Clean Things Come in Big Packages
We really need to talk about scale.
In Australia we’re seeing no deceleration in additions of solar- and wind-based capacity, which is steadily displacing fossil fuelled generation on our grids.
Rooftop solar aside — since it serves a rather unique role, with its own emergent challenges — what we have seen in recent years is a small handful of impressively large projects come online each year, while considerably more are built at the 100 megawatt (or lower) scale. The latter tend to be constructed fast, and integrate into the market with little issue, contributing to the largest renewable electricity share ever in Australia. Still, coal and gas remain dominant:
And the Clean Energy Australia Report notes:
The challenges associated with grid connection and transmission continued in 2020 as the need for additional investment in transmission capacity to allow more renewable energy connections became more urgent.
Furthermore, despite the ubiquitous belief that large battery projects are the answer to the gaps in electrical output dictated by weather conditions, this hasn’t actually been the reason for their construction.
Thus, it’s fortunate that the Clean Energy Council identified the Goyder South hybrid project in South Australia as an example of the “renewable energy zones [which] should help to ease pressure on the grid”. Here is a very large project, integrating sizeable storage in the way practically everyone expects it to function. Some statistics:
- Solar PV: 600 MW
- Wind: 1,200 MW
- Battery: 900 MW, 1,800 MWh
- Lifespan: 25–30 years
- Facility area: 31,000 hectares
- All-in cost: $3 billion
The Development Application Package states that at least 4,800,000 megawatt hours (MWh) will be generated annually, and the battery storage will “guarantee customers a fixed power price 24 hours a day, irrespective of regional spot price fluctuations. This reduces the risk of power price fluctuations to large, energy-intensive industries and businesses such as mines, smelters, manufacturers and retailers. …By combining energy production and storage, Goyder South would overcome the conventional critique of renewables that they are ‘intermittent’ and ‘unreliable’. Hybrid projects such as this are the natural ‘next step’ in the transition to a cleaner, cheaper, renewable economy.” The proponents illustrate a day of this reliability like so:
This project, along with the associated SA-NSW “EnergyConnect” interconnector, are large and exciting propositions for South Australia. Yet, for cautious context, big energy projects have floundered here in the past.
The southern hemisphere’s first “power tower” concentrating solar thermal power station was announced in 2016 after substantial campaigning and planning since 2011. Despite ~15% of capital costs covered by a federal grant, the proposal folded instead of coming online in 2021 as hoped.
In the meantime, South Australia’s remaining demand and system security needs not being met by solar and wind are still relying on substantial gas capacity. And the price of gas has an out-sized influence on price-setting of wholesale power cost. Clearly, even more fossil fuel needs to be pushed out of this state’s grid.
What might it look like?
By constructing a simplified simulation of the Goyder South project, we can potentially see what it will add. A real year of 5 minute generation data from the Bluff wind farm, 50 km north of the project site, was scaled up to approximate the wind power portion of Goyder South. The same was done with data from Bungala Stage 1 PV farm, sitting at about the same latitude. This gave a total output very close to what the developers expect.
Simple storage logic was then added, telling the battery to store power any time more than 900 MW was simulated, until full, then discharge at times under that value*. Assuming that 900 MW is the limit for supply from the site, this only results in a few percent apparent curtailment.
However, the simulated supply looks like this:
When windiest, with good sunny days, a two month period looks like this:
To restate, this is simplified storage behaviour, almost certainly not representative of real, profitable operation. It is however everything the chosen 1,800 MWh battery size can deliver in this simulation, based on maximised charging from real plant generation, accounting for 90% round-trip efficiency. Changing when it discharges may smooth output considerably, but doesn’t change the area under the trace.
Indeed, this result resembles the Planning Application Package’s example day much less than it does Yole Développement’s illustration of “Renewables firming” from this sample industry analysis. In both cases it isn’t clear how operation will “guarantee customers a fixed power price 24 hours a day”, thus power dispatch from Goyder South is bound to be significantly more complex.
Dare we compare?
- 12 unit Small Modular Reactor plant: 884 MW(net electric)
- Lifespan: 60 years
- Facility area: 14 hectares
- All-in cost: TBC
A NuScale power plant built and operated as designed (95% average capacity factor, 7,355,000 MWh) in the same market as Goyder South, accounting for staggered module refuelling, would hypothetically look like this at maximum output.
Output could be less where it shares demand with renewable sources (even ones firmed by co-sited storage) — that’s one of the advantages of a fully dispatchable, predictable generator supporting the adequacy and security of the system.
In response to questioning by sceptical committee members during the recent Victorian inquiry into the nuclear prohibition, Tom Mundy from NuScale emphasised:
If we were to start with a project developer [in Australia] shortly, it is very reasonable to believe that a NuScale power plant could be generating electricity by the end of this decade if not earlier in Australia.
Despite steady progress towards procurement and deployment, some might dismiss this as unreasonable. But for context, the development application indicates Goyder South is anticipated to only reach full operation by 2033.
Some might reliably move on to assertions of excessive cost, perhaps with the CSIRO’s cost study efforts in mind, and notwithstanding the willingness to invest $3 billion in a non-nuclear mega-project. As noted above, eventual nuclear project cost is TBC, depending heavily on regional conditions… to get just a basic idea, though, the expected all-in price of NuScale’s first facility in Idaho converts to $8 billion AUD. Since greenhouse gas emissions aren’t a concern, would such an investment be too high for an option with potentially dramatic advantages?
The point of comparing these sets of numbers is, firstly, to illustrate that large-scale clean energy projects — still relatively rare in Australia — involve time horizons and investments that can’t reasonably be cited as prohibitive and excessive in one case but not in the other, especially when speed of decarbonisation is of the essence. And secondly, it attempts to show that they may not be comparable at all as far as the grid is concerned: renewable energy coupled to storage doesn’t substitute for small modular nuclear, and vice versa?
Other energy technologies have had ample chance to partner with solar and wind as they continue ascending in Australia. As long as we’re thinking big, nuclear energy needs to be next.
* The 900 MW threshold is assumed based on page iii of the Development Application Package: “An energy storage facility (lithium-ion battery) with a capacity of up to 900MW/1,800MWh (2 hours)”, i.e. discharge from full at a rate of 900 MW is presented by the proponents themselves.
Discussion of alternative simulation results is encouraged in the comments section!
Oscar Archer holds a PhD in chemistry and has been analysing energy issues for over 15 years, focusing on nuclear technology for seven, with a background in manufacturing and QA. He helps out at Adelaide-based Bright New World as Senior Advisor (we want your support!) and writes for The Fourth Generation. Find him @OskaArcher on Twitter.