Location: Risø DTU, HH Koch Auditorium, on 12 July 2010, 09:00-17:00.
The background is that we got funding for a project looking at the applicability of small UAVs (two small planes, a helicopter plus a tethered lighter-than-air system) to be used as wind sensors to be freely placed anywhere in a wind farm, and/or to fly on given trajectories in any specified height and direction through a wind farm. The main aim of this workshop is to come with input to plan flight week at (hopefully) the Danish Test Station for Large Wind Turbines at Høvsøre.
The workshop had this agenda:
Challenges
Which scientific or technical problems can we solve with the new tools? And how do we use them best for our purposes?
Some of the ideas are the near and far field of the wake structure, possibly tip vortices, profiles high above what usual met masts can detect, but also more generic issues like the best way to get wind speeds out of the sensors, the obtainable accuracy and the general problem of how to compare a measurement along a line with the usual measurements of a time series from a fixed mast. More general boundary layer meteorology can be investigated as well, such as the development of the wind in different heights along a land-see boundary, the development of the night-time boundary layer or tall wind profiles as a mast extension.
Additionally, one should investigate the limits of the flight hardware: which flight patterns should be used for which purpose, how close to the turbines can we actually safely fly, and what are the upper and lower boundaries for speed for the aeroplanes.
Slides/Presentations:
Gregor Giebel, Risø DTU: The PSO project Autonomous Aerial Sensors for Wind Power Meteorology
Andreas Rettenmeier, Uni Stuttgart: Challenges - the LIDAR system on the turbine
Arnulf Knittel, hrafnkel: UAV in onshore wind power anemometry - challenges and hindrances
Capabilities
In this session, we will present our flight hardware: the SUMO aircraft, the M²AV Carolo, the Vario XLC helicopter and the Floatograph SkyDoc lighter-than-air aircraft. Additionally, the sensor package will be explained. We also will give a presentation on the test site and its instrumentation. Time permitting, a wish list for future sensor development might be added here.
Slides/Presentations:
Uwe Schmidt Paulsen, Risø DTU: Wind Sensing with a Lighter-than-air System
Colleagues’ Campaigns
I’m pretty sure that we are not the first ones to use UAVs for meteorology. In fact, there is the COST action ES0802 on the use of UAVs for general meteorology which we link to somewhat (one of the project partners is the project head of the COST action).
This is the session where we let you speak of previous experiences – please mail me a title and short abstract if you want to participate here.
Slides/Presentations:
Andreas Rettenmeier, Uni Stuttgart: The platforms of the IFC
Cerebral activity (fits better to the other C’s than Brainstorm…)
During this session, we should pull the previous sessions together in a discussion on various elements of the coming experimental campaign. Your input is needed for the preparation of Flight Week!
Additionally, ideas for the future beyond this pre-project could be discussed here.
Workshop minutes
After a round of introductions, the first session went underway with a talk of Arnulf Knittel, of wind power developer hrfnakel. His emphasise was on a better Power Production Estimate including uncertainty. To cover all relevant wind classes for a wind resource assessment, one has to get about 300 hours of measurements, which costs between 25 and 50.000 euro. If this time could be shortened with a better knowledge of the total wind field in the whole wind farm, clear savings would ensue, and the margin of error would be decreased, which would mean that the difference between the P50 (the most probable Annual Energy Production from a new wind farm) and the P90 (the value with only 10% probablility to not be reached, i.e. the value the banks look at when lending money) would be smaller. This additional knowledge could encompass certain meteorological phenomena such as cold air outbreaks in valleys in the south or zones of flow detachment behind hills. During the discussion it was made clear that manned vehicles are not an option, as they are too expensive (order of magnitude 2000 EUR per flight hour). The use case would be to have the planes on standby, and use them if there is some significant meteorological event.
Sven Erik Gryning of Risø DTU then discussed the use of tethersondes. The Civil Aviation Authorities (CAA) allow them up to 500m, provided that the line can be cut through by a plane. They work fine in no wind conditions, but have a hard time withstanding stronger winds. For a full Boundary Layer (BL) assessment, one would have to go beyond 1 km. For the aircraft, he proposed to use them for energy budgets. One difficulty in his opinion was the footprint problem: the higher the aircraft gets, the more surface on the ground is influencing the measured wind speeds. And it is not quite clear now to deal with it. One possibility would be to measure in the blended layer, where the very local influences already have been washed out - in about 2-300 m.
In the second session, Joachim Reuder of the University of Bergen presented the SUMO aircraft (Small Unmanned Meteorological Observer), a pusher model airplane of 580 g total weight. It can operate up to about 15 m/s. He has 5 of them (or had - they have recently been stolen), and recently (to the already existing pressure and humidity sensors) added a very cheap Pitot tube, which might get up to 100Hz readings (currently 7Hz due to the limitations of the circuit board). The absolute accuracy is in the order of 0.1 m/s. Due to the limitations in flow angle, it might not be able to give the correct mean wind speeds, but should be able to give the proper turbulence structure. SUMO does not have gyros, but does horizon keeping with an IR sensor. With three big turns of the aircraft, it can get up to 1500m. It can be flown as a swarm, but more than 5 planes get really difficult to observe for the safety pilot (which has to be able to take over manually at all times, according to CAA regulations).
Carsten Thomsen of DELTA finally presented the new data acquisition (DA) possibilities opened up by nano-synchronized DA boxes. They are working for the project on a small box containing a dual frequency GPS receiver, having a native accuracy of about 10 cm, with which data can be positioned accurately in space, and timed with a world-wide synchronicity of about 200 ns variation.
Ed Bervoets of Vestas R&D had some ideas for the use of such a technology, especially deep wakes behind large wind farms (especially offshore), a study in loads and the related winds, and a full wind farm optimisation for lifetime and power production.
Links:
Risø DTU
How to get to Risø
Copenhagen public transport guide