Comparative costs of power

In November last year a group of consultants tabled a report in the NSW Parliament with the results of some elaborate modelling work to generate  the  total System Levelised Cost of Energy (SLCOE)  which is defined as —

...the average cost of producing electric energy from the combination of generation technologies chosen for the system over its entire lifetime, discounted back to today at 6% per annum.

This includes additional transmission costs for various options and the resultant carbon intensity is also shown. All key technology performance data, costs, and other relevant information are listed in the Power System Generation Mix Model website .

There is a short report as well in PDF form is available on the website.

Barry Mur_ Reliable and Affordable Electric Power generation

Cutting to the chase, they recommended:

1.   Wind-up subsidies for intermittent power generation

2.   Add a capacity market component to the National Electricity Market

The current NEM is an energy-only market, which does not give clear signals when more or replacement dispatchable generation investment is needed. This weakness has been a key factor in the current absence of new dispatchable investment, i.e. power which can be delivered at the time it is needed by customers.

3.   Remove the ban on nuclear power

This ban is the result of a political deal done 20 years ago. It has no scientific merit, and is now an obstacle to much-needed decisions for the longer-term future. It prohibits by law the development of emissions-free, reliable, affordable nuclear power for Australia . The removal of the ban would allow more competition between various technologies to supply our future electricity needs.


This modelling shows that Base Case 1 (existing NEM approximation) has an average base load of 18,368 MW of constant electricity demand. This load plus daily peaks must be reliably supplied at all times. At present this is done using a system of 78% coal, plus combined-cycle gas turbine (CCGT), open-cycle gas turbine (OCGT), wind, solar PV, and hydro. Some battery storage is available for the provision of ancillary and other services as needed. The estimated cost is $68.87 /MWh.

Case 2 shows the effect of introducing 3,000 MW of nuclear power capacity into the Case 1 mix to replace brown coal. As expected, this adds+$ 3.61 / MWh (0.36 cents/ kWh) to the System Levelised Cost of Energy (SLCOE), making a total of$ 72.48 / MWh while reducing emissions by around 23%.

Case 3 shows the effect of replacing all coal in Case 1 with nuclear power. Emissions fall by some 93%, with SLCOE increasing to $ 90.23 / MWh.

Case 4 shows the effect of the combination of generation technologies projected by the Australian Energy Market Operator (AEMO) to 2040, as shown in its Integrated System Plan (ISP) of July 2018.

Case 5 shows the effect of replacing all coal in Case 1 with CCGT. Note that this shows an increase in SLCOE of + $ 6.49 / MWh versus Case 3 above, plus a substantial increase in emissions.

Case 6 shows a 100% renewable mix comprising solar PV, wind and hydro with support from pumped storage and some battery storage. Because of low capacity factors, solar PV and wind require a combined total of 110,000 MW of capacity.There is also a need for 30,000 MW of pumped storage capacity for 3 days. To this must be added high-cost additional transmission to get the power to points of high consumption where it is needed, making a total SLCOE of $ 415.50 I MWh.

Did anyone ever say it was going to be inexpensive to save the planet?


The suggested that for the moment the (then) proposed National Energy Guarantee (NEG) should be supported, but the expansion of intermittent generation and storage as projected by AEMO should be examined for practicability and cost. They also wanted to see  a capacity market component to provide clear signals for investment in dispatchable generation when needed.

For the longer-term, the existing ban on nuclear power development for Australia must be lifted. Given our role as an important supplier of uranium to the world, current national policy in this area appears confused. A bipartisan agreement in the Australian Parliament to lift this ban is now a matter of urgency, so that proper examination of this option can be undertaken.

August 2018

This entry was posted in Electric Power and Energy, Rafe. Bookmark the permalink.

18 Responses to Comparative costs of power

  1. EllenG

    Highlights one huge flaw in the AEMO planning. Biggest base load consumers are aluminium smelters which are virtually making ingots of power. In today’s market with big Russian and other cheap inputs, Australian smelters are losers above $50/mwh. Portland maybe worse. Tomato a little better.
    Point being we are headed for a train smash when base load users die – and they will. Barring a ludicrous subsidy. (Though they’ve had plenty of that already).
    As we drift onwards the likely outcome is fragmented industrial supply direct to big users. Off grid. Increasing grid costs rise to meet dispersed supply. Most commercial and industrial users get hammered by cost escalation. Grid becomes unreliable without gas firming.
    Welcome to the third world!

  2. herodotus

    current national policy in this area appears confused.
    It’s a disaster and has been ever since the ALP and Greens took a set against nuclear and the Libs/Nats didn’t feel like fighting back. Same goes for dams, new hydro, gas, and tinderbox national parks.

  3. BoyfromTottenham

    Thanks, Rafe!
    One small point – I clicked on the the link to the source, and there was no explanation of how, or whether the ‘100% Renewables’ option could or would actually work, just ‘modeling, modeling. modeling’. Whilst searching for such a network, I have not found a single example of a workable ‘100% renewable’ national-scale power grid design that does not require substantial traditional synchronous generation capacity to provide the necessary synchronous source of stable 50 Hz power to stabilise the grid, and more importantly allow recovery from blackouts.
    As many academics agree that traditional power networks become more unstable as the renewable power share rises above about 20% and the required grid stabilisation technology has yet to be invented, I consider that a ‘100% renewables’ power grid is decades away, and even then retro-fitting this new technology into any national network will be hugely expensive and take many years more.
    Is this ‘100% renewables’ option therefore just the power grid equivalent of the climate modelers’ unrealistic ‘RCP 8.5’ scenario?

  4. Tel

    Looking at the 100% “renewables” option, the wholesale cost comes in at 42c per kWh and then when you consider typical markup for electricity is at least 50% the consumer would be paying 63c per kWh or roughly triple what we pay now. If you are around $500 per quarter you would be facing something in the ballpark of $1500 per quarter.

    I think most middle classed households could afford to pay that … but they sure won’t be happy.

  5. Delta

    BoyfT – Will the 100% Renewables option work? As an electrical power engineer, the answer is a definite No. All the pundits pushing for this “solution” to the imaginary problem of “climate change” are ignorant of the complex dynamics of an electricity grid. Or they willfully ignore them.

    This SLCOE model was the basis of a presentation to the Australian Institute of Energy in Sydney nearly two years ago. And a pdf of the full presentation AIE Sydney – Achieving a Cost Effective and Reliable Electricity Supply for Australia may be viewed here on their past events page.

    You ask about the assumptions and whether the model is valid? Well you can read the assumptions behind the model in the presentation slides and judge for yourself. My understanding is that the model was built to determine the change to electricity power costs from moving to add more so called “renewable generators” in the NEM based on AEMO’s Integrated System Plan that mapped out significant change. Interestingly in the past two years AEMO and others increasingly talk about some natural transition to this type of generation determined and assuming that it will happen, and can be made to work. Actually it can’t but I won’t go into the technical details here of why that is so. Suffice to say, there are sufficient examples from around the world and here in Australia to demonstrate this fact: California, South Australia, Alice Springs to name but a few.

    Back to the AIE presentation: There was another case requested and that was what would the costs and emissions be from replacing all coal generation with High Efficiency Low Emissions (HELE) coal. That case can be seen on pages 14 and 15 of the presentation and provides the lowest SLCOE at $66/MWh with emissions of 0.68t/MWh (lower than the current NEM with all its renewables).

    In summary, the final results of the study can be seen on p15 – Summary of Generation Mix Options where the trajectory for each option show the cost route associated with the an HELE Coal Route, a Gas Route, a Nuclear Route or a Renewables Route.

    Once again, could the Renewables Route actually be built and made to work? No not really but hopefully that should become apparent well before we get too far down that path. E.g, you can’t start and electricity grid on renewable energy. Admittedly, that doesn’t seem to deter the NSW government wasting taxpayer money with its push to develop Renewable Energy Zones but I wouldn’t be surprised if they simply don’t understand what they are doing. A straight forward outcome is that the more Renewables are added to the grid, the greater will be the cost and environmental damage.

    A final comment before anyone comes back with AEMO’s flawed mantra …but Renewables are the cheapest form of new power generation, ask what those real costs would be on a SLCOE basis without any of the current massive government subsidies?

  6. Farmer Gez

    Keep up the good work Rafe.

    You may be short on comments but we’re reading and using your analysis to push the fight for a sane energy policy to a broader audience.
    My local members and the council have had your figures shoved under their noses.

  7. richardf

    Now recalculate the numbers with thermal coal price of USD50 / metric tonne and transmission infrastructure already installed!

  8. Rafe Champion

    Thanks Farmer Gez, a few words of encouragement make a difference! The good news is that the team has increased substantially by merging with another very active and organized gang of four (or more) in NSW with a similar agenda. On top of that we are recruiting experienced power professional, mostly retired, who will add a great deal of weight to our briefing notes and other output.

  9. RobK

    In case you missed it:

    #3582928, posted on September 12, 2020 at 11:53 pm
    Perhaps of some interest to Rafe and others regarding some simple modelling of RE storage requirements over a year. The penny is slowly dropping for some that there are gluts and droughts of energy seasonally. Never mind that no two years will be the same.

  10. Colonel Bunty Golightly

    I really don’t understand why the author of this piece bothers? The clowns who run this country are not interested in anything said here. Until the electoral pain From the insane rush to RE is transferred through the ballot box to politicians the madness will continue on its merry way. Then again, I really think the electorate stupidly believes the climate change crap they are fed – like lemmings they will follow our self interested leaders over the cliff. Leaders however will jump aside at the last minute. A civilisation that has passed its zenith and is preparing to make way for something better to rise from the ashes!

  11. Rafe Champion

    Thanks RobK, I put a comment on that piece.
    Given the current amount of storage that is feasible and affordable the monthly balance may be too coarse-grained to show the full extent of the problem. Looking at daily figures there are extended periods in June and July, up to 20 and 30 hours when the wind fleet is delivering less than 10% of installed capacity, down to 2 and 3% for shorter periods. This would not be such an issue if we could run extension cords to neighbours. For instance California has rolling blackouts but they won’t go completely black. In the Californian situation with the conventional power sources run down before the RE replacement is adequate, we would be in real trouble. This looks like a possibility post-Liddell.

    Just before Figure 9 he wrote
    So the Cumulative deficit is an energy deficit over time but we are also interested in how much firming power in Gigawatts is needed at any one time. The next figure shows the percentage of time firming power needed to be X GW or less. We can see that about 85% of the time firming power required is 15 GW or less. A small amount of time, ignored in the rest of this note, up to 27 GW is needed.

    I am looking forward to the time he looks at the high end of storage that is needed a small amount of the time. That is where the choke point factor enters! Like he is going to examine the small amount of time that it take for us to drown if we are submerged:)

  12. RobK

    Yes, I noticed that too. There’s 15% of the time he’s going to have to come up with almost the entire grid’s load from somewhere. As if that kind of horsepower is going to be sitting idly about waiting for an opening to fire up. Conversely, all the energy absorbing capacity is still going to spill 15 or more % regardless of H2 or other techniques. It’s unlikely to be economic to size the capital works for occasional return. This point seems to take a while for some to absorb. It knocks the supposed returns about a lot.

  13. RobK

    Sorry, i missed your comment on that piece earlier. His data resolution was half hourly, from memory, but your point still holds. Even given 10GWh of perfect storage in the model they’re in trouble 15% of the time . That’s a big number in that context.

  14. Aynsley Kellow

    Well done Rafe. I am too busy on other things at the moment to contribute much, but let me point to a risible response to the problem of excess rooftop solar generation in the middle of the day when it is not needed, and less in the morning and evening when it is.
    The ABC news on Saturday evening ran a story on a couple of bright sparks in Adelaide who had a solution: align the panels to the east and west, rather tan the north! Forgot to mention how much less the output would be. Risible.

  15. Ainsley:

    The ABC news on Saturday evening ran a story on a couple of bright sparks in Adelaide who had a solution: align the panels to the east and west, rather tan the north! Forgot to mention how much less the output would be. Risible.

    The ABC News team will be first into the furnaces when the lights go out, but they don’t seem to quite get that concept.

  16. Ubique

    The “100% renewables” can work in a supply-driven system. When the wind blows and/or the sun shines there’s power; otherwise not. It’s environmentally responsible to have life support systems and operating theatres on a supply-driven electricity system. Not so good for patients admittedly, but that’s a small sacrifice when the planet is at stake, no?

  17. Rafe Champion

    Lets face it, turning the panels away from the sun would be a great step in the right direction.

    Full marks to the ABC news people for suggesting it!

  18. Delta


    The “100% renewables” can work in a supply-driven system.

    I suspect your comment may have been a little tongue in cheek, but no they won’t. “Renewable” generators cannot follow the load and you can’t start actually start a “100% renewable” electricity system against any load.

Comments are closed.