About intermittency and being a net exporter

While writing the post on the upgrade of the Hornsdale Power Reserve, I became curious how South Australia balances its grid. Looking into the data, it became pretty clear that it aren’t the batteries that doing the balancing. According to the fuel mix data of AEMO, the battery storage output is insignificant compared to the huge swings in output of solar and wind power.

There are several balancing strategies possible. For example, in a previous series on the German energiewende, I found that Germany’s strategy is to use fossil fuels (gas, coal and even lignite) when there is not enough solar & wind and export the surplus to the neighboring countries when there is too much solar & wind.

South Australia also has a high share of solar and wind, so how do they do it?

There is a lot of data of the Australian grid on the AEMO website. Not only the fuel mix data that I used in the previous post, they also have other data like spot prices, scheduled demand, scheduled production and import/export balance. Unfortunately, the finer-grained (hourly) data is only for the last few days, so I say myself collecting this data in the last few weeks and this is an overview of almost one month’s worth of data (click to enlarge to get a clearer view):

South Australia fuemix vs demand

It is clear that there are huge swings in output of solar and wind. There are times when output of solar and wind is next to nothing (September 5, 15, 16 and 27), but there are also times when the output of solar and wind exceeds demand (September 4, 6, 17 20, 21, 22, 23 and 28).

As seen in previous two posts, the output of the batteries in South Australia is pretty pathetic and this is also clearly visible here (the blue colored blips glued to the x-axis). These batteries clearly have no function in the balancing of those swings in output of solar and wind (their function in the grid is frequency control and they make quite a buck doing so).

There is also a difference in how these situations are handled. When there is not much solar and wind, natural gas is jumping in, sometimes supported by import of electricity from Victoria. The production and import fit demand rather snugly. This is not the case when there is a lot of solar and wind. Although natural gas then powers down, these are the times when electricity is exported to Victoria.

The result is that in this small sample, roughly 21% from all electricity that was moved between South Australia and Victoria was imported into South Australia and 79% is exported to Victoria. That makes South Australia a net exporter.

South Australia apparently uses the same strategy as Germany. They both use fossil fuels (natural gas) to fill in the gaps and they export their surplus production to their neighbors when there is a lot of solar and/or wind.

This difference in import versus export in both countries is I think a consequence of the dynamics explained in previous post. On the one hand, solar and wind in South Australia can potentially provide next to nothing (even at peak demand) and this means that the need for backup doesn’t decrease much when adding more solar and wind capacity. If that need can be met by natural gas together with some import, then it doesn’t matter much that the need for backup doesn’t decrease much. On the other hand, solar and wind in South Australia can potentially provide all needed electricity (even at peak demand) and this capability will grow ever faster when adding more solar and wind capacity.

Combine a slowly decreasing need for backup (that currently can still be met) with an ever faster growth of peak production (generating surplus energy that they need to get rid of in order to protect their grid) and it shouldn’t be a big surprise that both countries became net exporters of electricity. They are net exporters, not because they are reliable energy producers, but as the result of adding more intermittent power sources to their grid without having the ability to balance that power.

9 thoughts on “About intermittency and being a net exporter

    1. trustyetverify Post author

      Hi Roger

      Yes, I came across his posts several times before. He also is collecting data from AEMO on a daily basis, specifically wind data from Australia (I restrict myself to the fuel mix and demand/import data of South Australia, in the context of posts on the battery power share).
      He is doing this for quite a while (October 2019) and has daily updates on wind power in Australia. This is the first post of that series:
      and this is the permalink of all the posts in that series:


  1. Chris Morris

    You have identified the issue very well. A lot of the power is generated when not needed and it can’t be stored. That is why you see negative pricing in South Australia to to tell the windfarms to shut down. http://www.wattclarity.com.au/articles/2020/09/new-minimum-point-for-scheduled-demand-in-south-australia-on-sunday-13th-september-2020/
    To add to the grid picture As well as the main SE 700MW AC line to Victoria, there is also a 200MW rated DC line coming in from New South Wales (it goes through a little bit of Victoria. and may be in their data). This line isn’t mentioned much because it has the same issues as the windfarms. The power is asynchronous.


    1. trustyetverify Post author

      Thanks, Chris. I knew about the Heywood and the MurrayLink interconnectors (both between South Australia and Victoria). I wasn’t aware that there is also one between South Australia and New South Wales. It is indeed not mentioned on the AEMO dashboard.

      [After reading the linked post, it seems something for the future. Another post that I found mentioned that the interconnector will be operational from 2022 (http://theleadsouthaustralia.com.au/industries/mining-resources/aemo-gives-sa-nsw-interconnector-project-top-priority/)]


  2. TonyfromOz

    I would like to draw your attention to a site that was put up by supporters of the renewable lobby, and having said that, because it is just the recording of data, it is something that can be used to make a point from any side of the spectrum, including those who are no real fan of these renewables, and that includes me.
    This is the site at this link. (https://opennem.org.au/energy/nem/)
    Now, while I include this an an overall ‘picture’ I know that you concentrate on just matters from South Australia, so this is in fact a good resource for just that, showing ALL power sources for that State.
    The site opens at the default of the overall NEM (AEMO) for the last seven days, and you can easily navigate your way around for so many things at the site.
    However, go to the menu bar at the top left where it is headed as NEM and if you click on that, a drop down menu appears, and that shows the five States, and you can then click on South Australia, and all the data opens for that State.
    I’ll leave it to you to navigate further from there with the menu tabs under that for days Months and Years etc.
    Down the right side are the colour buttons for each individual power source, and you can just rest the mouse on those colours to highlight just them, or even isolate them out on that main graph, and again, just work your way around, and you’ll see just how much data there is and what it indicates.
    One thing though, as you look at that data for the (blue square button) Battery (discharge) you mention in your text above, note that it never ever reaches one Percent of the total supply, and in fact, it only ever tops out at half of one percent on very very few occasions. That’s a half of one percent of a State only consuming 6.2% of Australia’s total power, so minute even in its own State.
    It may just be a data resource, but the whole overall thing can tell you so much.

    Liked by 1 person

    1. trustyetverify Post author

      Wow! Thanks a lot for the link, Tony!

      That site is a data junkie’s treasure trove. It would have made life so much easier. Currently, I downloaded every day several tables (with the data in different intervals) and manually put them in one summary file in the same format. Which was quite some work (and plenty of possibilities to make mistakes). Now I could have everything in one single table with the data in the same format.

      Liked by 1 person

  3. Chris Morris

    You are correct – the DC ends in Victoria at Red Cliffs on the banks of the Murray. However, the major connection from there is to Buronga just on the other side of the river that is NSW. That is why in the 2 year AEMO plan for Western Victoria they have “Option D1 – construction of a new 220 kV double circuit line from Red Cliffs to Buronga (operated as a single circuit initially) and a 400 MVA 330/220 kV transformer at Buronga, to be considered under the South Australian Energy Transformation (SAET) RIT-T.”



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s