POWER FROM THE SKIES

I

Producing wind energy efflciently and cheaply is a are modified and scaled up surf kites. They are 25 and 50m2 (figure 3), compared to challenge. The Laddermi/1 is a wind energy genera­ regular surf kites that are usually be­ tor that uses kites to produce energy. By eliminating tween 8 and 16m'. The cable is a 1 km long dyneema cable with a diameter of 4mm. heavy structural components and using the strong winds above conventional height, the A remote control unit that is suspended some 20 meters below the kite controls Laddermi/1 has the potential to produce energy cheap­ the steering and attitude of the kite (figure 4). This control unit is currently operated er than conventional wind turbines. Recent develop­ manually but all the provisions are there ments in the Laddermi/1 project have brought the mass for automated control. The control unit returns its control positions and status production of kite power a step closer to reality. back to the ground. This data is displayed on the telemetry computer together with TEXT Aart de Wachter MSc., Project Manager Ship, Institute for Applied Sustainable Science, Engineering and Technology (ASSET), TU Delft data from the inertial measurement unit (IMU) that is attached to the kite (figure THE LADDERMILL SYSTEM (figure 1). This cross-wind flying increases 5). The IMU measures GPS velocity, accel­ The Laddermill uses the pulling forces of the force on the kite by the square of the erations, attitude and rotation rates. The a kite to produce power. The kite pulls a lift-to-drag ratio. The kites that are cur­ winch is controlled by a separate com­ cable from a drum. The rotation of the rently used on the Laddermill are scaled puter. Data from load cells and encoders drum drives a generator, a process known up kite surf kites. These have a lift-to-drag is used to measure speeds and forces. A as the power phase. After the cable has ratio that is close to five, so the pulling PID controller, that controls the speed of been pulled out for a few hundred meters, force of the kite can increase roughly 25 the drum, takes care that the forces stay the kite is configured for low force and is times by flying cross-wind. within the limitations of the system. Final­ winched back to its starting altitude in the ly, a wind meter on the ground measures depower phase. This is called the pumping ASSET owns and operates an experimen­ the wind speed and direction at a height Laddermill because it works with a single tal setup of the Laddermill that consists of6m. kite that operates in a pumping motion. of a winch/generator with a nominal out­ During the power phase the pulling force put power of 20kW. It has a drum that can The data from all the systems is logged of the kite can be greatly increased by let­ hold up to 20km of cable (figure 2) . The together with data from cameras at the ting the kite fly figures of eight or circles kites that are used for power production groundstation and on the kite, and is used

I Leon,.doTime• DECEMBER 2010 J I

r--

.r ,- -. II'I ~ .._-~--'-- -- . ..-...... •I .. I }~ F1gure 1. Mov1e sequence of the Laddermlll k1te flymg figures of e1ght

F1gure 4 The k1te control un1t de­ veloped w1th the Fns1an F1gure 6. Wmd d1stnbut1on w1th he1ght averaged over compames a 20 year penod at De B1lt [Source KNMI]

to improve the system step by step. the shelf again. The technology got a kick­ RECENT DEVELOPMENTS IN DELFT start after Wubbo Ockels got a patent on One of the highlights of the Laddermill HIGH ALTITUDE the Laddermill principle in 2000: a contin­ project was the demonstration of the sys­ High Altitude Wind Power (HAWP) and uous rotating loop with multiple kites at­ tem in the cityofGroningen in 2007 where (AWE) are the com­ tached to it. The ascending and descend­ the Laddermill powered the electric gui­ menly used names for this kind of tech­ ing kites drive a generator on the ground. tar of guitarist Jan Akkerman during one nology where wind energy is harvested In 2004 research at the Delft University of of his rock concerts. Other highlights in­ above conventional wind turbine height Technology on this principle started. De­ clude the Laddermill featuring on the BBC (150m and up). The reason to go higher velopment of the rotating Laddermill was show "James May's Big Ideas" in 2008 and than conventional wind turbines for wind thought to be too complicated to start delivery of the brand new 20kW genera­ energy production is that the higher you with. A pumping Laddermill with a single tor. To operate the 20kW groundstation go the stronger and the more constant kite would be more suitable to develop some major changes needed to take place the wind becomes (figure 6). One of the the technology of automated kite control compared to the 3kW groundstation that big challenges is that you cannot build and the groundstation and would also be was used for the demonstration in 2007. a tower that high and economically pro­ a viable concept for power production. Bigger kites were needed. But bigger kites duce wind power. So going higher re­ The release of new scientific publica­ require also better and more reliable con­ quires a change of technology. Some SO tions triggered the interest of others and trol over the kites and over the forces. At groups worldwide, both commercial and since then AWE has seen a steady growth the end of 2009 together with the Aero­ academic, are now active in the field of of groups that started to do research of space Software and Technologies Insti­ AWE to develop this technology. The Delft their own. Another big breakthrough was tute (ASTI) special software for the speed University of Technology is at the fore­ when in 2006 Google invested $10 million and force control of the groundstation front of this development. in the American start-up company Makani was developed. With the new groundsta­ to develop a demonstrator of the tech­ tion ready for use an important missing Although the idea of using kites to pro­ nology. This showed that investors were piece was a suitable test location close duce power has been out there for quite a serious about the new developments. to Delft. The former naval airfield Val ken­ while, the development of the technology This opened the door for other groups burg, close to Leiden, met all the require­ only really started after the publication of to apply for funding. Not much later the ments for testing the Laddermill and has the article"" by Miles Kite Power group in Delft received €1 mil­ been the home base for Laddermill tests Loyd in the Journal of Energy in 1g8o. He lion from the Rotterdam Climate Initiative since the start of 2010 (figure 7). worked on the idea together with his sons. to develop a demonstrator of a pumping But after several experiments his sons got Laddermill to power a ship. The old kite control system was outdated other interests and the idea was put on and could not handle the larger forces DECEMBER 2010 Leonardo Times I that were required to produce 20kW. At visited the test site to report our progress. institute. This institute, which was formed the start of 2010 new prototypes were They saw the Laddermill reach altitudes at the start of 2010 is a cooperation of the built of a central kite control unit that was up to SOOm and produce a peak power of faculties of AE, IDE, EEMCS and 3ME. This suspended below the kite. The principle 13kW. It was a very successful test cam­ makes it possible to have students, PhD's worked. A partnership was setup between paign which got a lot of positive attention. and employees of different faculties work ASSET, ASTI and the Frisian companies Vi­ together on the development of sustain­ aferro and SMI. The four together received AWEC2010 able technologies. The Laddermill project a subsidy from the province of Friesland The second annual Airborne Wind Energy is run by a team of six researchers and to develop and build the new kite control Conference (AWEC) was held at Stanford PhD's (figure 9). They are assisted by sev­ units (figure 4) . These new control units University on September 28th and 29th. eral master students and work together will enter flight by the end of 2010 and The conference is organized by the Air­ with several companies that sponsor and will make it possible to start experiment­ borne Wind Energy Consortium, which support the project. ing with automated flight control. unites the stakeholders of the industry into a single, focused voice. The confer­ As AWE is a quickly developing field, students To develop the bigger kites a close co­ ence was attended by academics, indus­ who would like to do their master thesis or operation with dr. Henry Rebbeck, kite try representatives, investors and the PhD in this field are invited to contact the developer of Mutiny Kites and former Delft University of Technology. Speakers Kite Power group. Amongst others, topics developer of Flexifoil kites, was set up. included representatives from Stanford, for research are development of automated He designed a first and second genera­ UC Davis, UC Chico, UC Berkeley, the FAA, control for kites, automated launching and tion of 25m 2 leading edge inflated (LEI) ARPA-E, NREL, PG&E, venture capital firms landing of kites, fluid-structure-interaction kites that are specifically designed for the and Roland Schmehl of the Kite Power (figure 10), materials for large inflatable Laddermill. They differ from conventional group in Delft (figure 8) . Miles Loyd re­ kites, development of a JOOkW groundsta­ surf kites by their size, strength and abil­ ceived an award for his contribution to tion and integration of the Laddermi/1 with ity to depower (fly at low angle of attack). the field of airborne wind technology, current air traffic. The Laddermi/1 project In March 2010 the Laddermill produced speaking about his experiments on early offers a great team, interesting work and 20kW for the first time. kite technology in the late 1970s. "We a nice balance between design work and built computer simulations but ran into practical tests in the field. ?!' DEMONSTRATION FRIESLAND the limitations of early computers:' Part of the development of the control References unit with the Frisian companies was a The topics discussed included issues on demonstration of the Laddermill system how to integrate AWE technology with in Friesland. After the necessary permits current air traffic, legal issues, and how from the airspace authorities were ob­ to attract more investors. Furthermore, it tained the demonstration could take was a great opportunity for all parties to • • • place on a farm close to the Frisian village show their current state of development Jirnsum on the 9th and 10th of Septem­ and discuss technical issues. ber. The community had released a press release, which resulted in significant me­ STUDENTS AND PHD'S dia attention for the event. Several Frisian The Laddermill project and Kite Power TV and radio channels and newspapers group are part of the interfaculty ASSET I LeonardoTimes DECEMBER 2010