Purdue News
Purdue News
|
|
August 19, 2004 Three Purdue engineers win NSF early-career awardsWEST LAFAYETTE, Ind. – Three Purdue University engineers have won the National Science Foundation's most prestigious honor for outstanding young researchers. The Faculty Early Career Development awards range from $300,000 to $500,000 in research funding over four or five years. About 110 engineers win the awards annually, or roughly 20 percent of the total number of researchers who compete for the grants. Purdue's engineering recipients this year are Yung-Hsiang Lu, John A. Morgan and Jean-Philippe Richard. Details about the Purdue awardees and their research follow: Energy Management:
Lu, an assistant professor of electrical and computer engineering, will use the NSF grant for research to develop operating system software that enables computers to automatically save energy by managing how much power to use based on which jobs are being performed. The research, funded with a five-year, $419,775 career grant, will focus on three parts: creating universal interfaces between operating systems and all hardware components for automatically characterizing how much power each component consumes; building mathematical models that are better than conventional models at estimating how much power a component needs; and developing software techniques that determine which power states are needed for specific hardware components at a given time. Individual hardware components that are not being used at a given time can be placed in a sleep state, while other components are running normally. Software and models developed by the end of the five-year period will be open source. Metabolic Engineering:
Morgan, an assistant professor of chemical engineering, will use the NSF grant for research that could ultimately help scientists use plants and single-cell organisms like algae to create industrial chemicals, medicines and foods. Currently, most industrial chemicals are made using fossil fuels, which are being depleted, whereas plants and single-cell organisms that use photosynthesis are renewable resources. In research funded with a $400,000 five-year NSF career grant, Morgan will develop a mathematical tool and mass spectrometry methods to precisely measure the metabolism of a single-cell organism similar to algae. The ultimate goal is to make it possible to one day engineer such organisms so that they manufacture chemicals, pharmaceuticals and foods. Researchers must first, however, gain a better understanding of the organisms' complex metabolisms, which consists of hundreds of reactions, all having different rates. Using analytical methods known as gas chromatography and mass spectrometry, Morgan will study the organism's metabolism by tracing an isotope of carbon as it passes through numerous "metabolic pathways" inside the cells. Mixed Integer Programs:
Richard, an assistant professor of industrial engineering, will use his five-year, $400,000 career grant to improve software needed to "optimize" industrial operations. Such operations include scheduling the proper order in which to carry out jobs in factories, purchasing the right number and selection of parts for product manufacturing so that inventory space isn't wasted and creating more efficient shipping schedules for trucks and train cars. Such problems fall within two main categories: discrete problems and continuous problems. In the former, decisions are either made or they are not made. For example, company executives either decide to build a plant or they don't; there is no possibility of building part of a plant. In continuous problems, however, decisions can be made in increments that are fractions, such as increasing the production of a certain item by 50 percent. Typically, researchers have dealt mainly with the integer portion of problems, which is generally considered to be the most challenging. Richard, however, will take a different approach. Because both continuous and discrete decisions are interrelated, he will work toward creating software that tackles both types of decisions simultaneously within "mixed-integer" problems. Writer: Emil Venere, (765) 494-4709, venere@purdue.edu Sources: Yung-Hsiang Lu, 765-494-2668, yunglu@purdue.edu John A. Morgan, (765) 494-4088, jamorgan@ecn.purdue.edu Jean-Philippe Richard, (765) 494-5166, jprichar@ecn.purdue.edu Purdue News Service: (765) 494-2096; purduenews@purdue.edu
|
