AEMO explains caution on S.A. wind: We’re first in the world

One of the first fund managers to specialise in clean energy and ethical investments in Australia used to explain the challenge of his job like this: “It’s great to be ahead of the pack, buy you don’t want to be too far out front.”

It appears that the Australian Energy Market Operator has adopted a similar view. In explaining its recent caution on South Australia, and new rules that require new levels of gas generation, AEMO says South Australia’s grid is the first first large-scale power system in the world to deal with minimum quantities of synchronous machines.

In a new report explaining its actions, AEMO compares South Australia with other countries with high wind penetration – Denmark, Germany and Ireland, as well as the US state of Texas.

It says the penetration of wind in South Australia – 120 per cent of demand  on occasions – is far beyond any other, particularly those in “isolated” grids like Ireland and Texas.

And it uses this graph below to illustrate how. (The big blue bar indicates the maximum penetration of wind in the local grid).

Screen Shot 2017-09-08 at 9.42.35 AM copy

The implication is that AEMO – much like the rest of the industry – is forced to learn as they are going along. And because it has the ultimate responsibility of keeping the lights on, and got a B minus last summer, the grid operator is taking an ultra-cautious approach.

In early July, AEMO began implementing new rules that require a minimum about of gas generators to operate at all times, and at least four to operate when the wind output goes above 1200MW (although that can vary depending not the characteristics of the plant available).

If they don’t get that number of gas turbines, then wind will be curtailed back to 1200MW. It’s already happened on several occasions in the last two months, most notably earlier this week when the 1700MW of wind farms had their output clipped for three days running, delivering this extraordinary vision below.

renewable graph.

The move has caused surprise in the industry because it was unexpected, does not relate to the common issue of frequency and inertia, and was not directly prompted by the high profile “system black” last September or the equally controversial load shedding in February.

AEMO says it is an issue about system strength, which it says it is weakened by the lack of “synchronous generation”.

But many in the renewables industry think it’s a pretext for papering over some of the deficiencies in the state’s gas generators, and issues elsewhere in the network. In other words, they suspect that the issues would have emerged whether wind was in the system in great quantities or not.

AEMO is citing November 13 as the day of its epiphany about system strength. Nothing much happened that day, except that South Australia had a high level of wind generation, but only one gas unit from AGL’s Torrens B facility operating.

AEMO says the grid performance that day was satisfactory, but apparently it was spooked by what might have happened if that one unit tripped.

It deemed the situation to be “insecure”, and sought ways to boost “system strength” by having a minimum amount of “synchronous generators” – which in South Australia means gas-fired units.

Indeed the new rules, which were explained to industry participants in a telephone hook-up on Friday morning, after the release of a paper on Thursday, appear to be designed to protect the system in case of an unexpected outage of one of the state’s ageing gas generators.

Since the state-wide blackout there has been a increased focus on the performance of old coal and gas generators, and the so-called governor controls, which dictate how quickly the big synchronous plants can respond to system faults.

There is widespread concern, identified by the wind industry and echoed in the Finkel Review, that lax governor controls are weakening the strength of the grid and its ability to respond.

One observer said that if system strength  was such an issue, then why wasn’t more made of it in the study into the blackout. There is a suspicion that the issue is in fact related to the ability of gas generators to manage frequency and inertia, and their ability to do so may not be as strong as is assumed.

The impacts of the new rules imposed by AEMO will be several:

One, the need for multiple gas generators to be on line at times of high wind generation means that the wholesale price of power in the state will not fall as it usually does when there is a lot of wind. The price will continue to be set by gas, and that means it won’t fall far, barring constraints that stop exports to Victoria.

It also means that excess wind output – above 1200MW from the state’s 1700MW wind farms will be constrained, as has happened on several occasions already since the rules were first imposed in early July.

And it could have impacts on future installations. Already, a new 220MW solar farm is being built near Port Augusta and a new 215MW wind farm will follow soon. Numerous other large scale wind and solar projects are planned.

(It may at least partially explain the state government’s increased enthusiasm for hydrogen technologies to use the excess output from wind and solar farms. It announced a new tender today).

In its paper, AEMO says it will continually review its needs, and its modelling, and the new rules, to take into account any new installations and the emergence of new technologies, or the prospect of using existing technologies like synchronous condensers.

And it promises to be a vigorous debate, at least behind the scenes. There are strong differences of views about the fine details of electrical engineering and how it should apply in this situation – not surprising given the system is moving into virgin territory.

But the one aspect that frustrates many is the lack of independent engineering review – for the moment, most positions are coloured by the respective world views on variable renewables and traditional generation. It promises to be an issue of culture as well as technology.

 

 

Comments

53 responses to “AEMO explains caution on S.A. wind: We’re first in the world”

  1. Mike Westerman Avatar
    Mike Westerman

    It’s hard to believe that with 1200MW of wind and exports to Vic that the issue could be reserve. If it is inertia/reactive power control surely it would be cheaper and avoid a price impact if unfired gas generators were decoupled and run as synch cons.

    1. Peter F Avatar
      Peter F

      worth examining even better the old TI steam generators

    2. BushAxe Avatar
      BushAxe

      ETSA used to do that with Snuggery PS in the South East before privatisation.

      1. Mike Westerman Avatar
        Mike Westerman

        I hate to seem conspiratorial but it makes you wonder doesn’t it.

  2. Peter F Avatar
    Peter F

    Well “who’d a thought it?”

    They finally wake up after at least 20 years of academic papers, gas turbines are virtually useless for frequency support and for a big loss of generation actually have a negative response .i.e. their output initially goes down as frequency falls not up, to make up the shortfall,

    I suspect the SA government and customers would rather have a short term increase in power prices rather than more blackouts.

    1. Rod Avatar
      Rod

      Do you mean OC or CC turbines?
      Pretty sure the CC turbines offer plenty, if their governors are left to do what they are meant to?
      What really irks me about this is that wind farms are paying for FF unreliability.

      1. Peter F Avatar
        Peter F

        CC are better but not a huge amount, eventually when we fit most of the wind farms with synthetic inertia software and the batteries and solar thermal plants come on line the constraint can be relaxed if not eliminated

        1. Rod Avatar
          Rod

          I worked at TIPS for a while and those turbines are big heavy suckers with lots of physical inertia one would think.
          I can’t help but think fiddling with the governor response (probably for financial reasons) is at fault here somewhere.
          I will be happy if the Tesla battery causes a rethink about the system strength but to wait much longer will be costing SA untold millions.

      2. Richard Avatar
        Richard

        Except that renewables created the unreliability in the first place

        1. solarguy Avatar
          solarguy

          Oh, please tell us how?

          1. Richard Avatar
            Richard

            I love renewable and I’m all for 100% renewable. But I also realise that penetration of intermittent resources has taken away the viability of base load coal fired generators. In a privatized energy system this has created an unstable situation.
            You can argue about who is to blame and the failure of policy is probably the major issue but the reality is that if renewable hadn’t come along we would still be operating a pretty reliable cheap energy system.

          2. Rod Avatar
            Rod

            Except all those “cheap” assets wear out eventually and replacement cost is well above new renewables.

          3. Richard Avatar
            Richard

            Yes that is true and it doesn’t account for the cost of fossil pollution either.
            I guess we are in that awkward transition period. But most of the experts expect power prices to peak in the next year or two and then start declining. Hopefully, it will be a long decline as more and better renewable generation comes on line.

          4. Rod Avatar
            Rod

            These guys have done a lot of work on how we can make that transition at a similar cost to BAU. We just need Federal direction to take the awkward out.
            http://bze.org.au/

          5. solarguy Avatar
            solarguy

            Mate, I agree that policy is the major issue, but a privatized energy system is never cheap. Once that occurs they want the ROI of their investment over anything else and are prepared to use FF lobby groups and a FF paid off federal government willing to use lies and misinformation to protect that investment. That stops sensible policy and cooperation from the private generators who will also withhold capacity to gouge higher prices and blame RE as a scape goat.

            Make no mistake this a war of vested interests and truth is the main casualty. The collateral damage are the consumers.

          6. Richard Avatar
            Richard

            I have to agree with that. Privatising power was the dumbest thing Australian governments ever did. I’m sure the pollies and bureaucrats who gifted them the golden goose have some very nice plumb jobs at the top of power companies in Oz.
            We are just as corrupt as any tin pot third world regime.

          7. solarguy Avatar
            solarguy

            Privatising is mindless, they did it to pay off deficits, but lost a source of revenue. After they have pissed the benefits to the coffers up against the wall and into deficit again, they’ll need a magic wand, otherwise known as more taxation. Like we have an unlimited capacity to pay. The bloody morons!

          8. Chris Baker Avatar
            Chris Baker

            Hi Richard, another important consideration is governor dead bands. The relaxation of dead bands came as part of creating a market out of a natural monopoly, and is unrelated to renewable energy. Many generators have turned their governors off altogether, partly because the market operator fines them if they generate more than they are instructed to (when they were just responding to a frequency excursion and helping keep the system stable). In your later comments I see that you are not a supporter of privatisation, and the dead band issue may cement this even more. I think it would be closer to the truth to say that if the market had not come along we’d be still operating a reliable cheap energy system, including high penetration of renewables.

          9. Richard Avatar
            Richard

            Absolutely. However, I doubt that a government system would have been any quicker to move on renewables. It would have kept the prices down and the power on but it wouldn’t have helped in the transition, especially with the Libs periodically taking the reigns. As we have seen they probably would have built “clean” coal plants if they had control of the network.
            I actually think that the private system without a plan, we have atm will work to drive renewables far quicker than otherwise. Unfortunately, that will result in unnecessarily high prices in the short term.

          10. Chris Baker Avatar
            Chris Baker

            Maybe, maybe not. What we need for high renewable penetration is storage, and for significant storage we need pumped hydro. The action in the commercial sector we see is from Genex in north queensland, struggling though to finance the project because the market does not provide a clear signal about the value of storage. The other clear action is the Libs with the Snowy 2 initiative. They are pretty serious about it and currently spending $29 million on the planning phase. So this is government taking action outside the market, and surprisingly a Liberal government. How much quicker would it have happened if it was a fully nationalised operation? In China, an example of a government system, in 2016 they had 26 pumped hydro plants under construction for a total output of 32 GW which is about equal to the entire fossil fuel generation capability in Australia.

          11. daroiD8ungais7 Avatar
            daroiD8ungais7

            Chris, wouldn’t buying a bunch of synchronous condensers (big spinning lumps of mass) and chucking em on the
            grid also be a pretty quick and cheap way to solve the lack of inertia problems AEMO is trying to solve with these gas gen rules? Would expect it also put a fair dent in the +$30M pa spent on FCAS and
            regulation currently.

          12. Chris Baker Avatar
            Chris Baker

            you’re probably right. And maybe you don’t even have to buy them, just spin up some of the unused turbines already available. But also be sure that the governors are doing some work as well and not just sitting idly by as the system collapses.

          13. Rod Avatar
            Rod

            Thanks for the explanation on what is happening with the governors.
            I wonder if that had a hand in the State Black.

            I think Governor settings were one of Finkel’s or maybe AEMO’s complaints. In that they didn’t know what the settings were!

        2. Rod Avatar
          Rod

          I can live with the word “intermittent” but not “unreliable”
          My first lot of solar panels are 18 years old and still going strong.

          If you mean wind farm failures due to hurricanes then you have lost me.

          1. Richard Avatar
            Richard

            No, I meant overall system reliability. In the privatised system we have, renewables are driving coal stations and gas plants out of business and atm they are still the backbone of the system. I’m not saying that putting fossil out of business is bad, I’m all for it. But we need to manage it properly, the economy depends on it.

  3. David leitch Avatar
    David leitch

    what determines which wind generators are curtailed? Is it the pool price bid stack? It’s surelly creating winners and losers.

    1. Paul McArdle Avatar

      It’s a combination of the factors that apply to each DUID on the Left-Hand Side of a constraint equation, and the bid prices submitted by each generator (effectively*) for that dispatch interval (actually the trading period).

      For the SA_WIND_1200_AUTO constraint (one of the topical ones) the factors for all the wind farms on the LHS are all 1, so that means being constrained down in relation to bid price. Seems some sleepy wind farm generators initially did not have bids at exactly -$1,000/MWh at the RRN so were bearing the brunt of the constraint for a while…

      Snapshot here of the Constraint Equation Details widget in ez2view (www.ez2view.info) explains a bit.

      More on WattClarity next week, I hope (ran out of time).

      https://uploads.disquscdn.com/images/d560dfc06ab56031caac9cc8e8aa8f314e2d31bf861e9957687711546c7a6e19.png

      1. David leitch Avatar
        David leitch

        Thanks Paul.

    2. BushAxe Avatar
      BushAxe

      Lately they’ve all been limited to about 65-70% instead of curtailing the higher bidders.

      1. Greg Hudson Avatar
        Greg Hudson

        Personally, I still fail to see the logic of curtailing, when they could be exporting over the interconnector to Vic. To me, it’s totally illogical.

        1. Rod Avatar
          Rod

          We have an election next March and the current Govt. is very nervous about any more hiccups. Makes a little more sense when you view it from that angle.

          1. Greg Hudson Avatar
            Greg Hudson

            Maybe you are correct, but the AEMO is supposedly at arms length from Govt stupidity (well maybe) 😉

  4. AndrewATA Avatar
    AndrewATA

    Why weren’t the turbines from Northern Power station converted into a synchronous condenser? ETSA considered the idea, but said they didn’t have the authority.
    Too little planning, too late.

    1. Rod Avatar
      Rod

      There will soon be two spare turbines at TIPs A. Hopefully they will make use of these.

  5. David leitch Avatar
    David leitch

    Also Denmark manages a higher level of wind beause it can export and import easily. More transmission is surely part of the answer.

    It’s increasingly obvious that there are straightforward technical solutions to frequency control so it’s more likely to be a power/energy issue. I.e. The gas just can’t start fast enough to compensate for wind stopping. And it costs a lot to start and stop gas gens . I think it’s about 4 hours to run Osborne up from zero to full output for instance. Modern combined cycle gas units such as some in Texas can ramp up much faster, 30-40 minutes from memory.

    1. solarguy Avatar
      solarguy

      I think more storage is possibly the answer, cause as it is SA has SFA at the moment. Isn’t that why their putting in a 100MW battery?

      After all battery storage is working for me at a household level and very well in fact.

      1. Rod Avatar
        Rod

        There is a guy on another forum who was part of the recent trial at Hornsdale to help with FCAS. If/once this gets AEMO approval they may relax a bit.
        He also suggested giving AEMO control of the FCAS provided by the battery might help too.

        1. solarguy Avatar
          solarguy

          Thanks Rod. And when they get more storage it should become a sure thing. Going forward to 100% RE, I wouldn’t like the head ache of working out how much storage will be required for SA, but there is no doubt that it’s achievable.

          1. Rod Avatar
            Rod

            I’m not a fan on spending billions on an interconnector, but that is looking more and more like it is needed as well.
            The AEMO report said the DC interconnector is no use for system security. I’m not knowledgeable enough to understand why that would be.

      2. Greg Hudson Avatar
        Greg Hudson

        G’Day Solarguy. I’m in the process of analyzing my last years worth of power data on an hour by hour basis so I can ‘see’ how much money I could potentially save by installing a battery in the future. Maybe you have done this already and could give me some hints please ?

        I’m looking at installing a 5 to 10 kW PV array (still undecided), giving me 20 to 40 hours of generation time per day (average for Melb), a Tesla PowerWall 2, and allowing for the future arrival of a Tesla Model 3.

        Ideally I need to cover all my usage, fill the PW, and charge the car during the day, then export some to cover the grid charges (approx $1/day

        Your opinion/comments would be appreciated. Thanks.

        1. solarguy Avatar
          solarguy

          G’day Greg. To give you a full opinion would take a great deal of time and I don’t give out free information in that detail. However, I’m happy to give you some tips.

          If you have a true smart meter you can request that info from your retailer, otherwise you’ll need a note pad and time. What you need to find out is how much power you consume in certain time periods, from sun up to sun down and from sundown to bed time, then to sun up, all with normal loads running. This will tell you how many kwh in those time periods your using.

          Then some testing needs to be done to find out what heavy loads like A/C’s use over 2 hours, by turning them on separately and recording what their using in the fist hr and in the second hr. You will find with A/C they use the most power in the first hr of running.

          With things like toasters, kettles, TV’s the label is a good reliable indication of what they will use, including fridges to a point. Once you have this info you can plan what size PV you will need to drive all of the loads from base line to max loads and what size inverter you will need. Battery size, who knows. A trend these days is to over size PV and go less on the batteries as it’s cheaper.

          From the little info you gave me of what you want to achieve, I would be going 10kw PV if not bigger in reference to the model 3 charging. As the PW2 is just a battery and needs the correct inverter, I would be looking at another battery and an Imeon 3phase inverter which can output 12kw. Of course you will need a 3phase grid connection.

          In ending yes this is all time consuming but worth it and costs nothing, but to get a system that can do all you want be prepared to spend big, up to $40k perhaps. If you can afford an amount like this, then you can contact my colleague Glen Morris from Solarquip, near Melbourne. Just don’t expect free info.

  6. BushAxe Avatar
    BushAxe

    Will be interesting to see how they view the capabilities of Bungala Solar with 300MW of inverters and battery storage.

  7. JohnRD Avatar
    JohnRD

    The new solar farm (solar tower) near Pt Augusta produces power using a standard steam powered generator. This means that it will have similar characteristics in terms of synchronization, inertia and grid stabilization to a coal fired power plant. In terms of the grid it will allow more wind or solar PV to be used.
    Australia should be planning to build a lot more solar towers using molten salt heat storage with back-up salt heating.

    1. Mike Shackleton Avatar
      Mike Shackleton

      The new Tesla battery will also be able to provide short term backup (long enough for a gas unit(s) to come back online if required). That battery is supposed to be up and running in December I believe.

  8. Chris Baker Avatar
    Chris Baker

    That graph is indeed extraordinary. When I see that solid green band on the graph I can’t help thinking that sure looks like a base load power station.

    Clearly AEMO has the power to instruct the wind farms to limit their generation, and they clearly have the ability to respond quite accurately.

    Given that, why are AEMO being so dumb about it and just flatlining when they could be following the demand curve with the wind?

    Just eyeballing it the expensive gas fired power varies from around 300-400MW to as high as about 1100 MW. The proportion of synchronous generation varies from around 25 to 50% of demand.

    Why not keep the synchronous generators at a steady proportion of the total, so they still meet their system strength objectives, and have the wind do the bulk of the load following?

    Maybe its all too new for AEMO to get their head around, but this sure looks like a great opportunity to make better use of curtailed wind rather than this blunt instrument of a fixed constraint, unrelated to demand.

    1. Goldie444 Avatar
      Goldie444

      Good questions.

  9. Malcolm M Avatar
    Malcolm M

    There is another limit on wind output, which is the 500 MW export limit from SA to Vic.
    The Tungkillo-Heywood line has a thermal capacity of 1150 MW, but flows are constrained to avoid frequency issues. Why not convert this line to HVDC, then there would be no problems of SA becoming desynchronised from Victoria ? There would also be about a 30% increase in the thermal capacity of the line to ~1500 MW through conversion to DC. This upgrade is surely cheaper than other proposed links to NSW and Horsham.

  10. Richard Avatar
    Richard

    Even more reason to do it yourself and leave the grid or export more than you import.

  11. Richard Avatar
    Richard

    The great thing about this is that it will encourage wind farms to put in storage!

  12. Ray Miller Avatar
    Ray Miller

    Australian engineering? Missing in action? Yes good point Giles, the lack of quality engineering is what we all are paying for now. Instead of our top engineers being all over the new system format, probing all facets of both the new and old technology to may it play nicely, what do we get? Just suck it and see approach, while asleep at the wheel. This approach in not going to be a successful strategy.
    We let the outdated unreliable technologies be propped up with band-aids and poor standards while we spend all our resources on accountants and legals to see what they can extract out of the NEM for private benefit.

  13. howardpatr Avatar
    howardpatr

    It seems that if South Australia was so much on the ball they would have announced plans to bring ViZn battery technology for evaluation.

    https://www.viznenergy.com/

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