Update

The past week I've been working on an interesting project.  We were looking into why turbines weren't de-rating based on an algorithm created to de-rate a turbine when an inverter temperature reached a certain level so that the GCU would lower its power and the inverter temperature would decrease.   I pulled data from a site where this issue was occuring and looked at the three variables involved in the algorithm: Ambient temperature, Inverter Temperature (there are four of them) and GCU real power in kW (there are two of them).  They wanted me to find the relationship between the three variables to see if they need to change the parameters of the algorithm. 
After compiling and organazing the data I researched how to do a three-variable regression using Excel, where Ambient Temp and GCU power are independent variables and the Inverter Temp is the dependent variable.  I went into this not having any idea how it would turn out and skeptical I could actually pull it off! 
But to my surprise Excel came up with the summary data and I was able to determine the coeffecients and y-intercept for my equation.  A coworker looked at it and was surprised it was linear, so I decided to analyze it a bit more.  (Also the R squared value was in the upper 90's, another surprise!)
I ran the data using the regression function and then created a scatter plot to visually see how the actual Inverter Temperature compared to the Regression Inverter Temperature.  Here were the results for one of the turbines:

Once again I was surprised to see that the regression fit the data so well.  I gave myself a pat on the back that day.  :)
I decided to then use the same regression and use it for other sites to see if I would get similar results.  I did.
My supervisor asked me if I could write up some instructions for how I performed the multi-variable regression so that they could use it again later after I leave. 
Didn't I just mention this in my last blog?  Students may need to not only know how to do a process but also communicate how to do it others.  Case in point!