Turnbull right to fund energy storage: 100% renewable grid is within reach

Print Friendly

The Conversation

Pixabay/Wikimedia Commons

Pixabay/Wikimedia Commons

In a speech to the National Press Club yesterday, Prime Minister Malcolm Turnbull declared that the key requirements for Australia’s electricity system are that it should be affordable, reliable, and able to help meet national emissions-reduction targets. He also stressed that efforts to pursue these goals should be “technology agnostic” – that is, the best solutions should be chosen on merit, regardless of whether they are based on fossil fuels, renewable energy or other technologies.

As it happens, modern wind, solar photovoltaics (PV) and off-river pumped hydro energy storage (PHES) can meet these requirements without heroic assumptions, at a cost that is competitive with fossil fuel power stations.

Turnbull and his government have also correctly identified energy storage as key to supporting high system reliability. Wind and solar are intermittent sources of generation, and while we are getting better at forecasting wind and sunshine on time scales from seconds to weeks, storage is nevertheless necessary to deliver the right balance between supply and demand for high penetration of wind and PV.

Storage becomes important once the variable renewable energy component of electricity production rises above 50%. Australia currently sources about 18% of its electricity from renewables – hydroelectricity in the Snowy Mountains and Tasmania, wind energy and the ever-growing number of rooftop PV installations.

Meanwhile, in South Australia renewable energy is already at around 50% – mostly wind and PV – and so this state now has a potential economic opportunity to add energy storage to the grid.

Pushing storage

To help realise this potential, in South Australia and elsewhere, the Clean Energy Finance Corporation (CEFC) and the Australian Renewable Energy Agency (ARENA) will spend A$20 million of public funds on helping flexible capacity and large-scale energy storage projects become commercially viable, including pumped hydro and batteries.

PHES constitutes 97% of worldwide electricity storage. The retail market for household storage batteries such as Tesla’s Powerwall is growing, but large-scale storage batteries are still much more expensive than PHES. “Off-river” pumped hydro has a bright future in Australia and many other countries, because there are very many suitable sites.

Wind and PV are the overwhelming winners in terms of new low-emissions electricity generation because they cost less than the alternatives. Indeed, PV and wind constituted half of the world’s new generation capacity installed in 2015 and nearly all new generation capacity installed in Australia.

Recently, we modelled the National Electricity Market (NEM) for a 100% renewable energy scenario. In this scenario wind and PV provide 90% of annual electricity, with existing hydro and bioenergy providing the balance. In our modelling, we avoid heroic assumptions about future technology development, by only including technology that has already been deployed in quantities greater than 100 gigawatts – namely wind, PV and PHES.

Reliable, up-to-date pricing is available for these technologies, and our cost estimates are more robust than for models that utilise technology deployment and cost reduction projections that are far different from today’s reality.

In our modelling, we use historical data for wind, sun and demand for every hour of the years 2006-10. Very wide distribution of PV and wind across the network reduces supply shortfalls by taking advantage of different weather systems. Energy balance between supply and demand is maintained by adding sufficient PHES, high-voltage transmission capacity and excess wind and PV capacity.

Not an expensive job

The key outcome of our work is that the extra cost of balancing renewable energy supply with demand on an hourly, rather than annual, basis is modest: A$25-30 per megawatt-hour (MWh). Importantly, this cost is an upper bound, because we have not factored in the use of demand management or batteries to smooth out supply and demand even more.

What’s more, a large fraction of this estimated cost relates to periods of several successive days of overcast and windless weather, which occur only once every few years. We could make substantial further reductions through contractual load shedding, the occasional use of legacy coal and gas generators to charge PHES reservoirs, and managing the charging times of batteries in electric cars.

Using 2016 prices prevailing in Australia, we estimate that the levelised cost of energy in a 100% renewable energy future, including the cost of hourly balancing, is A$93 per MWh. The cost of wind and PV continues to fall rapidly, and so after 2020 this price is likely to be around AU$75 per MWh.

Crucially, this is comparable with the corresponding estimated figure for a new supercritical black coal power station in Australia, which has been put at A$80 per MWh.

Meanwhile, a system developed around wind, PV and PHES and existing hydro can deliver the same reliability as today’s network. PHES can also deliver many of the services that enable a reliable energy system today: excellent inertial energy, spinning reserve, rapid start, black start capability, voltage regulation and frequency control.

Ageing system

Australia’s fossil fuel fleet is ageing. A good example is the pending closure of the 49-year-old Hazelwood brown coal power station in Victoria’s Latrobe Valley. An ACIL Allen report to the Australian Government lists the technical lifetime of each power station, and shows that two-thirds of Australia’s fossil fuel generation capacity will reach the end of its technical lifetime over the next two decades.

The practical choices for replacing these plants are fossil fuels (coal and gas) or existing large-scale renewables (wind and PV). Renewables are already economically competitive, and will be clearly cheaper by 2030.

Energy-related greenhouse gas emissions constitute about 84% of Australia’s total. Electricity generation, land transport, and heating in urban areas comprise 55% of total emissions. Conversion of these three energy functions to renewable energy is easier than for other components of the energy system.

Transport and urban heating can be electrified by deploying electric vehicles and heat pumps, respectively. Electric heat pumps are already providing strong competition for natural gas in the space and water heating markets. Importantly, these devices have large-scale storage in the form of batteries in vehicles, and thermal inertia in water and buildings. Well-integrated adoption of these technology changes will help reduce electricity prices further.

So wind, PV and PHES together yield reliability and affordability to match the current electricity system. In addition, they facilitate deep cuts to emissions at low cost that can go far beyond Australia’s existing climate target.

Authors: Andrew Blakers, Bin Lu, and Matthew Stocks

Source: The Conversation. Reproduced with permission.The Conversation  

RenewEconomy Free Daily Newsletter

Share this:

  • Rob G

    Pretty much everyone can see the “dumbness” of ‘not very clean coal’. Today, it makes no sense – but think for a minute and imagine 5, 10, 20 years from now. Each year that passes will make this suggestion look more and more stupid. The rate of improvement is enough to scare any bank away from such an investment. It’s game over for coal and that is why we are seeing so much panic and talk around it by the coal loving government. Both our government and the Trump administration cannot save coal.

    • solarguy

      No they can’t but keep trying to hoping they will convince the uneducated and disengaged.

  • Ron Horgan

    Pumped hydro storage for renewables looks best as the lowest cost energy storage.
    The Hazelwood mine pit and associated cooling “pond” (a 32 Gl reservoir) could be developed into a large system capable of handling about 1Gw of peak capacity.
    The pit is problematic in that the coal seams are unstable. if such problems can be solved the existing connection to the grid, skilled workforce and water storage may be community assets for many years.

    • Cooma Doug

      When you think about pumped hydro in conjunction with the wholesale market rules, you will understand why the 8gwh available today is not used.
      If the market rules are not changed, there is no incentive for it and it would not happen. It was origionally installed to sustain base load PS at low loads over night.
      Then when we change the rules to accommodate evolving technologies,
      the cost of pump storage compared to load side response options, coordinated with load located storage, makes it a dum idea. The large scale systems especially.

      The cost problem is multiplied by the halving of the capacity factor of the large pump/gens. Also the removal of the HV load spikes and trouths.

      The flying duck profile is strongly emerging. It will end up do flat, you wont see a duck.
      So the existing large HV grid pump storage will be a major part of the transition. But a dud idea after.

      • Rod

        Are you saying the Snowy Hydro doesn’t bid and supply at spot prices?
        I would think NSW and Victoria are in a position to greatly increase variable generation and reserve Hydro to flatten the duck and stabilise the grid? And Hydro would and should be paid a premium for doing so.
        Tasmania too, especially if a second Basslink was to go ahead.

        I’m a keen watcher of the NEM generation widget and it is obvious to me in SA that at times of low wind (like most of Summer) we are very reliant on gas or the interconnector, even more once the sun goes down. Grid scale battery storage or built hydro storage makes much more sense here.

        • Cooma Doug

          Like I said, the market rules are such that pump storage is a low priority. More opportunity waiting for prices rule it out. Also it wont compete well until in the future when load side options are rewarded via the market rules.

          • Rod

            Not disputing, just trying to understand.
            Surely with more RE coming on line (In Vic at least) the quick start properties of Hydro will become more valuable?

          • Cooma Doug

            Yes. But in order to optimise this the rules need a tweak

          • Cooma Doug

            Load side responses will gradually make gas and large spinning plant too expensive. Not just because of falling solar and battery costs. The load side modern tech responses will be milli second response. This will have a big impact on how the market is designed and functions.

  • wmh


    Heating uses an annualised 60% of household energy in Sydney (Ausgrid data) and a lot more than that in winter. A well insulated house would need to have
    stored perhaps 20 or 30kWh of energy for central heating on a cold, cloudy
    winter’s day and so too expensive for batteries at $680 per kWh (the latest
    Powerwall battery cost).

    To cope with several such days in succession during an East Coast Low, figure on several times this amount. Hence going off both the gas and electricity grids
    in town doesn’t appear to make economic sense yet, particularly since almost all
    houses are so poorly insulated. However there is a solution which radically
    reduces the cost by reducing the amount of battery storage required.

    The solution is to store energy for heating as heat. Hot water storage (45
    to 90 Celsius operating range) can store 52 kWh per 1000 litres,
    water is cheap ($2 / 1000 litres) so most of the cost is the tank. This can be
    both a cheap unpressurised tank for hydronic heating and a normal mains
    pressure tank for the domestic hot water service. The tank needs to be well
    insulated of course but insulation is cheap. If the tank is mounted within
    the heated space then all the tank heat loss goes to heat the space and none is

    Hot water running through wall-mounted radiators is used all over the world
    for central heating. A small pump is required to push the hot
    water around with thermostats regulating the room temperature and there are no
    noises or drafts. Hot water at 90 Celsius is too hot to use in wall mounted
    radiators so a tempering valve mixes in cold water to reduce the temperature
    but this does not involve any energy loss.

    In summer, cooling rather then heating is required. Most cooling is required
    in mid summer and in the the daytime, particularly in the afternoon, but solar panels are producing their maximum output at this time so the solar panels can run air conditioners directly, particularly if some panels are turned to face west to provide power until sunset. Energy storage is free as once cooled, the house thermal mass will keep the house cool well into the night.

    Hot water energy storage systems are compact due to the very high heat capacity of water, can be expanded in capacity for basically the cost of the extra tank and store
    energy directly in the form that is most required in the cooler parts of the

    Batteries are still required to run your lighting and TV but such loads are a tiny part of your power bill compared to heating. These days LED lighting, the TV, fridge, microwave, induction cooktop and PC are all considerably more efficient than the equivalents of 20 years ago.

    Once installed, solar combined with energy storage as described, will fullfill the old nuclear energy promise of “energy too cheap to meter”.


    Anybody with an electric hot water service can heat all their domestic hot water using 2 or 3kW of solar PV. Most PV owners in NSW have now been net-metered and so need only fit a “Diverter” to control the power into their existing hot water service. Diverters are common in the UK and are available in Australia. Google “PV diverter” to see how it works.

    • Matthew Wright

      Hmm so you haven’t heard about the following.
      The average gas central heating system loses 50% of heat – 30% up the flue 20% through the ducting and 10% due to pressure equalisation mismatch. So your 20 – 30kWh figure is now 10-15kWh. Then combine this with a Daikin Ururu Sarara space comfort heatpump with a COP of 5.92 meaning 1kWh of electrical energy consumed in the motor delivers 5.92kWh to the space and suddenly you’re heating that Sydney (Ausgrid) house with 1.7 – 2.5kWh per day

      • wmh

        My calculation of winter energy use is not based on ducted-air central heating but on a calculation of wall/ceiling/floor heat loss based on thermal resistances and outside-inside temperature differential.

        Ignoring standby crankcase heating losses, heat pumps are efficient but I am taking about energy storage here. During a winter rainy period lasting days you may need lots of storage – only economic with hot water.

        • Matthew Wright

          No even if you need 3.4kWh – 5.0kWh – and you can reduce this with a bit of thermal envelope work. Then a few days of low output with a 20-40kW rooftop system ie Tesla roof combined with a battery and you’ll be exporting to the grid on those days.

          • wmh

            A 4m x 2.5m wall thermally insulated to R3 would need 0.8 kWh/day of heating if the inside temperature were 10 C above the outside temperature so four such walls would need 3.2kWh/day and you haven’t yet allowed for heat loss through the ceiling and floor plus a bit to heat the ventilation air or for the large heat loss through any windows. You can work it out for a complete house.
            Some people may need a 40 kW system (perhaps costing $28,000 without installation). It would generate 200kWh on a fine day and perhaps less than 10% of this on a rainy day. That’s why storage is essential.

          • Matthew Wright

            There’s some heat generated by other exercises in the house such as cooking as well and each human is adding 80-100W.
            20kWh is a lot and combined with a 13.2kWh powerwall will easily give you the required 5.0kWh a day to run the Daikin Ururu Sarara heat pump which through the refrigeration cycle delivers 30kWh to the space.

        • Rod

          One on the very few mentions of crankcase heating losses.
          These alone can waste 500 kWh per year per household.
          If someone could design a gadget to turn them off when not needed they would make a packet. I manage mine from the switchboard.
          I understand where you are coming from with water storage for heat but here in temperate Adelaide it makes more sense to me to get some solar gain happening as well as sealing the envelope, secondary glazing etc. and use the heat pump to heat the smallest living space possible when needed.
          We have a 3 x flat plate 400ish litre SHWS but during those Winter periods of low cloud we need to resort to electric boosting.
          The tank and pipes are double insulated but it still loses 10C overnight.
          Additional PV and boosting directly wouldn’t help IMO as just when you need it, the sun is not available.
          I’m also on PFiT so keen to export as much as I can.
          The ideal HWS would be evacuated tube solar with air sourced heat pump boosting but that gets expensive and complicated.
          As for grid stability, my HWS is interruptible and can also be remotely activated by the utility (assuming it is on at the board)

      • solarguy

        Your correct in what your saying Matt, but the only problem is, wmh isn’t talking about gas use what so ever. Conversely wmh isn’t correct on most of what he’s on about, except water storage being cheap.

        • Matthew Wright

          Yes I can see that from his clarification below and I’ve answered him. I said if a Tesla roof was installed at 20-40kW and it generated 10% of rated output on a terrible day therefore 20kWh then that would be ample to satisfy the 5kWh to drive a couple of Daikin Ururu Sarara’s delivering about 30kWh to the space that he’s after.

          • solarguy

            Sorry Matt, you didn’t mention the Tesla roof.

  • solarguy

    What has been written in this article is much my thoughts for years now, of how to go 100% RE. However articles like these, leave out a clean energy source that is going to waste and it’s sewerage, along with other bio waste, to make bio methane. This can be stored for use in times of east coast lows and burnt in gas turbines for back up. It is already being used overseas with much success and the dregs used for fertilizer.

    • Mike Shackleton

      The Melbourne Western and Eastern treatment plants already capture methane and generate electricity with it. My understanding is that the electricity generated is enough to cover the demand by the system itself, with a little excess being exported to the grid.

      • solarguy

        I knew that there were such plants in OZ, there is one in Sydney too. They were built as proof of concept mainly as a low cost option, that could reduce running costs of the plants. Bigger digesters = more power, they certainly have the feedstock to do so.

  • Ian

    The science of reliability is highlighted in this article and this issue probably needs deconstructing and re-analysing in the context of electricity usage. To the solar /battery home owner reliability may be 1. Reliability of supply for those few days a year when days of cloud drain batteries. 2.reliability of transmission – for the times when the grid fails and exports are impossible or more long term punitive tariffs frustrate the economic reliability of exporting electricity. 3. Financial reliability where prices remain stable and long term forecasts can be made when deciding on equipment investment. To the industrial customer such aluminium refining reliability means more of the traditional constant, regular , cheap electricity supply. To the hospital this might be a combination of regular supply and reliable standby supply

    The supply side generators have specific energy profiles. Coal is most economic producing power steadily without ups or downs. It has maintenance and breakdown downtime. It has a lifetime at the end of which its fuel source is exhausted or its social licence expires. Nuclear would have similar problems, hydroelectricity is dispatchable, cheap, can run through days of cloudy, windless weather but requires generally long transmission lines subject to failure and seasonally can fail due to prolonged drought. Wind has its idiosyncrasies including long transmission lines, but it is very reliable from season to season, not suffering the affects of drought,solar too has its profile. Similarly gas and liquid fuels demonstrate cost failings, environmental costs, dispatchability and baseload failings and benefits .

    The idea is very silly of an amorphous grid providing unlimited energy capacity, at any time, for every application from pool pumps to ICU respirators and from air-conditioning and hot water heating to data centre and aluminium smelters and underground mining , in the city and way out in the country for the same price and with the same reliability.

    The part-time solar/battery household generator/consumer will not be willing for very long to pay a premium for critical application standby reliability or never-off baseload energy supply reliability. These sorts of consumers would probably be quite happy with low reliability supply if very cheap. The Hospital customer may chose to accept slightly less than 100% reliability from the grid with in-house standby gas or liquid fuel generators, similarly mines with underground safety issues may use localised standby diesel generators.

    Perhaps the conversation should turn from how intermittent energy sources can provide perfect reliability to how much unreliability can customers accept in exchange for cost value and how should reliability be costed? If this is not addressed then solar/battery households will go off grid, large industrial customers will make alternative and private supply arrangements and mid to large sized commercial customers will follow the household customer’s lead and disconnect from the grid.

    • Chris Fraser

      Hospitals and transport hubs would need 100% uptime (thus the use of large backup generation), and business operations may need over 99% uptime, but dwellings take half the stationary energy load and can more easily sustain uptime risks. 99% grid uptime would allow 88 hours/yr of no grid energy. However given the common time length of grid outages a modest battery would reduce 88 hrs to perhaps 10 hrs, thus allowing 99.89 % supply. A small genset could then provide energy to power a respirator if needed, freezer, internet router and portable tablet for the remaining 10 hrs.

      • Ian

        Exactly, electricity can be very cheap but to have 100% reliability this can become extremely expensive. It’s the last 1% of supply that can add huge cost to the grid. A good illustration may be an exponential curve. Very easy to achieve the first 50% of supply the next 25% would require more cost and effort and so on. To ask those who do not have critical electricity requirements to pay the same as those that do require very reliable power is problematic.

        The final 1% reliability may best be supplied by distributed FF generators!

      • Ian

        The opposite of reliability and continuity of supply would be failure of supply. Here are some of the circumstances leading to failure of supply:
        1. Equipment or transmission failure SA and Tasmania have demonstrated that sort of failure spectacularly. mismatches Coal, for all its baseload capability could never supply the variable load adequately
        3. Limitations in the generator-type profile. This is a huge sticking point for those that love the baseload profile of coal,
        4. Failure to supply an extraordinarily high load demand. In the past large sums of money were spent on gas generators to be held in reserve for their capacity factor in an attempt to supply this very occasional high demand.

        Others might like to add to this list.