Thursday, March 3, 2011

Lycoming Physics students discuss current research

by: Aliza Davner

While most on-campus research remains hidden behind closed doors and locked away in labs, the students and faculty in the department of Astronomy and Physics are ready to share the results of their hard work with the rest of the campus community.

Lycoming physics Professor Dr. Christopher Kulp received his doctorate in physics from the College of William and Mary in 2004 and began teaching at Lycoming in 2008. His area of expertise lies in nonlinear dynamics with an even more specific emphasis on nonlinear time series analysis which, according to Kulp, is just a fancy term for data analysis.

Recently, Kulp has been studying combustion in open flames in industrial settings, with the “goal to develop a simple mathematical model of combustion … to ensure (the flames) are burning as efficiently as possible.”

In order to gather the necessary data, Kulp video records flames from a variety of burners, and then uses self-developed computer programs to get data from the image in order to understand how the flame is burning. Kulp prefers the use of video-recordings over other possible methods, as they not only allow him to observe the flames in real-time, but they eliminate the need to place a probe inside the flame, which could skew any data collected by affecting the flame itself. As such, the use of video recording allows for cheap, non-evasive research.

This combustion research is being conducted in connection with the Oak Ridge National Laboratory, which is a large national physics laboratory in Oak Ridge, Tenn. He hopes to apply his findings to industrial projects, such as the internal combustion engine and industrial boilers.

Kulp also is involved in several projects with Lycoming students. For example, he recently collaborated with Senior Suzanne Smith on a project which tackled what is known in physics as the “problem of characterization,” or, how can a researcher tell that a set of data comes from a random, chaotic, or periodic (regular) system?

A random system is just as the name implies — random. A chaotic system, however, is not random. What makes this system unique, Kulp said, is that there is some type of order imposed on the system which causes the data to behave in an unusual way.

“It is often related to a fractal,” said Kulp, but he emphasized that other possibilities exist. A periodic or regular system is also easy to characterize, as the data behavior repeats itself in regular intervals.

Smith explained the project in more detail, saying “Our research was a new approach in the field, we paired the 01 Test with the Test for Determinism … when you run (the 01 Test) on your series, a 0 as the output means it is periodic, a 1 as the output means chaotic.”

Smith added, “The Test for Determinism is what helps us interpret the results from the 01 Test. A pass in the test for determinism and a 1 from the 01 test means a chaotic system. While a fail in the test for determinism and a 1 from the 01 Test means a random … series. We modeled specifically a number of test systems that we knew how they behaved. Our ultimate goal was to find the threshold of failure within the 01 Test.”

Over the course of the project, Smith and Kulp developed an algorithm that improved upon preexisting algorithms in terms of being able to improve the ability of tests to distinguish between random and chaotic systems when data is “noisy.” According to Kulp, “noisy systems” result from measurement errors made when measuring data.

Smith and Kulp worked on the project for about 1 1/2 years and Smith presented their findings at the Zone 3 meeting of the Society of Physics Students, a national organization for physics students. The Lycoming College Chapter is part of Zone 3, which represents colleges and universities in Pennsylvania and New Jersey.

In addition to the presentation, Smith and Kulp co-authored an article based on their findings that was published in Physical Review E, a highly-respected physics journal.

For Smith, the opportunity to conduct such a research project was “exciting” and has lead to opportunities outside of Lycoming. “This research has helped me increase my knowledge of the world of computational physics and also helped me land an internship at NOAA (National Oceananic and Atmospheric Administration) over this past summer … (it) has really opened my eyes to one of the many fields of physics I could possibly go into … I have started to consider my path after graduation, and computational nonlinear dynamics is definitely on my list but so is nuclear physics, I am hoping to get another research position in one of those two fields this summer to help my decision along,” she said.

Some students, however, are looking to apply their research to fields outside of physics. For example, Junior David Surmick’s research could have effects on the fields of economics and finance. According to Kulp, Surmick is using data-analysis to detect bifurcations, or changes, in financial data. Surmick explained that he is using the closing figures from the Dow Jones market in order to specifically target a place where there could be a drastic change in the market, such as recession.

This project is one of the newest in the department of astronomy and physics and the research is still in the early stages. Surmick explained that right now he is busy entering data and determining how big of a window he needs, since “if it’s too big then our values will be too big, but if it’s too small, then we will get spikes.”

As such, Surmick is unsure in what direction his findings, if any, will take the project, though he noted that the research itself is his favorite part of the project so far, particularly “seeing that my preconceived notions were wrong. Research is fun and independent.”

While Smith’s project lead to publications and internships off campus, Surmick is not sure yet what the future holds, as this is what he called his first experience in the “adult physics world.” Right now, he is looking at undergraduate internships and hopes to go into experimental physics after his time at Lycoming.

Like Surmick, Senior Robert Brown is applying data analysis to fields outside of physics, namely that of computer science. According to Kulp, Brown is interested in image processing on computers, more specifically how a computer is able to recognize the boundaries of an object in a photograph. For his project, Brown is exploring the field of astrophotography by taking pictures of Jupiter with a webcam and using those images to develop his own edge-detection software.