Solutions in Radiocommunications. White Papers. Assessing the interaction of radar and wind farms. Enter. Author: Cyprien de Cosson, BEng MIET

Transcription

1 White Papers Assessing the interaction of radar and wind farms Enter Author: Cyprien de Cosson, BEng MIET December 2009 Solutions in Radiocommunications

2 me Assessing the interaction of radar and wind farms Overview: Wind farm developers are being given simple advice about how to assess whether their turbines will affect radars: get somebody to help you. While this is music to the ears of companies specialising in the assessment of such effects, it is also evidence of the changing nature of the wind farms issue for radio spectrum users. The wind farms and radar question used to be simple. In large measure this was because there were not many of them. Now, there are 93 onshore wind farms in the UK with approval for 52 more and, clearly, their physical impact on radio spectrum users grows every day. In the early days of the greening of energy production, defining the effects on radar of the turbines seemed simple, too, and the Civil Aviation Authority even issued a do-it-yourself manual on how to work it out for yourself in a report in July The six-page guide at the end of its CAP 764 policy document on wind turbines provided a process to calculate the magnitude of radar reflection from both the static and moving parts of a turbine. This process had some limitations. For example, it was silent on the issue of polarisation which can make a big difference on the impact to radar. wever, within its intended scope it showed developers and managers working out impacts how to proceed without external assistance. But, the CAA has since come to realise that, as with all else relating to wind farms, it isn t quite that undemanding.

3 me Apparent Bearing True Bearing Reflections may occur off any structure*. SSR systems include techniques for dealing with reflections but these may not function adequately with very strong reflections CAP 764 indentifies a range of a 15Nm of a prospective turbine as the trigger for further investigation Source: CAP 764, CAA Policy and Guidelines on Wind Turbines * Creation of false targets is unlikely in Mode 5/S where the access address will not be transmitted on an incorrect bearing Figure 1: Illustrates how false targets are created The CAA s revised version of CAP 764 published in February of this year now recognises that it takes an expert to plan and model accurately, and it advises: Since the inclusion of this methodology within CAP 764, further work has been carried out including a number of studies examining the operational impact wind turbines have upon the performance of radar. From these studies various assessment methods have been suggested; however, many of the methods derived assume a level of expertise and require specialist software and digital terrain mapping. In other words, only those with the right experience, knowledge and tools should be trusted with this vital work of ensuring radar gives an accurate reading of an aircraft s position. While this sounds like a welcome dose of common sense to companies like ATDI who specialise in these services, that is not the key issue here; rather, the central point is that the complexity of the wind farm question cannot be ignored or underestimated. Indeed, nor can anybody just hope it will go away because rather than fading from view, the issue is burgeoning: what was once just a blur on the horizon is now the elephant in the room. 1

4 me The CAA is clearly conscious of that growing complexity and scope in this issue, and, as can be seen, its report from February of this year rows back from suggesting there is a simple way of examining the impacts of wind farms. Having advocated the Line Of Sight (LOS) method as a means of making an initial assessment of the impact of a wind farm, the organisation now adds this caveat: The LOS method described is not intended to provide a definitive yes/no answer, but to assist in determining whether a further more detailed assessment needs to be carried out. Consequently, it was decided to simplify the assessment tool contained within CAP 764 and to provide a simplified methodology usable by anyone but without the need for specialist computer software. Which boils down to: this isn t as easy as it first appeared and you probably need expert help. These changes in approach beg the question: what has changed in the last three years? Given that radar reflections may occur off any structure, why is it that wind turbines present such a big problem? Firstly, not all the CAA advice has changed. It is important to make a distinction between primary and secondary surveillance radar (PSR and SSR respectively). The latter form of radar, called Identify Friend or Foe (IFF) by the military, uses ground systems to trigger a response from transponders on suitably equipped aircraft. The interactions between SSR and reflecting structures have been well understood for a long time. More than ten years ago, Vinagre and Woodbridge reported at the IEE the effects of a 1m diameter communications mast located 60m from the SSR antenna at Great Dun Fell. They showed that 2 degrees either side of the mast the SSR made measurement errors of a ¼ degree. wever, the effect decays rapidly as the bearing from the tower increases. Apart from these subtle effects, SSR systems include techniques for dealing with reflections but these may not function adequately with very strong reflections. CAP 764 indentifies a range of 15NM of a prospective turbine as the trigger for further investigation. This advice remains the same in the old and new editions of CAP 764. So what are the issues with primary radar? There are, in general, two principal concerns. Turbines can cause clutter on radar screens. Clutter refers to radio frequency echoes returned from targets which are uninteresting to the radar operators. Such targets include natural objects such as ground, sea, precipitation (such as rain, snow or hail), sand storms, animals (especially birds), atmospheric turbulence, and other atmospheric effects, such as ionosphere reflections and meteor trails. Clutter may also be returned from man-made objects such as buildings and, intentionally, by radar countermeasures such as chaff. Its effect is to cause confusion to the operator and it may also cause returns from aircraft close to turbines to be suppressed. These two effects of clutter are interlinked. More Subtle Effects Figure 2: Illustrates the effects of clutter from wind turbine 2

5 me Typically, radar operators need to be able to detect aircraft at long ranges, often as they first appear above the horizon. Inevitably this means the radar antenna must be pointed towards the horizon. But if the antenna is pointed towards the horizon it will see the sources of clutter on the ground in the vicinity of the radar. One technique to mitigate the effects of these sources of clutter is for surveillance radars to have two radar beams, one pointed above the other. At close range the radar listens to the higher pointing beam and at longer distances the lower beam takes over. Reducing Clutter Typically we need to be able to detect aircraft at long ranges. This requires the antenna to be pointed towards the horizon. But if the antenna is pointed low (towards the horizon) it will see all the sources of clutter illustrated on the previous chart. One solution is to have two radar beams. For close work use the higher pointing beam and for longer distances use the lower pointing beam. Implicit in this concept is the idea that there must be a changeover range where there is the switch from the higher beam to the lower beam. Operators may have some flexibility in choice of the switchover range. Figure 3: Illustrates the use of two radars to reduce effects of clutter Implicit in this concept is the idea that there must be a changeover range where there is the switch from the higher beam to the lower beam. Operators may have some flexibility in choice of the switchover range and this may be employed to mitigate the effect of wind turbines. If the turbines are close to the changeover range perhaps by changing this range it might be possible to make the radar look over the top of the turbines preventing their effects from becoming visible on the radar screen. But if altering the changeover range cannot remove the effects of wind turbines, then what? Techniques to remove clutter from the radar screen have been under continuous development since radar was first invented. One characteristic of clutter that can be exploited to remove its effects is that often the cause clutter does not move. Moving Target Indication/Moving Target Detection (MTI/MTD) systems in radar are able to filter out echoes from stationary sources. wever, when the wind is blowing, wind turbine blades rotate to extract energy from the wind meaning their radar returns can pass through these filtering systems and appear on the operators screen as clutter. Even when stationary sources of clutter are removed, whether or not an aircraft is present must be judged against background noise and any clutter remaining. In older radars this meant an operator having to make a decision. In more modern radars this decision is made by a computer. If it was only noise that was a concern, noise levels could be predicted in advance and the target present or not present decision could be made against a fixed threshold. But clutter confuses this process. For one thing clutter sources are often clumped together; it is said to be non-homogeneous. So, without mitigation, clutter might give rise to bursts of false target decisions, called false alarms. Bursts of false alarms are a potentially serious problem and Constant False Alarm Rate (CFAR) processors are one of several techniques that allow radars to deal with such effects. Simply put, CFAR processors remember what conditions were like on previous radar scans of the same vicinity or what they are currently like in the immediate vicinity within a single radar transmission. Based on these stored values the target decision threshold is adjusted accordingly. But CFAR development started in the 1960s long before wind turbines needed to be considered. Consequently CFAR algorithms were not developed to deal with the very strong but irregular echoes that may originate from a wind turbine. As a result, turbines may raise the target threshold artificially high and give rise to echoes from aeroplanes being obscured from view. For the future making the CFAR processors more sophisticated may mitigate this problem. Right now it remains a problem. 3

6 me Figure 4: CFAR processing detects aircraft against backgrounds of different and changing elements. But if these problems aren t new why is there a need to change the way business is done? As was noted, the number of wind farms in the UK is growing at a dazzling pace. For developers trying to identify sites suitable for wind farms the task is getting increasingly difficult. Therefore, the interaction between radar and turbines is becoming increasingly important so as not to further constrain site selection. For radar users the need to maintain safeguards remains paramount. Without such agreed solutions, conflict will arise. And in that conflict lies the potential to compromise aircraft safety. That leaves just one way to stop the wind farms issue going round and round in circles: plan, model and act in unison. But, action can come only from knowledge and knowledge is increasingly difficult to attain as the wind farms issue becomes ever more complex and extensive. If Government targets for renewable energy are to be met then the radar and wind turbine environment is going to become more complex. More and larger wind farms are being built, some closer to airfields and the development of radar technology to make the co-existence of radars and turbines is unlikely to be a panacea at least in the short term. Therefore, for the foreseeable future not only are radio propagation tools to help safeguard radars going to be important, their fidelity is going to have to increase. Modern radio planning tools, such as ATDI s ICS Telecom can help wind farm developers and radar users by modelling the effects wind turbines have on radar systems. The following example examines if a wind turbine will be detected by a radar. Detection by the radar can lead to problems such as clutter on radar screens and suppression of returns from aircraft as discussed previously. It is possible to specify all the necessary parameters for a radar and its antenna and a turbine or turbines. In this example some typical values were assumed for an S-Band radar and the coverage is shown in Figure 5 overleaf. 4

7 me Overview P1 P2 P3 P4 P5 Contact wever, to definitely state that the turbine will be detected, it is necessary to invoke the method from the previous version of CAP764, using tools such as ICS Telecom, to calculate the magnitude of the reflection from both the static and moving parts of the turbine. The resulting tabular outputs (not shown) from this method give a maximum reflected signal from the turbine. If the reflected signal exceeds the radar s minimum detectable signal then the radar will detect the turbine. So, in this example, the maximum reflected signal from the turbine is -71dBm, significantly above the radar s minimum signal level, typically -120dBm for S-Band radar, and therefore the turbine will be detected by the radar, potentially causing problems for the radar operator. This short example illustrates how radio planning tools, such as ICS Telecom, can play a valuable part in helping both developers and radar users evaluate the potential effects of a proposed wind turbine on radars. Figure 5. The blue shaded area illustrates the radar coverage at 1000m ASL for a typical S-Band radar with range rings at 10km intervals. The holes in the coverage are due to terrain effects. The position of the turbine is also shown in white on the image. Examining Figure 5 it is clear that the turbine falls within the radars coverage area, but that does not necessarily mean that the turbine will show up on a radar operator s screen. To make a better assessment one can study the profile from the radar and turbine to determine if there is a clear line of sight between the radar and the turbine. If, as is the case in Figure 6, there is a clear line of sight then it is likely that the turbine will be detected by the radar. But, as is always the case where the needs of two sides conflict, they key is communication. If wind farm developers talk to radio spectrum users before a single new turbine is built, and if both sides use the expertise available, all parties can find the solution that works best. With acknowledgement to Alan Collinson OBE, FIET for his technical contribution and creation of photos and images. Figure 6. Profile display between the Radar on the left and the turbine hub on the right. The blue ellipse indicates the 1st Fresnel zone and shows a clear line of sight between radar and turbine. 5

8 me ATDI Ltd Kingsland Court - Three Bridges Road Crawley - West Sussex - RH10 1HL - UK Tel. +44 (0) Fax. +44 (0) enquiries@atdi.co.uk Advanced Topographic Development and Images Limited Company Registration Number Copyright 2009 ATDI Ltd. All rights reserved. Specifications, terms and conditions are subject to change without prior notice. Solutions in Radiocommunications