimensional Structures for Control, Design, and Optimization
Boris Kramer
Postdoctoral Associate
Aerospace Computational Design Laboratory
Dept. of Aeronautics and Astronautics
MIT

In many applications in engineering and the sciences, the underlying dynamics of a complex system is found to be low-dimensional, and this can be exploited for the goal of control and optimization. We briefly present prior work, where physics-based reduced order models were used for system identification and control of fluids and mechanical systems. Following, we present recent work on indoor-airflow sensing through reduced order models and the compressed sensing method to detect flow phenomena. In this setting, sensing complex flow structures was possible by extending the dynamic mode decomposition basis to account for time series of measurements. We then discuss ongoing work on control of parameter varying systems, where the physical parameters are uncertain and unknown. In this setting, we leverage the availability of real data with the predictive capabilities of physics-based models.