Plunging cost of big solar in Australia

(AAP Image/Lukas Coch)

Australia still does not have a significant amount of utility-scale solar installed on a global scale, but its rapid short-term growth has already been enough to drive costs down faster than even the most optimistic project developers could have predicted.

Canadian Solar, one of the top five solar module manufacturers in the world, and a major project developer, last year predicted that the levellised cost of large-scale solar in Australia would fall by half from around $150/MWh in 2015 to $75/MWh in 2020.

canadian cost reductions 2015

This cost fall was likely to be helped by lower cost of finance, improvements in technology and capacity factors, and more efficient installation and operation.

But a range of initiatives, supported by the ACT government, the Australian Renewable Energy and the Clean Energy Finance Corporation, mean those achievements are happening more rapidly than expected.

Daniel Ruoss, the head of projects for Canadian Solar says Australia is now likely to reach that target by 2018.

canadian solar cost projections

His predictions appear to confirm the assessment of ACT energy minister Simon Corbell, who admits being taken aback by the prices offered in his government’s “next generation” renewable energy tender.

Corbell said earlier this week that the bids, filed for around 200MW of large-scale renewable energy capacity, showed that solar was already competitive with wind energy, which have won contracts with bids spread from $77/MWh to around $90/MWh.

Those levels are what are indicated in a graph presented by Ruoss at the Clean Energy Summit this week (box in green), and accords also with estimates by ARENA that project costs were already heading below $100/MWh.

It also confirms the assessment of Bloomberg New Energy Finance, that solar and wind energy are easily the cheapest form of new generation in Australia, a ranking that will be significant when Australia starts to replace its ageing thermal capacity over the next decade or two.

Ruoss credits the faster than expected cost reductions to the ground breaking investment of the ACT government, as well as the competitive tender run by ARENA, and the financial support from ARENA and the CEFC for the projects that have been built or are under construction.

He also says it is a buyer’s market because of an oversupply of panels and significant pricing pressure on balance-of-systems suppliers, and increased competition and capacity building in local EPCs (contractors).

Australia currently has around 270MW of operational large-scale solar projects – including the 103MW Nyngan project, the 52MW Broken Hill solar farm, the 57MW Moree solar project, and the 20MW Royalla solar farm – with others under contraction. All have been built under ARENA/CEFC programs or supported by the ACT.

ARENA has shortlisted another 20 projects with more than 700MW of capacity under its large-scale solar tender, and because the tender process has already elicited sharp falls in the cost of technology and the subsidy being asked, maybe half of these projects will get funding when the results are announced in the next few months.
Screen Shot 2016-07-26 at 11.40.13 am

In addition to this, there are numerous other projects in the pipeline or about to begin construction, including the first wind-solar hybrid project, to be built near Canberra by Goldwind, the first combined wind-solar-battery storage hybrid to be built by Windlab in north Queensland, and two solar-battery storage projects – one by Lyon Infrastructure in South Australia and another by Conergy in north Queensland.

And there are other mining-based projects, including with Sandfire Resources in Western Australia, and numerous off-grid hybrid projects such as the one in Coober Pedy and the completed solar farm at Rio Tinto’s Weipa mine.

On top of this, numerous investment vehicles are being created to invest in utility-scale solar, attracted by its falling costs and its guaranteed returns, given that the cost of generation is stable once the plants are built.

“The cost of delivering large-scale solar projects is just going down and down,” said Chris Lock, the head of Impact Investment Group, which has launched a $100 million solar fund.

“We’ve really got an asset class here that is a really predictable source of energy, and a really predictable cash flow over the next 20-30 years – not unlike fixed income or bond-like investments.”

Comments

16 responses to “Plunging cost of big solar in Australia”

  1. Malcolm M Avatar
    Malcolm M

    The big limitation with large-scale solar is its intermittency. If wind has been criticized for being intermittent, solar much more so. Its power feeds the grid at the same time as roof-top solar is depressing grid demand. Fortunately, most of the finalists are in Queensland and NSW, both of which have pumped storage hydro. Queensland has Wivenhoe (500 MW) and Kidston (~500 MW) under development, while NSW has Shoalhaven (240 MW) and Tumut 3 (600 MW pump, 1800 MW generation). These pumped storage hydros would then have 2 pump periods of ~10 am to 2 pm from solar and 2 am to 6 am from coal or wind. Build-out of solar capacity up to the capacity of these pumps plus grid demand should be relatively pain-free. Further development would require additional pumped storage or battery storage.

    1. GlennM Avatar
      GlennM

      Actually the numbers are far better than that, You do not need to have storage of 100% of your variable production, only enough to flatten the troughs. Therefore the storage you cite above would support 10X that much RE. Overseas they have reached 30-40% Wind without significant storage. NSW could go even higher.

      1. Chris B Avatar
        Chris B

        We have these wonderful machines called gas turbines that have already been built and installed on the grid, and in a lot of cases even amortized. Totally sunk costs. They can respond to demand in minutes and completely solve the intermittency problem. We’re talking about replacing big dirty coal that runs continuously, combined cycle gas plants that run when it’s cloudy. That’s an order of magnitude CO2 reduction even while we’re still burning fossilized algae.

        Solar benefits immensely from a few minutes of storage to ride through interruptions from cloud. Because grid operators hate it when a 100MW generator just drops dead without notice several times per day, they may start mandating it. The big upside of storage is that a “already paid for” solar inverter can also sell frequency control, spinning reserve and other ancillary services even at night and for quite big dollars.

      2. WR Avatar
        WR

        In a 100% renewable system where 80%+ of the generation would be coming from wind and solar, storage would mostly be used to meet evening peak demand periods. This would require a stored energy capacity of between 40%-50% of average daily energy demand and a maximum power capacity of 100% of the peak demand.

        Apart from meeting evening peaks, storage would also be used to occasionally address morning peaks and to smooth out small mismatches between supply and demand throughout the day.

        1. Analitik Avatar
          Analitik

          You need much more storage than that – there are periods of several days with very low wind. Add in some cloudy periods and the storage would need to hold about 200%-300% of daily usage as a minimum with demand shedding (700% would be required without periods of significant demand shedding)

          1. nakedChimp Avatar
            nakedChimp

            Solar is on par with wind now.. it will overtake wind.
            You will have more power during daylight from here on.
            Forget about wind as a ‘troublemaker’ in sunny Oz please.

          2. Alastair Leith Avatar
            Alastair Leith

            wind still heaps cheaper in Victoria, who will be building a lot of it to meet the VRET targets.

          3. TatuSaloranta Avatar
            TatuSaloranta

            Yes, but those events are better served by peaker plants, possibly powered by bio/waste-fuels, but on short term by gas; and eventually syngas from renewal energy. It makes no sense to have expensive storage to cover occasional 2-day low-production streak. So real storage needs can be capped based on economical value; and occasional shortages by thermal.

          4. sethdayal Avatar
            sethdayal

            Since solar plant must be backed up to 100% with inefficient peaker plant run inefficiently at most all it saves is a few pennies a kwh in fuel cost.

            Bio is a massive producer of GHG’s and deadly air pollution and wind/solar syngas is far more expensive than pumped hydro. Infrequent low wind/solar events can be weeks long adding a buck a kwh to consumer bills to pay for storage.South Australia power customers are now getting the shaft for their rather stupid reliance on wind power.

            China is accepting massive losses dumping second rate solar panels at a fraction of the cost of production just to keep cronies in business.,

            “an explosion of cheap, mainly Chinese-produced solar panels …….. the worst systems stopped working within 12 months, with others “falling apart” within two or three years”

            Google “Australian government orders solar install quality investigation”

          5. Alastair Leith Avatar
            Alastair Leith

            you think utility scale solar farms are running with anything less than A-class panels?

            It’s true that the bio-gas industry has issues, not least of which are a lack of feed stocks that don’t involve logging native forests. power2gas seems more likely to me. France commissioned a study of RE including 100% RE modelling and decided power2gas would serve as the storage of choice for them in addition to their existing 25% capacity Hydro.

          6. TatuSaloranta Avatar
            TatuSaloranta

            100% backed is only true by power rating (MW), but more relevant rating is actual relative sizes by energy produced by backup compared to primary (wind, solar). That is in low single-digit percents. The claim on weeks long events is not backed by evidence: even longer than 24h events are relatively rare, at least when generation is distributed along wide geographic area (such as is the case for Australia).

            Bio being GHG and pollution depends entirely on what is being burnt — when it is leftovers from forest industry (lumber, paper) it is very environmentally friendly, but opponents like to always dig up the least efficient methods (corn-based ethanol), then generalize.

            Finally, on Chinese panels; I am sure there are such things too, but existence tells very little on overall situation. I am not sure what that has to do with anything; high-quality panels exist and are not priced significantly higher to change actual cost calculations. But some buyers get greedy and hope that saving 10% is “free” without downside.

          7. Analitik Avatar
            Analitik

            “syngas” would just be another form of energy storage, albeit multiuse

          8. TatuSaloranta Avatar
            TatuSaloranta

            Yes, but it would be created using renewable energy, as opposed to exhaustible fossil fuels. So similar to batteries or hydro storage.

    2. Alastair Leith Avatar
      Alastair Leith

      it’s called variability, intermittency is the characteristic of dropping out, RE is very predictable over the short term and from day to day even which suits short term bidding markets just fine.

      And on variability and dispatch check this out:

  2. Brunel Avatar
    Brunel

    I wonder what currency these figures are in.

    Maybe everything should be in € or £.

    1. Chris B Avatar
      Chris B

      The bids are for an Australian tender and are in Australian dollars.

      The low bids are around 55USD/MWh, which adjusting for Australia’s lower insolation is about what Dubai is paying.

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