rss
29

AEMO: Shift to renewables is going to happen anyway

Print Friendly

Audrey Zibelman, the head of the Australian Energy Market Operator, has made a call for some sort of policy certainty, and new market mechanisms, repeating her view that the shift to renewable energy is unstoppable, but needs to be managed.

“The move to renewable energy is going to happen anyway, we just need to make sure we have the systems that produce the best outcomes for consumers,” Zibelman said at the AFR energy summit in Sydney on Monday.

Zibleman again cited the 21,000MW of wind and solat plants queuing for a place on the national electricity grid, but the “inevitable” transition she is speaking about is not on the shift in large scale generation – because that will depend on the shape of policy – but in small scale generation deployed by consumers.

Screen Shot 2017-10-10 at 11.44.32 am

She produced these three graph to illustrate her point and the challenges ahead. The first (above) represents the anticipated uptake of solar and storage by hosueholdes and businesses as they respond to the falling costs of those technologies, and the rising cost of grid power, which has reached absurdly high levels.

The new boom in household and business solar is already taking shape, heading to 1GW for the first time in 2017. AEMO’s figures, borrowed from last year’s CSIRO report, anticipate that to be repeated every year until 2030.

Battery storage will then follow.

The second graph (above) is how this uptake is transforming the grid and they way it is managed. This is the emerging “duck curve” in South Australia, or the “emu curve” as Zibelman sometimes calls it.

Screen Shot 2017-10-10 at 11.50.30 am

It is where AEMO predicts average midday demand will more than halve between now and 2030, from around 1800MW in 2009 (the top line in the graph) to below 700MW by 2030, all due to the huge uptake in rooftop solar.

AEMO has previously warned that “minimum” demand will fall to zero within a decade on certain days, and already it has experienced falls in demand to below 600MW.

The disconcerting aspect of this graph, however, seems to suggest that the electric hot water systems will remain untouched even out to 2030.

They were put there to make up for the deficiencies of inflexible baseload coal generators, but now that they have gone surely a smarter thing to do is to shift them to the midday “solar sponge” and introduce some flexibility into the way they are managed.

Screen Shot 2017-10-10 at 11.51.28 am

Flexibility will be the key, as this graph above illustrates, noting the huge fluctuations of wind power (the purple line) that occurred just recently in the same state. She said battery storage was important to minimise the “ramp rates” caused by the fluctuating renewables.

Flexibility is also at the top of the armory put together by Zibelman to cope with heat-waves this coming summer, where the greatest risk remains the unexpected failure of a major coal or gas generator.

Zibelman said she has assembled more than 1830MW to help deal with the summer peak loads, including around 830MW of previously mothballed gas capacity, the new Tesla big battery and 1,000MW of demand response from tenders managed by AEMO, ARENA and three state governments, details of which are also due to be released this week.

It is interesting to note that in Queensland, supposedly with the youngest and most ample fleet of fossil fuel generators in the country, has also issued a warning that households and businesses, such are its concerns about the ability of the grid to withstand heat waves.

The government has told consumers that households and business could be asked to set their air-conditioners at 26°C, and died loads such as pool pumped and hot water may also be varied to deal with the impacts of extreme heat, which caused coal and gas generators to lose capacity last summer.

This is not an issue about large scale renewables, because Queensland doesn’t have any to speak of (yet), although it will open its second large scale solar farm, the 100MW Ross River facility, some time over summer.

This is more about the capacibilities of an ageing grid. It is the issue that keeps the people at AEMO awake at night, and why Zibelman says she cannot give a cast-iron guarantee that there will be no outages this summer.

  

Share this:

  • trackdaze

    Midday demand won’t drop as much with energy storage uptake.

    • Charles Hunter

      Firstly, I think the graph shows that the peak demands to worry about are in the late afternoon and evening rather than around midday. Secondly, the extent to which midday demand is affected will depend on the amount of storage per household. For example, I’ve got 4kWh of batteries on a 5kW inverter with 6kW of panels (ie super sized). The batteries take almost bang on 3 hours to charge regardless of the time of year (around the winter solstice the batteries are fully charged by 10am EST; the summer solstice by 9am EDT). If I doubled my battery capacity to 8kWh (which is not yet cost effective) they’d still be charged by midday in summer and would still probably have some charge left by sunrise the following day (ie the time to full charge the next day would be shorter). I suspect that I’d need to be into a two PowerWall configuration and have had a run of cloudy days before there was any danger of anything affecting summer midday demand one way or the other. Thirdly, in my system at least, the batteries do not sit idle once even partially charged. I only ever draw from the grid where there is insufficient PV+battery to satisfy the house demand. If I had a midday load which exceeded what the panels could support, the batteries would be tapped to make up the difference before any demand was placed on the grid. Assuming similar behaviour for a large number of domestic systems it seems to me as if the uptake of storage should cause aggregate demand to DROP further than it would if storage was not installed widely. Yes/no?

      • AllanO

        No (if you’re talking about total energy drawn from the grid), because as far as centralised grid demand is concerned it doesn’t matter whether you have a battery to store and use your own excess solar production in house, or lacking a battery just export excess solar production for your neighbours to use. Net energy drawn from centralised sources is the same, although the timing changes. In fact, for a given aggregate amount of rooftop PV, centralised energy use might even be slightly higher with batteries if the recharge / discharge cycle losses are higher than the losses involved in exporting and distributing surplus solar within the distribution system.

        But the shape of the demand curve will change significantly and possibly maximum grid demands will fall as batteries supply some of that peak.

        • Francesco Nicoletti

          Your assuming the number of solar cells on the roof remain the same once a battery is added. In the current situation of high grid prices and low and falling solar cell prices, it might make economic sense to buy cells to cover the night time power consumption. With the aim of say using the grid only after two days cloud cover or other unexpected events.

          • Mike Shackleton

            That’s right, around 2020, 10 years after rooftop solar took off, we’re going to see existing rooftop systems being augmented as inverters start to die and maybe the occasional panel needs replacing. If you’re limited to 5 kW export to the grid, you’ll add as much as possible, plus install a battery so that anything over the 5kW limit goes straight to the battery and you continue to receive FiTs to help pay down the system.

      • trackdaze

        Would depend on charging protocols and any price incentives around peaks troughs.

      • Ian

        The duck curve is only partly based on facts , that is up to 2016 figures. The rest is conjecture. The deep midday belly of the duck is for 2030 – lots can happen in 13 years. The belly represents the fear that excessive solar ( namely, distributed solar) will ‘gut’ the midday grid market. For starters home storage could, most probably, become as ubiquitous as roof top solar. This will mean one thing: those households will be able to manage their demand at will. They will prioritise loads like pool pumping, hot water heating, even air conditioning to solar producing times and will maximise self- consumption of electricity, their contribution to the grid demand profile will not be a duck curve but a flat line – much like an ecg of a dead person. Anyone for a death- spiral? The utilities need to heed this ancient warning: “If you cling to your life, you will lose it, and if you let your life go, you will save it.”

        • Mike Westerman

          Ian your response is only partly based on facts: the midday belly is the result of Californian experience plus modeling in Australia by AEMO (and Prof Grace in the SWIS), certainly not “conjecture”. Their latest modeling is suggesting zero prices as early as 2025 in SA. Obviously there is considerable uncertainty as to how fast demand will shift to soak up excess capacity. However, it is reasonable to assume shifts in demand will be slower than capacity is being added because supply has a very strong incentive (in SA grid at >30c, LCOE solar <10c), whereas the incentive to shift demand is limited by capital costs to do so.

          • Ian

            The AEMO have obviously lifted their graph directly from the Californian experience. That is where the duck curve got its name. If behind the meter solar is part of the demand response side of the energy equation, then utility scale solar must be part of the electricity supply side of the same equation. If that’s the case then utility solar will not contribute to demand and will not be part of the duck curve. It will, however , be producing electricity at exactly the time when demand is at a minimum. Which could be a little problem for utility solar.

          • Mike Westerman

            I don’t think you can say anybody has “lifted” anything from anybody, though I’m sure AEMO are not so dull as to ignore what happens in other jurisdictions. Utility solar is going to take some time to catch up, but well before it does it will face the headwind of falling daytime power prices. The only way to make it pay once this occurs will be the Kidston solution, where solar is behind the meter and fed onto the PHES bus. It will only be exported when the upper reservoir is full, or when power prices are temporarily high due to plant failure elsewhere.

          • Ian

            Are you saying that utility solar will need to be coupled with storage such as PHES before it’s even sent through to the grid ?

          • Mike Westerman

            No, I’m saying it is a strategy to prevent utility solar from becoming curtailed due to low prices.

        • Jon Albiez

          Economics is likely to drive usage during the middle of the day. I already boost our hot water and run the dishwasher during the day to maximise self generation and limit imports. If people were more aware of spot pricing and charged accordingly we would likely see strong shifts in the market. One additional area I feel may become more useful with high levels of intermittent generation is flywheels and resistor banks to maintain frequency in a system with lower levels of inertia and synchronous generation.

      • Greg

        I’m thinking of putting in batteries. What batteries did you decide on and what are your round-trip (charge/discharge) losses? Have you been able to measure them? Thanks.

        • Charles Hunter

          My inverter is a SolaX Hybrid which manages the batteries and also implements emergency power supply facilities.

          I have two PylonTech Extra 2000 LFP batteries. They are nominally rated at 2.4kWh per battery (ie 4.8kWh in total) with an 80% depth of discharge. A full charge therefore represents a capacity of 3.84kWh at a maximum discharge rate of 2000VA.

          The batteries were commissioned on 10/Dec/2015 (676 days ago). In that time, 3010kWh of PV production has been consumed charging the batteries (averaging 4.45kWh/day), and the batteries have produced 2739kWh (an average of 4.05kWh/day). You can look at it from the perspective of either 3010/2739≈110% (ie you need to put in 10% more than you get out) or 2739/3010≈91% (you only get back 91% of what you put in).

          That the average daily discharge of 4.05kWh/day is more than the at-first-glance maximum of 3.84kWh/day gets back to what I was alluding to before, which is that batteries don’t just charge in the morning and sit there all day doing nothing until the sun goes down. The inverter satisfies house demand from PV first, then draws from the batteries, and only after that is energy taken from the grid. So, my house, with batteries, creates less grid demand while the sun is shining than it would without batteries.

          I hope that answers your questions.

          • Greg

            Thanks Charles. Excellent information. very much appreciated.

  • Joe

    “The move to renewable energy is going to happen anyway”…of course it is…. BUT try telling that to The COALition, the Rupert and all those radioland cowboys.

  • BushAxe

    A GW of utility solar in SA would make that graph even more interesting!

  • DogzOwn

    DMR to cycle off aircon is met with fierce protest. Yet looking for new aircon, bearing in mind recent report from ANU to expect heatwaves into low 50’s degC, looks like R32 aircon specs say max ambient 46degC so protest useless and inherent DMR mechanism already.

    Does industrial scale aircon max out past mid 40’s degC? Will pubs have cold beer?

    • Mike Westerman

      Industrial aircon have means of controlling head pressures to avoid tripping at high ambient temperatures, and units with cooling towers will possibly not be faced with higher WB temperatures so will simply see high ambients as more load. Either way, they may well max out in capacity so that spaces including freezers and coolers, won’t maintain design temperatures! So keep an icebox handy! And if you are worried about your aircon, put more panels on the roof and install a mist spray on the condenser.

  • Greg

    renewables aren’t yet reliable, and the battery options reduce the overall efficiency. Diversity will continue to be key. Coal, Gas, (and even nuclear) still have a part to play.

    • Mike Westerman

      Greg hydro plants typically are considerably more “reliable” than either gas or coal, and my rooftop solar is even more “reliable”, even on overcast days dutifully putting out some power, in a very predictable manner. A bit of research would show wind turbines are also right up there in terms of reliability, in excess of gas and coal, which being thermal plant, require periodic shutdown to replace components exposed to high temperatures, thermal cycling and aggressive internal environments.

      You are correct that diversity will continue to be key: geographical diversity in hydro, wind and solar, with DMR being properly rewarded as part of the mix.

      • Greg

        Thanks mike. I guess my comments were more about the energy source. wind only works when its blowing (climate change might actually help this), solar only works during the day (cloud does have a big impact), and hydro doesn’t work during droughts (ref Central Africa these last few years). Everything needs maintenance. even your solar panels could do with a clean at least once a year. Most plant is designed with maintenance in mind (eg redundancy, etc). Geographical diversity is a great idea for intermittent renewables. Super-grids are what is in view here. even inter-continental grid connections. but this is big dollars and a long time out. batteries are a short term answer but the cost is a loss in efficiency. Renewables (regretably) still has a very long way to go to improve reliability and reduce cost. In the meantime, it has to be coal, gas, etc. I’d like to see a lot more effort being put into exploring geo-thermal, as the potential here is enormous.

        • Mike Westerman

          Sorry if I came across as pedantic Greg but there is much obfuscation and inaccuracy coming from those who should lead that I believe it helps to be accurate. Yes wind only works when it’s blowing, but if you look at the stats for a particular good wind location you realise that calm days are infrequent. Amalgamated over long distances, say eastern Vic to the Flinders and they are very infrequent indeed. We have long weather records and good models that predict very well when these periods will occur. Likewise with solar in arid regions – very predictable. With predictions it is then not a big step to find the economically optimal level of capacity in intermittent sources so as to have firm power. Add pumped hydro into the mix (which is not drought affected or need be on a flowing river) plus behind the meter battery storage and excess intermittent power is stored to obviate peak demand periods. All of this is doable at very much less than the cost of gas, geothermal, nuclear and HELE coal. As Finkel has shown, the least cost path is to manage a transition as soon as possible, before aging coal plant start failing, as they have done in recent summers.

          • Ian

            Good concise answer! Talking of reliability, your points regarding the excellent reliability of wind and solar devices within the resource constraints is very pertinent. The storage devices such has hydro and batteries would also have extremely high equipment reliability advantages.

            In a future world where fossil fuel use is very much reduced to possibly just a standby function for the grid, you’d wonder if supply and storage of these fuels would be reliable at all. Consider this scenario. Grid domestic and industrial supply and transportation are for day to day purposes are run on wind and solar resources. Fossil fuel is retained for standby purposes only, perhaps called on to supply the whole grid and transport needs for a maximum period of three weeks long once or twice every two years. This would be an enormous amount of gas or liquid fuel stored for long periods. The industry supplying this would have atrophied to such an extent that there would be severe supply constraints – could fossil fuels reliably supply the whole grid for such short bursts without supplying it at other times. To be viable this industry needs a constant flow of fuel from the ground deposits to the burner.

          • Mike Westerman

            The only viable FF for standby would be LPG. Diesel “goes off” if stored without use for a length of time, becoming infected with a bacteria. A coal plant would need to be mothballed making recommissioning impractical. Gas including biogas would take up too much space. I would think the alternative of pumped hydro would be more attractive on every count.

    • Greg Hudson

      When was the last time you saw nuclear in the mix here in Australia? Doh… NEVER. Troll fodder.

      • Greg

        It is a natural power source with zero carbon emissions. think about it. coal currently supplies 80% of australia’s power needs. renewables are only now making a small dent at huge cost. solar only works a third of the time at best. australia already has a nuclear industry. there are several mines and concentrators and one reactor. perhaps a good place to put a new nuclear power station could be (eg) maralinga, which is already contaminated. another zero carbon energy source worthy of more serious attention is geothermal.

        • Greg Hudson

          Troll