Rooftop solar to cut total grid demand to zero in South Australia

See also Rooftop solar to overtake coal capacity before 2030

The Australian Energy Market Operator predicts that the growing uptake of rooftop solar by homes and businesses will reduce grid demand in South Australia on certain occasions to zero by 2023, highlighting the rapid change in the nature of energy markets, and the growing shift from centralised baseload generation.

The predictions from AEMO came in its 2015 National Electricity Forecasting Report, released on Thursday. It says that the near 575MW of rooftop solar is already accounting for one-third of total grid demand on certain days in the state.

But within a decade this total could treble, pushing minimum demand required from the grid in the whole state to below 0MW (zero) on some occasions in 2023-24, and for several hours at a time by 2024/25 – when AEMO expects 1864MW of rooftop solar.

It says zero demand from the grid could last from 11.30am to 2.30pm local time on some days.

ameo sa solar 2024 copy

“These observations from the 2015 NEFR reflect both the changing generation mix, and increasing consumer engagement when it comes to choice and energy supply solutions,” CEO Matt Zema notes.

But the impacts could be even more dramatic. The forecast is based on AEMO’s “medium scenario”. Given the propensity to underestimate the proliferation of solar deployment, it is worth mentioning AMEO’s high scenario, which suggests capacity of around 2,200MW of rooftop solar by 2024/25, and more than 3,000MW a decade later.

The high scenario suggests that rooftop solar will account for nearly one-quarter of the state’s total annual demand within a decade and nearly one-third of annual demand by 2034/35, when the rooftop solar market reaches “saturation”.

The state already sources one-third of its demand from wind generation, suggesting that on AEMO’s numbers the state will easily beat the state government’s 50 per cent renewable energy target by 2025, and that does not include the number of wind projects and large-scale solar projects that could be built over the next five years.

sa solar growth aemo

AEMO is investigating the implications of its solar scenario, and it’s impact on system security and reliability of having zero demand from the grid. The obvious answer lies in storage, both at household and grid level, to even out the output of solar and push more towards use when the sun does not shine.

For the first time it released a “minimum demand” analysis for South Australia, an extraordinary development considering that just a few years ago market operators were only concerned with “maximum demand” scenarios.

AEMO will also shortly release inaugural Emerging Technology Information Paper, which will look at the viability of battery storage solutions for Australian residential consumers over the next 20 years, and the uptake of other technologies such as electric vehicles.

South Australia will be a test case for Australia, and indeed the world, because of its high level of “variable renewables” such as wind and solar in its energy mix.

It already stands at more than 40 per cent and AEMO expects this to grow as the renewable energy target encourages more wind generation and households continue to take up solar. One in four homes in the state already have solar PV.

It explains why Alinta this week announced the imminent closure of its two base load coal-fired generators, giving the state the opportunity to move towards high renewable penetration, possibly 100 per cent.

Renewable proponents suggest that the two coal plants should be replaced with solar thermal and storage facilities, which have the flexibility to provide power when needed, and to store it when it is not required.

The International Energy Agency this week said wind and solar would play a critical role in reducing emissions from energy systems and helping to meet climate targets. It said variable renewables would account for 30 per cent of generation in Europe by 2030 and 20 per cent in the US.

In South Australia, the minimum load has already shifted from the early morning to the middle of the day, and peak demand has been shunted from the late afternoon to early evening, around 6.30pm. This is expected to move to 7.30pm as more solar is put into the grid.

South Australia is considered to be one of the prime markets for battery storage in Australia, given the high penetration of rooftop solar, and the difference between grid prices and payments for solar exports. Large-scale storage, at grid level, is also being considered in a trial co-sponsored by the Australian Renewable Energy Agency and AGL Energy.

See this story for state by state breakdown of AEMO solar PV forecasts.


24 responses to “Rooftop solar to cut total grid demand to zero in South Australia”

  1. Alan S Avatar
    Alan S

    We’re running out of fossil fuel power stations to close down. However we need transmission system upgrades to take advantage of installing more renewable power – including a decent connection with NSW

    1. Smurf1976 Avatar

      SA – NSW connection was proposed several years ago but died for various reasons. There was quite a lot of opposition to it on environmental grounds at the time and that helped kill the idea along with a few other factors.

  2. Finn Peacock Avatar

    First SA Power Networks need to remove the nasty little clause in their connection agreements that give you a 0c FiT if you have grid connected battery storage.

    1. RobS Avatar

      If you size your storage right then you’ll never need to use the FiT, that’s the whole point of storing the power for later use.

      1. Finn Peacock Avatar

        Couldn’t disagree more :-). The point of storage is not to reduce exports to zero. It is to minimise imports and use your exports to support the grid. If you size your storage as you recommend you’ll spend way too much and get a horrendous payback.

        1. RobS Avatar

          Absolutely disagree, supporting the grid is the role of utility scale storage, the role of home storage is maximising self consumption. The best thing a home can do to support the grid is go offline at peak periods, the best way to achieve that is to store daytime solar generation to use during the evening peak between 4 and 8pm. Sizing your storage to supply these few hours is by far the most helpful to the grid and will give you the most cost effective storage solution with the fastest payback particularly in TOU billing areas.

          1. Mike Dill Avatar
            Mike Dill

            RobS, I have to agree with Fin a little here, but I mostly do agree with what you are saying. Until the situation changes, it is not our job to support the grid.
            For the “average” day, you want to store all that you generate and consume it yourself. Unfortunately, there are a few days each year, when you will produce more than you need and can store. If you have extra, then dump it into the grid. There will also be a few days when you do not produce very much. On those days you will draw some and pay the grid price.
            IF you have TOU, and Net metering, then you want to put your extras into the grid when they pay you the most, DEPENDING on the tariff, and your cost to store. If you only cover your usage for the peak, you are probably close to being the most efficient and cost effective.
            I expect that the Feed-In-Tariff will change, and returning anything during high solar times will not get you anything. BUT there will be days when you have filled your storage and done everything else that you can do and still might have a few extra electrons.

          2. RobS Avatar

            I would love to see those times fairly reimbursed, however at some point I think the begging for it becomes demeaning. For me a huge part of the appeal of adding storage and going nearly totally or perhaps totally off grid is to give a big finger to the petty utility politics and quibbling over FiT’s and just look after myself without their arbitrary changes derailing my investment.

      2. Ronald Brakels Avatar
        Ronald Brakels

        Avoiding exporting electricity might be the whole point of having energy storage for you, Rob, but it certainly isn’t for me. My rooftop solar goal is to close down fossil fuel power stations – first brown coal, then black coal, then natural gas. And that means exporting as much electricity to the grid as possible. Home energy storage is currently not very useful for this goal, so my main interest in it is saving money. And I’m not going to be achieving my goal of shutting down fossil fuel capacity or saving money if I try to arrange things so I never export any electricity. What if my sister and 97 of her 300 children visit during a cloudy week? I’d need a stack of Powerwalls taller than the backyard termite mound to cover that. Nope, if I’m going to save money I’m going to need one Powerwall, used on-grid, with a time of use tariff. And even then I’d only save money if I increased my electricity consumption in the evening. Maybe if I took up arc welding as a hobby, or perhaps got a home electro-shock therapy system.

        1. RobS Avatar

          The primary point for me is not avoiding any exports, it’s really about avoiding drawing from the grid during the evening demand peak, that’s what will allow countless fossil fuel power plants to be mothballed and allow intermittent generators to penetrate the grid in unlimited capacities. The corollary of that is that solar generation excess to current demand will be stored during the day to be used in the evening and night, this will reduce the amount of power exported and render efforts to dissuade solar by cutting FiT’s ineffective. I agree that we could acheive even more if our storages not only supplied our own needs but also fed back into the grid at those times of peak demand however that requires larger more expensive storage capacity, additional grid interface electronics and a friendly compliant utility industry willing to facilitate the erosion of their primary source of profit. It is much harder for them to influence what we can and can’t do behind the meter than for them to influence what happens on the national grid. Ultimately I agree with you but the realistic side of me says lets at least pick the low hanging fruit by effectively taking ourselves off grid and hope the regulatory environment allows us to go further with time.

    2. Paul Roberts Avatar
      Paul Roberts

      Finn – SA Power networks only administers the scheme on behalf of the government. This clause reflects the legislation which deliberately does not include battery storage in the FiT scheme.

  3. WR Avatar

    I wouldn’t be in a rush to install a CSP storage plant. On good insolation days, its only possible benefit over PV+storage is that it might provide cheaper power. On poorer insolation days, it is worse than PV+storage for three reasons.

    Firstly, CSP relies on concentrated sunlight to generate heat. It therefore stops generating heat during any sort of cloud cover. PV systems will continue to generate between 10% and 30% of their maximum output even on completely overcast days, but a CSP plant is useless on those days.

    Secondly, the energy in thermal storage has to be used within 6-12 hours of when it is generated. Otherwise, heat loss causes it to dissipate. So again, the stored energy is only there for use at night after average-good insolation days. It won’t be there on the hard-to-supply winter nights that follow overcast days such as those being experienced in southern Australia this week.

    Thirdly, the storage is inflexible in that it can only be recharged by the CSP plant itself. Other storage technologies such as batteries or pumped-hydro can be recharged by any source. So there might be days with poor isolation but good wind energy where the CSP storage is not providing any benefit to the system but where other storage technologies would be working well.

    For these reasons, CSP basically works well only on days when there is plenty of sunlight but is useless on overcast cays. PV+storage works just as well on the sunny days as CSP and continues to provide benefits on even the most overcast days.

    CSP is less flexible than PV+storage on grids with high RE penetration. So the LCOE of CSP needs to be significantly lower than the alternatives to justify its place in a RE system.

    1. Pedro Avatar

      Your arguments are quite good which could apply equally to wind energy. Australia at this point has no really large CSP plants at present and I think a few need to be installed to prove the concept. Chances are they could work quite well in certain parts of the country where there are very few cloudy days.

      1. WR Avatar

        Wind’s downside is that there is a big gap between its highs and lows. However, its upside is that it supplies energy at night and on overcast and rainy days when solar is mostly sleeping (especially during short winter days).

        Wind’s biggest upside is that it is still much cheaper than the competition. At current and foreseeable prices, any cost-effective RE system for southern Australia should have wind energy as its single largest component.

    2. Alastair Leith Avatar
      Alastair Leith

      Technically that’s not true that solarCST can’t be recharged with non-sunlight sources. At Ivanapah even though there’s no storage that’s exactly how they get the salts to the ‘cool’ molten temperature to start cycling the fluid up to the collector — by burning gas (which happens to be cheap there ATM). If there was an electrical element the salts could be also melted to the cool or hot operating temperatures that way — heat is heat after all.

      Efficiencies of energy transfer obviously come into it in the real world, but as with PH even 50% efficiency is fine when you buy at 12c/kWh and sell at $10,000/kWh, the limiting factor becomes your capacity for storing the molten salts multiplied by the efficiency (losses) on the conversion of heat to electrical energy side of the process. The electricity to heat efficiency side of the cycle is less relevant in terms of finding a market for solarCST with molten salts (or steam) storage.

      Be interesting to see if the French pursue the ammonia storage option they have proposed to get to 100% RE in that leaked report.

  4. Beat Odermatt Avatar
    Beat Odermatt

    How can a business like SA Power Network punishing people with solar? The banks were taken to court over excess fees. It maybe time to challenge the validity of anti-environmental fees to be taken to ACCC or to a court to find out if fees which are clearly harmful to the community are valid or not. Maybe a class action could do no harm.

  5. lin Avatar

    Might be a good time to start encouraging the use of smart electric hot water heaters and other non-time critical high power use appliances to use power when production is high and demand is low. It may also be worth looking at some alternative high capacity storage methods such as pumped hydro to assist in load balancing.

    1. Greg Avatar

      Yes. Look at this graph of off peak hot water usage.

      Also Adelaide should look at daytime Electric car charging as standard, not overnight.

  6. Alen T Avatar
    Alen T

    Does AEMO have anything interesting to say about Qld? I would think with the 1 million solar roofs by 2020, the 40 MW solar proposal and the 50% target in general, Qld would at least get some mention by AEMO.

    I still think Qld is the sleeping giant of RE in Oz about to wake up and shake things up. With minimal GOV support since halting the FiT in the RE state industry it has still managed to grow at a ‘healthy’ pace. Now with backing and support from the government it should begin to boom, I would think.

  7. RAG Avatar

    900% to go then.

    That’s the target homes are going for where I live. Can’t have too much current applies ‘ay. Expand, expand, expand.

    Too easy.

  8. Ben Rose Avatar

    ……’Zero demand from the grid could last from 11.30am to 2.30pm local time on some days’. Yes but the hardest part will be still to come
    The problem is how to supply power overnight . Increasing wind and PV to cover the whole of the stillest darkest winter days would mean huge oversupply for most of the year. So modelling how much wind and solar is important. But if we want near 100% renewables on power grids, modeling how much and what technologies of renewable dispatchable power and storage are most economic will be the hard part. i.e. Providing wind and PV is easy and becoming steadily cheaper – the harder and more expensive part is the backup. In the interim of course gas can be used to supply disptachable backup (about 1/3 of total generation) that is needed.
    In my view future RET / subsidies should favor dispatchable renewable generation and batteries.
    PS I agree with most of what WR says about CSP. The key with it is a very cloudless, sunny site, ideally like the Great Sandy Desert. Modeling will soon be able to how if there are less remote sites that would be suitable. There is no reason why the tanks couldn’t be enlarged to supply say 4 overcast days (tank temperature drops by less than a degree per day), but even this may not be enough in winter or during cyclones

    1. Mike Dill Avatar
      Mike Dill

      I expect that in 10 years the costs will come down enough that solar PV and storage will make sense for nearly everyone. When the Feed-In-Tariff is zero at midday, it will make sense that those electrons will be captured on-site for use in the evenings.

      My biggest worry is that they will do something stupid with the other charges and I will need to disconnect from the grid.

  9. Alastair Leith Avatar
    Alastair Leith

    The first graph indicates that demand in SA peaks between 23:30 and 01:30? How can that be possible?

Get up to 3 quotes from pre-vetted solar (and battery) installers.