If developers can show that wind turbines have a long lifetime, financiers will be more inclined to advance loans for longer periods, possibly with favourable interest rates.
The lifetime of a plant has a strong influence on the generation costs of wind energy. A wind farm with capital costs of $1,600/kW would deliver electricity at around $64/MWh on a 15-year loan, falling to $56/MWh with a 20-year loan, and $52/MWh for a 25-year loan.
The database provides a valuable source of information on wind-turbine characteristics. There are more than 55,000 records. Just under 10% of these — 4,978 to be precise — were commissioned in 1998 or earlier, and so are at least 20 years old. More than 4,000 are at least 30 years of age, and the oldest were commissioned in 1981.
California Energy Commission reports from the late 1980s and early 1990s provide data that enables the robustness of the technology to be assessed. At the end of 1990, 14,910 wind turbines were operational in the state (around 95% of all turbines in the US), and the latest database suggests that 4,475 of these are still extant. The growth in average wind-turbine ratings may be gauged by comparing the average in 1990 (97.5kW) with the average rating across the latest database (1.633MW).
The table (above) gives details of the oldest machines that are still operating. They were commissioned in 1981-84, all in California.
Figure 1 (below) shows the age distribution of the wind turbines and reflects the varying fortunes of the US wind energy programme. In terms of numbers, the peak year for commissioning was 2012, when 6,747 turbines were installed and started operating. Few, if any, machines commissioned in that year are likely to have been taken out of service.
By looking at the average ratings of turbines commissioned year-on-year, it is possible to follow the development of turbine technology, as successively higher-rated machines were brought to market. The trend is continuously upwards (see figure 2, below), and shows no sign of slowing.
The average rating of turbines installed in 1981 was 76kW; by 2000 it had reached 803kW, climbing to 1.76MW in 2010 and 2.32MW in 2017.
Although the average rating of turbines commissioned in 1984 was just under 100kW, 500kW machines had begun to appear, 14 of which are still operational. The first 1MW and 2MW turbines came to market at the end of the 1990s, after which the average rating climbed steeply.
The most popular machine rating of the existing fleet is 1.5MW, with just over 13,000 turbines operating. Another popular size is 2MW with more than 10,000 machines. The distribution of ratings is shown in Figure 3 (below). The most powerful turbines (off the graph) are the five 6MW GE Haliade machines that comprise the Block Island offshore project.
There have been various changes to turbine configurations over the years. In 1990, 95MW of wind capacity in California was accounted for by vertical-axis wind turbines, but no machines of that type appears in the latest database.
As turbines have increased in size, towers have grown taller, but the old rule of thumb, "hub height equals rotor diameter" is no longer relevant. Figure 4 (below) shows the relationship between hub height and rotor diameter for the wind turbines commissioned up to 2005, and it may be observed that the rule of thumb fits quite well up to that point. By 2017, the picture had changed (see figure 5, bottom chart). Although there is range of hub heights for any given rotor diameter, there appears to be a preference for 80-metre hub heights, and maybe a reluctance to go above 100 metres.
Figure 5 applies to wind turbines commissioned in 2017 and so does not include the Block Island offshore project, commissioned in 2016. The 6MW turbines — the US’s only offshore machines — have rotor diameters of 150 metres and hub heights of 105 metres.
One quite substantial change to wind-turbine design philosophy has been the move away from high ratings — for a given rotor diameter — to far lower ones. This is exemplified by comparing the specific ratings (rated power/rotor swept area) of the early machines in the database with the most recent ones. The change in philosophy has been driven by a number of factors, including a desire to minimise grid-connection costs and to utilise gearboxes and generators more effectively, as their costs depend on their rating.
The average specific rating for turbines commissioned up to 1990 was 491W/m2, compared with 230W/m2 for machines commissioned in 2017. The change in specific power rating means that present-day capacity factors are not comparable with earlier ones.
The American Wind Energy Association put the average capacity factor for the US in 2016 at about 43%. The figure for 1990, from the California Energy Commission, was about 21%.
However, a better comparator is the energy generated per unit rotor area, which is 835kWh/m2 in 2017, against 720kWh/m2 in 1990. This comparatively modest increase of 16% can probably be attributed to the fact that the 1990 figure for California was derived from turbines installed in the windy passes, whereas the 2017 figure applies to the whole of the US, taking in some less windy areas.
At a glance — This month’s report conclusions
United States Wind Turbine Database, US Geological Survey, American Wind Energy Association and Lawrence Berkeley National Laboratory, Hoen, B et al 2018, https://eerscmap.usgs.gov/uswtdb Comprehensive review of all US wind turbines operating in 2017