Understanding inflow angles

So, we've been using WAsP for a while now, and we've come up with some great layouts optimised for energy production - however what is this pesty inflow angle that keeps coming up in the turbine manufacturer's siting reports? Is this something that really has to be considered?

The inflow angle is the angle (off the horizontal) at which the mean flow comes into the rotor, and we want to keep this as close to the horizontal as possible. It affects both the turbine loading, and the aerodynamic efficiency of the rotor (the power curve). As per IEC 61400-1 edition 3, the turbine manufacturers' design requirement for inflow angles are +/- 8 degrees (from the horizontal). These inflow angles are used as the design basis for the aeroelastic load calculations, so they really drive the fatigue loads. In practise, the more critical inflow angle is positive (coming up into the rotor), and this is usually the only one calculated by the turbine manufacturer as this is conservative. The reason for this is predominantly due to the rotor tilt being positive, and therefore a slight negative inflow angle (coming down into the rotor) can sometimes be a beneficial thing for fatigue loads.

Here are some outputs from WAsP Engineering 2.0 of inflow angles as it comes over a North-to-South ridge (wind from the west), it shows the positive inflow angle as it climbs over the ridge (green) and negative as it descends back into the valley (red). For this theoretical project, you can see the turbines right on the crest of the ridge where the flow is relatively neutral before it descends again - coincidentally this is where the high wind is and where we should be aiming for.

Sometimes it can't be avoided putting turbines in higher inflow angles, particularly for more complex sites where we are trying to optimise capacity, however in practise you should avoid exceeding inflow angles of +/- 4 degrees to minimise loads and maintain aerodynamic efficiency (power curve). If you exceed this you'll have to dive into more detail on the loading, and the accuracy of the power curve will be called into question (the manufacturer's power curves are also usually linked to a narrow range of inflow angles).

I'll devote a future post to the aerodynamics of inflow angles, and how this affects turbine loading and the power curve. For a taste, just think positive inflow into a rotor with rotor tilt - we're starting to venture into helicopter aerodyanmics!

Melbourne offshore wind: WAsP in the southern hemisphere

Following on from my last review of WAsP 10, Risoe have now fixed up the southern hemisphere issue for their map editor, and more importantly for their Google Earth plug in. Now even guys in the southern hemisphere can visualise their latest layout in 3D rendered Google Earth brilliance!

Check out a preview below, where you can see the purely ficticious offshore project that I quickly knocked up in my hometown of Melbourne, Australia. Purely a figment of my imagination, however it would be rather nice...

WAsP 10 review

The latest version 10 of Risoe's wind-analysis software WAsP has been released for a few months now, and after working with it for a while I thought I'd post a few thoughts and comments on how it's working.

WAsP is a micro-scale wind modeling software package used to estimate wind speeds over a region based on measured wind speeds at discrete points (typically met masts). WAsP also forms the backbone for other wind packages such as Windfarmer and Windpro. The orographic-flow model used by WAsP is the 'BZ-model' of Troen (1990), which in simple terms places a high-resolution 2D polar grid over the measured wind source to estimate the flow perturbations over the region - so this isn't a CFD package. Importantly, the resolution and accuracy of the grid reduces the farther away from the initiation point. The flow model used to estimate the perturbation also assumes that there is no flow separation, and therefore the validity breaks down with more complex terrain.

One of the biggest changes in this release in my opinion is the new Google Earth integration, and it works very well (in the northern hemisphere at least - see below). In a few simple clicks, WAsP will send your live turbine layout and met mast locations into Google earth so you can visualise the turbines and masts, as well as your topographic and roughness assumptions, and any resource grid you run. It's a well implemented feature and actually models your turbine tower height and rotor diameter based on your .wtg model. Eye candy aside, this can really help in layout development as it allows you to have your land boundaries and other GIS data in Google Earth (as .shp files for example) and develop your layout around it. It also allows you to bring the satellite imagery back into WAsP as an overlay image.

Official demo site in Portugal - wind resource grid and turbines visible.

There are also some great user friendly changes in WAsP, such as the turbine diameter circles around your positions (this was my favourite), and the added reference locations (which allows you to check how the wind modeling is handling certain features of interest, such as other masts).

workspace view with orography, turbines, and spacing visible

Unfortunately a minor bug in WAsP still hasn't been fixed, and that is that it assumes all map projections are in the northern hemisphere. What this means is that for projects in the southern hemisphere, the WAsP map editor can't convert between datums and projections, and the Google earth integration doesn't work! The actual wind modeling isn't affected however, so it's more of a cosmetic issue that can be overcome with another GIS package. Risoe have told me they are working on a fix for this however.

Overall however I think this is an excellent and constantly evolving software package, and is well
worth the one-off EUR 3300 license fee. For users of Windfarmer and Windpro, I thoroughly recommend looking into your base flow model again!

note: I have no affiliation with Risoe/WAsP and have fully paid for my licensed copy! :)

LIDAR-based wind turbine control system

There's been a little bit of interest in laser based anemometry, or LIDAR, wind turbine control systems. It's been the precept of Risoe's Windscanner program below, and several Risoe-DTU papers at EWEC 2009. It even made this month's Economist technology quarterly. As rotors get bigger, I believe this concept will become more and more import - both for energy production optimisation, and for loads reduction on the blades and drive train.

So what is LIDAR? LIDAR is simply the use of a laser in the same way a RADAR is used, by measuring the time it takes for pulses to return after they have bounced off minute particulates carried in the air, a profile of the wind speed can be developed. If used in an array, a three-dimensional picture of the column of approaching wind can be more accurately estimated.

How does this help a wind turbine? At the moment, wind turbines estimate the wind speed based on a nacelle-based anemometer - typically ultrasonic, or a traditional cup anemometers. The wind speed measured here is used as the basis for the controller to estimate the wind speed over the entire rotor and set the appropriate pitch setting of the blades to optimise the angle of attack. Wind turbine blades operate most efficiently at only one angle of attack setting - too much, or too little reduces the aerodynamic efficiency and therefore the energy output.

The bigger rotors get, the less accurate it is to estimate the entire wind speed of the column of wind based on one single anemometer input. So, a LIDAR can help by providing a more accurate distribution of wind speeds of the incoming wind. Armed with this, the controller can then anticipate more accurately the pitch settings to use (possibly even individual blade pitch settings). In the case of an extreme gust, the blades could then feather and dramatically reduce the loads on the turbine, where as a traditional system will only know a gust after it's experienced it.

There are some definite practical engineering problems with integrating this system however. If you base it on the ground, how do you yaw it efficiently to align with the turbine? If you base it on the nacelle, you have to place it in the rotating hub and allow for the complications of rotating a delicate laser array continuously. Whether or not the increased cost and complication of this system will be balanced by the production increase and possible cost reduction in structure optimisation: only industry experience will tell.

350.org international day of climate action this 24 October 2009

350.org are organising numerous actions across the globe this coming Saturday 24th of October 2009, you can see a full map below of where you can find them.

For those of you unaware, 350.org is trying to raise public awareness of the science of climate change, and prompting open public debate about our CO2 emissions - specifically returning from our rapidly increasing 390+ ppm of CO2 to 350ppm to avoid the serious impact of climate change. The key message here is that we have already passed the safe limit, and we need to get back - not argue about how much further we can go.

Key to returning CO2 levels to 350ppm is the future energy sources we use as a planet, a significant part of this mix will be wind energy - the most cost-effective and globally accessible source of renewable energy we have at the moment. It's a great message, and well worth getting out and showing your local support!

View Actions at 350.org

extreme cyclonic weather

Based on some recent experiences, I thought I'd recap on extreme cyclonic weather and how it influences wind-turbine site analysis.

Cyclonic weather relates to weather systems that rotate clockwise around low-pressure systems in the Southern hemisphere, and anti-clockwise in the Northern hemisphere. Once they exceed a set level (and there are various classification systems) then they are called: typhoons in the Pacific, Hurricanes in the Atlantic, and tornadoes when they are land based. Tornadoes can be spawned from typhoons/cyclones/hurricanes when they hit land.

Research in the field is fairly limited, however they seem to be characterised quite well by the Holland model: a warm centre surrounded by colder air with no fronts. Extreme winds occur in the outer eye wall. As the cyclone rotates, centripetal acceleration is balanced by the pressure gradient (suction internally), therefore the lower the eye pressure the higher the wind speeds.

example cyclone passage over a weather station

So, how do we deal with them? In my opinion they affect wind turbines in three ways:

• extreme wind speeds
• extreme changes in direction as the eye passes
• grid-outages, and the turbines ability to 'center' with no external power

On the extreme wind side, the Vref (10-minute average) and Ve50 (3s gust) should be investigated closely, difficult to get data however any cyclone tracking station data should have estimates of these speeds. These are typicaly at 10m height, so you have to extrapolate them to hub height, a wind shear of around 1.2 is typically representative. However how do you estiamte the one-in-fifty cyclone strength? To do this, I did a Gumble analysis on peak cyclone wind speeds in the area. Failing this, the local building codes may also give you an idea of the extreme wind speeds (however not recommended!).

Extreme wind direction changes can be estimated by looking at the cyclone's speed along the path, and estimating the angle change over time. The peak rate of change can be compared to the IEC standards.

Grid outages are typically coincidental with cyclone passages, as they knock down power lines in the area. Although a turbine might be rated to class I extreme winds (DLC 6.1) the load case with loss of power (DLC 6.2) doesn't use safety factors as it's an accidental load case - therefore the loads are significantly lower. For maximum survivability then, it's worth investigating a UPS (a diesel backup generator for example) to kick in when a cyclone is coming to keep the turbine centering into the wind. These are really 'soft systems', where the turbines are manually stopped switched over to the UPS, and not put back into run again until the cyclone has passed.

For further reading, I strongly recommend checking out Risoe report RIS-R-1544.

power-line mounted turbines?

Check out this winner of the 2008 red dot design awards by Nils Uellendahl, it's a concept for a helical turbine that mounts directly onto over-head power lines. It's quite interesting in two ways:

• it generates electricity directly into the line using induction
• It's actually horizontally mounted, however is actually a traditional vertical-axis turbine

I'm sure there are some large issues with the electrical connection and grid stability (to say the least!), and with the additional loading of the power lines, however it's a great concept.

24 October 2009 - an international day of climate action

In preparation for the December '09 Copenhagen round of international discussions on climage change, 350.org is busy preparing local actions around the world on the 24th of October to reiterate the goal of a return to 350ppm of CO2 - a worthy aim for any climate change negotations indeed!

The Copenhagen round will be keenly watched by the renewable-energy industry, as it could well provide a huge boost to meet any meaningful CO2-reduction target. It's fair to say that the wind industry will be set to receive the lion's share of any boost, as it's commercially the most ready technology we have available at the moment. A 2010 boost would be welcome news to an industry that is starting to feel the effects of the GFC through hard-to-find financing.

Big rotors king for 2010+

A few years back, it was quite common to see various turbine manufacturers releasing their latest Class IA turbine, followed rather quickly with a class II and III rotor. This was typically just a longer blade with minimal nacelle changes. The Vestas 2WM, Siemens 2.3 MW, and GE 1.5 MW platforms are good examples of this.

This was understandable at the time, as there were numerous easily exploitable IA sites around, with demand in the market for suitable class IA turbines. However now in 2009, as we are rapidly shooting past 100 GW of installed capacity around the world, the number of high-wind sites that are also close to the grid and easily developable are becoming quite rare (although there is no shortage of high-wind sites in the world) - a quick chat with any developer will tell you that! The dominant sites for 2010+ are looking to be for class II and class III.

Preempting this, we are seeing the latest generation of turbines being released with this in mind. No longer are turbine manufacturers jumping straight in with their smaller-rotor IA machines. The first release looks to be the II/IIIA specific turbine, with the IA coming later as an after thought (if at all). Have a look at the new Repower 3.xM, Vestas V112-3MW, and GE 2.5XL. Adding to this the last iteration of existing nacelles with new class III rotors: Siemens 2.3-101, Vestas V100-1.8, and the Nordex N100-2.5 ; big rotors are set to be king for 2010+.

Picture of the SWT-2.3-92 I took in the US.  Progression from Class IA (82m rotor) to class IIA (92m rotor) to class IIIA (101m rotor) in around 5 years.

density correction of stall-controlled power curves

For energy estimates at non-standard IEC conditions (p=1.225 kg/m3), the power curve should always be density corrected. For stall-controlled turbines this can be quite tricky.

If you follow the IEC 61400-23 recommendations (which are for power-curve testing) then a simplistic ratio approach is recommended. However, this standard is intended for normalising datasets at 'near sea level' conditions - not for detailed energy estimates. In reality, a stall-controlled turbine will be tuned for site, with the pitch setting feathered for lower density so that rated power is always hit.

For example, if we take a 2000m site (p=0.99 kg/m3) and correct it using IEC measurements we see a large decrease in energy (around 25%) - whereas the actual power curve will be around 10-15% decreased after adjusting the pitch settings. In this case, the pitch settings were feathered by around 2 degrees to achieve this power curve.

Therefore, the site-specific power curve should always be used for energy estimates, and this should be sourced from the manufacturer. For sites at higher altitude (I would say above 1000m), then tested power curves should be insisted on as the aerodynamics of the turbine will change (particuarly for stall-controlled turbines) which invalidated the original testing. I'll go into more detail on stall-controlled turbine aerodynamics in a future post.