Dissertation Simulations

As of this morning, I have begun what I hope to be the official simulations for the work on my dissertation. I have 33 separate cases to study and this morning I queued them all up to run on my school computer over the weekend.  For exactly this reason, the machine I get to use has eight cores and 48GB of memory so it should be able to handle the workload; nevertheless I don't expect it will be done for several days.

I've been working since summer 2013 on the development of the computer code that is being run in these simulations.  I started with a model of the electrical grid in the western US and have been working on a way to add energy storage (a battery) to a specific location in the grid.  It was important for my research that the simulated behavior of the battery should wear out over time (like batteries do) so a bit of my work was getting that implemented. A week or two ago I started putting the individual pieces together and began testing the system as a whole, making sure everything was behaving as expected.

When the final development test results came back this week with the system appearing to function I was faced with a startling realization: I was ready to run the simulations with the specific conditions I wanted to study.  Though the difference between these "real" simulations and the test ones I had been doing involved only changing a few lines of code, the psychological hurdle was immense.  I'm actually doing it; I'm almost there.

These simulations will take a while to complete.  They have to figure out, for a given set of demand for electrical energy in the western US, the output level for each generator in the western US that will provide this energy for the lowest cost.  And it has to do this for every hour of the year (8760 of them).  In half the cases I'm simulating, it has to do this for five years in a row.  

I hope everything will be done when I come in on Tuesday.  When I left, all eight cores on my computer were maxed out and I expect they will stay that way for a number of days.

The completion of the simulations is a significant milestone but I still have plenty of work even after they are done. Analyze data, make graphs and tables, and write the several hundred page document that PhDs traditionally have to write. I won't be done this semester but I'll be very close.