On the world energy market wind's greatly improved economics are largely masked by the dramatic fall in the price of fossil fuel -- by some 50% in the past decade. But they can be tracked through the pages of a timely report from the European Union's Eurowin data base, the most comprehensive of its kind in the world. Wind plant registered by Eurowin produced just over 2 TWh in 1993 -- contributing some 0.5% of all domestic electricity consumed in the European Community. Including a conservative estimate of production from machines not registered by Eurowin, wind energy now meets the electricity needs of more than 1.5 million Europeans. Grounds enough for taking a serious look at the trends which an analysis of the Eurowin data brings to light.
Better technology
The reasons for the steady decline in wind energy's costs are many and varied. Firstly, better technology means higher energy output. The effects of the improvements in technology are strikingly illustrated by the increase in output from machines in the database. This has improved steadily from 1988 to 1993 (fig 2). The performance of the top 20 machines shows a dramatic upward trend -- the output from the 20th machine in 1992 was higher than the most productive machine in 1985. The trend in the overall average is also upward and there appears to be scope for even greater improvement. Further reductions in the economics of wind energy should be achievable.
The improvements in technology stem from investments in research and development, principally in the fields of meteorology, aerodynamics, fatigue and control. Wind turbines, as a result, are now better understood and can be designed to be reliable and safe. Reliability is reflected in the declining number of failure reports which Eurowin receives. In 1990 there were reports from 1013 wind turbines, 27% of the total. In 1993 just 839 faulty machines were reported, a decrease to just 13% of the total. Although the absolute numbers are probably understated, the trend is clear.
The investment in R&D also means that today's machines are not often over-engineered. Possibly the most significant advances have been in the production of lighter and more efficient blades. Lighter blades are cheaper blades and lighter blades also lead to lighter and cheaper hubs, nacelles and towers. In Eurowin's 1989-90 report, the lightest nacelles in the 40 m diameter size range weighed around 20 kg per square metre of rotor area; by 1993 this figure was down to 15. As the nacelle accounts for about two-thirds of the cost of a turbine, a 25% reduction in the cost of this component yields a 16% reduction in turbine costs.
Economies of scale
Although the manufacture of wind turbines is not yet akin to that of cars, several builders do now have production lines, which helps to reduce costs. Productivity in the factory has increased markedly as a result. German wind institute, DEWI, reports that turnover per employee in the German wind industry has more than doubled since 1990, during which time about 1500 machines were installed in the country, 600 in 1993. The total number of machines added to the Eurowin database during 1993 was 1021, taking the total to 6318 (fig 3). The database reflects a steady increase in machine numbers of about 15% a year -- a doubling of machines since 1989.
Significantly, as fig 3 also shows, the rate of increase of capacity -- at over 20% per annum -- is even faster than the increase in numbers of turbines. This is because they are getting larger. Capacity in Europe reached 1050 MW by the end of 1993 and is now about 1400 MW. The trend towards larger sizes is very marked. Eurowin shows that the number of machines installed in the 150-400 kW size range increased from 604 in 1989 to 1986 in 1993.
By tracking the size and cost of the largest machines commercially available in Germany, currently the world's most active wind market, it can be seen that sizes doubled and costs fell by 10% (uncorrected for inflation) between 1990 and 1994 (fig 4). Allowing for inflation, the cost reduction is at least 20%. This assessment is consistent with a recent report by DEWI, which shows a 35% reduction in the cost of energy from machines in the size range 32-45 m diameter (400 to 750 kW output), during the same period.
Several other cost benefits stem from the use of the larger machines:
Â¥ annual charges (per machine) for maintenance for 600 kW turbines in 1994 are no higher (some are less) than those for 300 kW turbines in 1990
Â¥ the use of larger machines means that fewer roads and electrical connections need to be installed
Â¥ less land is needed per kilowatt of installed capacity
Â¥ taller towers means wind speeds are higher, leading to higher energy yields
Â¥ small, but useful increases in the efficiency of the blades, transmission and electrical equipment are achieved
Â¥ and fewer machines, further apart, perhaps gives a better visual impression, smoothing the way for the acquisition of building permits and reducing the time spent on the process.
The value of wind energy
Europe's political leaders often stress the need for "convergence" between the price of wind and that of conventional energy sources. However, there is a related issue -- the value of wind energy. And there is an increasing awareness that it is undervalued. There are several aspects to judging wind's value, including its contribution to reducing (costly) pollution, the usefulness to a utility system of distributed (as opposed to centralised) generation, and the correlation between generation and demand (“uåX˜äŠÊ˜·³Ç, September 1993). In Europe, the value of electricity is greater during the winter, when lower efficiency, higher cost generating plant needs to be pressed into service. Fortunately the seasonal variations of wind mean that energy production from wind installations is higher during these periods (fig. 5). The energy generation during the winter months is typically 30-60% higher than during the summer. This fact is starting to be reflected in wind energy payments. The payment systems in both the Irish Republic and Northern Ireland reflect the seasonal variations in the value of energy.
A related topic -- the capacity value of wind -- continues to be a hotly debated issue (“uåX˜äŠÊ˜·³Ç, September 1993). Is the existence of wind capacity useful to a utility; and can wind power capacity replace conventional plant capacity on a power system? Submissions to a recent UK Parliamentary Committee investigation of wind energy (“uåX˜äŠÊ˜·³Ç, September 1994) have provided tangible evidence that wind installations are, in practice, likely to be generating at times of system peaks, when energy may be worth as much as four times its average value. During the 1993-94 winter peaks in Wales, for example, the 30 MW Llandinam wind farm was generating at 90% of its rated output and would have attracted significant capacity related payments from the UK power pool.
Scope for further improvements
All the Eurowin data relating to performance costs and energy output indicate continuing improvements, with little sign, yet, of a levelling out. Spectacular steps in cost reduction or performance improvement are unlikely, but steady progress seems assured. The trend towards larger machines will even out, but no one is quite sure at what level of power output this will happen. Several European manufacturers are firmly committed to the development of machines in the MW size range (around 50 m diameter) although the benefits of ever bigger turbines will become more marginal, due to difficulties in building and handling larger components. On the planning side, citizens have voiced doubts about the merits, visually, of continuing increases in size, but others point to the fact that it is very difficult to distinguish between MW and kW machines at a distance beyond a kilometre or so.
The economics of wind energy continue to improve. Generating costs may not yet be on a par with those of gas -- the cheapest fuel for centralised generation -- but they are already equal to, or below, those of nuclear. When the increased value of decentralised generation is taken into account, the gap between cost and value is closing rapidly.
.png)
.png)
