Renewable Energy: How much PV has the UK installed to date?

61 to 74 of 74

- CommentAuthorGavin_A
- CommentTimeJul 27th 2013
Posted By: SteamyTeaThe point I am making is that, and you can see it on the charts, for any given week the fraction of 4-10 kW and 10-50 kW when compared to the 0-4kW is consistent. So if domestic installs go up, then the other 2 do, if domestic installs go down, the others go down.
Does that not strike you as odd?
Why should these very different markets be linked in this way?
Is there any reason why sales/installs in the domestic market should have any influence on the smaller commercial market.
I know that some domestic installers can fit a 4 kW system in half a day, but I don't know of any 50 kW systems fitted as quick.
Just seems a bit odd that all systems installed and all installed capacity tracks each other so well.

I'd dispute that they track each other that well - as the graph below shows there's significant variability in the weekly percentage of 10-50kWp vs 0-4kWp installations, as well as a long term trend to increasing proportions of 10-50kWp systems.

But I don't think it's that strange that they do track each other in terms of rises and falls at the same time, given that they're all vulnerable to the same weather conditions, and changes to the FIT scheme in that period that have been by far the biggest driver in the main peaks and troughs in installations.

10-50kWp systems do see a much bigger peak proportionally in the last week before the FIT cuts though, probably due to the longer timescales involved in planning, permissions etc
- comparison-10-50kWp-vs-0-4kWp-solar-installs.JPG
- CommentAuthorSteamyTea
- CommentTimeJul 27th 2013 edited
If you run a Student T-Test on it you find that, even though the raw numbers are different, that the variance, skew and, for want of a better term, the general trends, are statistically identical.
That is what I find so odd.

Posted By: Gavin_Agiven that they're all vulnerable to the same weather conditions
I figures are installed capacity of number of installations, don;t think the weather conditions affect it as it is not generation. That woudl probably show a very different picture.
- CommentAuthorGavin_A
- CommentTimeJul 28th 2013
try installing in the snow or pissing down rain on a slippy roof, takes at least twice as long, therefore the rate drops in those periods and vice versa in sunny weather. Same with bank holidays as well.

The figures are for the registration date, not the installation date, which is important to consider.

As for the student T test, I really can't work out what you've done that test on specifically, but all the elements I've looked at seem to be about as variable as I'd expect them to be. The most likely explanation I can see is that you've done it wrong, or are analysing something specifically that's not really surprising that it's very similar then generalising from that... for example

"The point I am making is that, and you can see it on the charts, for any given week the fraction of 4-10 kW and 10-50 kW when compared to the 0-4kW is consistent."
well no it's not, the graph above proves it's not consistent, it varies from 8% to 90% for 10-50kWp vs 0-4kWp install capacity, or 1-8% on install numbers.

"7So if domestic installs go up, then the other 2 do, if domestic installs go down, the others go down."
While that's generally true, as would be expected due to the factors stated, there are enough weeks where it's not true to make your hypothesis that there's something odd about the figures because they're too similar seem pretty silly.

too much time on your hands anyway I reckon tbh.
- CommentAuthorSeret
- CommentTimeJul 28th 2013
Posted By: SteamyTeaNot such a silly way to test it. Shall have a look, think it is a while off for the UK.

Wouldn't work, you'd lose the signal in the massive fluctuations in demand you get at solar eclipses. The eclipse in '99 caused the largest swing in demand the grid's ever seen.
- CommentAuthorSteamyTea
- CommentTimeJul 28th 2013 edited
Posted By: Gavin_Ait varies from 8% to 90% for 10-50kWp vs 0-4kWp install capacity, or 1-8% on install numbers.

That is the point I am making.
Take Week 6 on your Chart, both peak that week, as they do Week 10, 32, 41, 50, 60, 75. Troughs align nicely too. And those are just the ones that are easy to spot.
That is what worried me about the data.
This is what the Student T-Test is for. It looks at variations and if there is any difference between between data sets.
If we take your figures for variation, the 8 to 90% and the 1 to 8% for installations, once transformed so that they are on the same scale they are pretty similar, not identical, but similar.
Add that to the the variations peaks and troughs happening at the same times, just seems strange to me.
I would expect that larger installations have grown at lesser rate than domestic as the market has been concentration on the domestic side more (was basically set up for smaller systems), but the slope of the trend line is not the point I am making.
The point I am making is that I don't think these two markets are dependant on each other and that is what the DECC data seems to be showing.

I do understand that there are times when nothing much will happen, extreme weather and holidays, and the commissioning date may be important, but I struggle to believe it affects both markets evenly.

Posted By: SeretWouldn't work, you'd lose the signal in the massive fluctuations in demand you get at solar eclipses. The eclipse in '99 caused the largest swing in demand the grid's ever seen.
Yes the signal to noise ratio may be hard to spot, but we have good data on what would happen if there was no eclipse and some data when we have an eclipse, just be a fun experiment. I am sure National Grid have a plan in place already.
- CommentAuthorGavin_A
- CommentTimeJul 29th 2013
you've reminded me why I gave up posting on here tbh steamytea.

"That is the point I am making."
no it's not, it's the polar opposite of the point you're making.

"Take Week 6 on your Chart, both peak that week, as they do Week 10, 32, 41, 50, 60, 75. Troughs align nicely too. And those are just the ones that are easy to spot.
That is what worried me about the data."
seriously?

Of course the 2 lines peaks and troughs correlate well, as they're 2 sides of the same coin - both comparing the weekly installation rates of 10-50kWp systems vs 0-4kWp systems.

One graph is related to the total number of systems installed, the other to the total rated capacity of the systems installed each week.

So obviously the peaks and troughs correlate, but there is also a lot of variation within that correlation.

If what you were saying was right, then both graphs would show a virtually flat line

"If we take your figures for variation, the 8 to 90% and the 1 to 8% for installations, once transformed so that they are on the same scale they are pretty similar, not identical, but similar."
The only way 8% is similar to 90% is that they're both below 100%, so yes it's true that the capacity of 10-50kWp installed was always below the capacity of 0-4kWp installed, that's about it.

"The point I am making is that I don't think these two markets are dependant on each other and that is what the DECC data seems to be showing."
They're not dependent upon each other as such, but are very significant factors that apply to both.

whatever tree you're barking up here it's the wrong one.
- CommentAuthorSteamyTea
- CommentTimeJul 29th 2013 edited
Posted By: Gavin_AOne graph is related to the total number of systems installed, the other to the total rated capacity of the systems installed each week.
I did not spot that, my error.
But my charts don't mix units, they are either installation numbers or installed capacity.
An analogy would be comparing car sales with lorry sales or annual car mile with lorry car mile.
I would not (though I admit I missed it on your chart) compare number of installation with capacity, that would be like comparing car sales with annual lorry miles (possibly not the greatest analogy, count have been lorry price).
If you go back to my charts (starts middle of page 2), you will see that I compare like units with like units (all you have done is got half way to working out the average system size (if you want this then plot installations against installed capacity to get the mean MW/Installation).

Do you think that questioning data is a good or a bad thing?
- CommentAuthorSteamyTea
- CommentTimeJul 30th 2013 edited
I had a quick word with DECC and the person I spoke to there thought it odd too.
Not had a chance to investigate more yet though.

They get the data from Gemserv
- CommentAuthorted
- CommentTimeJul 30th 2013
I still don't understand why you are surprised. From your first chart the correlation I see is that the peaks align with major tariff changes and would, as such, be expected. The number of 4-10kW and 10-50kW systems is so small (5% of all 0-50kW installs) that all you are graphing, in the non=peak periods, is 'noise' compared to the 95% of 0-4kW systems.

How about adding vertical markers to that chart at the date of each tariff change?

3rd March 2012 - 21.0p - 16.8p - 15.2p
1st August 2012 - 16.0p - 14.5p - 13.5p
1st November 2012 - 15.44p - 13.99p - 13.03p
1st February 2013 - no change
1st May 2013 - no change
1st July 2013 - 14.9p - 13.5p - 12.57p

(tariffs are for 0-4, 4-10 and 10-50 at each date)

and also show Christmas 2012 (a week when almost no PV has historically been installed).
- CommentAuthorted
- CommentTimeJul 30th 2013
So remove those completely explicable 3 peaks and 1 trough, account for the fact that the 0-4kW systems (blue line) is 95% of all 0-50kW installs and that the green and brown lines only represent 2.5% each and what 'strangeness' are you left with?
- CommentAuthorSteamyTea
- CommentTimeJul 30th 2013
I may try that when I have stopped doing the analysis on the latest PV system I am monitoring. A few more hours.
- CommentAuthorjms452
- CommentTimeSep 26th 2013 edited
As of the end of August the total installed PV was 2469MW (they have even produced a new month by month total installed pv table - see 6.4 of below link).

https://www.gov.uk/government/publications/renewables-section-6-energy-trends

2013 Q2 was also quite impressive for renewables in general exceeding 15% of electricity demand (on average).

Corrected - thanks Ted

light blue - PV
purple - offshore wind
light green - onshore wind
red - hydro
dark blue - bioenergy
- 2013Q2 renewables.jpg
- CommentAuthorted
- CommentTimeSep 26th 2013
That figure was to the end of August. We haven't got to October yet
- CommentAuthorjms452
- CommentTimeSep 26th 2013
Cheers Ted - this is getting to be a bit of habit.

I should learn that I really don't have the brain for accurate multitasking

61 to 74 of 74

Add your comments



	Site Map \| Home \| View Cart \| Pressroom \| Business \| Links	Contact Us
Logout
© Green Building Press

Green Building Forum

Categories

Feeds

Renewable Energy: How much PV has the UK installed to date?

Username	Password