Data Catch-up!

Several months have gone by without me posting my data - sorry about that. Actually the last posting was 1 September - shame on me. Sorry again!
Our system has now been running for almost 1.5 years and we can now compare data between two months of different years to begin to see how variable the solar energy received is. Since our system was commissioned August 2010, we can now compare September to December (and in a few days January). We know that weather and solar energy will be different between the years, but I didn't know by how much. Also, the effects of weather will affect the solar generation - I'm primarily referring of snow and its persistence. So, on to the data...

Daily Power Generation

The data is downloaded from my Fronius datalogger and creates two M$ excel files: a daily summary file and a instantaneous (15 min average) file. The files report the kWhr, AC and DC voltage, and error messages (never had any error messages. I'll have to do a posting on the AC power but I don't know enough about it to say anything intellegent about it (yet). I add the daily data to a spreadsheet file where I do calculations and the figures.
The annual trend in solar radiation is clearly visible now that we have more than a full year cycle in Figure 1. There is considerable daily variability - as much as 30 kWhr for the system or 6 kWhr per kW per day. The highest variability is during the spring months of March and April. In August and September the variability ranges about 20-25 kWhr or 5 kWhr per kW per day. One reason for the larger daily variability in the spring months is because the overall potential output is highest (35 kWhr) and there is still the good chance that snow covers the panels or there is a very dark overcast day. The days are also still a little short therefore the total accumulated solar radiation will be less than the longer summer days. This is why we see the lowest generation days in mid-summer still seldom being less than 10 kWhr. This means the variability per kW to be closer to 5 kWhr/kW/day and not 6 kWhr/kW/day.
December was an interesting month to compare between the two years. December 2010 received a quite a few snow days and this resulted in many days with very little power generation (see all the points clustered around the x-axis during the second week in Dec with virtually no days over 10-15 kWhr until mid-January. I've (begrudgingly) reported this in an early posting. Even though we love the snow and winter, the good news for December 2011 was that there was virtually no snow fall this year and there were many days where the system generated over 10 kWhr. Several good days over the month makes a big impact on monthly revenue.

Figure 1. Daily Power Generation, 4.9 kW DC PV System