Easier said than done
The AWEA recommended practice, produced jointly with the American Society of Civil Engineers, is just that - it is not a standard, although it makes frequent reference to appropriate national and international standards. The difficulties with producing international standards are partly logistical, partly technical. If they are to be of value, they need to be acknowledged by as many countries as possible - and securing international agreements takes time.
Historically, standards have been written nationally by bodies such as the British Standards Institution in the UK and the Certification Committee for Wind Turbines in the Netherlands. In the early days, certification authorities such as the UK-based Lloyds Register and Det Norske Veritas in Norway formulated their own guidelines, drawing on their experience in shipbuilding and the construction industry.
One of the early bodies to be involved internationally was the intergovenmental International Energy Agency (IEA) through its wind-power programmes, now combined under the title IEA Wind. The IEA has formulated a number of recommended practices, many of which are still valid, while others have been superseded by standards of the technical body, the International Electrotechnical Commission (IEC).
Much of the current activity around harmonising standards internationally revolves around the IEC's Technical Committee (TC) 88, which oversees the IEC 64100 series of wind turbines standards. These cover everything from wind-power performance to lightning protection to terminology. Twenty-four nations are represented on TC88, with hundreds of professionals and academics working on numerous projects setting standards - these can take up to seven years to produce.
Meanwhile, there are many national standards and many government agencies require these to be used. Harmonisation will inevitably take a long time, with national representatives on the various IEC committees tending to champion their particular needs, for valid technical reasons. Wind turbines designed for South East Asia will need to be sufficiently robust to withstand typhoons. But those designed for such conditions would be over-engineered for the markets of northern Europe, where storms of similar intensity are less frequent.
With the IEC taking up to seven years to produce its standards while new international wind markets open up every year, some in the industry have criticised the IEC standards for being outdated. The industry is "constrained by existing standards", Andrew Garrad, chief executive of global wind-energy consultancy GL Garrad Hassan, said on a Windpower Monthly webcast aired in September 2011, adding that they were based on a limited geographical picture of wind speeds. "The IEC classes actually describe pretty well what happens in northern Europe, but are pretty bad at describing what happens elsewhere," he said.
A time lag between what the standards say and what the industry is doing is "inevitable", according to one member of the TC88 standard teams, who argues that this is the case across most technology-based industries.
A second technical difficulty in creating universal standards is that wind-turbine design is complex, and framing the design criteria involves a degree of simplification. This could be why government agencies tend to prefer standards devised by their own test laboratories, which they may see as tried and tested.
Yet, there is a strong case for global standards, as demonstrated by the huge variation in the two existing methods of categorising the wind speeds for which a wind turbine is designed (see table above).
In the US, wind power class 1 usually refers to a site where the annual average mean wind speed at 50 metres is no higher than 5.6m/s. At the other end of the scale, wind power class 7 refers to a site where the annual mean wind speed is 8.8-11.9m/s at 50 metres.
The wind-turbine standards formulated by the IEC, on the other hand, grade wind sites in reverse. The IEC has two standards that define wind-turbine classes: 64100-1 for large turbines and 64100-2 for small machines. Under 64100-1, a Wind Class 1 site has a wind speed of 10m/s, while a Wind Class 4 site has a wind speed of 6m/s. To confuse matters further, the IEC winds are measured at the hub height of the turbine measured, yet the US standard takes measurements at 50 metres. The US system originally used wind speeds at 10-metre height, which is of little use to turbine designers.
Although international standards that can be used on a worldwide basis is a desirable objective that would enable turbine manufacturers to work to common design criteria, in practice it is likely to be difficult to realise. It therefore seems likely that national and international standards will coexist for some time to come.
A TYPICAL STANDARD THE MEASURE FOR WIND TURBINE POWER PERFORMANCE
Most standards are quite lengthy documents and so only a small snapshot from one can be used to give an indication as to how they work. The International Energy Agency's (IEA) first recommended practice for power performance testing was formulated in 1982. It was then revised in 1990 and now exists in the form of IEC Standard 61400-12.
The practice lays out a procedure for determining the power curve (see graph) of a wind turbine, which shows the relationship between wind speed and power output over the full operational wind-speed range.
This is vitally important but difficult to establish, because the variable nature of the wind makes measuring wind speeds under consistent conditions challenging.
The IEA standard thus lays down practices to ensure power-curve data is gathered consistently across the world so turbine performances can be accurately compared.
For example, a copy of the site layout, which should be free of obstructions, must be part of the test report. The anemometers - instruments that measure the wind speed - should preferably be calibrated by an independent laboratory and placed at the turbine's hub height or within a 10% range of that height. They must also be fixed within two and six rotor diameters' distance from the wind turbine.
The IEA specifies a minimum sampling rate of 0.5Hz - one power reading every two seconds - and averages are taken over a period of ten minutes. The averages are then computed at increasing wind speeds of 0.5m/s intervals, starting 1 m/s below the turbine cut-in speed up to the rated wind speed and 2 m/s above that point to 20 m/s.
While many wind turbines have cut-out wind speeds higher than this (typically around 25m/s) the recommended practice acknowledges the difficulty of acquiring data at these higher wind speeds. At a typical site, with a mean wind speed of 7.5 m/s, wind speeds above 20m/s occur for less than 1% of the time.
The IEA recommended that at least three measurements of average power are taken at each interval. Before drawing a power curve, it is necessary to correct the power measurements so that they apply at an air density of 1.225 kg/m3.