Developing a computer approach to graph designs for better suspension bridges, more efficient chemical processing plants or other systems is the focus of a $400,000 National Science Foundation Early Career Development award received by a mechanical engineer at The University of Texas at Austin.
The CAREER awards are prestigious grants for young teacher-scholars expected to be future academic leaders.
Assistant Professor Mathew Campbell will use advances in engineering, artificial intelligence and mathematical programs to improve the engineer’s design process.
“Optimization has been used in engineering for a long time,” Campbell said, “but it’s been used to provide simple solutions, such as finding the lightest material for a structure, rather than addressing more complicated design requirements, such as determining the most efficient steps for assembling something.”
Campbell has proposed that engineering designs could be represented graphically, providing a better way to review designs and allow computers to be used earlier in the process. Most engineering systems have components that can be represented as nodes on a graph, linked in ways that can be represented as arcs. For example, the motor, springs and lever in a household scale would become nodes, while their interconnections would become the arcs.
Campbell will develop the graph topology optimization method by studying designs for the body-weight scale and other structures, applying tools such as object-oriented programming and artificial intelligence in the process. Other designs he will test include creating a bracket by cutting and folding sheet metal, and developing beams that flex between two curved positions to serve as on/off switches in devices.
Initially, he seeks to make the new approach match traditional computer optimization approaches. The process will involve graphing the design concept under study, generating solutions, evaluating their worth and using information from poor solutions to refine better ones.
A more advanced graphical computer method will ultimately be developed that could select the ideal parts or configurations in a structure, suggest new parts that could be added or offer other design innovations. The method would also have broader applications, such as selecting the best distillation and other components to include in a chemical processing plant, and the optimal physical connections for the processing equipment.
“I really think this will open up what creative power a computer can have,” Campbell said.
He is also developing a graduate course for engineering and fine-arts design students, and continuing to develop a Web tool he helped create for students to post their educational and other information online (www.polaris.engr.utexas.edu/).
A photo of Dr. Campbell can be found at: www.engr.utexas.edu/news/action_shots/pages/Campbell_2005NSFCareer.cfm
Email this article to a friend
About UT's Cockrell School of Engineering:
The University of Texas at Austin's Cockrell School of Engineering ranks among the top six public engineering schools in the United States. With the nation's fourth highest number of faculty elected members of the National Academy of Engineering, the School's more than 7,000 students gain exposure to the nation's finest engineering practitioners. Appropriately, the School's logo, an embellished checkmark used by the first UT engineering dean to denote high quality student work, is the nation's oldest quality symbol. The School maintains a Web site at http://www.engr.utexas.edu