Rafe Champion, Independent Scholar, Sydney
The Australian electric power supply is on the edge of a cliff because one more substantial coal-fired power station is scheduled to close in 2023. Then for the first time we will need input from the wind at the evening peak, certainly in very hot and cold weather. If this input is not available there will be blackouts of greater or lesser extent.
The bottom line of this argument is that RE cannot replace conventional power in Australia because at the lowest level of supply of sun and wind they deliver no power. That is the choke point for the grid, like the point where the air supply to our lungs is cut off by choking or drowning. To keep the lights on there must be enough conventional power available to meet 100% of the demand, 24 hours a day/365 days a year.
In recent years Australia has led the world in the rate of building RE capacity. Inspired by this rate of progress the Federal Opposition (Labor) party and the state governments (Labor and Liberal alike) have committed to push for zero emissions by 2050.
Given the choke point problem this target is not realistic because we have no neighbours to help with power if we are short and we have no nuclear power despite holding a third of the known reserves of uranium in the world. That is a result of the international campaign against nuclear power in the 1960s and 1970s that was successful in Australia http://www.the-rathouse.com/2011/Grover-Power.html
The introduction of intermittent energy into the grid doubled the retail price of electricity and hastened the demise of 12 coal-fired power stations since 2010. Almost 6GW of capacity left the integrated grid covering South Eastern Australia (excluding Western Australia and the Northern Territory). That is a significant loss in comparison with the high point of demand at dinnertime that is around 25GW for most of the year and over 30GW in high summer
After the last coal power station closure in 2017 the Energy Market Operator (AEMO) issued a warning that we have next to no spare capacity to handle peak loads. At the same time the CEO, Ms Audrey Zibelman, was giving talks around the country to say we can have cleaner, cheaper and reliable power all at the same time. Ms Zibelman was imported from New York by the ex-prime minister Malcolm Turnbull to oversight the transition to green energy.
Blackouts in some Melbourne suburbs in January 2019 demonstrated the lack of spare capacity when a coal-fired generator failed and took some 500MW out of the system on a hot summer afternoon.
Another station providing almost 2GW of coal power is scheduled to close in 2023. Then for the first time the intermittent providers will be called upon to supply energy regularly to keep the lights on, unlike the situation where they were surplus to requirements and just displaced conventional power when the sun shone and the wind blew.
This discussion focuses on the capacity of the wind system because the lack of solar power after sunset is a given at the current state of storage technology. The question is: How much wind capacity is required to substitute for 1.8GW from Liddell Power Station?
We have 7GW of installed wind capacity at present with some 7GW under construction. On the face of it, 14GW looks like a good cover for 1.8GW of lost coal power but everyone in the industry knows that the supply over a year is about 30% of installed (plated) capacity. That shrinks 14GW to 4.2GW and that still looks good compared with 1.8GW but that is not the end of the story.
The critical number for the viability of continuous, secure power from the wind, in the absence of mass storage, is the lowest point of the wind supply. The average is no more helpful than the installed capacity if the grid depends on wind power as a vital contributor, not just an addition to an adequate supply from other sources.
Several analogies can be used to reinforce the point. A chain is only as strong as the weakest link. Your car will stop running when the petrol tank is empty regardless of its capacity. The supply of air that your lungs received in the previous years of your life will not help when you are choking or drowning; that is the choke point and death soon follows.
Warning signs. The low points in the wind supply.
Did the planners and politicians take any notice of the warning signs that there is not enough reliable wind to make the RE transition without feasible and affordable mass storage? The Bureau of Meteorology (BOM) does not publish in its information on cloud cover and wind but it surely the vital data would have been made available on request from the Government or statutory authorities planning to spend tens of billions of dollars to subsidise intermittent energy.
Extended adverse weather events have been reported in the past , for example in the Murray Mallee region of South Australia where:
“A drought of a very different kind occurred in March and April of 1934. Because Lameroo sits above our underground water supply, windmills (wind pumps) were used to draw water to the surface for stock water and personal use – windmills paved the way for Mallee agriculture. The period from mid-March to the end of April was almost completely windless; therefore no water. Farmers were soon desperate for stock water……. http://www.malleehighway.com.au/html/lameroo.html
In recent times the AEMO with a continuous record of the power delivered from all the wind farms attached to the South Eastern Australia grid. Paul Miskelly used that data for the calendar year 2010 to find that the total wind output across the entire grid fell rapidly to zero or near zero on many occasions in the year. Some of the episodes involved multiple falls and rises in quick succession.
- During the first 6 months of the year, there are 58 intervals where the output falls below 2% of the installed capacity. The longest such interval is 229 consecutive 5-minute time steps, or 19 hours approximately. This event occurred in May.
- During one such event, on May 18, on two occasions, the total output actually dropped to slightly below zero, the first starting at 2.50 am and lasting 40 minutes, the second commencing at 4.35 am and lasting for 70 minutes.
- For the entire year, there are 109 such intervals of varying length, comprised of a total number of 1867 5-minute intervals, totalling 155.6 hours, or nearly 6.5 days.
The net result is that a fleet of new-build fast-acting OCGT plant, of comparable capacity to that of the total installed wind capacity, constantly operational in standby mode, is required to balance wind’s mercurial behaviour. (Miskelly, 2012).
The wind fleet in 2010 consisted of 23 wind farms with a total capacity just short of 2GW and as the number of sites increased there was an expectation that the supply would become more reliable. John Morgan reported that the situation was much the same in the 12 month period from Sep 2014 to Sept 2015 when the capacity of the wind fleet was approaching 4GW. He found 29 days in the year with the fleet delivering less than 10% of capacity. The lowest was 2.7% and there were seven sets of successive low wind days. https://bravenewclimate.com/2015/11/08/the-capacity-factor-of-wind/
The collective capacity of the SE Australian windfarms has now passed 7GW (in early 2020) and the problem of reliability persists, even to the extent of occasional wind droughts. Mike O’Ceirin, an independent analyst, has unpublished information collated from the AEMO records over a seven-year period that show an average of 15 episodes per year when the delivery is 6% or less of the installed capacity.
The AEMO data in recent months show that the supply frequently falls below 10% of installed capacity. 10% is by no means the lowest level of supply but it provides an instant guide to the amount of power coming from the installed capacity as the supply falls towards the minimum.
In December 2019 there were eight episodes in the month when the supply fell below 10% of installed capacity and one where it was below 5%. In January the numbers were twelve and three, for February nine and two, March six and two, and April so far nine and three. The longest durations for the low periods were 16.5 hours on 18-19 April, 13.5 hours on 13 April, 9 hours on 29 Feb, 6 April and 7 April, 7.5 hours on 31 March and 7 hours on 29 December.
All of the long “dry” spells were potentially disastrous for the power supply if wind was required to make a contribution to meet the demand in the grid and this is a point that has so far eluded the wind proponents who look at the good days and announce that we are making progress towards a green future.
RE supporters pin their hopes on storage by batteries and pumped hydro to fill the gaps between the peaks of wind and sun. Clearly battery storage on the scale required is out of the question for the foreseeable future and much the same applies to pumped hydro in Australia.
The ambitious Snowy2.0 pumped hdro project is under way and it could possibly function in partnership with the existing windfleet to provide 2GW of continuous power. Going on to replace the rest of the coal fleet would require some six or seven times the current capacity of wind power and six or seven Snowy2.0 equivalents without any suitable sites available for that scale of development.
In any case Snowy2.0 will not be completed for many years and urgent action is required to find almost 2GW of reliable power from other source before the next coal station closure that is scheduled for 2023.
The proponents of wind and solar power apparently either cannot or will not appreciate the logic of refutation. A single (true) fact can refute a universal proposition. One first black swan refutes the notion that all swans are white. The proposition that we can run the grid on RE is refuted by a single period when there is not enough RE to run the grid.
How often are we prepared to have the power supply fail? Infrastructure like drains and flood levies are built to withstand events of various frequency – 10 years, 20, 50, 100 years. Facilities like major bridges and dams presumably have to be designed to handle just about the most severe events that can be envisaged and the electricity supply should have the same evel of reliability. It can’t be allowed to fail three times in a month which is what would have happened in SE Australia in April.
Anero site with data from the Australian Energy Market Operator https://anero.id/energy/.
Paul Miskelly (2012), Wind farms in Eastern Australia – recent lessons. Energy & Environment 23 (8). https://journals.sagepub.com/doi/abs/10.1260/0958-305X.23.8.1233