South Australia – in a disconnected state

The government of South Australia has announced a program for insulating South Australia from the uncertainty of the electricity links to Victoria. The plan involves the construction of a 250 MW gas fueled generator for $360 million and the purchase of some $100 million of battery storage. In addition a further 250 MW power station is needed and industry would be encouraged to build it.

The basic assumption presumably, is that both the Heywood and Murraylink interconnectors are no longer operating. It certainly was the case during the recent spate of blackouts in South Australia, when the Heywood interconnector tripped on overload supplying South Australia in the statewide blackout of 28 September 2016 The loss of these interconnectors at any time remains a plausible scenario. The importance of these links is shown in Figure 1 for the period January 1 to March 31 2016. This 3-month period was chosen as being typical of the period where the likely summertime peak in demand would occur. Here Victoria supplies an average 232 MW each day with extremes of 705 MW flowing to and 583 MW from South Australia.

Figure 1:Victorian interconnector supply in 1 hour intervals from two interconnectors, Heywood (600 MW) and Murraylink (200 MW) for 1 January to 31 March 2016. At times an oversupply from wind farms is sent into the Victorian electricity market as shown by the negative-going excursions above.

The South Australian plan is to add a 250 MW power station. This does not remove the great variability in supply but it would provide surplus energy for storage. These surpluses and deficits are shown in Figure 2 for the period January to March assuming that the power station operates continuously at full capacity.

Figure 2:Variations of surplus and deficit from a 250 MW power station operating continuously to meet the demand formerly supplied by the interconnectors for January to March 2016

The South Australian plan calls for battery storage so with one 250 MW power station it is possible to estimate the storage needed from the surplus energy and the discharges necessary to meet demand.

The battery storage behaviour is shown in Figure 3 for the period January to March. It is clear that 50 GWh of storage is necessary. The cost would be some $10 billion at $0.20 per Wh.

Figure 3:Battery storage behaviour to match demand with a 250 MW power station.

This storage cost is not a match for the $100 million storage cost in the South Australian plan and explains why there is a need for another new power station so that the storage cost can be greatly reduced.

The result of adding another 250 MW power station is shown in Figure 4 in matching the demandformerly supplied by the interconnectors.

Figure 4:Variations of surplus and deficit from two 250 MW power stations operating continuously to meet the demand formerly supplied by the interconnectors for January to March 2016

There are very large surpluses of energy if the two power stations run continuously at maximum output. However the battery storage performance shown in Figure 5 indicates that matching the energy deficits requires only 2.5 GWh of storage so one of the two power stations would operate as a peaking generator. A peaking station is used to make up for sudden increase in demand and to recharge the battery storage to a maximum, another intermittent function. On the January to March period the utilisation would be 36% for the peaking generator and 68% for the "baseload" generator.

In fact 2.5 GWh of storage comes at a greatly reduced cost of $500 million at $0.20 per Wh compared to the $10 billion of storage costs with one 250 MW power station.

Figure 5:Battery storage behaviour to match demand formerly supplied by the interconnectors with two 250 MW power stations.

As an aside, if a complete shutdown of the sort experienced on 28 September 2016 occurred again then some 7 GWh of storage would be needed as well as the two 250 MW power stations.

The cost of one power stations of 250 MW is $360 million (government estimate) and $500 million for the batteries gives a total initial spend of $860 million from the South Australian government on behalf of its taxpayers. But even worse the government funded power station would run as a peaking station and would have little chance of making any profit as a result of low utilisation and government constraints on bidding into a wholesale electricity market where a high price would be expected. How battery discharges would be priced is an interesting question, Meanwhile the industry funded 250 MW power station might need to be financially encouraged.

The choice of batteries is "quixotic" as an additional 250 MW power station would need no back up storage and at $360 million would be an apparent bargain.

So the South Australian government plan has a mad logic to it but it requires two power stations in addition to its fashionably-chosen, unspecified amount of battery storage. As shown in this analysis, the proposed amount of battery storage, an amount that can only be gleaned from the cost estimate provided, and seems to be a touted 500 MWh, (0.5 GWh) to be consistent with the government plan but not the requirement of 2.5 GWh. In fact the battery need is probably larger as there are limits to battery discharge levels that might double the storage size and cost.

Of course there were two power stations that could have supplied this power. But they have been closed and a baseload power station cannot operate commercially with 40% intermittent wind and solar power.