Computational Co-Designs for Controlling Very Large-Dimensional Power Systems

Abstract:

In this talk I will present a set of new results on the problem of wide-area control of very large-dimensional power systems. The first half of the talk will focus on the computational aspects of the control design, where I will present a new design approach called "control inversion". By this approach, large-dimensional power system models can be projected conveniently into lower dimensional spaces by exploiting the inherent clustering properties of the network dynamics; then, a reduced-order controller is designed for this simple model, and, finally, this controller is projected back to the full-dimensional network for actual implementation. The method not only improves the tractability of the design, but also provides significant savings in the number of communication links needed for feedback. In the second half of the talk, I will shift my attention towards two of the most important challenges in data communication arising in wide-area control- namely, sensitivity to delays and data sparsification. Using the concepts of modal participation factors and relative gain arrays, I will propose a distributed communication architecture by which control centers can implement a sparse realization of wide-area controllers with very little loss in the overall response. I will also describe a co-design strategy by which one can spot the most important generators in the system for the purpose of power oscillation damping after any disturbance, and, thereafter, prioritize the communication of states from these special generators to minimize the overall delay in the feedback path. I will present simulations to illustrate the pros and cons of such data prioritization, and their associated protocol designs. The talk will end with some final remarks about the resilience of these wide-area protocols against denial-of-service and data manipulation attacks.  

The overall goal of the talk will be to pinpoint some of the most challenging research problems for today's power grid where power engineers can largely benefit from collaborations with control theorists, communication engineers, computer scientists, and numerical analysts. The content will highlight my recent works with PhD students Nan Xue and Abhishek Jain at NC State, as well as my work with Dr. Anuradha Annaswamy and her postdocs from MIT.

Biography:

Bio:

Aranya Chakrabortty received his PhD degree in Electrical Engineering from Rensselaer Polytechnic Institute, Troy, NY, in 2008. Following that he was a postdoctoral research associate at the University of Washington Seattle, for a year. From 2009 to 2010, he was an Assistant Professor at Texas Tech University. Since 2010, Aranya has joined the Electrical and Computer Engineering department of North Carolina State University, where he is currently an Associate Professor. His research interests are in the general area of power system dynamics, modeling, stability, and control, with a special focus on wide-area monitoring and control using Synchrophasors. He is a senior member of IEEE, and currently serves as an Associate Editor of IEEE Transactions on Control Systems Technology. He has held visiting positions at University of Illinois Urbana-Champaign and Tokyo Tech University. He received the NSF CAREER award in 2011.