Lies & Statistics

Australian Financial Review
Saturday 19/8/2006, Page: 63

Wind power can be relied on, argues Mark Diesendorf.

Prime Minister John Howard's energy initiative this week allocated $124 million for renewable energy projects in remote Australia. Howard cited Bremer Bay in WA, which gets 40 per cent of its electricity from a wind turbine.

Ironically, it's fashionable to label wind power as an intermittent energy source, making it unsuitable for replacing coal power.

Superficially this sounds plausible. Everyone knows a single wind turbine may start and stop many times in a day. Fallacies in the intermittency notion were originally refuted by a CSIRO/Australian National University research group I led in the 1980s.

The term intermittency is misleading when applied to several dispersed wind farms, because their total power output varies smoothly and very rarely drops to zero. Graham Sinden of Oxford University analysed over 30 years of wind data from 66 sites spread out over the UK. He found low wind speed events affecting more than 90 per cent of the UK had an average recurrence rate of only one hour per year.

Large-scale wind power is more appropriately described as "variable". The variability associated with breakdowns of conventional power stations is better described as "intermittent".

There is no such thing as totally reliable power. Because electricity cannot be stored cost effectively on a large scale, the electricity system is a continuous balancing act between two fluctuating variables, demand and supply. Integration of wind power merely adds a third fluctuating variable, which can be predicted quite well on an hour-to-hour and day-to-day basis.

The better models of the integration of wind power into electricity grids find that the additional system costs incurred by quite large penetrations of wind energy are small.

Even these studies overestimate the net costs, because they do not take into account the economic benefits of optimising the mix of conventional base, intermediate and peak load plant in the presence of wind power. Brian Martin and I showed that wind power can replace base-load power stations with the same annual energy generation.

For example, 2600 megawatts of wind turbines, with an annual average power output of about 850 megawatts, can replace a 1000 megawatt coal-fired power station with the same annual average power.

To maintain reliability of the generating system at the pre-wind level, a little additional peak-load plant (for example, gas turbine) has to be installed.

In at least three isolated small Australian grids (Denham, Hopetoun and Mawson Antarctic base), wind power supplies more than 40 per cent of annual electricity generation.

Technically, it is easier to integrate wind power into a large electricity grid, which offers both dispersed sites for wind farms and a mix of conventional base, intermediate and peak load plant for balancing supply and demand. The issue is economics: as the wind penetration increases beyond about 20 per cent, the additional costs of balancing and of lost wind energy during off-peak periods start to become significant.

Nevertheless, Denmark already generates 20 per cent of its electricity from the wind and so could Australia.