Thank you John.
Earlier today I was reviewing my 120 MWh solar electrical energy power generation for the lifetime of my microgrid and you have given me the opportunity to share it with you.
I invite you to squint your eyes a little to better see the pattern over the more than 7 years. Perhaps it is best to leave out 2017 as that was the project startup and was only half the solar array. Of course winters, especially January stand out but I see the shape of the yearly curves are not symmetric with spring boldly stepping forward in production, maximizing in July and a more gradual decline into November. The little bottom curve brings these comments out rather clearly.
For those not familiar with the term "nonlinear", you can now see the nonlinearity (low predictability) of this data with an underlying cyclic trend. However, as a rough guess, one can use what happened last year to predict what will happen this year and subsequent years but there is uncertainty associated with this due to the nonlinearity when you look at details.