a year in the sun

We’ve had our solar PV system for (nearly) a year now, so here’s a review of 2014.  We generated just shy of 9000 kWh: that’s 9 MWh!

daily power generation by month

violin plot of daily power generation, showing notched box plots of mean (dot), median, quartiles, and outliers, overlaid on a density plot

You can clearly see the effect of the year: much more in summer than winter.  Interestingly, December, containing the winter solstice, was better than November, which, as usual, was a miserable month; December has some lovely sunny days.

No sun – no moon!
No morn – no noon –
No dawn – no dusk – no proper time of day.
No warmth, no cheerfulness, no healthful ease,
No comfortable feel in any member –
No shade, no shine, no butterflies, no bees,
No fruits, no flowers, no leaves, no birds!
November!
     — Thomas Hood, 1844

We can look at the energy generation across each day for the whole year, and can clearly see the changing sunrise and sunset times, and the darkness near the end of November.

Each pixel represents the energy generation at the sample point, one every 5 minutes. Time is in GMT throughout. The colour indicates the energy generation in the relevant interval: darker colours indicate more energy.

It’s even easier to see the effect of the seasons looking at the generation through the day.  The curve gets higher and much broader.

Energy generation over the day, averaged per month. Data is gathered at 5 minute intervals. The horizontal time axis runs from 3:00am to 9:00pm GMT. The vertical axis runs from zero to 8kW. The orange regions indicate the minimum, lower quartile, median, upper quartile, and maximum generation at that time, over the month. The line indicates the monthly mean, The number in the top right is the monthly mean generation in kWh.

Of that power generated, we use some, and export the rest to the grid.  In March we got an extra device to measure this, too.  (March was anomalous in that our hot water boiler was out of action for about a week, so we were using more electricity than usual.)

Energy usage over the day, averaged per month. Data is gathered at half minute intervals from the Wattson meter (which we didn't get until mid March. The horizontal time axis runs from midnight to midnight GMT/BST. The vertical axis runs from -8kW to 8kW. The region above the axis represents our usage: orange is generated usage, red is imported from the grid. The green region below the line is surplus generation exported to the grid.

In the plot above, when the green area below the line is greater than the red area, we exported more than we imported, so are nett generators (month name shown in green). There are days where we import more than we export, taking more from the grid than giving back: on “orange” months we still generate more than we import, but use enough of it that we are not nett exporters; on “red” months we import more than we generate, but even so, may still export a little during the day.

The situation is actually even greener than this implies: some of the orange usage of generated power is being used to heat our water, thereby saving gas consumption, too.  That is why there is such a large oragnge lump in December: nearly all excess power is heating the water for the central heating.  In the summer, the only water heating is for hot water, as the central heating is off.

Next year we’ll have data to compare across years, not just months.  I’ll need to write more Python scripts for that.  Great fun!