Been comparing my 2023 generation with 2022 they are the only full years data I have. 2022 generated 6281Kwh and 2023 5347 Kwh. We peaked in May but we had a wet June so not surprised followed by very wet autumn into winter. I have 6kw of generation. Efficiency was 69% in 2022 and 58% in 2023 and used 200 Kwh more from the grid in 2022. How have others on the forum done?

Compared to the PVGIS projection, 2023 has only delivered 94.5%, the worst year since installation in 2011. Second worst was 2017 with 97.9%. The 2011-2022 average performance is 104.2% (yes, it means the projection needs tweaking!)

From 2016 to 23 our figures from a 4kw system have been
16 3732
17 3547
18 3740
19 3680
20 3557
21 3661
22 3774
23 3555

MCS forecast was 3342 for the first year with a small but steady decline in output but annually we've produced between 6-13% more since installation. We went for a solaredge system with optimisers which I think accounts/contributes to the over production

My PV numbers, 4.2 kWp system in the South West

2019 3,728.82
2020 3,968.28
2021 3,617.17
2022 3,816.17
2023 3,510.23

against an installers prediction of 3,467.52 kWh/annum. Ours is also Solar Edge with optimisers, due to intermittent overshadowing by mature trees. Seems to be doing its job, but we also saw a dip last year.

2020 wasn't good for me but good for you and DT. I'm in NW Wales so I'm gusseing it's regional??

Looks like 2023 poorer than 2022 for all. I am 14.8% down which seems a lot, Cannot find sun hours for 2023 so far.
I am N Wales (Anglesey)

Posted By: revorLooks like 2023 poorer than 2022 for all.

True, but for us that's more a case of 2022 was our best ever year, so far! 2023 was just below average (which is 3889). 2021 was our worst year.

To be honest I stopped recording my PV generation a while ago - building my house is more of a priority. All I do is check my Leiderdorpe occasionally to see that it is producing near or over 2500w at midday on a sunny day. Anything else and I'd be suspicious.

[The main array has terrible panels - I would never have sold them to a customer - but I was offered them brand new at £25 each, (the whole installed system cost £600 including scaffold when the average at the time was £3500) so I'm just happy to get any generation out of them!]

I do monitor the PV system at work for our carbon reporting, but only been doing that for the last two years:

2022 - 8563
2023 - 8307

This is a 9.75 kW system and one of the strings was shut down for the first 2 1/2 months in 2022 so 2023 was a bad year for generation. Given that we are likely to have more rain and more heatwaves in the coming years, (both are bad for PV panel generation) I can only see this trend continuing.

PV generation and performance are affected by many factors. Location and average sun hours are just one aspect.

Panels next to the sea generate as well or slightly better than inland, despite experiencing more morning mists that you would expect to reduce output. This I believe is due to more reflected light reaching the panels from the water.

Panels in flat areas, (Norfolk, Cambridgeshire) generate more than expected – our theory was that they experience more uninterrupted wind flow, and the panels are operating slightly cooler on sunny days.

Microinverters obviously help increase generation, (about +5% on an unshaded roof from figures we gathered), but I never considered the extra initial cost and poor reliability to compensate enough for this. Well-designed strings using tier 1 panels and a top-notch inverter is almost as good unless shading is a major issue.

It should be noted that (AFAIK) all MCS figures are still based on a single panel test they conducted when they were set up, (funded by Sharp using some (fairly average) poly panels when the trend was already towards mono cells). If this has been updated since - I hope someone can correct me on this.

That ‘test’ took no account of the variation in panel performance for different technologies or manufacturers, (they only tested a single panel from a single manufacturer) and was seen as highly unfair by the PV installation companies, (and other panel manufacturers) that existed before MCS came along to widdle in our tea.
Whilst it was very true that there needed to be some checks on the Cowboy installers who were popping up like rats from a McDonalds drive through dustbin at that time, they ignored the years of data we had compiled and took a backhander from a company with a vested interest in the outcome. It’s hard to sell a more expensive and much better-performing Sanyo HIT panel if the only figures you are allowed to quote show it as equal to any old Poly panel.

This trend continues today. I watched a video on YouTube recently that showed them removing 250w HIT panels for some ‘more efficient’ 300w mono panels. I’d love to see the comparison between the original and the new panel outputs at the end of a couple of years… I doubt there would be much difference between the two.

But it’s a win for the salesman who has gulled another sale from an unknowing Joe public.

Posted By: bhommelsCompared to the PVGIS projection, 2023 has only delivered 94.5%, the worst year since installation in 2011. Second worst was 2017 with 97.9%. The 2011-2022 average performance is 104.2% (yes, it means the projection needs tweaking!)

I'd a thought 94.5%, 97.9%, 104.2% was pretty good projecting, when it's based on 10yr or whatever rolling average of a mixture of largely indirect data.

<blockquote><cite>Posted By: Mikel</cite>Valley station data available from here

https://www.metoffice.gov.uk/pub/data/weather/uk/climate/stationdata/valleydata.txt</blockquote>

Thank you. Have looked at their figures which are provisional at moment and it looks like we had about 10% less sun hours. My generation was 15% down so might warrant some investigation.

@revor

I am in west Cornwall and we get very localised weather, probably due to being surrounded by sea. In summer, we can start out with a clear sky, have clouds by lunchtime and back to clear sky by late afternoon. I dont know how our weather differs from that where the Met Office weather station is located, even though only a few miles away.
I dont know whether you get similar localised weather in Anglesey.

This year the sun hours for March, April, July, August, October and November were all down significantly (approx 30-50%). July's generation was about 60kWh down from the PVGIS figure. However, although the July sun hours were 70% of average, the generation was 88% PVGIS. May was the top month with 132% of average sun hours and 120% PVGIS generation.

I only look for gross anomalies, e.g. if the percentage sun hours are higher than average but the generation is lower than expected.

It was optimiser/micro inverter data logging I was asking after as its a god send for diagnostics. I guess dual MPPT inputs are logged individually so a bit more helpful than a single input.