The study of synchronization of networked systems has attracted the attention of several disciplines including biology, chemistry and physics,to name a few. In engineering, synchronization is a fundamental requirementof most networked engineering applications. It enables the necessary coordination among agents required to implement several communication systems as well as network protocols. In this talk, we present a systematic study of synchronization on distributed (networked) systems that spans from theoretical modeling and convergence analysis to distributed controller design and implementation.
We first focus on developing a theoretical foundation for synchronization of networked oscillators. We study how the interaction type (coupling) and network configuration (topology) affect the behavior of a population of identical oscillators. Using models from the physics community, we show that, under mild conditions, phase consensus (common phase value) can be achieved for arbitrary network topologies, something only known before for very restrictive network topologies and coupling. The underlying theoretical machinery developed also allowed the study of oscillator heterogeneity, extending our results for populations of non-identical oscillators with different natural oscillation frequencies.
We then focus on more practical aspects of synchronization in two different areas: information networks and smart grids.
Within information networks, we examine the synchronization of computer clocks connected via a data network and propose an algorithm to synchronize them. Unlike current solutions, our algorithm neither estimates the frequency difference (skew) among clocks nor does it introduce offset corrections for synchronization, which makes it more robust to noisy measurements. Moreover, we show that the recommendation of current standards (e.g. NTP and PTP) to avoid loops is not only unnecessary but even detrimental for achieving further noise reduction. We implement the algorithm on a cluster of IBM BladeCenter servers running Linux and we experimentally verify that our algorithm outperforms the well-established solutions.
Finally, we study synchronization on smart girds. We propose a load-side frequency control scheme that can rebalance power and resynchronize frequencies after a disturbance (a.k.a. primary control), while restoring the frequency to its nominal value (a.k.a. secondary control). Unlike the generation side secondary frequency control that is centralized, our load-side control only requires each bus to communicate with its neighbors. Our scheme also provides a fair distribution of the load corrections by minimizing the total disutility of the controllable loads. We prove that such a distributed load-side control is globally asymptotically stable and we illustrate its convergence with numerical simulations.
Enrique Mallada is a Post-Doctoral Fellow in the Department of Computing and Mathematical Sciences and the Center for Mathematics of Information at California Institute of Technology. He received his Ingeniero en Telecomunicaciones degree from Universidad ORT, Uruguay, in 2005 and his PhD degree in Electrical and Computer Engineering with a minor in Applied Mathematics from Cornell University, Ithaca, in 2014. From 2004 to 2007 he was an IT-Specialist at IBM, and in 2008 he worked as an Engineer at the Traffic Engineering Department of ANTEL, the main telecommunications operator in Uruguay. He was also a Teaching and Research Assistant in the Department of Telecommunications at Universidad ORT, and a summer intern in IBM T. J. Watson Research Center during the summer of 2011. Dr. Mallada was granted a scholarship by the Organization of American States in the 2008-2009 and 2009-2010 academic years, and he received Cornell University’s Jacobs Fellowship in 2011. His research interests include control and dynamical systems, networks, and optimization.
