Researchers developing next-generation energy models with eye toward smart grid
During the summer of 2007, electricity prices shocked the Southern Hemisphere.
In Victoria, Australia, demand for electricity skyrocketed - and so did the price, to $10,000 per unit, many times higher than the average of $50. In neighboring Tasmania, prices crashed to -$1,000, with power companies paying the tiny island south of Melbourne to take electricity.
How could such disparity exist in communities just a few hundred miles apart?
Researchers from the University of Illinois at Urbana-Champaign, including three from the Coordinated Science Laboratory, recently received a three-year, $1.03 million grant from the U.S. Department of Energy to tackle complex questions regarding energy markets. This research will allow for a deeper understanding of the competitive interplay between firms, power marketers and consumers in the face of uncertainty in both demand and resource availability.
Ultimately, the work will contribute toward the development of the smart grid---a system that is expected to lead to increased efficiency, lower power costs, and greater reliability.
“A major challenge offered by the power market is that pricing is the result of a complex bidding framework,” said Uday Shanbhag, a professor of industrial and enterprise systems engineering and principal investigator on the project. “Together with the uncertainty in electricity usage and fuel costs, this leaves consumers with unpredictable energy supplies and prices. As the smart grid evolves, existing market designs are inadequate.”
During the past decade, deregulation has played a role in wildly fluctuating prices. Consider, for example, the 2000 California blackouts, during which power companies chose to "game the system" by reducing production during peak usage times to drive prices higher.
However, strategic gaming is not the sole culprit. For instance, a few years earlier during a brief period of deregulation, prices in Illinois shot to $5,000/megawatt per hour, at least one hundred times greater than the price before deregulation. The price spike in this case was not the result of manipulation, but rather of plant operators being unable to meet increased demand, perhaps due to suboptimal generator technology or the limitations of the electric power network.
In the markets of tomorrow, utilities will have access to a broad power portfolio including wind power and other renewable energy sources. Generation will no longer be restricted to conventional power plants but will be distributed in nature and include residential solar panels and small wind farms. In addition, consumers will play an important role in the market by responding to prices. For example, “smart” refrigerators and air-conditioners will dynamically curtail usage in response to increasing prices.
The Illinois team plans to develop the next generation of models that seamlessly accommodate new energy sources, distributed generation, and the activities of consumers. These models will be the basis for research on electricity markets of tomorrow.
“On paper, you'd have perfect efficiency through deregulation, but unfortunately it doesn't always play out that way in the real world,” says Sean Meyn, professor of electrical and computer engineering and a co-principal investigator. “We're looking at what kind of restrictions you can place on deregulated markets to help achieve a system that works better for everyone.”
The research will draw on the team's significant interdisciplinary strengths, including dynamical systems and control, numerical and stochastic optimization and game theory, stochastic networks and probability theory, and machine learning. Other team members include Prashant Mehta, a professor of mechanical science and engineering, and Swanlund Chair Tamer Başar, a professor of electrical and computer engineering.
While the Illinois team will primarily focus on electricity markets, their work could apply to many complex economic systems, including communication networks, health care, air and ground transportation and more.
“In addressing the challenges associated with electricity markets, we will also draw upon the knowledge base we have developed in connection with our recent work on pricing and incentive issues that arise in other areas, particularly communication networks,” says Başar. “In the same vein, we expect the new methodologies to be developed within the scope of this project to find immediate applications in other complex economic systems as well.”