2015 Weather Statistics

I've recently finished my quick and dirty sun analysis for 2015 based on the Met Office data. Here I take the data for the Heathrow weather station and apply a few quick tricks to understand the amount of sun we've had in the years relative to previous years available on file. In this case there are now 59 years of data available.

I get the following results:

The Max, Min, Mean and Stdev stats are based on the last 59 years of data by month and are the numbers of hours of sun reported.

First, the good news. The year started well with fairly average Jan and Feb - Jan was actually quite good, but offset by a slower Feb. March and April were excellent, with April the 7th sunniest in the 59 year sample. April, it turned out was to be the sunniest of the year and my generation statistics back that up.

And then the averages started to bite. May, June and July were ok - but very average. I hope to have at least one monster month each summer when the sun comes out to play and stays there. In 2015 this did not happen, I suppose we were paying for April.

And then, the rot really set in. August was unseasonably dull the 55th sunniest out of 59 - pretty close to the bottom. Then September, which can often be a great month was dull and 57th out of 59. October was closer to average but without making a big contribution.

And then November. What can I say? The least sunny November in 59 years with only 41% of the average for the month, my lowest ever solar panel performance on record (in terms of % against model) and a grey 30-day period that never seemed to do anything. At least it kept unseasonably warm which kept the heating bills down.

Thankfully December was a little better and actually produced more kWh than November. But still there were too many dull days. 

Overall the year was not that sunny (45th out of 59) and 7% down on the average. Thankfully this was enough to hit the annual target. Although as expected not as much as the previous years.

Hopefully 2016 will be a bit better.

2014 in a nutshell

So, 2014 came and went -  a great year in theory for the PV system, but in reality it lacked any real outstanding performance.

On paper it was a very sunny year, but this was not the case in all months and in it failed to appear in the key months that mattered (April, May, August  and September for example). So whilst it was a good year, it wasn't a great year.

However there were a few milestones - November saw 10mWh appear on the meter. So just under three years since the system went in and the numbers look good. So, as a rule of thumb and subject to decent weather it looks like the system will create nearly one kWh of power per year for each watt of solar panels I have. So that's not a bad figure to aim for.

Money wise, the FIT payments keep coming in - £1600 a year is not to be sniffed at - and it looks like I'm using 70% of the power being produced. I've worked the latter out based on 2 methods:

1) the reduction in my electric bills

2) detailed analysis of the OWL data based on previous usage levels compared with the minute by minute data over a year (i.e. what amount of energy would be used vs exported each minute assuming a 1000w continual household consumption).

Both converged to the same number, so that's good enough for me.

Outlook for 2015... well hopefully more sun, more free power and I suspect a year end total close to 14mWh.

September 2013

September was dull.

It felt very grey from the start of the month and so it continued. I had to scrape back every last kWh to hit the target this month (this meant checking the generation meter under the stairs and finding my Owl estimates were a couple of units light).

101% of target achieved which was not bad given the we only had 81% of the average hours of sun during the month. September 2013 was the 47th sunniest in the last 57 years - you get the idea, it felt like Autumn all the way through and the PV system reflects this. Still, to continue generating under such conditions shouldn't be underestimated.

Unlike September 2012, I did not complete the year's generation target in September, falling about 20 kWh short by the end of the month. This will easily go during October, but the stats don't lie - I'm down on the previous year and the weather feels like that too. Maybe 2014 will give us a treat instead.

August 2013

August was good without being great - good kWh figures most days but what it lacked was the solid weeks of good weather that really help the numbers stack up when you hit 120+ kWh in seven days.

Still no complaints, 425 kWh produced, in what was a distinctly average month - we had 104% of normal hours of sun during August, with the months being the 23rd most sunny since 1957. Very average, but lots of watts anyway.

July 2013

Ok, I admit it, I'm late with the updates. It's been that type of summer - too much to do and not enough discipline to sit down and write some stuff up. 

But now, with a few hours to think and a load of data in Excel it's time to fill some gaps in.

July, well what can I say? Extraordinarily good weather, a severe drought that turned the lawn brown and lots of sun, although quite a bit of hazy weather that reduced the ultimate output of the PV system. According to the weather stats July 2013 was the 6th sunniest July since 1957 with 134% of usual hours of sunlight. The result was my best month ever with 499 kWh produced for 136% of target, that's over 130 kWh ahead. I could take that every month.

June

Well... June was a bit odd. It started well (as seems to be the pattern lately) but then the clouds descended and stayed put. 

At one point it looked like I'd just scrape home over the target, and then it looked like being a struggle. And then, a slight perk up towards the end of the month and all was well. 

So, still ahead of the cumulative target, but far behind where I was at the same time last year. But maybe the predictions for great weather in July will come true. Beating the target by 100kWh in a single month would be superb - but I really cant see that happening. Fingers crossed.

Mega time

So the latest from the meter under the stairs. Five megawatt hours of production have now been logged since the system was installed.

That's an awful lot of (energy efficient) lightbulbs run for free. With a good summer (unlikely, the way things are going) we could well exit the year close to 8 mWh. Well, maybe 7 (fingers crossed).

May 2013

So that was May.

It started nicely enough with some really nice weather (just in time for the bank holiday) and finished in a similar manner for the final bank holiday, but there was some real grey days in between. The weather this year has been odd to say the least - too cold, too wet and too grey - you can see this in the garden (everything is a few weeks later than usual), and in nature (fishing is very slow this year) and it even seems to be effecting some people (a sort of SAD if you like).

Output wise, too many mid-single figure days diluted down the good days and led to a good performance (112% of predicted output) but one that should have been better. Still, I supposed I shouldn't complain - another month done a dusted and another target beaten, however output was lower than April 2013, which indicates just how mixed the weather has been.

Let us hope that the good start to early June carries on through the rest of the month.

Spring, finally.

It took it's sweet time coming, but spring has finally arrived in the UK.

It's still not quite right (the garden is well behind where it should be) but suddenly getting up in the morning doesn't seem so difficult - blue sky rather than dull grey cloud helps somewhat in this regard.

So the highlights of the month:

(1) passing the target with ease - 407 kWh generated vs a prediction of 311 kWh

(2) best ever single day's production on 20th April - 23.56 kWh, beating the previous record of 26th May 2012 of 23.24 kWh. On it's own this was not a memorable day, but clear skies and a good elevation angle might have squeezed out some additional watts. Who knows? Maybe we'll get a few good summer days this year (without the haze of 2012) and even better results might be observed

(3) finally seeing some separation in cumulative generation for the year

Here's hoping the nice weather continues for a little while.

Worst March Ever

According to today's papers March (and Easter for that matter) have been the worst ever. I will wait until I see the stats from the Met office to be sure, but I can well believe it.

Cold weather, clear nights, threats of snow, long dark grey days all featured too often in the month. Despite a few good days, the PV system just couldn't do it's thing often enough and the month finished on 213 kWh against a predicted value of 225 kWh, giving 95% of prediction. This is the second month this year that I've missed the target and the second time ever - I just hope this isn't the pattern for the year. (If I'm honest I suspect I've missed by a little more than that, I think the OWL CM160 is over-reporting, but I'm ignoring that for now.)

The killer issue is the lack of average days in the month - there were far too many days when the weather never got going and 3 kWh or less was the output. It's hard to compensate for that when a good day is 15 or so and you need to average 7.

It's a big contrast to last year when the weather suddenly turned on for two weeks and all the March targets were well and truly smashed, such a thing was extraordinary at the time, but it'd be nice to have a good week for a change.

Looking ahead the weather looks to be brighter for the next few weeks. Although this will keep things cool, the PC should at least get a good work out. Here's hoping.

It's been a while

So it's been a little while since my last post - I've kept on top of the tracking, but not the blogging bit. So hopefully a chance to put a few things right.

So the highlights:

December 2012 was amazing, the target completed in just 12 days, thanks to some unusually sunny weather. At one point it looked possible to complete about twice the quota, but the weather was not so consistent come Christmas, but you cannot really complain.

Lowlights:

Err... that would be January. Awful weather, deep grey skies, snow etc and too many days when virtually nothing happened. All-in-all, this is the first month that I've not beaten the forecast - that's how bad the weather was.

Eagle eyed readers will spot a few negative days in there - this is not quite accurate. There was some nominal production but once I take into consideration the background noise levels on the Owl CM160 in order to track the meter under the stairs, this is roughly what drops out. Either way the overall trend and values are close enough. So that works for me.

February on the other hand, looks a damn sight better, with ten days to go I'm nearly there. Hurrah.

As a change for this year, I've adjusted the generation targets to be slightly phased. This reflects the day length/suns angle better and back weights the months in the first half of the year, and front weights the second half. It's just a minor adjustment, but it pleases me. I may explain how I've done this another time - it's not perfect, but its close enough for me.

Sun vs PV

Since the system went in, I've been watching the output of the PV system (no surprise there then), but I've been wondering how closely this follows the weather stats. So, now, with a good few observations to look at, it was time to have a proper look.

Courtesy of statistics published on the met office website, I've downloaded the observational data from London Heathrow, which is not too far from me, for the analysis. This contains various data, by month for the last 50 years or so, most importantly this includes the number of hours of sun per month.

The first chart, shows the hours of sun reported, based upon the monthly records going back to 1957, with the mean, max and min shown as the lines. I've then superimposed the 2012 actual figures (diamond markers) on the chart.

The overall shape of this chart is no real surprise and looks familiar to the shape of the PV predictions, however the variations are quite interesting. Indeed in a good month it is possible to have 50% more sun than average, and in a bad month 50% less.

2012, as anyone who has been watching the weather will tell you, has been another strange year. March was unusually warm and sunny for the last few weeks, after which the weather took a distinct downturn, and it was only August and September that provided any semblance of Summer. June in particular is worthy of note - the least sunny June for over 50 years. No wonder my chilli plants didn't do well in 2012. 

Which brings me on to the real point of the analysis - does the PV system performance vary with the relative hours of sunshine? Intuitively this is obvious, but it's always good to run the analysis properly. So let's look at the results (chart below). 

 

Here we see each month (plotted as a diamond) comparing the monthly sun hours normalized against the seasonal norm, plotted against the PV system output (indexed against the PV output prediction), with a simple linear fit applied. Visually this appears to be a good correlation (r=0.957), although there is a little noise. This is not surprising - not all of the sun would hit the panels at a good angle, (if it's at the wrong time of the day) and not all of the sun would be at the same strength, but overall, this would appear to be a good fit. More sun = more PV ouput, is hardly surprising but it does seem to suggest a linear relationship. 

A quick t-test on the above data provides a value of p=0.00209 which suggests a fairly high degree of significance, it'll be interesting to revisit this in a year or so's time to see how the correlation and p-values develop, especially as the system starts to degrade.

One important observation on the above is there is a natural bias in the system performance - whenever the sun hours were on the seasonal norm (i.e. 100% on the X-axis), the PV system appears to be producing 120% of predicted output. This is good news from a generation and FIT payment point of view and back up the anecdotal views of the PV system installer. They stated that PV systems seem to out-perform the prediction, so hopefully as the system degrades, I will still see a good rate of output and system return. One lives in hope.

 

October

Well that's October out of the way. The month started really well, with plenty of clear days and strong sunlight. After looking like I would pass the forecasted output with ease, the weather suddenly changed and ten days of poor weather followed. Heavy cloud, rain, shorter days meant that output of 2-3 kWh per day were common and that really started to erode the output.

Two good days towards the end of the month finished things out nicely and that as they say was that.

Done and done

A momentous day has past - two major milestones.

Firstly, the September total has been met already - a series of good weather days at the start of the month, followed by good mixed weather afterwards has made the daily target of 8kWh per day very doable. Job done and still 12 days to make a profit.

Secondly (and probably the most important) that's also the point at which the total for 2012 has been reached. After double-checking the meter under the stairs, I can confirm I'm comfortably past the 2952 kWh annual target with over 3 months of the year to go. Admittedly these will not be the most productive months of the year, but it's all pure profit now, which is a nice feeling to have. It helps that so far this year I've beaten the target every single month, which given the poor weather is no mean feat, so I wonder what the output would be in an unusually good year. Maybe 2013 will show.

On an aside, I've now had to change my home laptop, which has been an experience in itself. So I've reinstalled the OWL software, but need to transfer over the database at some point. Hopefully no big deal. However, the interactive charts on the blog have been interesting - works in Firefox but not Explorer. So now I've no idea what's going on. I shall have to ask the nice man at work for his thoughts.

Watts going down

Recently I received an updated set of power consumption figures from my electricity supplier and I was interested to see by how much the solar PV system had reduced my imported electricity levels. Based on the figures supplied (which are far from ideal, being spaced at 80 day intervals) I got the chart below.

It's not a perfect way to work, but it does go some way to show the impact of the system. At the moment we are relatively high power users - part of this is the garden pond which can use 500 watts of continuous power in the summer (less in winter when some of the devices can be switched off) plus the air-conditioner in the summer can get the meter whizzing round too. (Although the latter has seen much less usage this year).

Based on a few rough calculations this would seem that I'm using about 7 kWh less of power a day since the panels went in - which seems a little high since this would suggest only 20% of power was being exported. I suspect we're being a little more economical than before and running the dishwasher and laundry during peak sunny times probably helps make the best of the system, hence not all of the drop is from the panels, some of it may just be our changing behavior.

7 kWh a day are equivalent to about £25 a month off the electricity bill, which is not to be sniffed out - the equivalent of £300 a year of additional return on the system (in addition to FIT and export payments). Not bad at all.

August

So that's August done and dusted. 

Weather-wise this was a strangely good month - without ever really getting going, it kept fine and bright and produced many good days that meant it felt like exceeding the prediction was a real possibility. By finishing the month 20% ahead of prediction on 411 kWh means it was the best month yet - here's to an Indian summer and a chance on completing the annual target by the end of September. Who knows, it might happen.

A few good days

A few good days of weather and we're back on track. In fact the last two days have been about as good as it gets with 20kWh a day flowing from the panels under clear blue skies. This has improved the generation stats and closed the deficit against the predicted generation levels - hurray. 

Although the weather has been near-perfect, the peak output is not as good as I'd hoped - more in the upper 2 kWhs rather than in the lower 3's. I suspect this might be the slightly hazy nature of the sun, but I cant really complain after all this rain we've had. It's great to see some output.

Overall, the meter under the stairs says 2237kWh to date, which means I've passed the £1000 point since the system was first commissioned. It's taken a little time, but I'm still ahead and it's nice to see the investment starting to pay back. I look forward to submitting my next meter reading in mid-August.

Weather, Owls and Spaniards

So they tell me that the weather is about to get better. Apparently, the run of poor weather we’ve been seeing of late is courtesy of the jet stream being too far south and hence the mixed weather to the north is fairly and squarely ensconced over the UK. The smart people in the met offices of this world now tell us that this is about to move, by the weekend the jet stream will move northwards and the lower part, at least, of the UK will start to see the benefit of improved weather. (This could be a major boon in time for the Olympics which might otherwise have to be held underwater).

You’ll have to pardon my scepticism at this point. It’s not that I don’t believe the science behind weather forecasting, it’s that based on recent experience, and the chaotic nature of weather (that’s chaotic in the mathematical sense) I think I’ll believe it when I see it. It would be nice to think we’ll get a summer at last – which would be handily in time for the school holidays, my solar panels and the contents of my greenhouse. But I hope to be amazed shortly.

The solar PV system is, however, slowly doing its thing and generating power, albeit somewhat slower than you might expect for this time of year. We’ve made it through the equinox, and are on the slow run down to winter (makes me sound like a right pessimist) but with the balance of July and the whole of August to come, one hopes that we might get to see something special at this peak weather time of year.

July so far has been pretty mixed, the odd nicer day mixed in with lots of cloud and many showers. I’ve had a couple of good days (mid teen production) but no wall-to-wall days when the kWh’s really started to flood in. It’s a little disappointing given the day lengths available, but on the other hand I should be pleased that even with these limited days we’re fairly close to the generation prediction. I know I’m running about 10% behind where I should to be, and a lot more behind where I’d like to be, but given the conditions, I still find it amazing that I’m able to generate so much, so easily.

As an update the Owl seems to be behaving itself again. Two problems connecting to the transmitter in three days was a little worrying, but all seems to be good. Talking of the Owl, I see on their website that there is a new gismo (or should that be toy) available that allows automatic transmission of readings to the web for access via iphone apps and the like. I haven’t seen any reviews of this, but I might be tempted to play with this at some point, since it appears that it may be possible to have this as a retro-fit using the existing transmitter. I think I will hold fire until I see some independent reviews.  

In order to share data more effectively, I’ve added some additional ‘pages’ to the blog, these are on an extracted page per month based upon generation stats courtesy of google docs. This was fairly simple once I’d worked out what to do, but it does lack a professional look to it (no colour scheme and no links). However, it is a start, since it appears most browsers are unable to see the java charts on the top and right hand side of the blog. (I’ve still no idea why this happens).

On a completely different tack, I’ve just returned from a couple of days in Barcelona, where the sun shone, the humidity rose and it was all-round too damn hot for my liking.  However, I was surprised not to see a single solar panel all the time I was there. Given the excellent natural climate and strong light I would have expected to see something. From my hotel room window I could look out over a fairly large swathe of residential housing (a good few hundred properties) and there was nothing to behold. Maybe the subsidy is not there, or the planning is too draconian – who knows? I would spend some time investigating if I was really that bothered, but as a point of interest, I would expect some panels to be blasting away given the suitability of the conditions.

Owl on the blink

So for the second time in three days the OWL CM160 unit has stopped recording any power. Thankfully this is not the fault of the system, more an issue with the OWL. 

I've changed the battery in the sender unit, and reestablished the connection, only for it to cut out again a few days later. I shall have to have a go under the stairs again to try and work our what is happening - but without waking everybody up.

I've also noted another potential issue with the unit - it tends to stop recording internally for short periods. It still displays a reading, but sometimes when you come back to it, it does not seem to have advanced as far as one would expect. So on a bright day when it's showing in excess of 2kW you can come back a couple of hours later to kind the cumulative count has not moved as far along as you might think. I thought it was just me, but a recent check of the meter suggests that I'm 40kWh behind what the meter under the stairs says, and that's accumulated over 6-8 weeks since the last time I aligned the readings by hand. I already have an error correction built in to the readings spreadsheet, so I don't think that that's the cause, but I suppose I shouldn't complain too much if the difference is in my favour.

I can feel a replacement unit on the way.

What do you see?

Whilst you're reading this there should be a table at the top of the page and a set of charts/graphs down the right hand side. if you cant see them then I'm sorry. I'd love to know the reason - I suspect it's a security setting of some type, but I'm not sure how you fix it. 

Answers on a postcard...