Read Der Ampelmann text version

Der Ampelmann

Safe and easy access to Offshore Wind Turbines ----- Patent Pending ----J. van der Tempel, D. Cerda Salzmann, D-P. Molenaar, Th. J. Mulder, S. Hoonings, G. Etna, J. Koch, F. Gerner Duwind Delft University of Technology [email protected]


Offshore wind turbines require maintenance. But to reach the turbine to apply the maintenance is difficult and costly offshore. The main problem is to transfer personnel and equipment from a moving vessel to a fixed turbine. To overcome this problem, this paper explores a new transfer method: the Ampelmann. This system consists of a vessel mounted Stewart platform. By measuring the vessel motions and real-time control of the actuators in the platform, the top plate becomes stationary compared to the fixed world. For this concept a design was made for a typical offshore wind farm and scale model test are being performed. This paper gives a summary of the design, some preliminary outcome of the ongoing scale model tests and an outlook to future test plans.

Maintenance: ball and chain Offshore wind energy may be on the threshold of large-scale commercial exploitation, it has only just escaped infancy. And in the process it wandered into the very sea that proved to be the Nemesis for traditional offshore 40 years ago. The North Sea is a hostile place, not nice to sailors, offshore platforms or turbines. Although the installation of Horns Rev was favoured by continuous calm weather, the operation of the farm has required many, many more maintenance attention than foreseen. In the first 1.5 year 75.000 times an engineer was transferred to a turbine, requiring 7 vessels, regular helicopter service and a crew of 66. There are three main reasons for this high maintenance requirement [1]: · Unfamiliarity with the remote detection system (what is this alarm for?) · Poor planning of maintenance teams (several specialists following each other around) · Not fit for purpose design (either in design philosophies or specifications). Offshore wind turbines will continue to grow in size. This means that the newest turbines will be continually applied offshore (bigger seems better), opening the pitfall of encountering design flaws where they are most expensive: offshore. Maintenance will remain high priority for successful offshore wind exploitation in the early stages of farm development. Access: the key Onshore wind turbines have availability of over 98% because service teams are able to access (almost) at any time a faulty turbine. Normally, the service team is alerted by the SCADA (Supervisory Control And Data Acquisition) system and subsequently a route past faulty turbines is planned. Most problems can be fixed within 2 days [2]. Offshore, this scheme can only work when a turbine can be accessed as easily and safely for a large amount of time. Using a helicopter not to be dependent on waves is a workable but expensive and not entirely safe option. Access by ship becomes unsafe for a significant wave height of 1-1.5m depending on vessel and access method. On a typical North Sea location, this limits access to 50-75% of the year. In both cases the amount of tools and spare parts that can be transferred is limited to what a man can safely carry. The goal of this project therefore is to develop a vessel mounted transfer system that can be operated safely for more than 90% of the year. Requirements Several research groups at the Delft University of Technology joined forces to analyse and solve this problem. Offshore wind structures, in contrast to offshore oil & gas structures, always appear in numbers, even great numbers. This means that any adaptation required on a turbine needs to be on every turbine. The first requirement was therefore to create a system that can operate without interface on the turbine. In order to create a safe transfer system, it would be ideal to have a platform on top of the vessel cancelling all motions to make it stand still in comparison to the fixed world; the turbine. These systems exist in a reversed form: flight simulators. The moving part of these simulators consists

of a contraption of 6 cylinders, known as a Stewart platform. These platforms can move in all six degrees of freedom and are therefore ideally suited to cancel all ship motions. The requirements: · No adaptation requirements on the offshore wind turbines · Vessel mounted system · Use a Stewart platform · Safe · Easy · Fast · High workability: > 90%. Introducing the Ampelmann The concept was first envisaged during the 2002 World Wind Energy Conference in Berlin. After an intensive day of listening to scientific palaver, German beer and open minded brainstorming lead to the accidental birth of idea to put a Stewart platform on a ship. The concept should be so reliable that the maintenance engineer only needed to watch the pedestrian traffic light change to green to signal him that he could safely transfer. As the concept needed a codename, the solution was quickly found: "Der Ampelmann". Ampel is traffic light in German and at the time East Berlin could only be recognised by the fact that the figure in the pedestrian lights is wearing a hat, as opposed to his hatless colleague applied in West-Berlin (and the rest of the world). Ampelmann has since become the trading name of both the concept of Stewart platform on ship and of the Stewart platform itself. Figure 1 shows the "Mann mit Hut geht's gut", the Ampelmann.

Figure 1. Der Ampelmann, name giver and mascot to the concept

Conceptual design The 100 MW Near Shore Windpark (NSW) off the Dutch coast was selected as a reference site. For this location the wave scatter diagram is well known through several studies [3, 4]. To give the system a required workability of more than 90%, the system must be able to work up to and including a sea state with a significant wave height of 2.5m. At the NSW site this means that this system can work 93% of the year, which is a monthly downtime of a little over 2 days. Figure 2 shows the wave scatter diagram and indicates the workability limit. Hs Tz 6.25 m 5.75 m 5.25 m 4.75 m 4.25 m 3.75 m 3.25 m 2.75 m 2.25 m 1.75 m 1.25 m 0.75 m 0.25 m Sum 0.5 s 1.5 s 2.5 s 3.5 s 4.5 s 5.5 s 6.5 s 0.04 0.38 1.18 3.61 4.83 0.38 7.5 s 0.08 0.34 0.15 Sum 0.08 0.38 0.53 1.18 3.61 8.90 19.21 37.96 68.95 118.51 225.95 381.19 133.49 999.93

7.1 %

0.76 0.76


1.10 1.10

0.1 5.9 237.2 117.1 360.5

26.51 113.6 219.4 143.0 14.22 516.81

4.07 18.83 37.96 42.44 4.83 0.57 0.87 0.11 109.68

92.9 %

0.11 10.5


Figure 2. NESS-NEXT scatter diagram for the NSW site with workable sea states in the dark cells [5]

For the design, 2 reference vessels were used: a 20m and a 50m long offshore supplier. The motion characteristics of these vessels were determined with the program DELFRAC and are expressed as response amplitude operators (RAO) for every degree of freedom. Figure 3 shows the typical layout and dimensions of the vessels.

Figure 3. Layout and dimensions of the vessels used for the conceptual design

To design the Ampelmann, a program was written in Matlab. Based on the input wave spectra from the scatter diagram in figure 1 and the RAO's from Delfrac, the response spectra for the vessel motion in a particular sea state can be calculated. By transforming the response spectra to the time domain through inverse fast fourier transform (IFFT) time signals of the vessel motions were generated. These vessel motions could then be transferred to any location on deck to acquire the envelope of motions of that particular point in the sea state under consideration. As the Ampelmann needs to counteract the motions of the deck, the inverse of the motion envelope, is the envelope the Ampelmann must reach. A final parameter to take into account is the cylinder length: the design envelope must be reached from a neutral position of the platform, but by increasing the cylinder length to reach a larger envelope, also the neutral position of the platform changes, affecting the design envelope. This iterative process resulted in the fitting dimensions of the platform. Figure 4 shows a design of the platform and its envelope.

resulting 3D-envelop

10 8 6 4 2 5

Heave 2


-2 0.5 0 1 0 -0.5 -1 Surge









Figure 4. Design envelope for the centre of the stewart platform top plate

The small reference vessel was quickly abandoned as the required maximum leg length exceeded 11m, which makes a structure of unrealistic proportions compared to the 20m long vessel it should be installed on. For the large reference vessel, the maximum leg length is 7.6m, with a neutral position of 5.8 and minimum length of 4m. A series of optimisation tests were done to find the ideal location on deck. Further, it was tested whether the system would work in sea states beyond the design. For a sea state with Hs = 2.75m only a

few overshoots of the system occur in the order of centimetres, at Hs = 3m the overshoots become more numerous and in the orders of tens of centimetres. The basic assumption is that a vessel is used which keeps its position by DP (dynamic positioning). These vessels are available in all sizes and shapes. However, the accuracy of the DP system varies: more accurate systems are more expensive. The eventual requirements to the DP system have not yet been explored. The design was based on excellent station keeping capabilities. Future design studies and market investigation have to turn out which vessel-DP-Ampelmann configuration is most economically viable. Based on the Matlab time domain registrations, the required platform velocities and accelerations were calculated. These were used to further design the required cylinders. The cylinders required for the Ampelmann are all within current industry standards. The required velocity and acceleration fit easily within the cylinder specifications. Finally, the amount of oil required in the hydraulic system was estimated. The system requires some 2.5m3 of oil, which fits easily into a standard 12ft container. This container can be put on deck, together with a power pack container to make the system independent of the vessels engine and power supply. Conceptual design conclusions The main conclusion of the conceptual design is that an Ampelmann can be constructed to meet the requirements. The structure will be larger than typical flight simulators but can be constructed with off-the-shelf components. The biggest issue still to explore is the connection between a real measurement system and a real Stewart platform. Although computer programs can simulate all subcomponents of the systems, the connection between real measurement and control mechanisms can only be solved in the real world. It was therefore decided to start a testing program of a scale model Ampelmann. Scale model testing The goal of the scale model testing in threefold: 1. Acquire experience in connecting measurement system and Stewart platform 2. Demonstrate the working of the Ampelmann 3. Create promotion material: pictures and video. Although the university comprises all required fields of science to approach this problem, contact was sought and found with commercial companies for back-up, experience and equipment. This resulted in the consent to participate of Rexroth Hydraudyne, a company specializing in the construction of Stewart platforms. Their Micro Motion System has been made available for these tests for 3 months. Boskalis, the world's largest dredging company with a fleet of over 300 specialized vessels and extensive knowledge of ship movement measurement systems donated two measurement systems: the MRU and the newer and more accurate Octans. As the project progressed, also the manufacturer of the Octans, IXSea became involved. Finally, Shell Wind B.V., as developer of several offshore wind farms, provided financial sponsoring to cover transport and insurance costs for the borrowed equipment. The Delft University of Technology has a history with constructing Stewart platforms. To create a flight simulator, the mechanical engineering and aerospace engineering faculties co-operated to create the "Simona" flight simulator. As the project neared completion, the platform was transferred for the mechanical the aerospace department. The platform was so dearly missed that a smaller version was created not long after: the "Simonita". The platform is used as a research tools and is also available for Ampelmann testing. The following test will be performed. · Reliability testing of the measurement systems on the Simonita (June & October) · Fine-tuning of the filters for the measurement system (October) · Communication between control computer and the MMS (November) · Dry-testing of MMS on top of the Simonita (December) · Wet testing of the MMS on a vessel in the wave tank (January)

The first tests with both the MRU and the Octans proved that the systems show some drift: slowly increasing deviation of the measurements. The Octans was much more accurate but did show coupling between the roll and pitch motion. Figure 5 shows the Simonita with the Octans mounted on top (the small blue box).

Figure 5. Testing of the Octans on the Simonita

A new series of tests was performed in October. The assistance of the manufacture IXSea with liningout the Octans and tuning the software proved very valuable and resulted in highly accurate measurements. Figure 6 shows the results of input and output signals for all 6 degrees of freedom, excited at 0.2 Hz [6]. Deviations in measurement are all less than half a centimetre. This will be very acceptable when the Octans is applied on a real size (4-8m) Ampelmann but is on the edge for the scale model system with a maximum platform reach of 18cm.

Figure 6. Input and output for all 6 degrees of freedom at harmonic excitation at 0.2 Hz

To create Bode diagrams for the Octans, the Simonita was fed a multi sine signal. The outcome proved that the Octans is a very stable and accurate system, provided that the control software is operated correctly. A minor drawback of the testing device came to light: it seems the assumed input to the Simonita is not fully reliable in some degrees of freedom. With this test it turned out that the testing device was less accurate than the device tested. The conclusion is that the accuracy of the Octans is sufficient for scale model testing. For a full scale Ampelmann, the results will only be better. Dry testing At this moment the real-time communication between measurement system and computer and computer and Micro Motion System is detailed further. The MMS is to arrive in Delft the first of

December 2004. To further fine-tune the controls, the system is first fixed to the concrete floor and tested solitarily. A week later, the MMS is mounted on the Simonita. Together with the Octans and the control computer, the working of the Ampelmann concept is "dry-tested". The Simonita is fed with scaled vessel motions; the MMS needs to counteract these to create a stand-still top-plate. The contraption will simulate a large number of motion sequences to test its capabilities. Although scientifically the Simonita-MMS combination will be able to prove the working of the concept, a final testing phase is also envisaged: the wet testing Wet testing To illustrate the working of the Ampelmann in its natural environment, the MMS will be fixed to a scale model supply vessel (3.5 by 1m) and tested in the wave flume. These tests are by no means meant to further detail the design. Only the general dimensions are used for vessel and Ampelmann to make a live example of its function. No high-detailed scaling is used to make sure the contraption simulates real life. Figure 7 shows an artist impression of the MMS on a supply vessel.

Figure 7. Artist impression of the MMS mounted on a supply vessel

Future developments Access to offshore wind turbines remains an issue and the Ampelmann could be a workable tool to improve it. Although the current testing takes nearly all the time of the project team, future steps are subject of continuous discussion. A first step has been taken to further analyse the available fleet and check the impact of vessel size and configuration on the required Ampelmann dimensions. 10 different vessels will be used to analyse this impact: small to large mono and twin hull. Next to these parameter variations, the main goal will be to arrange for a full size prototype to be manufactured. Current sponsors to the project already revealed their interest, but concrete plans will only be drafted after successful testing of the scale model. Acknowledgements The scale model testing is being made possible only through the willing co-operation of a number of companies. Their contribution is gratefully acknowledged: · Boskalis for providing the MRU, the Octans and valuable hands-on experience on both systems. · Eye 4 U for creating 3D prototype animations and graphics support · IXSea for providing the Octans for the later tests and on-site assistance with lining-up and control parameter settings. · Rexroth Hydraudyne for providing the Micro Motion system and software engineering assistance. · Shell Wind for sponsoring insurance and transport of the equipment. The project proved to be so multi-disciplinary that the 7 disciplines within the DUWIND inter-faculty institute did not cover all. The authors would therefore also gratefully acknowledge the input and assistance of the Control Group and the Hydromechanics Group both the faculty of OCP. Finally, this project is sponsored by the Dutch Government through NOVEM under the DEN framework.

References [1] Vestergaard, S., The Horns Rev Project, Operation and Maintenance", IEA R&D Wind Annex XI Topical Expert Meeting: Critical Issues Regarding Offshore Technology and Deployment, Skærbæk Denmark, March 2004 [2] Personal communications with Mainwind, 2003 [3] Ferguson, M.C. (ed.) et al., Opti-OWECS Final Report Vol.4: A typical Design Solution for an Offshore Wind Energy Conversion System Delft University of Technology, Institute for Wind Energy, 1998. [4] Tempel, J. van der, Lifetime Fatigue of an Offshore wind Turbine Support Structure, Section Offshore Technology, Delft University of Technology, May 2000 [5] Hoonings, J.S.P., The AMPELMANN Design of a motion-compensated platform for offshore wind turbines maintenance operations, Offshore Engineering, Delft University of Technology, September 2004 [6] Etna, G., The Ampelmann, summary of testing the 'Octans' Offshore Engineering, Wind Energy, Delft University of Technology, November 2004


Der Ampelmann

7 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


Notice: fwrite(): send of 202 bytes failed with errno=104 Connection reset by peer in /home/ on line 531