This NSF award is Sanfelice's second grant this year related to hybrid systems research. The first was an award of $360,000 from the Air Force Office of Scientific Research to research the design of control algorithms for autonomous vehicles in adversarial environments.
Sanfelice was awarded the 5-year NSF grant under its prestigious Faculty Early Career Development Program. The program supports junior faculty who exemplify the role of teacher-scholars through outstanding research, excellent education, and the integration of education and research. The award also recognizes junior faculty with great potential to become future leaders in their areas of research.
Sanfelice received the AFOSR award through the agency's Young Investigator Program, which awards about $18 million in grants to young researchers yearly.
Under the NSF grant, Sanfelice, who is director of the Hybrid Dynamics and Control Research Laboratory, will research the application of hybrid systems theory to the design of smart grid systems.
His AFOSR grant will support development of new hybrid systems theory that enables autonomous systems to predict how adversaries might act or how situations might change, and how those systems can be designed to make the right control decisions during uncertain and rapidly altering circumstances.
Hybrid systems are so called because they exhibit two distinct types of behavior, which are embodied in two different types of mathematical equation.
Simply put, hybrid systems involve both jump and flow.
Assistant Professor Ricardo Sanfelice: "This is a topic of high importance to society worldwide and, in particular, to our nation."
Hybrid systems include walking robots, colliding billiard balls, genetic networks, and electrical circuits. Perhaps the simplest example is a bouncing ball, which exhibits periods of flow, during the smooth arc of a bounce, in between impacts, or jumps in velocity.
Sanfelice believes the mixed jump-and-flow behavior of hybrid systems theory lends itself particularly well to the testing and design of smart grids, which are characterized by periods of steady conditions continually interrupted by jumps in currents and voltages.
Hybrid systems theory is also well suited to controlling multiple autonomous systems that have to make rapid changes in their actions and parameters. "These are networked, interconnected hybrid systems of utmost relevance to our nation and worldwide for which analysis and design tools have to be developed," Sanfelice said.
Smart grids are basically remote-controlled computerized electric utility grids with components that can communicate digitally with one another and with the grid control center. Smart grids aim to make electricity supply more efficient and more responsive to rapidly changing supply and demand, especially on grids that include wind and solar energy, generation of which ebbs and flows with the elements.
Jump and Flow: The break in a game of pool exemplifies a hybrid system in which periods of flow (when balls roll smoothly across the beize) mingle with jumps (when balls impact).
Examples of multiple autonomous systems, Sanfelice said, are aerial and ground vehicles used in disaster relief and hostile environments. To complete their missions, these vehicles must coordinate their own actions while cooperating with other vehicles and control centers, all under highly uncertain and adverse conditions.
Sanfelice's approach will be to interpret these systems as a series of interconnected hybrid systems that can be partitioned into small, more easily manageable components, thus enabling them to be analyzed and designed in a modular fashion.
"This divide and conquer approach needs a theory of interconnections of systems and some design tools that guarantee robustness and optimality," Sanfelice said.
The tools developed by Sanfelice through the NSF grant will ensure that smart grid components work under real-world conditions, such as those simulated on the smart microgrid tested at Sandia National Laboratories, where Sanfelice will conduct testing.
The Photovoltaic Systems Evaluation Laboratory at Sandia National Labs provides expertise and test support for evaluating photovoltaic and other distributed energy hardware. Sanfelice's work will be tested on arrays such as that pictured.
Photo courtesy of Sandia National Labs.
"Collaboration with Sandia National Laboratories will be instrumental in the validation of the new tools and will introduce new problems emerging from the smart grid field to the control community," Sanfelice said.
In December 2007, Congress passed, and the President approved, Title XIII of the Energy Independence and Security Act of 2007, which mandated several smart grid initiatives aimed at modernizing the national grid.
"This is a topic of high importance to society worldwide and, in particular, to our nation," Sanfelice said. "It will also impact other areas of science and engineering, including autonomous transportation systems, genetic and biological networks, and cyber-security systems."
The tools developed by Sanfelice through the AFOSR grant will be used in the design of reconfigurable satellite systems as part of a collaboration with the Air Force Research Laboratory's Space Vehicles Directorate in Albuquerque, N.M.
The results of this research will not be confined to the laboratory, said Sanfelice, who intends to increase awareness among young students of the future applications of hybrid systems.
"The proposed research plan is deeply integrated with teaching and training at middle and high school levels," he said. "We will be training instructors and students on control engineering and its applications to smart grids and autonomous systems."
On a local level, Sanfelice wants to encourage Hispanic students from low socioeconomic backgrounds to participate in smart grid studies, and on a broader level he will disseminate his research via workshops, collaborations with local industries, and international partnerships.
Sanfelice's research on hybrid systems was published by Princeton University Press recently as a textbook titled Hybrid Dynamical Systems: Modeling, Stability, and Robustness, by Rafael Goebel, Ricardo Sanfelice, and Andrew Teel. Sanfelice teaches this material to graduate students at the UA department of aerospace and mechanical engineering, and internationally at various institutes and conferences.