Skip to main content

UK Statistics and Energy Cabinet Forecast Kentucky's Energy Future

by Keith Hautala

(April 15, 2014) — A team of students and researchers from the University of Kentucky Department of Statistics and the UK Center for Applied Energy Research worked last summer with the Kentucky Energy and Environment Cabinet (EEC) and the Pacific Northwest National Laboratory to evaluate energy and environmental policy under a range of potential carbon dioxide regulatory scenarios.

The UK team assisted with the enhancement of the Kentucky Electricity Portfolio Model which was developed at the EEC and responds to highly variable factors such as weather, fuel prices, and federal environmental policy, to identify the optimal electricity portfolio and forecast electricity prices, demand, emissions, fuel consumption, employment, and economic growth.

The project report, available online (PDF), discusses the results of the study including the potential economic implications of changing Kentucky's electricity-generating portfolio.

The UK team worked directly with Energy and Environment Cabinet Secretary Leonard K. Peters and his staff. Peters spent nearly two decades at UK as a professor of engineering and administrator. During his time at the university, Peters served as chair of the Department of Chemical and Materials Engineering, vice chancellor for research and graduate studies, and as acting vice president for research and graduate studies.

The model will help the Commonwealth to predict and prepare for the economic effects of projected greenhouse gas emissions regulations, Peters said.

"We were looking at things like what would happen if we had a carbon tax," Peters said. "What would happen if natural gas prices doubled or tripled? If coal prices would change, or if in fact we were able to generate power via nuclear? And what roles do renewables play?"

The model can also help inform policymakers at the U.S. Environmental Protection Agency, by highlighting the specific energy needs Kentucky has, as a state with an intensive manufacturing sector.

"Kentucky is different than New York, we’re different than California, and we’re different than Utah," Peters said. "If you’re doing something that’s going to hurt the manufacturing industry, it’s going to hurt Kentucky, but it’s also going to hurt the nation because those jobs probably aren’t going to go to another state, they’re going to go overseas.”

The collaboration was the brainchild of Aron Patrick, assistant director for the Cabinet, who saw the potential for the Cabinet and the university to work together. The work of the team was two-fold. One part consisted of data analysis, looking at the relationship between variables (such as comparisons between energy sources). The other part, called an optimization problem, is more mathematical, and it involves looking for a particular solution given data with constraints.

“It was our job to determine potential constraints by the EPA: What is the best strategy for building power plants, what to do with existing power plants, et cetera," said Woody Burchett, one of the graduate students who worked on the project. "So it was kind of a fun problem to tackle, because there are a lot of ins and outs that need to be controlled for."

Stromberg said the project is a source of pride for his department.

"We’ve been able to show others across the Commonwealth and across the nation that this type of analysis produces useful information, and that’s exciting," Stromberg says. "And we would hope that it leads others as an example of the kinds of things we can do. This was a bit outside the box for the Department of Statistics."

The UK team included several students, Department of Statistics Professor Arne Bathke, and department Chair Arnold Stromberg. Students on the project were Adam Blandford, Edward Roualdes, Woody Burchett, Matt Rutledge, Shaoceng Wei, Zhiheng Xie, Joel Perry, Yang Luo, and Michael Skapes.  From CAER, Shiela Medina served in leadership role for the project team.

You can also listen to an A&S podcast about it here: