Rooftop solar plays key role in reducing, deferring peak in heatwave

Print Friendly, PDF & Email

Rooftop solar was providing more capacity in South Australia than the old Northern coal generator at the height of Thursday’s peak. Solar pushed the peak grid demand into early evening.

share
Print Friendly, PDF & Email

Households and businesses played a key role in reducing peak demand and capping wholesale electricity costs in South Australia last week, with new data showing that rooftop solar played a major role in reducing and deferring demand peaks in the midst of the heatwave.

On Thursday, says Solar Citizens, rooftop solar was providing more capacity at the demand peak than would have been offered by the now closed Northern coal fired power station in Port Augusta.

Because of the rooftop solar capacity, and because a large chunk of it is consumed on site within the household or business that has installed the solar, grid demand (the part that needs to be actively managed by the grid operator) did not actually peak until after 6pm on both Thursday and Friday.

On Thursday, solar owners produced more power than the capacity of the now-closed Northern coal-fired power station in Port Augusta (see top graph, courtesy of Dylan McConnell of the Climate and Energy College) when grid demand would have peaked just after 3pm on Thursday without the state’s solar.

“Without solar, demand on the grid would have peaked just after 3pm on Thursday,” Dan Spencer, the South Australia campaigner for Solar Citizens said in a statement.

“Instead, rooftop solar provided more power than the now-closed Northern power station in Port Augusta, keeping the lights on with clean energy.”

And on Friday, he noted, rooftop solar reduced the peak by 358.6MW “far exceeding the capacity of the government’s new backup generators.”

Indeed, as energy analyst Simon Holmes a Court notes in this separate piece, the emergency back-up diesel generators ordered by the South Australia government have yet to be switched on, despite claims by federal energy minister Josh Frydenberg.

The results from the rooftop solar capacity in South Australia are not unusual. State network operators and grid owners have reported for the last few years that rooftop solar has pushed back the peaks from mid to late afternoon to early evening.

And, for the first time in more than a century of grid power supplies, it is the customers who are having a major influence on this – a phenomenon that is set to grow as rooftop solar and other “distributed generation” grows to near 50 per cent of all supply within a few decades.

“Everyday people generating their own clean power meant that peak demand was pushed back by more than 4 hours until after 7pm on Thursday and by nearly 3 hours on Friday,” Spencer wrote.

Spencer said the failure of Victorian coal plant Loy Yang B during the heatwave on Thursday highlights the need for South Australia to continue investing in renewable energy with storage.

“The failure of Victorian coal during the heatwave meant prices went up for South Australians. With a state election just around the corner we’re calling on all parties to increase South Australia’s renewable target to bring more local renewables with storage to SA.

“Along with raising South Australia’s renewables target, we’re calling on all South Australian politicians to do more to make cost-cutting rooftop solar accessible to all South Australians,” Spencer said.

“More accessible solar means that more people will be able to take back control of their electricity bills while helping to safeguard the grid.”

South Australia has some 730MW of rooftop solar across the state, and is about to become a major player in utility scale solar, with the opening of the 6MW Whyalla solar farm, and another 500MW being built or planned at Bungala (Port Augusta), Whyalla steel, and Tailem Bend.

Spencer has been a key campaigner for the Aurora solar thermal project, which will deliver 150MW of capacity, and more than 800MWh of storage, from a new solar tower and molten salt storage facility near Port Augusta.

The Port August office for US developer SolarReserve, is to be formally opened by premier Jay Weatherill on Tuesday.

 

Print Friendly, PDF & Email

44 Comments
  1. GlennM 10 months ago

    Brilliant, just need more solar and household batteries and the peak will be even lower. Roll on 2019

    • George Darroch 10 months ago

      Batteries and west-facing solar would definitely give that peak a big downwards push.

      • solarguy 10 months ago

        Sure would. In fact my solar is working a treat for me right now.

        • juxx0r 10 months ago

          Need to get three strings in an inverter happening, two is not enough any more

          • solarguy 10 months ago

            Hasn’t for a long time Jux.

          • juxx0r 10 months ago

            Can get three string inverters in the US, sadly not in AU.

          • solarguy 10 months ago

            Oh well, just have to get a second inverter and have an extra tracker to play with.

          • Mike Shackleton 10 months ago

            Or you could go down the micro inverter route!

          • solarguy 10 months ago

            Not a great fan of micro inverters Mike. Tigo DC optimisers can be used just for panels that will be shaded and does the same job and cheaper. MI’s are at least 35% more expensive per array, than a DC string inverter and if you want batteries later your shackled. Plus AC coupling is less efficient.

            There are other detractions that I haven’t got time to go into.

  2. Charles Hunter 10 months ago

    “South Australia has some 730MW of rooftop solar across the state, and is about to become a major player in utility scale solar, with the opening of the”

    The paragraph stops at “opening of the”. Should there be more?

    • Giles 10 months ago

      oops, this appears to be an earlier half complete version. We been having problems with site today. now update, hopefully. thanks for heads up.

  3. George Darroch 10 months ago

    Let’s not forget that this doesn’t just reduce the generation peak.

    It also reduces the transmission peak, saving us billions.

    • Andy Saunders 10 months ago

      …and the distribution peak, ditto.

  4. Chris Fraser 10 months ago

    Deferring the warm day peak also gives AEMO plenty of time to react for the evening load. Go Rooftop Solar … an unexpected partnership for better NEM management.

  5. Joe 10 months ago

    Why hasn’t the Joshie jumped on Solar being cause of that vicious recent price spike last Thursday. I mean he makes stuff when he feels like it to demonise the good state of SA and renewable energy. Why not give SA Solar a kicking.

  6. Andrew Inglis 10 months ago

    would all face west or would a range of orientations give a smoother power curve. surely must be afuntion of localized grid robustness and what if any storage capacity 7 rapid dispatch exists. as “stationery embedded energy supply is offered at “least cost price” is available then it comes down to all-in carbon pricing, and a redeinition of “base load”. The AUSRA work done at Lidell power station (per David Mills) circa 2009 and later in California (2013) suggests renewables with molten salt or large volume steam pressure storage back-up / smoothing will satisfy this along with a mix of wind, hydro and wood pellet fired Tri-gen or “waste “stream burning. again use of EV’s forms a large “battery storage on standby , after morning & evening travel commute. has anyone done the marginal cost study to see the difference. Such transparency of data will influence crucial investment decisions including power generation in built infrastructure and key national security planning. dispersal means a much harder target for anyone seeking major “disruption”. Dust storms, maintenance factors, materials failures?

    • Mark Roest 10 months ago

      Can someone come up with capital and levelized cost of the technologies mentioned? Batteries will keep falling in price; expect $US180/kWh capital cost within 2 years.

      • Feynman Anon 10 months ago

        Me too. And what, for instance, would the levelized cost of energy (LCOE) be after a %20 reduction in battery cost? (Cost of batteries being only 1 component of total system cost).

    • Greg Hudson 10 months ago

      I’ve read that wood pellet burning = even worse pollution than brown coal.
      Maybe not such a good idea ?

      • Mike Shackleton 10 months ago

        The best way to generate energy from wood waste is to produce a bioethanol. It means you can use other organic waste sources as well.

  7. Ray Miller 10 months ago

    A clue to our problems is the temperature sensitivity of the load on the NEM. Many of the problems (and peak load) stem from the heat ingress into our buildings coupled with the laws of physics on the operation of heat pumps.
    Without heat pumps many of our builds are hazardous on extreme heat days which is known and we also know how to make them less hazardous and consume less energy especially lowering the collective peak. Yet we hide form the real problem and the solution?
    If you really want to see a very positive and significant cost impact on the NEM, improve our overall efficiency and reduce the temperature sensitivity and by the way resilience will improve and just maybe also save many lives especially the frail.

    • Ian 10 months ago

      Good points Ray, but there is a serious difference between insulating a house against cold in winter to insulating it from heat in Summer. Any ingress of sunlight is converted to heat and any waste heat – like fridges, TV’s, cookers lights, human bodies will turn a well insulted house into a solar oven! If you open the windows then all your summer insulation is for nought, and you allow those hot winds into your house – damned if you do, damned if you don’t.

      • nakedChimp 10 months ago

        How many kWh does a normal house turn into heat (from consuming electricity) vs. how much solar heat (kW) is incoming during summer?
        There for sure is a >10 fold difference there.

        Just napkin: 10×10 roof = 100m2 .. this means the roof is heated with ~100kW.. during full sunshine hours.
        I dare you to match that with your normal home appliances all running at once. Normal houses are fitted with a 15kW connection at best per phase.
        So no, no insulation oven that cooks the tenants.

        Also, if you get some thermal mass into the walls that are insulated TO THE OUTSIDE, your house will become a buffer.
        This works all over Europe, just the new world(s), with their cheap stick construction has problems in this regard.

        • George Darroch 10 months ago

          Australia’s housing is idiotic.

          We spend huge amounts on energy (paid by tenants and uninformed home-owners) and almost nothing on making sure our existing housing stock doesn’t just pour energy in and out of the walls and windows. A few R3.5 batts under a dark metal roof and “she’ll be right”.

          • vibrantage 10 months ago

            Too true. We built a custom house with one of the mainstream builders and what a hassle it was. I wished we didn’t go down that route but we did. The builder was more interested in offering us “extra inclusions” like bigger rooms, better styling, high capacity a/c and list goes on. What we wanted was greater insulation, large eaves, thermally broken double glazed windows, ceiling fans in all rooms, evacuated tube (on a tilt frame) HWS plus other energy saving measures. WELL, it was like we were talking another language. In the end we threatened to walk away from this builder and it was only after this threat that we got all our energy saving measures built into our house. I’m glad we did too. On the couple of 40 degree days last week our home only heats up to the mid 20’s and with no a/c at all. We have to careful though and not leave doors/windows open etc.
            It just shows with a little planning you can build a house that is energy efficient and does not heat up like a rusty tin shed on a summers day.
            And the cost was not a great as you would imagine. About 15% of the building cost, since recovered, with the ROI.

          • MaxG 10 months ago

            Fully agree… and had the same experience, and ended up building myself!

        • Ian 10 months ago

          Good answer, not saying insulation should be avoided in hot areas, just saying that without managing the ingress of heat through windows you might end up being worse off than if you had a very thermally leaky house. Hettie is right of course – just shade windows so that direct sunlight does not come inside.

      • Hettie 10 months ago

        External shading of all windows, to exclude the sun from the Spring equinox Sept 22 to the Autumn, March 22, is pretty simple. On the north side, eaves of appropriate depth to the longitude will do the job. Pergolas with deciduous vines will enhance the effect.
        East and west, verandas with vines at the outer edge like blinds will block even the low morning and evening summer sun AND create a cool cavern for out door breakfast or dinner.
        It’s not rocket surgery.
        Construction and insulation requirements vary with the location, of course. What suits in Canberra would be horrible in Darwin, but these things are well known and documented.

    • Feynman Anon 10 months ago

      Ray, you say that “we know how to make them less hazardous” but you haven’t said how or pointed us to any resources. I’m reading this thinking that effectively you are hiding your vision of the real problem and solution from the rest of us. it’s ironic because I’m wondering what you think the “real problem and the solution” is?

      What I “know” and what you “know” might be two different things. The problem gets much worse with the reference to “the real problem”. Real problems presumably being different from just the apparent problem? It’s gets real interesting to compare different versions of real problems.

  8. Feynman Anon 10 months ago

    The article is lacking citations and sources.
    ** What is the source for the data behind the charts?
    ** Are they your charts — did you make the charts from tables of data or did you import the images from elsewhere?

    • Mike Shackleton 10 months ago

      See top graph, courtesy of Dylan McConnell of the Climate and Energy College

      • Feynman Anon 10 months ago

        So I take it you don’t know anything more than I do other than the graphs are said to be “courtesy of Dylan McConnell of the Climate and Energy College”??
        I’m hoping for a quality citation. Meaning a URL / internet link.

        • George Darroch 10 months ago

          What more do you want? Dylan McConnell of the University of Melbourne
          (https://theconversation.com/profiles/dylan-mcconnell-1602)
          put these together from published data from Australia’s energy agencies.

          If you don’t trust him, or think he’s lying, say it outright.

          • Feynman Anon 10 months ago

            What I want is a citation to the source. My editor will not allow me to use the information without it. At this point it seems clear that information will have to come from Giles Parkinson the author.
            I am hoping that the information is available without having to track down the sources, extract the data and generate my own graphs.

          • George Darroch 10 months ago

            There are plenty of links to the author. If you need to contact him directly your best bet is to contact Dylan McConnell by email, phone, or by tweet – he’s active on Twitter and today is a weekday at the university. You’ll probably get a reply promptly.

          • Giles 10 months ago

            As the people above have made clear, Dylan sources his data from AEMO.

          • Feynman Anon 10 months ago

            Giles, you aren’t going to say where the graphs used in your article came from?

          • Feynman Anon 10 months ago

            Sorry, meaning you won’t give a link or cite the name of the publication?

          • Giles 10 months ago

            What publication? This is contemporary data. It was sourced from Dylan McConnell from the Climate and Energy College, using AEMO data, which can be found on the AEMO website. Not sure can be much clearer.

        • BushAxe 10 months ago

          You get dispatch data from AEMO’s website and solar PV from AVPI.

  9. Ian 10 months ago

    Looking at these graphs, the demand curve is probably fairly typical of an average day, a minimum of 1500MW at about 4 or 5am and a maximum of 3200MW at 3 to 5 pm. Solar peaks at 11am to 2pm and produces about 600MW. You’d need another 1 GW of solar before the effect of optimum solar output flattens most of the midday peak – still plenty of scope for solar installations before the “duck curve problem starts to occur.

    Why exactly is there a demand of 1500MW at 4am? How much of this nighttime demand is a legacy effect of the Baseload- creating ‘off-peak tariff’. This was originally intended for discretionary loads, like water heating, to shift energy consumption to the night to match the unvarying output of coal power stations. If 1/2 of the minimum load is domestic water heating and this occurs most of the night, then shifting this load to the solar peak hours could allow another 1500MW of solar installs before demand dips below a new minimum baseload datum point of 750MW. That’s 1500MW plus 1000MW.= 2500MW additional solar before Baseload is compromised.

    By managing demand to encourage the daytime peak and discouraging the nighttime use of electricity, the sinusoid daily demand curve could possibly have a minimum of 750MW and a midday maximum of 4000MW . A huge expansion of solar from 730MW as currently installed, to a total of 3200MW would match this new demand curve very well without the need for much time-shifting battery storage.

    This little analysis would suggest demand management is critical to maximise solar installations and could be highly effective in creating a robust renewables based grid in this SA experience.

    Perhaps Dylan McConnell who kindly shared these graphs can do a little update with an analysis of the potential to modify demand superimposed on these graphs.

    • Mike Westerman 10 months ago

      Good analysis Ian. It illustrates the order of magnitude of storage that would make the system very robust. If there is 3100MW of solar, on a cloudy day the output would reduce to perhaps 600MW. Normally in SA when it’s overcast it’s also windy, so 2-3GW of wind will normally cover the demand. So in most situations, only several 100MW of storage would cover the over night demand if wind is not available (tho’ it normally is), support the evening peak without gas, and support exports to VIC. In the rare events when it is both overcast and still during the day, biodiesel or gas fired generation would meet the shortfall.

    • BushAxe 10 months ago

      This is a peak hot day for SA normally demand peaks at around 1500MW maybe a bit more depending on rooftop PV output, overnight dips to @1000MW with a brief peak at midnight when the HWS kick in. 1GW of utility solar, a 250MW PHES and HWS demand management would replace a huge chunk of gas generation. It would put SA up to about 80% renewables.

  10. derekbolton 10 months ago

    It would be interesting to turn these data into an estimate of the savings for customers. It is made tricky by the erratic relationship between demand and price. You need to ‘postdict’ what the price would have been if demand had not been suppressed by PV. I once did an analysis for NSW January 2015 and 2016 and found an exponential fit was reasonable, but you need a lot of data to get reliable parameters.
    With a simpler analysis, just using the data for one day in this post, I get $10m a day. The true figure would be quite a bit higher, and does not take into account the long term savings in infrastructure from a reduced peak.

Comments are closed.