Australia’s biggest wind farm is also its least productive

Print Friendly

The Macarthur wind farm in Victoria– at 420MW – is Australia’s biggest. But in the last 12 months it has been the worst performing of any wind farm in the state, and among the worst in the country.

When it was first announced way back in 2010, it was tipped to deliver a capacity factor of around 35 per cent. It opened in early 2013, but in 2016/17, it delivered a capacity factor of just 23 per cent.

“The performance of the Macarthur wind farm in FY2017 was primarily affected by planned outages and poor wind conditions,” an AGL spokesperson said in an emailed response to enquiries by RenewEconomy.

True enough, there were shutdowns of around 3 weeks that may have explained some of the lost output, and wind farms across Victoria and South Australia were all affected by weaker than normal wind conditions in the last quarter.

But that doesn’t appear to resolve the difference between Macarthur and other wind farms in the state. Nor does it explain its poor showing the previous year (26 per cent) when there were no apparent interruptions.

vic wind farms cap factor

The performance of Macarthur has been the subject of much speculation in the wind industry, with no one too sure what is the reason for the lower than expected performance.

Screen Shot 2017-08-11 at 12.41.52 PMIndeed, according to this data from the Climate and Energy College, it has never produced anywhere near its hoped-for capacity, and never reached above 28.6 per cent.

By contrast to Macarthur’s output, Stockyard Hill, which at 530MW will dislodge Macarthur as the biggest wind farm in the country, is expected to have a capacity factor of around 45 per cent. Little wonder, then, that it is able to deliver electricity to its client Origin Energy at less than $55/MWh (free of subsidies).

Hornsdale in South Australia, the next biggest at 309MW, is expected to also have a capacity factor in the 40s when its third stage is finished in the next few months. Snowtown has performed as high, reaching 44 per cent in 2014/15.

The 207MW Collgar wind farm in Merredin in Western Australia, is also a strong performer with capacity factors in the mid to high 40s.

Of AGL’s new wind farms, the 200MW Silverton wind farm near Broken Hill is expected to have a capacity factor of around 44.5 per cent, while the 460MW Cooper’s Gap in Queensland is expected to have a capacity factor of around 35 per cent.

To put Macarthur’s recent performance in context, here is a list of the capacity factors of all the wind farms in South Australia over the last five years, from last year’s AEMO analysis of renewable energy in that state.

wind farm capacity factors.It shows that no wind farm in South Australia had a capacity factor a low as 23 per cent in the five years surveyed by the market operator.

Of course, wind farms are not the only energy source to either not perform to expectations, have low availability, or not get used at all.

AGL’s Liddell coal plant operated at a capacity factor of just 51 per cent last year, despite increasing its “availability” from just half the time when it had to shut down for repairs to just over three-quarters of the time. The Torrens gas plant was not available nearly one-quarter of the time.

The Somerton peaking plant was used just 2 per cent of the time. But at least it was switched on. The Merredin peaking plant in WA still has not been switched on, despite being paid for by consumer subsidies.  

Share this:

  • George Darroch

    Canunda and Lake Bonney also had a poor year last year.

    Long term wind patterns and variability are important considerations. I hope other operators are doing their homework.

    • Matthew Wright

      One poor year is fine – it’s part of the averages. It’s even better if poor years don’t effect all wind farms at the same time so much though as they get taller and the blades output will be much more consistent across sites.

      • George Darroch

        Well yes. And you would have seen if you looked at the graph above that these perform poorly year-to-year.

  • Just_Chris

    Doing some very rough calculations MacArthur wind turbines have an area of 9850 m2 and Stockyard hill has an area of 15840 m2. The power of MacArthurs wind turbines are 3 MW and Stockyard hill are 3.5 MW. So MacArthur works out as 3280 m2/MW but Stockyard hill is 4524 m2/MW. My understanding is that modern wind farms feather the blades past a certain wind speed to stop the blades rotating to quickly. This allows them to have more efficient blades that cut in at lower speeds. So basically the Stockyard hill wind farm should cut in at a much lower speed but be much less efficient at higher wind speeds but still get the full 3.5 MW per blade.

    Does this make sense? I am really not sure about wind farms it’d be interesting to see a comment from a more knowledgeable person. What is the final point on this? could we get to 50% capacity factors if the blades were oversized accordingly?

    • Matthew Wright

      Yes spot on – the trend in the industry is for oversizing the swept area to generator capacity. The trend is towards 6m^2 / kW

      But despite that Macarthur is still performing poorly. So either it has technical problems, transmission constraints or the wind resource was modelled very poorly.

    • Peter F

      Chris feathering implies completely turning into the wind so there is no power at all. Most wind turbines have pitch control so they can optimise the power for a wide range of wind speed. Two developments have made modern wind turbines more efficient.
      1.Stronger blades which can be longer without breaking or hitting the tower so if McArthur were built today, it would probably have 136 m diameter turbines instead of 112. At wind speeds over about 13m/sec the power is limited by the generator to about 3.1MW so once wind speed exceeds 45km/hr at the hub a large turbine will produce no more power than the smaller one. However at 6m/s the current wind turbines only produce 550 kW. With 136m diameter blades that would increase to 810 kW

      2. they would probably be mounted on taller towers probably around 110m instead of 85m. In rolling country like that near McArthur an extra 25m height in turbines will probably result in an 6-8% increase in average wind speed at the hub. That is 20-25% more power because power is related to the cube of the speed so now the output of the same generator has effectively doubled at low wind speed.
      However as you say the efficiency at higher wind speeds is reduced. Improved electrical design has increased the peak power to 3.45MW but it is reached at 10m/s(36 km/hr) so the extra energy in wind speeds over 36km/hr is in effect wasted.

      The capacity factor is obviously related to average wind speed but it is more related to the shape of the wind speed curve. For a simple example you could have a situation where the wind alters between two regimes 18-22m/s for 4 hours and 0-5m/s for 8 hours. The average wind speed would be 8.3 m/s so that would in theory give a very high capacity factor of around 65% for the 136m diameter turbine, but in fact the turbines would only be generating for around 5 hours in 12 and actual generation would average 1300kW or 38% capacity factor.

      The same exercise with the existing turbines would result in average generation of around 1220kW so the larger blades would make no difference. In fact they could be worse because the old turbine cutout speed is 25m/s vs the new one at 22.5 so the old one would keep generating at full power when wind speed is between 22.5 and 25m/s while the new one would shut down.

      On the other hand if you had a typical wind speed varying smoothly from 5m/s to 20m/s at 85m hub height, the old turbine would probably run at around 30% and new ones close to 40%

      • Just_Chris

        Thank you to both Peter and Matthew for your considered responses. One other thing I have been wondering is if there is a power loss associated with how the turbines sync with the grid? The reason I ask is because I am wondering if the battery at hornsdale will lead to higher output. The theory being that if the battery smooths the output that the wind turbines can be push harder which might lead to less stable but higher output. Does that make any sense? I really haven’t got any idea about a modern wind turbine and how it functions, I am just trying to figure out if there is any major advantages to basing storage at a wind farm rather than distributed throughout the network.

  • Malcolm M

    I suspect it wasn’t modelled properly in the design stage, and that there are some equipment problems. Most of the SA mid-north wind farms are on hills about 200 m above the surrounding landscape, which would cause some speeding up of the wind flow as it passes across the hill, whereas the MacArthur and the SE SA wind farms are on only slightly above surrounding areas. The SA mid-north wind farms are also exposed to both easterlies and westerlies, whereas MacArthur and the SE SA wind farms have poor easterly exposure, and too far south for a good easterly flow in summer. The site for the MacArthur wind farm was selected because of landholder interest, and because it was close to the 500 kV transmission line. The site looks windy on low-resolution wind maps, but high-resolution modelling would have revealed much better sites on the hills around Coleraine (50 km to the north-west of MacArthur) that are 50-100 m above the surrounding landscape. The high-resolution maps were probably not available at the time the site was chosen. The designers were probably asked to make it work financially, which meant requiring scale. Two problems with scale is that some less-than-suitable wind tower sites are included to make up the numbers, and that in its grid layout some generators are in the turbulent zone of upwind generators.

    Its capacity factor is seldom over 90% even when surrounding wind farms (Mortons Lane and Oakland Hill) are close to 100%. There must also be a technical problem that AGL are keeping quiet about.

    • Matthew Wright

      Just switching the turbines to the Enercon E-141 on the Macarthur site (as an academic thought excercise) would give the performance of a good site. Taller tower, longer blades and more sqm swept area to generator rating.

  • David K Clarke

    A couple of points. Snowtown, rather than Hornsdale, 309MW, is effectively the second biggest wind farm in Australia in installed capacity: 371MW. The Snowtown 2 turbines are mixed in with the Snowtown 1 turbines; calling them two separate wind farms is very artificial.

    I remember seeing a graphic at one point which indicated (to me) that the turbines on the margins of Macarthur wind farm were ‘shadowing’ those to the leeward (can’t find it now). I note a drop in capacity factor in Snowtown 1 following construction of Snowtown 2 indicating similar, but not so severe, shadowing.