Search: go

Aviation

Wind Farms Impact on Radar Aviation Interests

This study was 100% funded by the DTI Sustainable Energy Programme

Objectives

A study has been completed by QinetiQ to provide a detailed understanding of the interactions between wind farms and radar systems. The main objectives were as follows:

  • To determine the effects of siting wind turbines adjacent to primary air traffic control radar;
  • To determine the extent to which detailed design of wind turbines influences their effects on radar systems;
  • To determine the extent to which the design of the radar processing influences the effects of wind turbines on radar systems;
  • To provide text suitable for inclusion on the UK guidelines on Wind Energy and Aviation Interests.

Summary

This study has focused on the development and validation of a computer model that can be used to predict the radar reflection characteristics (Radar Cross Section, which is measured in square metres and is normally presented on a logarithmic scale) of wind turbines and understand the complex interaction between radar energy and turbines. The scope of the model includes:

  • The affects of the radar propagation over the terrain between the radar and the wind farm;
  • The dynamic radar scattering from the wind turbines;
  • The signal processing in the radar;
  • Display of results via a simulated radar display.

The model was validated through a full–scale trial, using a QinetiQ mobile radar system to collect data for a single operational wind turbine at Swaffham in Norfolk. The model was then used to perform a detailed sensitivity analysis and to compile a list of the key factors influencing the radar signature of wind turbines.

The following are some of the results generated by the project:

  • The design of the tower and nacelle should have the smallest Radar Cross Section (RCS) as possible. The RCS of these components can be effectively reduced though careful shaping and choice of construction materials;
  • Large turbines do not necessarily lead to large RCS (i.e. tower height does not greatly affect RCS);
  • Blade RCS returns can only be effectively controlled though the use of absorbing materials;
  • Spacing of wind turbines within a wind farm needs to be considered in the context of the radar cross range/down range resolutions.
  • Spacing the turbines such that only one turbine can appear in any range cell has advantages in identifying the wind farm, filtering out the turbines and in tracking aircraft over the farm area;
  • Single wind turbines do not create a significant ‘radar shadow’. Any shadow region is only dark to a distance of a few hundred metres behind the turbine. Beyond this there is some reduction of the radar power, and a time-variation, but these will not prevent detection except possibly for very small targets.

This study complements the recently the completed study by AMS which looked at the feasibility of modifying radars to remove the effects caused by wind turbines (report number W/14/00623/REP).

Contractor details

QinetiQ Ltd, Malvern Technology Centre, St Andrews Road, Malvern, Worcestershire WR14 3PS
Tel: +44 (0)1684 894118

Report Number: W/14/00614/00/REP

Background

The process of obtaining planning permission to build a wind farm involves many considerations, including consultation with various aviation stakeholders. These parties may raise objections for a variety of reasons, with a known source of objections being that the wind farm may appear on the display of air traffic control radar.

Decisions made regarding the likely impact that a wind farm may have upon radar operations are currently based upon assumptions. The electromagnetic interactions between a wind turbine and a radar signal are complex and there is currently limited understanding in this area and no accepted method for quantifying this potential impact.

A conflict of interest currently exists between the desire to encourage wind farm development as a renewable energy source and the desire to maintain the operational safety of air traffic. The practical manifestation of this conflict is that the UK has seen objections against a significant proportion of proposed wind farms on the grounds of aviation safety.

The DTI has established a "Wind farms, civil aviation and defence interests working group" to tackle these issues. As part of the activity of the working group, QinetiQ were commissioned to undertake this study.

Field measurements of a single wind turbine


Figure 1: CAD model and photograph of the Swaffham turbine

Field measurements of a single wind turbine at Swaffham in Norfolk were successfully completed over a five day period during July 2002. Figure 1 includes a photograph of the turbine along with the CAD model used for RCS modelling. In total, nearly 250 RCS measurements of the turbine were recorded covering a range of different turbine operating parameters (yaw angle, blade pitch etc.). The yaw angle is the orientation of the turbine relative to the radar, with 0° being when the plane of the turbine blades is normal to the boresight (pointing direction) of the radar. Many of the measurements were made in both wet and dry conditions. For each of the RCS measurements we also collected rotor revolutions per minute, blade pitch and yaw angle data from the turbine control system.

A typical set of results are shown in Figures 2 and 3. Figure 2 shows how the RCS of the wind turbine varies with time. The figure shows data for just under 2/3 of a revolution, with the RCS profile repeating three times per revolution (ie. Every 120° of rotation). Analysis of the Doppler spectra results in Figure 3 shows a large expected zero Doppler spike containing all the reflected energy from the stationary parts of the turbine, and significant returns out to the Doppler frequency equivalent to the tip speed of the blades.


Figure 2: Measured RCS of the Swaffham turbine versus time


Figure 3: Measured Doppler spectra versus time

To give a sense of scale to these results Figure 4 shows the range of RCS values for a number of different objects. Data from the radar at RAF Marham, which has line of sight to the Swaffham turbine was also collected during the trial. This was later analysed to assess the validity of the radar display simulation part of the model.


Figure 4: RCS value for a range of different objects including wind turbines

A selection of the measured data from the trial is available on CD from the Renewable Energy Helpline (tel 01235 432450, email NRE-enquiry@aeat.co.uk).

Model validation

Validation of the model was carried out in two separate stages:

  • The RCS data and radar propagation predictions were checked against data collected from the trial;
  • The simulated radar display results generated by the model were compared to the display recorded from the primary radar at RAF Marham.

The RCS predictions showed good correlation to the measured data as can be seen in Figure 5 below. When the simulated radar display results were compared with the recordings from RAF Marham a level of agreement between 93% and 98% was achieved.


Figure 5: Typical set of measured (green) and predicted (black) RCS data

Wind turbine RCS sensitivity

The important parameters that influence the RCS of a wind turbine include:

  • Blade design (shape, size and construction materials)
  • Tower design
  • Nacelle design

The shape of the turbine blade defines how the scattering changes in time. However, it is not a parameter that can be changed easily, as it must be designed to be aerodynamically efficient. The material the blade is manufactured from affects the amount of energy scattered and, hence, how easily the turbine is detected by radar. Turbine blades are mostly made of fibreglass, and this could be modified with specially designed Radar Absorbent Materials (RAM) to reducing the RCS of turbine blades.

The turbine tower RCS return is a function of frequency, shape and material, and is not affected by pitch or yaw, as it is the stationary part of the turbine. We have looked at the variations with frequency and basic dimensions. The tower contributes a constant return, which should be minimised to aid in the filtering of the tower return by standard MTI radar processing. This processing is designed to suppress stationary objects, but if the RCS is very large (typically greater than 40dBsm) the MTI can fail.

Important points to note from this study are that for a tapered cylinder tower the key parameters are the taper angle and the radar frequency. The tower diameter and height are not as significant so long as the tower is many wavelengths high, which is always true for radar frequencies above 1GHz. A small change in the taper angle can have a large effect on the RCS (at 0.88° the RCS is ~20dBsm, at 2.88° it has fallen to 10dBsm, and by 5.88° the RCS is ~3.5dBsm).

The nacelle design affects the turbine scattering in the same way as the tower. Therefore, minimising this return makes the returns from the nacelle easier to remove by radar filters and reduces the possible detrimental effects to the radar. The nacelle rotates only on one axis, so this leaves considerable scope for its signature to be reduced by altering its shape. The nacelle modelling considered two designs, one egg shaped and one rectangular. The RCS values as a function of yaw for the rectangular design is shown in Figure 6.


Figure 6: RCS for a rectangular nacelle as a function of yaw angle

The large spikes in the RCS pattern occur at a yaw of 90° when looking normal to the nacelle side panels. At 3GHz these spikes are 43.8dBsm which is potentially large enough to defeat many radar MTI filters. To remove these spikes the side panels should be angled so that the surface normal cannot point at the radar. This can be done by tilting the face upwards, with an angle of 10° reducing the spikes from 40 to 50dBsm to below 20dBsm.

The RCS for the egg shaped nacelle was found to be more uniform as a function of yaw angle, with values typically in the range 5 to 20dBsm. This results from the curved surfaces, which reflect the incident radar energy in all directions quite evenly. With careful design the nacelle of a wind turbine can be designed to give low returns in all yaw directions.

Conclusions

The results from the project have enabled us to provide a much more detailed quantification of the complex interactions between wind turbines and radar systems than was previously available. The key turbine factors influencing the effect of a wind turbine on radar are:

  • The design of the tower and nacelle should have the smallest RCS signature possible;
  • RCS of the tower and nacelle can be effectively reduced though careful shaping;
  • Large turbines do not necessarily lead to large RCS (i.e. tower height does not greatly affect RCS);
  • Blade RCS returns can only be effectively controlled though the use of absorbing materials;
  • A low probability of detection, but a large clutter return can be expected when the yaw angle of the turbines is close to 90° from the radar direction;
  • A high probability of detection, but a smaller area of clutter, can be expected when the yaw angle of the turbines is close to 0° and 180° from radar direction;

The key factors influencing the effect of wind farms on radar are:

  • Spacing of wind turbines within a wind farm needs to be considered in the context of the radar cross range/down range resolutions. Spacing the turbines such that only one turbine can appear in any range cell has advantages in identifying the wind farm, filtering out the turbines and in tracking aircraft over the farm area;
  • In a circumstance where a single wind turbine in clear line of sight to the radar is undetected, it is highly likely that a wind farm of similar wind turbines would also be undetectable;
  • No optimal layout or format has been prescribed as each wind farm will have its own specific requirements dependent on many factors.

Key terrain and shadowing factors include:

  • In non line of sight situations, the level of detectability of the wind farm is dependent on the frequency of radar and the distance from the wind farm to the point of diffraction and the distance below the line of sight horizon, where the wind farm is located;
  • Single wind turbines do not create a significant ‘radar shadow’. Any shadow region is only dark to a distance of a few hundred metres behind the turbine;
  • Beyond this there is some reduction of the radar power, and a time-variation, but these will not prevent detection except possibly for very small targets.

Potential for further use of the model

The model has the potential to be a valuable tool for the wind energy community and can provide a validated route to:

  • Generate the detailed data required for more sophisticated initial screening of potential wind farm sites than is currently available;
  • Support the development of mitigation and solutions, including: siting optimisation, control of wind turbine RCS and enhanced radar filters (able to remove the returns from wind turbines).

Full report

For further information the complete report can downloaded in pdf format: