May 12, 2008

Tokyo (JAPAN)

Running Tests in the Laboratory for My System Dynamics Research

My SOTSURON research project, which is a mandatory element of my year long program at Tokyo Tech, has been moving forward at full tilt over the past month. My topic revolves around dynamic modeling of complex systems, and I am working on a new algorithm to model the behavior of such complex systems without having to run physical tests - thereby saving time and money in actual practice. However, before I can run my algorithm, I have been accumulating response data from a structure so that I can prove the method works.

The picture to the right shows the force input mechanism, whereby I hammer the structure using a hammer with an attached force gauge. The force applied is stored using the machine shown below, as are response functions at designated locations on the structure (via attached accelerometers).

Most of my testing was performed with Okuma-sensei's assistance (sensei means something like teacher in Japanese), the professor for whom I work for, and the head of the System Dynamics Department at the Tokyo Institute of Technology. I am fortunate to have Okuma-sensei as my advisor because he is not as rigid as many Japanese research scientists, and yet he still produces significant results. Given that I visit the lab on a near-daily basis, and have constant contact with Okuma-sensei, there is no under-estimating how grateful I am for his affable nature.

As a side note, this weekend Okuma-sensei will even be featured on an NHK talk show (NHK is Japan's national broadcasting network) along with a famous movie actress, another "expert", and two well-known hosts. The topic has something to do with the dangers posed by mothers bicycling with babies in tow, and my professor will discuss the dangers of impact as well as ways to make this safer.

Anyway, now that I have collected all the necessary data, I can turn to the MATLAB side of affairs (an engineering program used to manipulate data), which is where the real "fun" resides. Over the next couple of weeks I will be analyzing my data and comparing the actual results with my simulated results (using Finite Element Analysis software and data extrapolation).