Guided Question #2

After meeting with my supervisor regarding the experience so far, we decided to get going on some more "mathy" projects.  I'm working in the PHM department where they work closely with RMDC in trying to increase wind turbine availability.  Their role is to take on a more proactive approach and detect issues with turbines that can be dealt with before there is a major fault that needs to shut the turbine down, therefore decreasing availability.  This is where I see a lot of math being used!  Currently I'm working on collecting a set of data from random turbines to help establish a norm.  To be honest, these turbines are so complicated I can't even competely describe what the numbers I'm dealing with relate to, but its something to do with the the high speed cartridges that are in each of the four generators.  There is data that is measuring the temperature vs. time and using a co-variance matrix (5 variable) a t-squared value is derived.  I'm basically pulling those t-squared values from random turbines to find a norm value.  The goal is that if we can determine a norm value than eventually they want to create a program that will recognize when a turbine is approaching a value outside the norm so they can fix the problem sooner.