It does have the virtue of providing long histories, and the methodology ties back the estimates to satellite weather data, with supposedly a correlation to real world data too. More recent years are available if you sign up. The data are given as fraction of capacity generation hourly. It's perhaps a shame that they haven't explicitly published the data used by the Royal Society. I ought to write to Iain Staffell and ask him to do so.
It's important to remember that the Gridwatch data only includes the transmission connected windfarms that are visible to the grid control room (live "metered"), so it leaves out quite a bit of onshore wind that is connected at the distribution level, plus some small offshore projects. A rising complication now is from curtailment - some paid for, and some simply when prices are negative and the subsidies don't apply to some of the CFD holders and those who have yet to commence their CFD.
However, the older data is less affected, and should allow comparison between them. A very useful technique is to look at the cumulative sum of each series relative to its mean. By subtracting the series mean from all points and cumulating you get a clear idea of the behaviour, and whether the modelled data really does look like the actual data when you chart them. I know when I did this for Andrew Blakers' study on Australia I was easily able to show that his wind modelling was entirely phony. Periods of Dunkelflaute and runs of windy weather show up as negative and positive slopes respectively. Seasonal variations will result in significant cumulative deviations.
Thank you for your comments. The thing I *really* like about GridWatch data is its transparency. Since late 2022 / early 2023 I've been applying adjustment factors to GridWatch Wind and Solar records to align with DUKES 6.2. I've explained this in a bit more detail in my latest post "Diminishing Returns".
I also like recent years' data because much of the 'renewables' fleet has been live and working since 2020, meaning my simple extrapolations are less likely to be massively off.
[Similarly, I really like recent EIA data for US states & US48. It's recorded hourly but provides plenty sufficient granularity to see intervals when only 'renewables' would get you blackouts.]
If you have contacts you can ask about some of the observations I've made, please feel free. If, however, their response is along the lines 'Well our modelling shows...' then unless their modelling is equally transparent vs my extrapolations I'm unlikely to be convinced.
This is a great advance for me in planning the parameters for Seraph Power wind-turbine-based hybrid power systems. Generation 1 is predicated on a small system intended to provide 24/7 grid-independent power for a modest household (under 1KW). Integrating a wind turbine with battery storage and fossil-fuel backup, this analysis helps project how much fuel should be kept on hand in case the individual owner opts to keep the lights on during Dunkelflaute events. I wish I had power systems ready on the shelf, but a successful crowdfunding event in the near future should get us closer to achieving our goals of responsible stewardship while bringing power to the people.
You can actually get the renewables.ninja data (or at least a version of it) here:
https://www.renewables.ninja/downloads
It does have the virtue of providing long histories, and the methodology ties back the estimates to satellite weather data, with supposedly a correlation to real world data too. More recent years are available if you sign up. The data are given as fraction of capacity generation hourly. It's perhaps a shame that they haven't explicitly published the data used by the Royal Society. I ought to write to Iain Staffell and ask him to do so.
It's important to remember that the Gridwatch data only includes the transmission connected windfarms that are visible to the grid control room (live "metered"), so it leaves out quite a bit of onshore wind that is connected at the distribution level, plus some small offshore projects. A rising complication now is from curtailment - some paid for, and some simply when prices are negative and the subsidies don't apply to some of the CFD holders and those who have yet to commence their CFD.
However, the older data is less affected, and should allow comparison between them. A very useful technique is to look at the cumulative sum of each series relative to its mean. By subtracting the series mean from all points and cumulating you get a clear idea of the behaviour, and whether the modelled data really does look like the actual data when you chart them. I know when I did this for Andrew Blakers' study on Australia I was easily able to show that his wind modelling was entirely phony. Periods of Dunkelflaute and runs of windy weather show up as negative and positive slopes respectively. Seasonal variations will result in significant cumulative deviations.
Hello Idau, and Happy New-ish Year!
Thank you for your comments. The thing I *really* like about GridWatch data is its transparency. Since late 2022 / early 2023 I've been applying adjustment factors to GridWatch Wind and Solar records to align with DUKES 6.2. I've explained this in a bit more detail in my latest post "Diminishing Returns".
I also like recent years' data because much of the 'renewables' fleet has been live and working since 2020, meaning my simple extrapolations are less likely to be massively off.
[Similarly, I really like recent EIA data for US states & US48. It's recorded hourly but provides plenty sufficient granularity to see intervals when only 'renewables' would get you blackouts.]
If you have contacts you can ask about some of the observations I've made, please feel free. If, however, their response is along the lines 'Well our modelling shows...' then unless their modelling is equally transparent vs my extrapolations I'm unlikely to be convinced.
This is a great advance for me in planning the parameters for Seraph Power wind-turbine-based hybrid power systems. Generation 1 is predicated on a small system intended to provide 24/7 grid-independent power for a modest household (under 1KW). Integrating a wind turbine with battery storage and fossil-fuel backup, this analysis helps project how much fuel should be kept on hand in case the individual owner opts to keep the lights on during Dunkelflaute events. I wish I had power systems ready on the shelf, but a successful crowdfunding event in the near future should get us closer to achieving our goals of responsible stewardship while bringing power to the people.