Wind turbines and noise: potential risks and opportunities



One of the most important environmental constraints for wind turbines, especially in relatively densely populated areas, is the noise aspect.

A number of mitigation strategies are available for wind farm project developers to reduce this effect, and in any case to comply with applicable noise regulations. These strategies range from keeping a certain distance from dwellings, carefully selecting the adequate turbine type, going for blade add-on elements to reduce aerodynamic noise or applying noise curtailment modes. However, these strategies often impact the project cost, and pretty much of all them would impact expected revenues, making it all the more important to apply them carefully and correctly.

Wind turbine noise: where is it coming from?

Wind turbine noise emission are generated by:

  • Wind turbine blades: this noise is mostly generated by the last third of the blade (when going from the blade root to the blade tip), as the blade tip passes through the air at rather high speed (250km/h). A lot of focus has been given to this issue over the years and this noise has been reduced by clever design and blade add-ons. However it remains the main source of noise.
  • Machinery such as gearboxes and generators: these heavy pieces of rotating machinery convert the rotor torque into electrical power. The noise is kept inside the nacelle by efficient noise proof material.
  • Convertor, cooling fans, auxiliaries..: these components are rather standard across industries, and are located in the nacelle or in the tower.

How to reduce noise?

From a sound emission point of view, all wind turbines are not the same. For a given rated power, different turbines will produce different noise levels to start with. This is mostly due to the actual turbine, or better said, blade design. From a wind turbine control point of view, it is also possible to reduce noise emissions by slowing down the rotor, and/or pitching the blades in a less aggressive way, when the wind blows from a certain direction and at a certain speed, depending on the time of the day. This solution is software-based only, and various “noise modes” can usually be implemented in a wind turbine. Although it is a fairly easy solution to implement, it tends to have a direct impact on production as reducing rotor speed and playing with blade pitch angle has as a direct consequence over the turbine power curve, hence production. The following graph shows the equivalence between noise curve and power curve for a 2MW turbine type.


Recommendations from noise consultants and wind turbine capabilities: are they always compatible?

As discussed previously, noise modes have a potential impact on production, which will be taken into account in the estimated annual energy production of the project (typically the P50, P90 estimates), resulting in lower revenues for the project to be used in the financial profitability analysis. It is therefore very important to use the correct and realistic noise reductions in these estimates. However 3E has noticed that quite often, a discrepancy occurs between wind turbine capabilities and noise strategy. This noise strategy is most often the result of the noise impact assessment, usually performed as part of the building permit process. An independent noise expert is contracted by the project developer to assess the potential deviations from the authorized noise level limits at the surrounding dwellings. If any exceedances can be expected, a noise curtailment is proposed: a schedule of different noise modes at different times of the day, or for different wind speeds and direction. Based on this noise curtailment schedule, the production loss can be calculated.

As part of due diligence processes 3E is often charged to carry out a full technical risk assessment, over the full range of available project documents, namely building permits, Environmental Impact Assessments (EIA), design documentations, turbine technology, etc… Through these assessments, 3E has noticed that the curtailment scenarios proposed by the noise specialists are not always applicable in practice, due to technical limitations of the selected turbines. Eg wind turbines cannot necessarily switch to 3-4 modes in a day, or cannot be set up the take into account wind speed & direction and time of day all at the once. Assuming this unrealistic schedule would wrongly estimate the impact of the noise restrictions, which can have a big difference in the assumed revenues in the financial model.

In the example below, the noise specialist recommended a certain noise day/night strategy for a given turbine type, and the resulted estimated noise curtailment loss was estimated to be 2.1% of the annual production (P50). As can be seen, the turbines were due to run 3 different modes (standard, type B and type D), and were to stop in some occasions.


Table 1 – Original noise curtailment strategy foreseen for the wind turbines

However, when looking into the actual capabilities of that specific turbine type, this recommended strategy was not applicable. After discussion with the wind turbine manufacturer, the following strategy has to be used to comply with the permitted noise levels. As can be seen, the same modes apply, including turbine stop, but in a very different way.


Table 2 – Final noise curtailment strategy foreseen for the wind turbines

However, the estimated losses due to this realistic noise curtailment schedule amounted to 5% (1100 MWh/year), instead of the 2.1% (470 MWh/year) initially foreseen.

Noise mode implementation in wind turbines: is it always set up as expected?

The next issue wind turbine developers/operators can face is the potential wrong implementation of a noise curtailment strategy in the wind turbine controller. Also this is unfortunately a situation commonly observed by 3E in many projects. The example below shows a wind farm that had been running for 2 years before serious doubt arose over its production, which was well below the P90. A thorough analysis performed by 3E revealed that 2 noise curtailments were implemented when none of them was supposed to be.


Table 3 – Power curve and torque curve analysis revealing the presence of several running modes.


Table 4 – Wind speed analysis

Moreover, as can be seen in Table 4, 3E’s analysis showed that the standard mode (0) was mostly used in the low wind speed range while the reduced modes (4) and (6) were used at high wind speeds, making the production loss even more dramatic.

Potential wind turbine control software upgrade: is this worthwhile?

In some occasions, production losses can be decreased due to turbine software upgrades from the wind turbine supplier. Very often new turbine types are released with 1-2 noise modes available and more modes become available as the development of the turbine design goes on. This can actually bring interesting benefits as it may result in lower production losses for the same noise emissions. Depending of the terms and conditions of the Turbine Supply Agreement, such upgrade should come for free of charge. Production improvement can reach 2% per year, so 3E recommends to keep on talking to the wind turbine supplier on a regular basis to find out if new software is available and if it has new noise modes.

Conclusions & best practices

Wind turbine noise impact can be decreased quite substantially, but often at a cost in terms of production, therefore:

  • Ensure noise curtailment scenarios from the noise consultant are compatible with the specifications of the actual wind turbines to be erected, to avoid wrong (usually underestimated) loss estimations in the P50,P75 and P90.
  • Ensure a correct noise curtailment strategy is implemented in wind turbine controllers when commissioning the turbines in order to avoid applying the noise modes in wrong wind conditions.
  • During turbine contract negotiations, ensure that wind turbine control software upgrades are included in the O&M contract, so that any potential “improved” noise mode that may become available in the course of the O&M contract, is free and implementable right away, in order to reduce noise curtailment losses.