Archive for the ‘wind resource assessment’ Category

Wind farm layouts are pretty controversial. The bare fact is that putting turbines in the most lucrative positions which catch the most wind generally means putting them on top of hills. Which makes them visible for miles around.

There’s not really much that can be done about this conflict.

Besides, developers of wind farms don’t, as a general rule, actually buy the land they’re building on. Usually they rent it under contract from the landowner. And although the area of the wind farm is usually large, there’s usually a fair bit of spare ground around the turbines which can continue to be used for livestock or crops. In Scotland, we have laws protecting the right to access land; this means that if you want to go mountain biking at an operational wind farm the law is on your side (up to the point where you do any malicious or criminal damage etc obviously).

Once a contract has been drawn up with the landowner or landowners for a particular wind farm, it’s time to design a layout. This remains a challenging issue.

There are a number of criteria which are likely to restrict your options before the wind can be taken into account. These will include bird and wildlife surveys; land use and availability for roads; waterways and steep valleys which restrict access to heavy plant; planning restrictions on tip height; noise considerations; nearby residents; ground suitability; and local considerations such as archaeology, sites of scientific interest, and so on.

From there, the best practice is to use actual wind measurements to model how the wind flow changes across the site. Because you need at least a year’s worth of data from a met mast before you can really use the data (to cover all seasons), the reality of this part will vary substantially depending on how far into the project we are. If the project has two years’ of measurements at one or more masts on site, then great. Otherwise there are other sources of wind information we can use: bought data from a Met office measurement station; a virtual met mast built from a model; reanalysis data based on satellite measurements; extrapolation based on a combination of measurements. If the worst comes to the very worst the rule of thumb that “higher elevation = windier” would provide at least a guide.

Once you have an idea of the wind flow, you need to decide where to put the turbines. There are a number of things to take into consideration when doing this.

Each individual turbine removes a little of the energy from the wind it encounters, resulting in a slower wind speed for those turbines behind it. It also increases the turbulence, which further reduces the effectiveness of the turbines behind: it’s harder to extract energy from turbulent air. The combination of these is called the “wake” effects in the industry. To reduce the impact, it’s considered best practice to leave between 4 and 7 rotor diameters’ worth of gap between the turbines. Larger spacing is generally left in the predominant wind direction so that the overall wake effect is lower. (Offshore the spacing is larger, because wakes travel further offshore for reasons to do with atmospheric effects. Best practice will also vary from region to region based on the appropriate climate drivers.)

Trees and slopes will have several impacts on your positions. The top of a hill will be the windiest location, but steep slopes can provide huge challenges for accessing the turbines for construction or maintenance. Steep slopes also tilt the wind to an angle, and above 17° or so start to cause real problems for accurate wind flow or turbine performance modelling. Forestry increases turbulence directly above the forest, and can have other effects on the wind flow (increased change in wind speed with height, for instance, and boundary effects at the edge of the forest) which reduce the efficiency of the turbines.

Dwellings should generally be avoided as far as possible. I think the guideline in Scotland is 500m (note: there are experts on these constraints, and I’m not one), but a much larger buffer zone is wise. The issues of noise and shadow flicker are only relevant with regards to nearby homes. The danger of ice throw from blades or of blade throw is not thought to be a risk beyond tip height of the turbine (so if the turbine is 160m tall and you’re more than 500m away the risk to your property from these things is vanishingly small). To be honest I think the main driver here is the good will of the community. Big wind farms are generally built by bespoke developers, and there is much to be lost in appearing to trample over communities.

You want to maximise both the number of turbines and their output. Developers (or the banks who lend to them) take on the financial risk of a project when they sink their money into constructing the wind farm; they get nothing back until they start to produce electricity. If the costs of building and maintaining the wind farm turn out to be more than the wind farm can generate, the project is a failure. So the energy output is actually critical to project success.

Ultimately, then, from an industry perspective, the challenges of layouts are as follows:

  • Comply with all planning restrictions
  • Keep the local community on-side as far as possible
  • Space the turbines 5 by 3 rotor diameters, which for an 82m rotor diameter machine (about average for large wind farms at present) is 410m by 246m
  • Keep the turbines away from steep slopes, forestry, and dwellings as far as possible
  • Install as many turbines as you can to increase your maximum production
  • Put your turbines as high up as you can manage

I’ve often seen the accusation “poorly sited” levied at wind farms in newspaper letters. Reading between the lines, I suspect that this is because the writer objects to wind farms on hills where they can be seen, rather than that they know a secret way of establishing the best place to put wind farms that the industry hasn’t stumbled on yet.


Read Full Post »

As I said earlier, I’ve been reading the Geek Manifesto lately. This is the sort of thing you don’t read if you want a quiet life of rolling your eyes when people are wrong rather than gathering friends to overdose on homeopathic remedies outside a chemist.

It has come to my attention that one Mr Griff Rhys Jones has recently weighed in on the wind farm question. He has been quoted as claiming that wind farms are “green tokenism”, and being “randomly deposited” across the country. (Apparently this is from a column in the Radio Times, which is not online.)

Fresh from my perusal of the Geek Manifesto, it occurs to me that there are some parallels between this occurence and the British Chiropractic Association vs Simon Singh libel case. In the BCA case, Simon Singh had claimed Chiropracters made bogus claims; the BCA then sued for libel claiming damage to their reputation. In the case of Mr Rhys Jones, a celebrity has similarly made comments which are highly damaging to the reputation of an industry.

Only I can provide all sorts of evidence that wind farms are not “randomly distributed”, and that far from being “green tokenism” they actively contribute to our electricity networks saving on fossil fuels.

(I’ve not linked to much for the first statement about not being randomly distributed because to be honest it’s a bit of a blog post in itself and I don’t think anyone’s written it yet. Basically I need to demonstrate that there are financial incentives to build in the windiest places, that there are well established procedures in the industry for establishing windiness before construction, and that these procedures are generally followed. Some of the evidence may be commercially sensitive, but certainly there’s a solid case there.)

Has the reputation of the wind industry been libelled? Well, let’s be honest, Mr Rhys Jones is no more guilty of that than dozens of journalists, editors and commentors in print and even more random people online. But maybe they are all guilty of libel. Because I see far more accusations of bad practice from random people than I’ve seen any evidence of it. I’m not saying that the industry as a whole should start suing for libel when critics make rash statements which aren’t backed up by even a modicum of evidence. Neither am I saying that the wind industry is perfect. But no industry is perfect.

I do wish that celebrities, whose opinions are magnified in today’s culture, would try to remember that if it isn’t backed up by evidence it’s only an opinion.

Read Full Post »

I talked here about the job that myself and my colleagues actually do, in terms of why it would be complex to predict how the wind will flow across a potential site. I’d like to continue that topic a little and talk about how we actually measure the wind.

The first rule of good science is to make good measurements. In the wind industry, that means we want to know as much as possible about the wind which will encounter the rotors of the planned turbines. The most common instrument to use for this is a cup anemometer, which is actually a fairly simple mechanical device. Most commonly it’s made up of three cup-shaped objects which sit horizontally in the wind and spin round a central axis. By counting the number of spins we get a measurement of the wind speed. You’ll most likely have seen them; you see small ones quite commonly by the roadside, near small wind vanes.

Of course, a cup anemometer can only measure the wind speed that it’s actually sitting in. To measure at the heights of a wind turbine we need to mount it on a mast, preferably one which reaches to the turbine hub height. We also need several instruments at different heights, so that we can see how the wind flow changes with height. Then there’s the wind vanes, which give the wind direction, generally we want two of them. Multiple instruments have several advantages: they provide measurements at different locations, they can be used to sanity check measurements, and there is redundancy built in if something fails.

In fact the wind industry has generally had far higher requirements for measurement accuracy than the Met Office when it comes to wind speeds. There are wind industry professionals who visit masts in various locations and assess how accurate the measurements are, how consistent across the dataset and whether the data could be used as a reliable indication of how the wind behaves. The accuracy of the eventual dataset will depend on whether the mast is correctly sited, how the instruments are mounted on the mast, what sort of instruments are used and at what height, and whether the data are regularly checked and maintained.

Recently, lidar and sodar technology have started to really take off in the wind industry. These are alternatives to a mast, to an extent, and they work in a similar way to radar: by bouncing a wave off a moving target and looking for the reflection. Lidar units use light, generally infrared wavelengths, and sodar units use a sound-based wave. They’re collectively termed “remote sensing”, because they can sense the wind speed without sitting in the wind flow.

As it turns out, met masts with their instruments and the less intrusive remote sensing units are complementary technology rather than competitors. Met masts are large and unwieldy and cost a lot to install, but once they’re properly installed they continue to take data and require very little maintenance or additional expense. Temporary met masts which are installed for a wind project often take data for three years or even longer. Lidar and sodar can be bought outright or hired. Their huge strength is that they’re comparatively portable: lidar units in particular can generally be moved across a muddy field by two people, and they are not generally mentioned in planning requirements. A resource analyst might have two or three places on a site where the wind will be challenging to model or which are a long way from the mast but a mast can’t be installed there — in this case a short lidar deployment can really help in forming a full overview of how the wind is behaving.

Met masts are relatively simple things. Wind industry met masts are generally much smaller and less intrusive than the big telecommunications masts you see. However they have their challenges. They can be deployed in incredibly remote locations, which can make getting the required construction vehicles to the required location challenging; sometimes helicopters are required to transport the mast to site. (Note that one of the first things done when constructing the wind farm itself is building the roads. The turbines can then go along the roads. The met mast pre-dates this step, though.)

Remote sensing is obviously also a huge advantage offshore — the wind can be measured from the surface of an oil rig or even on a nearby shore rather than expensive and time consuming offshore met masts being required. The taller wind turbines of today, and the challenging terrain they’re sited in onshore, also benefit from remote sensing measurements which can be made far higher than a mast would support without any increase in cost.

There are technical differences between met masts measurements and remote sensing measurements which the wind industry as a whole are starting to get a handle on. It’s one of the more interesting elements of my job, watching the techniques and the technology changing and evolving. In many respects the wind remains something of a mystery to us.

Read Full Post »


I often read anti-wind letters to newspapers, which get linked to in various round-ups of renewables news which land in my inbox each day. Most of the time, they simply annoy me. The one above, on the other hand, made me want to clap my hands when I read it. It hits quite a few of my own personal list of things that make a good argument: it’s based on fact, it doesn’t rely on hyperbole or insult to make its point, and it bases most of its point on things which can be objectively verified or argued.

This is a good paragraph, for instance:
“Whilst [the assertion that wind turbines produce energy for 70-85% of the time] may well be true, it is grossly misleading. It should be pointed out that for much of this time the turbines will not be producing much electricity at all. For example when the wind speed falls by half, the generation falls to a level of around 10% of capacity (a cubed power law). And additionally we already know that the timing, or volume, of electricity generated is seldom matched to demand, often high when we don’t need it and low when we do. And don’t mention the shut-downs when the wind is too strong.”

The energy available in the wind does indeed follow a cubic relationship with the wind speed, meaning that if you multiply the wind speed by half, you get an eighth of the energy available in the wind. It’s not quite as simple as that, however. The turbine of course doesn’t actually remove all of the energy in the wind and convert it to electricity; in fact the Betz limit tells us that no theoretical machine ever could. Rather, it follows a power curve which remains at or near zero until the wind reaches a certain level, rises steeply (but not as steeply as the Betz limit) until a given point and then generates at capacity above that. That means that the energy in the wind is reduced by an eighth if the wind speed drops by half, but the power generated by a turbine will not drop by the same amount.

“Grossly misleading” is an interesting term when applied to a true statement. It suggests the statement was intended to detract attention from some other idea or issue. Without access to the original content I can’t really comment on that. However it is true that wind turbines generate electricity most of the time. Personally, I consider it misleading to equate the behaviour of a wind turbine power curve with the energy in the wind, but I suspect the writer is talking from knowledge of general engineering principals rather than specific knowledge of wind turbines so I think it’s forgivable.

A second point is made that Scottish wind farms generally quote a capacity factor of 30% whereas in the first part of 2011 Irish wind farms only managed about half this.

Again, this suggests limited knowledge of the intricacies of wind energy. It is not currently possible to predict how windy it will be next year, or the year after that (when your planned wind farm will be available for generation). Weather systems are chaotic, and resist any attempts at accurate predictions on a timescale longer than a few days. Before construction, the capacity factor quoted is usually based on what’s known in the industry as a “P50” value. This a target value which we expect the wind farm will meet about half the time over a long-term period. Essentially the pre-construction capacity factor is an expected annual average. Comparing it to a short-term value is not going to yield meaningful results; it’s an apples-to-oranges comparison. Really if you want to compare how realistic it is you’d want to go back ten years and calculate the average yearly capacity factor over the whole ten-year period for a few sites at similar exposure and elevation, and for a similar turbine model. (People like me earn their living doing that for developers, for banks lending money, for potential purchasers and others with a vested interest in knowing how much a wind farm is likely to generate. We are needed because it’s not straightforward to predict wind farm output.)

In fact this map suggests the Scottish wind resource is slightly better than the Irish wind resource. We have the same climate drivers, but wind speed increases with elevation and Scotland is higher than Ireland. Still, the actual output will vary from location to location.

I disagree with the letter. I think there are a couple of things the writer got wrong. Still, it’s a letter which I feel I can meaningfully answer; where there are misconceptions these can be addressed. No vitriol, no hatred, no assumption that everyone-thinks-like-me. A letter which leaves some potential for compromise, for discourse and even to agree-to-disagree. I think that’s worth savouring for a few moments.

Read Full Post »

Wind resource assessment is something of a dark art. It’s not taught in schools, you can’t study it in university: it’s handed down from one professional to their trainee bit by bit. In fairness, that’s mainly because the growth of the wind industry is still a fairly recent phenomenon and it takes a bit of time for schools and universities to catch up.

Still, it does mean that when people are asked to comment on a met mast planning application on some prospective wind farm site, they don’t really know much about why it is needed, what benefit it brings to the developer, or what it is attempting to measure. What will it look like? How much space will it take up?

I’m going to begin by thinking about the wind. Chances are that the wind is considerably more challenging and complex than you realise, unless you are an atmospheric scientist or an expert in fluid dynamics. In fact, the wind is a constantly-changing phenomenon, interacting with itself and with the ground in interesting and difficult-to-predict ways.*

An example I particularly like which illustrates this is a pedestrian walking through a town centre with an umbrella on a windy day. Town centres are full of right angles, rarely found in nature, which redirect gusts of wind in strange directions, sometimes providing shelter and at others channeling a powerful blast. Such a pedestrian would find that they need to keep changing the position of the umbrella as they walk, to avoid it being blown inside-out.

On a bigger scale, that happens with even relatively simple landscapes. Hills force the wind upwards and around, making for more wind at the top of the hill than the bottom. Buildings block the wind, diverting it around themselves and creating a turbulent wake behind themselves. Forestry provides a very challenging environment for the wind, resulting in shelter within the trees and turbulence above them. The ground itself slows the wind, so that as you move away from the ground it gets windier.

We are small fry to the wind. It covers the whole landscape, and our tiny area is relatively small. We experience only a tiny fraction of it at a time (unless we’re carrying an umbrella…) Buildings, bridges, lorries and wind turbines are far larger and more likely to feel its devastating effects.

Considering all this, it’s perhaps not surprising that there is a dark art to measuring the wind. Deciding where to make the measurements, working out how the wind will vary across the landscape, establishing the best positions for turbines, and providing a confident estimate of how windy the site can be expected to be over the next ten or twenty years are all genuine challenges faced by wind farm developers. It plays a crucial role in establishing whether the money spent on building the wind farm can be recouped, and how long it might take. This in turn helps to convince investors to provide loans, and insurance companies to provide insurance.

It’s an art, and a science, yes. But there’s no need for it to be a dark art. I’ll share some of it with you over the next few months.

* In this case I’m simply talking about the wind as we experience it near the ground. There are large scale atmospheric winds higher up, but these are not really of much interest or use to the wind industry at present, save perhaps as a potential future resource. (Back)

Read Full Post »