Don't be fooled: there is no such thing as wind speed

OPINION | Lidar technology reveals the complexities of the wind, and why turbines perform differently in the lab than on site, writes Peter Clive

Ask yourself a question: if there was no wind power industry, and we had to start one from scratch today, what would it look like? How would we do things differently? With all the tools now available to us, which of our established procedures would turn out to be matters of consistency rather than correctness? Which objectives do we pursue through habit rather than relevance?

Wind is a complicated phenomenon, describing the advection of air in accordance with atmospheric physics, terrain conditions, fluid dynamics, and so on, a vector field varying in time and space. It can't be described in terms of a scalar quantity. The term "wind speed" is an oxymoron. The ways we accommodate this contradiction in how we think about wind are responsible for many limitations in our techniques for analysing it. So why do we use it?

"The term 'wind speed' is an oxymoron. So why use it?"

We have been fooled by the simplicity of our instruments. We have confused this with what we use them to measure. Yes, in a trivial sense, I can talk about the magnitude of a velocity vector at a particular time and place. But the reason I do so is because I am accustomed to using a cup anemometer mounted on a met mast. Wind speed is a convenient name for the output of such a device, nothing more. It's like talking about the average colour of an art gallery because I'm blind and have a light meter that reports ambient conditions, or describing a painting in terms of a single pixel. Trivially these may be accurate, but they overlook everything significant, and wind is the significant part in this instance.

World-first for on-turbine lidar at Fred Olsen wind farm

Read more

Lidar has opened our eyes. Its versatility has forced us to elaborate measurement use cases where previously the limitations of cup anemometry restricted our thinking in ways we didn't even acknowledge. Now we must ask ourselves not how best to replicate met mast capabilities with lidar, but what measurements lidar enables which were not previously possible. We must ask "what do I want to measure?" on the basis of project outcomes, not "what can I measure?" on the basis of assumed limitations.

Why does this matter? As we drive down wind-energy costs to fully mobilise market mechanisms in the decarbonisation of our economies, we need energy assessments that are as realistic as possible, to reveal every opportunity for improvement. We can no longer afford convenient fictions. We are already facing the consequences of doing so.

Energy estimates are revised because of blockage effects. Induction zones raise the possibility that turbine compliance tests and performance tests cannot be reconciled. These are consequences of procedures based on the capabilities of met masts and the spurious distinction between pre- and post-construction phases they introduce.

‘The choice of data collection technology isn't a zero-sum game’

Read more

Pre-construction conditions are not representative: the circumstances determining project output arise after the wind farm is built. It modifies the wind. It is part of the terrain that needs to be modelled. We need digital representations of the wind that are compatible with all phases of project delivery. "Wind speed" cannot do this.

Instead lidar datasets validate site-specific computational fluid dynamics used with aeroelastic and finite-element models to predict turbine behaviour on the one hand, and couple with mesoscale models on the other to forecast long-term yields. At no point are we concerned with what "wind speed" would have been measured in some location.

So, the next time you are measuring wind speed, remember: there is no such thing as wind speed.

Peter Clive is principal wind energy consultant at EPC firm Black & Veatch

Read Next

Data – wind power's wasted opportunity

IN DEPTH | ‘Digitalisation saves’ is becoming a mantra in the wind industry, but is enough being done with the data available? asks Darius Snieckus

29 Apr 15:48 GMT