Sunday, July 7, 2019
By:
This past week felt like an intermission to a play or musical. Just like in most shows, it came at a point that was just beyond the technical halfway point; it split our 10-week journey into Act I and Act II with the former representing the first 6 weeks and the latter representing the last 4 weeks. Allow me to elaborate.
The United States of America turned two thousand and nineteen years old this past Thursday! Can you believe that? As a result of the holiday, we were given a nice 4-day weekend, and let me tell you, it was much needed. It was a chance to cool off from work for more than just a regular weekend. I got to catch up on some sleep, do some extra exploring in the city, watch some tv (wooo Stranger Things 3), write some letters, etc. The intermission felt like a vacation!
Before we could get to the intermission though, we had to work a short 3-day work week and I was determined to wrap up a lot in that time span. Knowing I had a nice break right in front of me, I set my sights on tying up many loose ends that I didn’t want to have to worry about by the time I came back on Monday. For the past 2 weeks I have been coding my butt off trying to come up with simple yet sophisticated ways to extract specific data from my computational model of Rayleigh-Lamb waves in a thin plate. More specifically I’ve been trying to come up with a method to extract eigenvalues and eigenvectors from a dispersion curve given an arbitrary wave frequency. If you’re confused and wondering how the heck would anyone figure out how to do that, let alone why we’d want to do that, well those are all perfectly reasonable questions I’ve been trying to answer for the past 5 weeks. You see, I’ve known since the beginning that I would eventually have to do all this, but it was going to take a lot of understanding of the underlying physics and math before I could tackle the problem. That’s why for much of the first few weeks I was reading and asking tons of questions to my advisors, rather than just jumping into something that was so daunting. I had to set the stage and develop the plot before I could conclude Act I and achieve that thunderous applause as the curtain closed.
With that being said, I did it! On Wednesday afternoon, at 4:30, right near the end of the day, my code spit out everything I had been after for so long. Given a user-defined wave frequency in a plate, my code can tell you the corresponding eigenvalues and their 9-dimensional eigenvectors (3 displacement components and 6 stress components). Doesn’t sound too impressive yet? It computes this information for over 15,000 coordinate points inside and along the plate! Remember, this is a wave we are sending into the plate so at different points in the plate, it is going to cause varying displacements and stresses to our material. Lastly, the code does this not just for one user-defined frequency, but for as many as you want. These eigenvalues and eigenvectors change as our frequency changes, so if you want to know all this information for every integer frequency between 0 Hz and 5000 Hz, my code can do that too! To top it all off, it computes all this data and outputs it nicely organized in just over 1 minute. I don’t know about you, but I think that’s pretty good. I’m sure you’re all so proud of me.
Now if that still all sounded confusing and I sort of lost you along the way, that’s okay. The point is that I accomplished everything I wanted to going into our mini-vacation. I, of course, have plenty more to do in the remaining weeks, but this week was a huge breakthrough and a great way to conclude Act I of the SPS internship. Here’s to hoping Act II is just as much of a success! Terry Schuh, signing off.
Terance Schuh