Closer look at daily theoretical output

March 1, was a record day.
27.7 kwh produced.
All sun, no clouds.

I took a deeper dive into the PVWatt model and tried to extract the daily output for that day of the year.

The expected array power output was:
8,446 Wh (east)
15,605 Wh ( south)
or 24.1 kWh

PVWatts has the day estimated at 87% of actual peak output. Whats the difference?

I dont know, yet, how PVWatts gets to their annual output vs. their hourly output. Maybe the hourly csv file is approximated from a continuous solar curve?

Well, the PVWatts data file is broken down by  the hour. To get my daily PVwatts number, I summed up the power output for each hour to get a daily Wh estimate. Doing this assumes that the power output stays constant for that hour...which it does not. 

This causes an approximation error. Think Riemann sums from Calculus 1. Before you could integrate, you approximated the area under a curve using a sum of rectangles. Like below:

This is basically what my summed up PVWatts estimate is doing.  So if the rectangles are too few, the estimate is not as accurate as having more rectangles....as if we had PV output at the 15minute or 5 minute intervals.

Or, the difference is the weather correction / solar irradiance factor that is built into their model.... which means 13% positive daily variance can be possible on an ideal sun day vs. the average solar irradiance database.





Comments

Post a Comment

Popular posts from this blog

Solar panel journey.....on to Tesla

Too busy to shop around? Energysage is a central market place that can pull many quotes from nearby installers and it provides lots of information about going solar.   My quotes from Energysage (Front Range, Colorado) #1 $ 3.10/  watt #2 $3.05/ watt #3 $3.19/watt #4 $3.15/ watt (all prices before tax credit) Several of the follow up calls were aggressive, pushy....salesman.   Some were laid back. I knew what DIY costs were amd Tesla. So these were way too expensive...payback too long. Most played dumb when asked about Tesla pricing. Tesla  was $1.49/watt (after tax credit).  Thats all inclusive (hardware, installation, permitting). How can you not know your competition pricing? One salesman acknowledged and played the local, customer service angle.  Tesla does have quite a bit more negative reviews out there mixed in with some postives.....and poor customer service is a big complaint.... But the price difference is too compelling.... I placed an order with Tesla Online in lat
Read more

More on the Delta Inverter

Image
To log into the M-4-TL or M-8-TL inverter,  Open the app while standing very close to the inverter. I stand inside by the window nearest it, maybe three feet away and it finds it automatically. It will prompt you for a code. Enter the 4 digit code to get you logged in. Once you in, this what you see: -3.2 kWh produced today so far -2.776 kW current power output -1732.5 kWh produced since system install (all data in the Tesla app) After that though, its all new stuff. We get the system level AC voltage and AC  current. 247 V and 11.2 {Power=I  * V} P= 11.2 A x 247 V=2.776 kW The string level detail is next. PV1 and PV2 are the DC voltage and current produced by each string. In this case, the real time power from each string is: P1= 7.4 A x 229 V= 1694.6 W P2= 4.4 A x 244 V=  1073.6 W Total P=  2768.2 w or 2.768 kW This data was about 9am. So PV1 must be the Eastern array. Here is the same data near mid day when we are at peak output.  You can see the inveter is putting out a
Read more

2022 Summary

Image
Went through 2022 without any updates. But I have another year of data, so we can do year by year comparison to see if anything changed significantly. According to NREL (recall that is the government agency that made the PVWatts software), the median solar panels lose 0.5% efficiency per year. Hmm, that doesn’t sound like too much. Below is my 2021 total output of 7,132 kWh.  Taking 0.5% of that, is 36 kWh, which is just over a day of peak production in the summer or 1.5 days in the winter. Not much. A rainy day or two.   No way we could confirm efficiency loss in year 2. Weather is a much bigger uncertainty. But we have?  data now, so let’s crunch the numbers and see if we showed any loss in production. Below is is 2022 monthly and annual production.  Not much to crunch, 2022 output was higher than 2021 by 1.8%. So much for degradation analysis by looking at annual production. Going back to our original PVWatts output prediction, we can see if year 2 was any more or less agreeable tha
Read more