Page 1

Annual report

2013

Norwegian Centre for Offshore Wind Energy www.norcowe.no

page 1


3

Welcome

5

Dear reader

6

Mid-term evaluation

8

Transforming Energy Systems

List of partners and committees

10

7

Energy transition and offshore wind

13 Planning of Operation & Maintenance (O&M) 14 PhD student Mihai Florian 16 PhD student Masoud Asgarpour 17 Interactive design of wind farm layout using CFD and model reduction 18 PhD student Siri M. Kalvig 20 New partner: Acona Flow Technology 22 Slamming waves 23 Opening of the Norwegian Motion Laboratory 24 Dr. Tore Bakka 26 PhD student Torge Lorenz 27 Obtaining site climatologies for use by the offshore wind industry 28 New partner: the Norwegian Meteorological Institute 32 NORCOWE strengthens its capacity for wind condition monitoring 35 Measurement campaigns at Sola and ECN 37 PhD student Valerie Kumer 39 New partner: Leosphere 40 PhD student Mostafa B. Paskyabi 42 PhD student Martin Flügge 43 The Gwind story 44 National and international cooperation 47 Centre Management Group 48 Summer school 52 Organization 55 NORCOWE Reference Wind Farm 

Photos and illustrations by: Marit Hommedal - Petras Gagilas - Andrew Scholbrock, NREL - Simen Malmin, Prekubator - Øystein Eldholm Christian Michelsen Research - Geofysisk Institutt, University of Bergen - University of Agder - University of Stavanger Aalborg University Shutterstock Editors: Kristin Guldbrandsen Frøysa, Centre Director - Annette Fagerhaug Stephansen, Centre Coordinator Layout: Per Gunnar Lunde, Communication Officer

page 2

www.norcowe.no


Welcome

Bjørn H. Hjertager, Professor at the University of Stavanger, Chair of NORCOWE executive board

The research activity in NORCOWE in 2013 has been extensive, and the reorganization undertaken last year has had a positive impact, improving the collaboration between the six research institutions involved in the centre. Highlighted projects are naturally the measurements campaigns and the opening of the Norwegian Motion Laboratory at the University of Agder, which are featured in this annual report. Five new PhD students finished their studies in 2013, working on topics regarding the Weather Research and Forecasting model (WRF), control of bottom fixed and floating wind turbines, turbine reliability and installation of turbines. The centre has also had several master students during the year, and has welcomed a number of new PhD students, of which one is an industrial PhD in collaboration with ECN. Many projects have been generated based wholly or partly based on work within NORCOWE. This is positive and in line with the objectives of the center to contribute to increased

activity outside NORCOWE itself. These projects make it possible to continue work and ideas from the center and will hopefully contribute to increase the Technology Readiness Level of the research. The project “Decision support for installation of offshore wind turbines”, which was kick started in the autumn of 2013, has already gathered international interest, while a model of the vertical axis Gwind was tested in the harbor of Stavanger. You may read more (and watch videos) at www.norcowe.no. An important part of the work of the center management and the board in 2013 has been to strengthen the ties with the industry and increase the number of user partners. NORCOWE has a result obtained five new partners (Acona Flow Technology, Aquiloz , Leosphere, Norwegian Meteorological Institute and StormGeo) from January 1st 2014. The new partners have strengthened NORCOWE within such key areas as meteorology, LiDAR technology and modelling. Some of them are presented in this report.

Executive board, from left: Bjørn H. Hjertager, Gudmund Olsen, Birgitte R. Furevik, Kristin G. Frøysa, Annette F. Stephansen, Eirik Manger, Svein Winther, and Harald Rikheim. Anne Marie M. Seterlund, Peter M. Haugan, Frank Reichert, Jostein Mælan and John D. Sørensen were absent.

www.norcowe.no

page 3


page 4

www.norcowe.no


Dear reader This annual report presents our scientific work and selected results together with the Centre Management Group, PhD students and scientists from NORCOWE. The report also presents topics like the NORCOWE summer school, international cooperation and public outreach. The list of publications is not included in the report, but is available at our website. Setting the scene are two texts: The first one is written by Professor Arild Underdal. He is the director of CICEP, a FME-centre addressing environmental friendly energy from a social science point of view. The second one is written by Professor Peter M. Haugan, who was chair of NORCOWE’s executive board in 2013. The current (2014) NORCOWE work packages are: WP1: Met/ocean data - Measurements and database WP2: Wind energy estimation - Wind resource assessment, energy yield and layout for offshore wind farms WP3: Design, installation and operation of offshore wind turbines This annual report is not a complete overview over NORCOWE, but is meant to give you a grasp of NORCOWE. Please visit our website for news from NORCOWE. You may also contact us if you want more information. It might be a good idea to attend our two “Science Meets Industry” conferences in 2014. The first conference is in Stavanger April 2nd; the second is in Bergen September 9th. I hope you will enjoy reading our annual report! Best regards, Kristin Guldbrandsen Frøysa Centre Director NORCOWE

www.norcowe.no

page 5


Mid-term evaluation NORCOWE has now been in operation for more than four years. Results have been produced and presented, but it is always an issue how to increase the impact of the work in NORCOWE. Joint projects between user partners (industry) and scientific partner are one way of doing so. Workshops and scientific meetings with exchange of knowledge is another tool. There will be a stronger focus on commercialization and utilization of the results in 2014, bringing some of the results from NORCOWE into practical application in the industry. NORCOWE and seven other FME-centres went through a midterm evaluation in spring 2013. The mid-term evaluation was a thorough evaluation of each FME-centre by an evaluation committee appointed by the Research Council of Norway (RCN). The evaluation report for all eight FME-centres is available at RCN’s website.

“NORCOWE has developed a scientifically highly valuable and productive research program well on its way to receive international recognition and covering an important value chain related to offshore wind power.” The evaluation committee

The evaluation committee visiting NORCOWE/CMR in March 2013

page 6

www.norcowe.no


The evaluation of NORCOWE The overall evaluation of NORCOWE was positive and the centre will get funding for the last three years of the centre periode. The evaluation committee highlighted that “although the Centre has been active for less than four years it has already created considerable international visibility. In this respect the Centre is well on its way to receiving identity and recognition.” Furthermore, “the Centre can be commended for an excellent training of the doctoral students. These emphasized that the summer school in particular is very supportive, and they feel very positive about the interaction among themselves.” NORCOWE was asked to strengthen its international cooperation and to attract new partners to the centre. These topics have been addressed. Five new user partners joined NORCOWE on the 1st of January 2014, and the centre management is still in dialogue with potential partners. Three partners left NORCOWE by end of 2013. The international visibility is strengthened, in particular by joint projects like the measurement campaign together with

ECN at ECN’s test site at Wieringermeer in the Netherlands. A new motion laboratory was opened at the University of Agder on the 27th of November 2013. Current status The new partners in NORCOWE have strengthened the centre on key topics like meteorology, oceanography, modeling, LiDAR measurements and operation of offshore wind farms. There are joint projects with international partners and new projects are under development. An offshore measurement campaign is planned for 2014/2015. Thus NORCOWE is in a good position to add value to the international offshore wind industry in the future.

Key figures 2013 PhD students: 17 Completed PhDs: 7 (of which 5 during 2013) Post Docs: 1 Master students: 9 Number of publications: 34 Posters and presentations: 26

List of partners and committees Partners Christian Michelsen Research Uni Research University of Agder University of Bergen University of Stavanger Aalborg University Statoil Statkraft Acona Flow Technology Aquiloz Leosphere Norwegian Meteorological Institute StormGeo

www.norcowe.no

Internal Scientific Committee Angus Graham, Uni Research Birgitte Furevik, Norwegian Meteorological Institute Finn Gunnar Nielsen, University of Bergen/Statoil Geir Hovland, University of Agder Hamid Reza Karimi, University of Agder Jasna Bogunović Jakobsen, University of Stavanger Kristin Guldbrandsen Frøysa, CMR Thomas Bak, Aalborg University International Scientific Committee Finn Gunnar Nielsen, University of Bergen/Statoil Bill Leithead, University of Strathclyde Cecilie Kvamme, Institute of Marine Research Line Gulstad, Vestas Trond Kvamsdal, Norwegian University of Science and Technology page 7


Transforming Energy Systems Arild Underdal, Professor, University of Oslo and CICERO, Director of CICEP, www.cicep.uio.no A well-functioning energy system should meet at least four basic requirements. First, provide universal access to the energy required to satisfy basic human needs. Second, be (environmentally) sustainable, i.e. meet the needs of current generations without leaving future generations disadvantaged. Third, provide stable supply for consumers and workplace safety for all involved in the production and distribution of energy. Finally, the system should be cost-effective in the sense of minimizing the costs incurred in achieving the other three requirements. Each of these objectives raises intriguing technological, economic, and political challenges. Combining all four objectives may seem to leave us with a mission impossible. The Centres for Environment-friendly Energy Research (FMEs) are established to help build a knowledge base for transforming carbon-intensive energy systems into environmentally sustainable energy systems. Each centre focuses on specific pieces of that transformation puzzle; NORCOWE on energy

production from offshore wind resources, CICEP on determining the political feasibility of alternative policy options. This kind of differentiation is a sensible strategy; cutting-edge research requires high specialization and concentration of efforts. Yet, energy systems are – albeit to varying degrees – complex and dynamic systems, and several attempts at transforming such systems (including the German Energiewende) demonstrate the importance of understanding how components interact. As the technological centres enter their second phase and the social science centres come up for mid-term evaluation, time may be ripe for exploring what we might be able to do together. In essence, the intellectual challenge of understanding what can drive or block energy system transformation is one of coupling (a) system components and (b) contributions from different disciplines and research units. The main system components are production, distribution, and consumption. In functional terms, these components are sequentially ordered, meaning that production precedes distribution and distribution precedes consumption. From the perspective of systems transformation, however, other modes of interplay are equally important. First, change in any of the three components may trigger change in any (or both) of the other two. For example, energy producers can be expected to respond to changes in market demand, and they may do so in anticipation of change that has not (yet) occurred. Moreover, as illustrated by the US ‘shale gas revolution’, improvements of production technologies can facilitate access to huge energy resource deposits and/or enable producers to supply certain types of energy at significantly lower costs. Second, and equally important, attempts at promoting change by intervening in one component may well fail unless some ‘matching’ intervention is made in at least one other component. Some of these links are obvious; electric cars require a network of facilities for recharging batteries. Other links may easily be overlooked until they materialize. One example is the relationship between local ‘green’ initiatives on the one hand and centrally managed development of high-capacity grids to transport ‘green’ energy from (remote) production sites to main centres of consumption. Other things being equal, success for local initiatives will lead to lower demand for long-distance grid services (and, presumably, to lower investment in the infrastructure required for

page 8

www.norcowe.no


delivering such services). Particularly in large urban settings, however, local initiatives often rely more or less heavily on external supply of ‘green’ energy. Under such circumstances, precise coordination of efforts can help both parties achieve their goals. Each of these components (or functions) includes economic, political and cultural mechanisms as well as natural resource potentials and technologies (see figure on the left page). For example, energy distribution is partly a matter of geographical parameters and physical infrastructure and partly determined through markets and politics. Similarly, patterns of consumption reflect material factors such as climate conditions, human-made infrastructure, income levels and relative prices, but also human beliefs, values and lifestyles (and the importance of the latter may well be underestimated). The quote from the Global Energy Assessment report (on the right) is a reminder that some of the most demanding transformation challenges are found in domains studied in economics, political science, law and related disciplines. More importantly, the quote is a reminder that building a knowledge base for energy system transformation is not merely a multi-disciplinary project to which each FME can contribute within its own particular domain; it also requires coupling models and integrating findings across disciplinary divides. www.norcowe.no

Risk-averse researchers stick to their own domains. A recent initiative by the Research Council invites us to explore more integrative approaches.

“Humanity has the capacity, ingenuity, technologies and resources to create a better world. However, the lack of appropriate institutions, coordination mandates, political will and governance structures make the task difficult.” Global Energy Assessment: Summary for Policymakers, 2012, p. 8

page 9


Energy transition and offshore wind Peter M. Haugan, Professor, Geophysical Institute, University of Bergen

The main political motivation for stimulation of new renewable energy research, development and demonstration in Europe presently is the concern about climate change and other effects of CO2 emissions including ocean acidification. Other motivations such as energy security have probably been more important in the past, notably in the USA. Concerns about local pollution and health issues may be more important in the future, notably in China. But at present in Europe the energy policy discussion is primarily driven by climate. However, while nearly all political parties and interest groups admit that anthropogenic climate change is real and will be costly if not mitigated early, there is a varying level of commitment to decarbonizing the energy sector rapidly. The famous two degree target requires global emissions to approach zero within 20-30 years from now. This is comparable to or even shorter than the typical life time of energy infrastructure. Yet e.g. Norwegian policy is to expand oil and gas exploration to new areas, increase recovery from established fields and build new pipelines and systems for export of petroleum. There seems to be a disconnect between what is sometimes called normative goals with the two degree target as a prime example and the more “realistic” goals which guide day to day operations and allow traditional fossil fuel businesses to thrive. This disconnect between global long term ideals and what we actually do, probably often motivated by one’s own short term economic benefit, is perhaps at its most obvious in Norway. We state commitment to the two degree target, engage in global climate negotiations and protection of rain forests. At the same time we plan new infrastructure that will lock-in to fossil fuels. Yes, burning methane gives less CO2 than burning coal, but after 2035 we should not emit CO2 at all, at least not in Europe. And no, we cannot trust Carbon Capture and Storage (CCS) to solve the problem. That option has considerable energy penalty and costs. In addition injectivity and long term storage integrity is hard to prove in advance and there are unsolved local environmental and permitting issues. In later years I have become more and more fascinated with various non-technical i.e. human and societal factors and processes that determine our energy future. Perhaps such

page 10

factors have greater importance than science and technology? I say this because we now have an increasing amount of evidence that a global transition to a decarbonized energy supply is not only possible but also affordable. It would have many positive side effects, including making WHO health and air pollution targets less expensive to reach, and improving energy security. A compilation of evidence and projections was presented in the Global Energy Assessment (GEA) in 2012. The main conclusion is that it is indeed possible to serve a population of 9 billion people in 2050 with decarbonized energy supply providing electricity to all as well as reaching the other targets mentioned. It would require emphasis on energy efficiency to enable more than the present 1/3 of the world’s primary energy to make it all the way through the system to useful energy services. And it would require increased investments in new renewable energy supply infrastructures for a period of 2-3 decades. But not increased investments thereafter. And the added costs would be comparable to present world subsidies of fossil fuel. This becomes even more true if we include the so-called external costs or social costs of CO2 emissions. So where does this leave offshore wind? With a huge potential. We now see the industry producing 8 MW turbines and expect to see 10 MW soon. Some of these may be placed onshore, but in contrast to small turbines, handling and installation of really large turbines is at some point likely to become easier and cheaper offshore than onshore. Offshore areas are available with good wind resources. Even more areas become available when (I say when, not if) we can reduce the costs of floating turbines to compete with fixed installations. There are definitely challenges ahead. But wind and solar are likely

“What to do is a policy question that cannot be answered by research alone. But it should be informed by research.” Peter M. Haugan

www.norcowe.no


to remain the two most important renewable sources on a 20-30 year time scale given the present market penetration, the resource base and the level of industrial practices that are established. The big question is to what extent the new renewable energy transition will be nurtured sufficiently by a sufficiently large number of countries for a sufficiently long period of time. Sweden has shown the way for bio-energy, Denmark for wind, Germany for solar and the UK for offshore wind. In Norwegian waters we still have only the one single offshore wind turbine that Statoil put up in 2009. We have offshore wind research, but no stimulation for demonstration or deployment. Can we expect a transition of the Norwegian offshore industry from petroleum to wind without a home market? The interest is there, but it sure would go faster with schemes resembling those of the UK. In addition we would then have the possibility to develop new industries utilizing the energy.

www.norcowe.no

What to do is a policy question that cannot be answered by research alone. But it should be informed by research. In this connection it is a problem if the public debate is not well informed about the status and possibilities for energy transition as described e.g. by the GEA. We have seen over the years that respected organizations like the International Energy Agency (IEA) have severely underestimated the growth of renewable energy. Now solar and wind are becoming cost competitive and have achieved industrial scale. Should we not expect an even stronger growth in the future? Yes, unless too many new investments are made in fossil fuel infrastructure demanding payback way into the future, and unless those who set the agenda and provide the content of public debate are predominately players with a special vested interest. We need to contribute to that debate, too. On the way, social and human sciences can help shed light on decision processes, human behaviour, economic and political forces, the barriers and driving forces that determine the energy transition including the role of offshore wind.

page 11


NORCOWE Reference W Thomas Bak, Professor, Aalborg University

Offshore wind farms are complex systems, influenced by both the environment (e.g. wind speed distribution, waves, atmospheric boundary layer, and sea bed) and the wind farm characteristics (e.g. turbine type, foundations, layout, and distance to shore). These topics determine the capital expenditure, the operating expenditure, which divided by the energy produced determine the cost of energy. A better understanding of wind farms is hence of critical importance to the wind energy industry. NORCOWE finances conceptual studies and research within a number of the key topics in wind farm design and operation, such as wind resource assessment, energy yield and layout for offshore wind farms. Design, installation and operation of offshore wind turbines are also an important part of NORCOWE research. To obtain useful information from such studies, use of a realistic and representative baseline wind farm is required. This encourages clear communication between experts and allows assessment of the effect of concept studies.

page 12

The reference wind farm NORCOWE is developing has three purposes. First, the reference wind farm serves as a database containing all the data needed to represent a wind farm. Researchers can then use information from this database in their research. Second, the reference wind farm serves as a baseline for researchers to compare and discuss results, and to quantify the benefits of one technological concept or solution compared to another. Third, the reference wind farm can be used to communicate work to the industry, the research community and the general public. The reference wind farm is representative for a current large-scale offshore wind farm with multi megawatt turbines. A set of key parameters has already been defined, driven by the rationale behind the wind farm. Details will be worked out during the first part of 2014, and the first public revision of the wind farm is expected by mid-2014.

www.norcowe.no


Wind Farm design options are explored. Users can contribute with new signals and designs by suggestion to NORCOWE, expanding the available database. The key parameters reflect a realistic scenario, and all key parameters will be defined using best practice. The design will however not be optimal. A simple calculation of the key expenditures will also be provided for assessment of the concepts’ relative cost of energy. As the reference wind farm serves as a baseline, this allows experts to individually assess their contributions and provides a common platform for discussion. Data, parameters and other information about the reference wind farm will be made publicly available, encouraging other groups to use the same baseline.

NORCOWE Reference Wind Farm - a few of the key parameters’ nominal values:

The researchers behind the definition of the reference wind farm have consulted industry partners on the values of key parameters. The aim has been to define a realistic wind farm while not adapting an existing design. Avoiding the latter removes the possibility of an unproductive discussion with the designers of the existing farm in question, and allows focus on comparing concepts. Another driver behind the definition has been the availability of relevant measurement data for assessment of e.g. met/ocean conditions. Also the use of a publicly available turbine model has been important, to simplify the implementation and increase the impact. The NORCOWE Reference Wind Farm allows users to change designs and parameters in specific areas, while retaining the nominal values for the rest of the parameters. For example the layout can be fixed as prescribed while different electrical

www.norcowe.no

Installed capacity: 400 MW Number of turbines: 80 (specific layout will be determined) Reference wind turbine: NREL offshore 5-MW baseline wind turbine Reference zone: Area in vicinity of FINO3 (a German research platform located about 45 nautical miles (80 kilometres) west of the German island of Sylt) Climatologies: Normal conditions are the baseline, both in terms of wind, waves and ocean currents Water depths: Around 30 meters Foundations: Bottom fixed monopile design (with defined eigenfrequencies) Electrical design: HVDC connected electrical design Farm controller: Basic farm control

page 13


Planning of Operation & Dr. Erik Kostandyan, Aalborg University

carry out the repairs. Identifying the optimal times for inspections, maintenance and repairs as well as what type of action is needed requires knowledge of the condition of the wind turbine components/system, especially the accumulated damage or degradation state as well as the level of uncertainty. For offshore wind turbines, interventions can be delayed due to weather restrictions and result in damage or increased degradation of the remaining components. Thus, reliability models should be developed and applied in order to quantify the residual life of the components. Damage models based on physics of failure combined with stochastic models describing the uncertain parameters are imperative for development of cost-optimal decision tools for O&M planning, and these procedures are implemented in Aalborg University’s research.

Reliability modeling of wind turbines, exemplified by power converter systems, as basis for O&M planning

In Erik Kostandyan’s PhD research the focus was on reliability modeling of critical subsystems of wind turbines, such as the power converter system. Reliability models for the electrical subsystem and its components were developed. The models are aimed for use in the planning of O&M strategies and could be integrated with non-destructive evolution techniques (e.g. remotely obtaining information on fatigue measure evolution without damaging the component). Reliability of structural components such as tower and blades has been considered, taking into account possible fault events due to the grid loss resulting from failure of power converter components. Such situations are quite common and have vital importance, especially for pitched controlled off-shore wind turbines. For a representative structural failure mode,

Planning of inspections and operation & maintenance of offshore structures are common themes for the oil & gas and the wind energy industry. At Aalborg University a key focus area during the last two decades has been research on and practical application of a reliability- and risk-based approach to these activities. The research has mainly concentrated on fatigue critical details and on formulating a framework for optimal planning of operation & maintenance. The PhD research project performed by Erik Kostandyan has concentrated on the application of these generic methods for electrical components in wind turbines and exemplified by power converter components which have quite high failure rates. O&M costs account for a significant share of expenses related to offshore wind farms, involving mainly replacements and/or repair of the wind turbines and the actions needed to page 14

www.norcowe.no


& Maintenance (O&M) a probabilistic model has been developed that incorporates grid loss failures. Many systems including electrical systems can be represented by dependent and load-sharing systems, so models for such systems reliability estimation are needed. Advancing the topic of dependent and load-sharing system reliability estimation, a theoretical background has been developed based on sequential order statistics and structural systems reliability methods. Erik Kostandyan defended his PhD thesis successfully on the 28th of October 2013 at Aalborg University. His supervisor was Professor John Dalsgaard Sørensen.

O&M planning at the University of Stavanger (UiS) PhD student Ole-Erik V. Endrerud is working on a simulation model regarding O&M within offshore wind, drawing experience from the oil and gas expertise at the University of Stavanger. In order to validate the software he is participating in a project with researchers from EDF, Marintek, Sintef and the University of Strathclyde who all are developing their own simulation models. A base case has been developed using data and information from Statoil and Statkraft, giving the possibility to validate and compare the models. Ole-Erik’s supervisor is Professor Jayantha P. Liyanage (UiS) and his co-supervisor is Dr. Nenad Keseric (Statoil).

www.norcowe.no

page 15


Mihai Florian is doing his PhD at Aalborg University with Professor John Dalsgaard Sørensen as main supervisor.

What is your scientific background? I graduated at the Technical University of Bucharest with a degree in civil engineering. Afterwards I followed a M.Sc. program at the Aalborg University where I sustained my thesis in Uncertainty Assessment of Wind Turbines. I am currently continuing my studies as a Ph.D. at Aalborg University. What topic is addressed in your PhD? Operation and Maintenance (O&M) activities account for a large part of offshore energy cost, thus making development in the field a rational solution for increasing the feasibility of offshore wind farms. Development in condition monitoring systems and inspection procedures together with increasing field experience offer decision makers a growing amount of information on offshore turbine behavior, information which cannot be fully accounted for in the decision making process with currently developed O&M strategies. This can potentially be corrected with a risk-based approach, capable of accounting for variability in input parameters, accounting for component reliability and failure consequences and including outcome of future inspections in the decision process using a pre-posterior decision analysis.

page 16

How does your topic relate to previous/other on-going research within NORCOWE? A probabilistic model for estimating the reliability of wind turbine (WT) converter system critical components (insulated gate bipolar transistors (IGBT)) for the goal of maintenance activities establishment has been developed by Erik Kostandyan (see previous page). The model can be used for O&M analysis in converter systems and for other WT components (e.g. structural) O&M analysis by coupling it with appropriate physics of failure models, i.e. fatigue crack propagation modeling. You participated at the NORCOWE summer school before the start of your PhD. Could you tell us a bit about your experience there? I saw that being a part of NORCOWE and working in collaboration with researchers from different fields of engineering/science is a great opportunity for getting to know the offshore wind industry as a whole and a great chance for professional growth. The friendly social environment, coupled with interesting lecture topics and outdoor activities have made it a memorable experience. I participated in the summer school as a research assistant for Aalborg University and decided on following the full Ph.D. program afterwards.

www.norcowe.no


Masoud Asgarpour is doing his PhD at Aalborg University with Professor John Dalsgaard Sørensen as main supervisor.

What is your scientific background? I studied Mechanical Engineering as my Bachelor study with focus on structural analysis and fluid mechanics. In 2012, I completed my Master study in Wind Engineering at Kiel University of Applied Science in Germany, with focus on structural analysis and load calculation of wind turbines. Currently, I’m working as a wind energy researcher at Energy Research Centre of the Netherlands (ECN) and also working on an industrial PhD in cooperation with NORCOWE. What topic is addressed in your PhD? The goal of my PhD project is to develop a risk and reliability based decision support model for operation and maintenance of offshore wind farms. Existing O&M models for offshore wind farms are mostly focused on estimation of corrective maintenance costs based on the defined failure rates of components and not their actual state of health. The idea of this project is to use probabilistic deterioration models of components in a risk based decision model taking into account all available information from inspections and condition monitoring systems. Through this model an integrated system approach for O&M planning of the entire wind farm will be developed, which takes into account the relevant systems effects.

www.norcowe.no

How does your topic relate to previous/other on-going research within NORCOWE? This PhD project is a part of WP 3.1 of NORCOWE (Farm operation and maintenance) and is connected with two other NORCOWE PhD projects, which are ‘reliability analysis of wind turbines basis for O&M planning’ by Erik Kostandyan and ‘risk-based operation and maintenance of offshore wind farms’ by Miahi Florian. During this project, the reliability and probabilistic deterioration models defined in the mentioned projects will be used to establish an integrated system approach for O&M planning of the entire wind farm. You are working at ECN and are doing an industrial PhD. What was your interest in starting a PhD at Aalborg University? I deeply believe that international scientific cooperation is an essential step for more efficient utilization of the wind power and overcoming its challenges. During my PhD project, I will use ECN’s in-depth knowledge and experience in cost modeling of O&M activities and Aalborg/NORCOWE’s profound knowledge in risk and reliability based planning of O&M to develop an optimal decision support model for operation and maintenance planning of offshore wind farms.

page 17


Interactive design of wi CFD and model reductio Chad Jarvis and Yngve Heggelund, CMR - Marwan Khalil and Lene Sælen, GexCon AS Current solutions for optimizing wind park layouts often use CFD coupled with models for incorporating wake interactions in order to diminish computational efforts. An alternative approach is to use the technique of model reduction. At CMR, prototype software using model reduction of the steady state Reynolds Averaged Navier-Stokes (RANS) equations is under development. Using this technique it is possible to arrive at an interactive wind farm layout design tool which reduces the computational time drastically. For the test cases run so far, the model reduction technique provides accurate approximations of the CFD results in seconds rather than hours. Both offshore and flat terrain conditions can be considered. The overall objective is to reduce the cost of energy of offshore wind farms by more optimal placement of turbines by diminishing power losses due to wakes and maintenance costs resulting from wake induced fatigue loads. The first step has been to develop an interactive tool for layout design, which can later interact with other software tools for layout assessment and optimization. By basing the method on Computational Fluid Dynamics (CFD) and the full RANS equations, we believe the model reduction technique can lead to more accurate results than the current state of the art for fast flow field assessment.

The method explained in 4 steps: 1. Run multiple RANS CFD simulations for varying setups (turbine positions, wind speeds). Extract snapshots for each simulation.

2.

Apply Singular Value Decomposition (SVD) to produce a reduced space solution basis of orthogonal modes.

3.

Interactively move tiles into arbitrary configurations. Solve the RANS equations and boundary matching in the reduced space spanned by the solution basis.

4.

Compare solutions to RANS CFD solutions. If necessary improve the solution basis by running more CFD simulations and repeating steps 1-4.

Tile – a subdomain of the wind farm Snapshot – the CFD solution within a tile for a given simulation. Modes – the set of orthogonal vectors/functions representing the reduced space

page 18

www.norcowe.no


ind farm layout using on Results obtained so far One of the test cases that has been tried was based on a row of ten turbines with a uniform distance aligned with a neutrally stratified ambient flow over a surface with roughness length of 3 cm. The aim of this test case is to show that the model reduction technique is able to simulate the multiple wake effect for a long row of turbines from a relatively small set of basis modes. The turbines were of type BONUS 2MW with a hub height of 76 m. The CFD simulations were performed with CMR-Wind. Solution bases were created from a CFD simulation with turbine distance of 5 rotor diameters using only snapshots from the first N turbines. These bases were used to test how well they could predict the power production of each of the ten turbines. As seen from Figure 1, the deviation of the total production is less than 3.5% compared to CFD when using 3 or more modes.

Fig 2: The power production of each turbine with a distance of 7 rotor diameters. The basis constructed from the first 3 turbines of each CFD simulation is labeled 3 + 3 modes etc.

cases where the user can interactively move the turbines in the crosswind direction. In all cases, full three dimensional fields are computed, including the turbulent kinetic energy. Future plans include verifying the model reduction technique for more general setups and for more wind speeds and wind directions. There are also plans to assess the wake induced fatigue loading on turbines by coupling the flow field and turbulent kinetic energy field to an external tool.

Fig 1: The power production of each turbine. The basis constructed from the first 3 turbines of the CFD simulation is labeled 3 modes etc.

In a variation of the test case, the solution bases were created from two different CFD simulations with turbine distances of 5 and 9 rotor diameters respectively. Again, the bases were constructed from the first N turbines of each CFD simulation. These bases were used to test how well they could predict the power production of each of the ten turbines for turbine distances of 6, 7 and 8 rotor diameters. The deviation of the total production was found to be less than 3.3% compared to CFD when using 4 + 4 or more modes (see Figure 2).

Fig 3: 21 turbines, simulated in the reduced space within a second.

All the test cases have been verified with corresponding flow field solutions from CMR-Wind, a Computational Fluid Dynamics (CFD) simulator.

Earlier test cases have presented the ability to simulate www.norcowe.no

page 19


Siri M. Kalvig What is your scientific background?

What are your main results?

I have a master degree in Meteorology from the University of Bergen. This is quite a long time ago, back in 1995. In those days we graduated as Cand. Scient. Since then I have been working in TV2 and StormGeo for over 15 years, so for me it was quite a change to start on a PhD!

Fast moving swell in low wind will influence the wind field several hundred meters above. The logarithmic wind profile often used in offshore wind engineering will then not be valid. When swell is opposing a wind field there will be increased turbulence levels. My simulation results have compared well with results from other literature but it would have been best to compare with measurements.

What topic is addressed in your PhD? My main topic is to investigate how the waves will influence the wind fields and how this may affect a wind turbine. My approach is to use computational fluid dynamics (CFD) in order to study the dynamics between the waves and the wind. I also use CFD to model wind turbine performance. We have developed a new set up where you can study a wind turbine operation in “wave influenced wind” directly. I use an open source CFD tool called openFOAM.

Siri M. Kalvig is doing an industrial PhD at the University of Stavanger in collaboration with StormGeo AS. Her main supervisor is Professor Bjørn H. Hjertager. She also collaborates with Acona Flow Technology, having Eirik Manger as a co-supervisor (see page 22).

page 20

I have tested two different turbine models (actuator disk and line) in the NOWITECH/NORCOWE wind tunnel blind test project. This gave me results about how the models perform. Together with my supervisors we have made a new and very flexible set up that allows you to study how waves influence the wind fields and how this “modified” wind field plays with the turbines. I call this CFD set up for wave influenced wind turbine simulations (WIWTS). I hope that someone in NORCOWE will do more work on WIWTS after I have finished my PhD. I believe there are a lot of opportunities to use this set up to investigate a lot of interesting features that I cannot cover in my PhD. When we made WIWTS we used parts of the NREL developed SOWFA tool. But we have changed it to run with a less computationally requiring turbulence model and now WIWTS can handle a moving wave surface as well as topography. www.norcowe.no


I have also made a review of the different standards used in the field of offshore wind energy and summarized how boundary layer effects over the ocean are handled. Of course - a lot of simplifications are necessary to make, but I think we can say that there is a gap between ‘best knowledge’ and ‘best practice’ in boundary layer meteorology. In my opinion all the NORCOWE activities contribute in bridging this gap!

Are there any international or industrial collaborations you would like to highlight?

A snap shot of a WIWTS animation. WIWTS is an openFOAM based set up that uses parts of the NREL developed SOWFA tool. It allows you to study wind turbine performance together with moving wave surfaces.

Eirik Manger, Acona Flow Technology, and Siri M. Kalvig, StormGeo/UiS at NREL.

I’m collaborating with Matthew Churchfield in NREL in Boulder Colorado. He made the SOWFA code and is a very talented and experienced researcher so I am very happy that I have the opportunity to collaborate and learn from him. I stayed at NREL for one week and learned a lot! I hope we will write an article together where we compare some simulations done with different turbulence models.

“It all began with some young enthusiastic weather loving people who wanted to communicate their fascination of the weather and the forces of Mother Nature.” Siri M. Kalvig on StormGeo

StormGeo AS was founded by Siri Kalvig and TV2,

Norway’s largest commercial broadcaster, in 1997. Celebrating their 15th anniversary in 2013, StormGeo is now a global company whose main asset and focus is operation support 24/7 for marine activity in shipping, oil/gas and offshore wind. StormGeo Renewables deliver development and services to the offshore wind industry. The last years they have supported the majority of the offshore wind operators and stake holders with decision support during installation in UK and Germany. They have also paid much attention to wind resource assessment and hence developed and set up both wave and atmosphere models, SWAN and WRF, in hindcast mode. The focus of one of their projects has been the coupling of WRF and SWAN to better understand the wave www.norcowe.no

interaction with wind over ocean. Wind production forecasting is also in their portfolio. StormGeo has previously been members of NORCOWE, and decided to rejoin the centre in January 2014. One of their main activities within NORCOWE will be met-ocean data and verification studies, sharing their expertise on WRF and SWAN. Olav Krogsæter from StormGeo defended successfully his industrial PhD at the University of Bergen in December 2013. The title of his thesis was “The marine atmospheric boundary layer and ocean waves under the aspect of offshore wind energy applications”, having Professor Joachim Reuder (UiB) and Dr. Gard Hauge (StormGeo AS) as supervisors. page 21


Acona Flow Technology A new partner in NORCOWE Acona Flow Technology (AFT) has joined NORCOWE as of January 2014. The company is located in Skien, Norway, and is a subsidiary of Acona as well as a member of the Acona group. Acona Flow Technology AS is an ISO 9001:2008 certified company which has senior personnel with extensive academic background and long industrial training. Our focus is on complex fluid flow calculations, especially tailored against oil and energy. The company activity spans several industries, including renewable energy. Participating in NORCOWE seems like a natural part of this ambition. A large part of the simulation activity in AFT is focused around Computational Fluid Dynamics, often denoted CFD. In short these are methods enabling numerical solution of conservation equations for mass, momentum and energy. With increasing computer capacity these techniques have become widely accepted and popular, and they are today used in many industries for investigating various flow phenomena. AFT has licenses and uses state of the art codes such as ANSYS/FLUENT and ANSYS/CFX, but the company is also looking into Open Source codes such as OpenFOAM. AFT has worked closely with StormGeo and the University of Stavanger (UiS) in developing tools and methods for investigating the effect of waves on the Marine Boundary Layer (MABL). The company seeks to continue this development as well as performing studies within this topic. The goal of this work is to quantify the MABL’s impact on important wind turbine properties such as power extraction, variable loads, operation window etc. In NORCOWE AFT will continue the work together with StormGeo and UiS on investigating the effect of the MABL. Studies with different wave profiles, wave ages and wave heights will be conducted, while monitoring thrust and energy extraction for the turbines. Also other studies and the coupling towards meso-scale are of interest. AFT is looking forward to the cooperation with NORCOWE and the partners.

page 22

www.norcowe.no


Slamming waves Ove Tobias Gudmestad, Professor, University of Stavanger

Laboratory test of wind turbine jacket foundations against breaking wave loads

The tests were carried out in the Large Wave Flume (GWK) in Hannover in June 2013, utilizing a grant from EU Large Scale Facilities. The tests were conducted as collaboration between UiS and NTNU and supported by Statoil, DNV GL, NORCOWE and NOWITECH. The experimental campaign was managed by the staff from the University of Stavanger (Ass. Prof. Charlotte Obhrai) and from the Norwegian University of Science and Technology. In parallel with wave conditions, global loads on the jacket were monitored together with the loads on the individual jacket members. The data from the experiments will be analyzed and also used by a NORCOWE funded PhD student who will start working at UiS in August 2014. At the University of Stavanger, PhD student Sung-Jin Choi will, furthermore, complete his PhD work on the analysis of the breaking wave loading utilizing other data sets.

Breaking water surface waves Breaking water waves can be grouped into four basic types: spilling, plunging, collapsing and surging.

Wave slamming tests performed from 3rd to 30th of June 2013

The offshore wind industry is concerned that breaking wave loads may cause large loads on offshore wind turbine foundations located in certain shallow water areas, like on shoals and in waters with complex bottom geometries. Such areas could lead to spilling breakers and even to plunging waves. The combined effects of a number of breaking waves could cause early fatigue damage of the foundation structures and in the worst case structural overload caused by large breaking waves. In order to improve tools for estimating wave loads on jacket types of wind turbine foundation structures installed on shoals and in other areas where breaking waves could occur, experimental data basis for the load modelling was gathered in a wave flume at a relatively large model scale (1:8). www.norcowe.no

Spilling breakers occur in areas with gradual slope, and are characterized by a slow dissipation of energy as they break for a longer time than other waves. Plunging breakers occur when the ocean floor is steep or has sudden depth changes. Collapsing waves are a cross between plunging and surging in which the crest never fully breaks. Surging waves originate from from long period, low steepness waves and/or steep beach profiles, and are characterized by short, sharp bursts of wave energy as the base of the wave moves rapidly. (Extracted from Wikipedia - Breaking wave)

page 23


Opening of the Norwegi On Wednesday the 27th of November the new motion laboratory at the University of Agder (UiA) was officially declared open by Rune Volla, director of the Department of Energy at the Research Council of Norway. Around 50 invited guests were present. The equipment in the laboratory has a price tag of around 6 million NOK (approximately 720 000 Euros), while the extension of the building made ad hoc to host it all has cost around 10 million NOK (1,2 million EUR). All the invited speakers at the opening expressed their admiration at how quickly the laboratory has been built and equipped. The laboratory is unique in Norway, and several of the tests that can be performed here will be close to impossible to carry out offshore. Since the laboratory hosts two motion platforms (so-called Stewart platforms), it is possible to simulate for instance heavy lifts from a boat to a floating wind turbine performed in presence of large waves. During

Examples of possible experiments: Heave-compensated crane solutions for transfer of personnel/loads Heave-compensated drilling systems Accuracy performance evaluation of measurement systems, such as LiDARs Dynamic drop tests, for example of scaled-down lifeboats Playback of time series (6-dof), for example waveinduced vessel motions www.motion-lab.no

the opening PhD student Magnus Berthelsen Kjelland used a hydraulic lift to place a load onto the largest of the two motion platforms. During the operation the platform was programmed to undergo movements corresponding to that of

Three generations of UiA mechatronics personnel in front of the new motion platform: From the left Professor Michael Rygaard Hansen, previous headmaster and doyen of hydraulics Knut Brautaset, PhD student Magnus Berthelsen Kjelland, laboratory director Eivind Arne Johansen and Professor Geir Hovland.

page 24

www.norcowe.no


ian Motion Laboratory a large floating wind turbine situated in the North Sea. NORCOWE hopes that the motion laboratory will lead to an even closer collaboration with the industry. In the laboratory one may work with projects that can be identical to those encountered in the industry. This is a vantage point also for the students, something that Magnus Berthelsen Kjelland highlights. The laboratory thus has a two-fold mission: to aid the development of the industry and to educate new personnel to the businesses that work within the sector. The equipment is sponsored by the Research Council of Norway and NORCOWE. During the opening NORCOWE was represented by the Centre Director Kristin Guldbrandsen Frøysa. She emphasized the importance of creating a culture where knowledge is shared. Informal mingling can be of great importance, leaving space for the germination of new ideas. The idea behind the motion laboratory was in fact born during one of these informal sessions at a NORCOWE meeting. Bjørn Vedal from Ugland Eiendom is naturally very proud of the new building. Little more than a year passed from when the Research Council of Norway gave the project green light

www.norcowe.no

to when the building stood finished. Rune Volla from the Research Council of Norway is of the view that the new laboratory will be an important element in wind energy research. Wind will be an important energy resource in the future, and it is therefore crucial to have qualified research institutions that are prepared for the new challenges.

Equipment in the Norwegian Motion Laboratory Stewart-platform EMotion-1500 Stewart-platform EMotion-8000 Vehicle Loader Crane HMF 2020-K4 FARO Laser Tracker (X,Y,Z), R=36 m Siemens 300-series and ET200S PLCs ABB 800xA DCS RT LabView PCs/CompactRio’s Profibus, TCP/IP, UDP Interfaces Visit www.motion-lab.no for more information. Organisations which are members of NORCOWE receive a 50% discount on the operational costs.

page 25


Tore Bakka finished his PhD in November 2013 at the University of Agder with Professor Hamid Reza Karimi as main supervisor. His topic was turbine control. You have recently finished your PhD. Congratulations! What where your main findings? In its core my thesis deals with pitch control of a floating wind turbine – more specifically the identification of a control scheme for the operation at high wind speeds. I try to use modern model based feedback design techniques in order to maintain a good performance of the floating wind turbine. Especially the usage of linear design methods for a nonlinear system, by making use of linear parameter varying models, is an important development. The models are extracted from a known and validated model, and simplified to enable control design. Simulations of the full model are used to show the potential.

page 26

Where are you working now? Now I am working at National Oilwell Varco (NOV) as a hydraulic design engineer. In my PhD I have worked a lot with modelling and simulation. This is something which is very useful in my new job.

National Oilwell Varco was a NORCOWE partner from the start in 2009 untill 2013 (included). In 2013 they supervised two master students at the University of Agder whose topic was motion compensation for a crown mounted compensator. A scaled version of the compensator was placed at one of NORCOWE’s Stewart platforms to test out an active system to compensate the motion. Read more about the Norwegian Motion Laboratory on page 24.

www.norcowe.no


Torge Lorenz is doing his PhD at the University of Bergen/Uni Research with Angus Graham as main supervisor. What is your scientific background? I graduated from the University of Hamburg with an MSc in Atmospheric Physics in December 2011. For three years during my studies I worked as a student assistant at the German Climate Computing Center in Hamburg. My master thesis was about numerical modelling of the marine atmospheric boundary layer (MABL) in the Weddell Sea, Antarctica. This project was a cooperation between the University of Hamburg and the Finnish Meteorological Institute (FMI) in Helsinki. I continued my model experiments at the FMI for a few months before I moved to Bergen to start as a research assistant at Uni Research. I began my doctoral studies in January 2013. What topic is addressed in your PhD? The topic of my PhD program is numerical modelling of the MABL on the mesoscale. The first part of my work addressed the wind energy sector’s need for high-resolution climatological statistics on the wind in the North Sea. Since in situ meteorological measurements in the North Sea are too sparse to obtain such statistics, we use a numerical model of the atmosphere to compute wind data of sufficiently good spatial and temporal resolution. We apply the Weather Research and Forecasting Model (WRF) to downscale dynamically the reanalysis ERA-Interim (a leading set of large-scale estimates www.norcowe.no

of the past state of the atmosphere), while assimilating QuikSCAT satellite surface wind measurements (see page 29). My future work will be about short-term wind forecasting linked to marine operations in the North Sea (e.g. turbine placement and maintenance). We will develop a downscaled forecast ensemble, where several forecasts are computed in parallel with slight differences in the model parameterisations and initial conditions. One of the main goals here will be to find the right algorithm to create the perturbations in a meaningful way and tailored to the conditions in the North Sea MABL. What are your main results? The result of our downscaled reanalysis in the North Sea is a data-set which allows us to compute high-resolution annualised statistics of meteorological characteristics like wind speed and direction, turbulence intensity and atmospheric stability. These characteristics are of utmost importance for the wind energy sector, for example to calculate turbine power output, optimize the wind-farm layout or assess wake effects. In comparison with observations, our 10 m wind speed estimates show an accuracy of Âą0.3 m/s for most of the open North Sea. In addition, the verification against radiosondes and ASCAT satellite winds allows us to quantify the benefit of QuikSCAT data assimilation on the simulated MABL winds. page 27


Obtaining site climatolo offshore wind industry Angus Graham, Idar Barstad, Torge Lorenz, Alla Sapronova and Alastair Jenkins, Uni Research

important mesoscale and coastal effects, involving the transition from flow over land to sea and vice versa; lee waves; or topographic jets and wakes. Local seabed bathymetry may modulate the incident wave field through refraction, with some associated wave breaking being possible, together with diffraction at islands and headlands. Accommodating these effects to predict site climatologies accurately is scientifically and computationally demanding: a dynamical downscaling from hindcasts of the large-scale synoptic state of the atmosphere and ocean is involved. Outputs from the downscaling must also be carefully specified and post-processed, converting from the time domain to obtain distributions and spectra of direct interest. Accurate load calculations require joint directional energy spectra of the waves and hub-height winds, obtaining over at least the planned lifetime of the wind farm (typically twenty years). Calculations of the total energy yield from the farm must take account of the dynamical interaction of turbines through their downwind wakes, and thus require the joint distribution over the farm lifetime of hourly hub-height wind speed and direction, turbulence intensity and atmospheric stability parameter (such as Richardson number).

Introduction It is essential to know the relevant environmental climatologies at the site of a potential wind farm, to the statistical standard necessary for accurate calculation of extreme and fatigue loadings and of energy yields. The investment risk, and the associated financial premium, can thus be minimised. Installing a measurement mast extending to at least the planned hub height of turbines is an expensive proposition offshore, however, and carries significant risk as the option not to proceed on the basis of the findings must presumably be retained. If model predictions, post-processed as necessary, can approach the accuracy of site measurements, such masts need no longer be recommended in the standards. Some modelling is, in any case, unavoidable, as investors are seldom in a position to wait for a statistically robust set of measurements to be obtained over the lifetime of the envisaged farm before reaching a decision. Offshore wind farms are necessarily sited on the continental shelf or shelf break, and winds there may be subject to page 28

FIG. 1. Outer and inner domains used in the WRF wind-climatology simulations. Red shows the land mask, and the sea-ice mask adopted during a thirty-day wintertime period of calibration to satellite winds. www.norcowe.no


ogies for use by the

(a)

(b)

FIG. 2. Joint histograms of stable atmospheric flow parameters at a North Sea location 57.26N, 7.00E, from a ten-year, 3-km gridded mesoscale model simulation, a) wind speed at 2 km and buoyancy frequency, NBV, between heights of z1=500 m and z2=2500 m, b) wind direction at 2 km and NBV. Ordinate shows the number of cases.

Joint histograms of relevant environmental variables may be obtained fairly straightforwardly from time-series outputs following downscaling. Manipulating these to deliver distributions in a form suitable for use by the industry requires rather more work. The sample size of the histograms may be limited, and not climatologically robust, as a result of computational cost and run-time constraints. There may also be associated statistical fluctuations that are best smoothed out. There is also the need to parameterise and reduce joint distributions, so information can be supplied in as economical a form as possible.

NORCOWE research Downscaled wind climatology - model The Weather Research and Forecasting model (WRF) is being used to generate a downscaled climatology of the northwest European shelf. A fine-scale inner domain of horizontal resolution, 3 km, is telescopically nested within an outer domain of horizontal resolution, 9 km, as shown in Fig. 1. The boundary

www.norcowe.no

forcing is that of the ERA-Interim reanalysis dataset supplied by the European Centre for Medium-Range Weather Forecasts (ECMWF). This has a horizontal resolution of about 70 km. A two-way coupling exists between the inner and outer domains, so the boundary conditions initially imposed on the inner domain are relaxed, and do not limit the accuracy achievable within. The Mellor-Yamada-Janjic (MYJ) scheme for Reynolds-averaging the boundary-layer turbulence is used. There are fifty-one vertical levels, specified using the model’s terrain-following hydrostatic pressure ordinate, and solutions are output at least every six hours from the outer domain and every three hours from the inner. Downscaled wind climatology - simulations and results A period spanning the first millennial decade has been simulated (corresponding to 30552 three-hourly samples). Statistics of simulated flow parameters at a specimen offshore location some 70 km south of the southern tip of Norway have been obtained. Joint histograms of the mean wind velocity at 2 km height, in the free atmosphere, and of atmospheric stability between reference heights of 500 m and 2500 m

page 29


(a)

(b)

FIG. 3. Degree of overlap of distributions of predicted and measured 10 m wind speeds. a) 3-km gridded model vs. QuikSCAT b) Era-Interim reanalysis vs. QuikSCAT. The degree of overlap is given by 1−∫|pp−pm|dU10/2, where pp and pm are probability densities of occurrence of U10 according to predictions and measurements, respectively.

are presented in Fig. 2. The histograms concern times when the characteristic fractional bulk gradient in virtual potential density between these heights is negative, and flow in the layer can thus be deemed stable. The form of the stability distributions appears uncorrelated with wind speed or direction, and the associated joint probability densities may thus be supposed products of univariate functions. The lack of correlation may be indicative of a choice of heights comparable to or exceeding boundary-layer depth, with geostrophic winds in the absence of turbulent mixing forced by shear stress at the boundary tending merely to advect the stratification. Modelled distributions may be validated against a number of observational datasets. Predicted 10 m wind speeds have been validated to values derived from satellite scatterometry over a 30-day period from February 19th – March 20th 2008. Twice-daily measurements from the SeaWinds scatterometer on the QuikSCAT satellite are involved, along with output from each model grid point over sea in the inner simulation domain (at an output time nearest the output time stamp of the QuikSCAT cell in which the grid point lies). In Fig. 3, the resultant histograms are compared via a metric of their degree of overlap (see caption for more details): a value of zero would correspond to no predicted wind speeds lying within the measured range, and a value of one to the histograms being identical. Predictions made according to the large-scale boundary forcing are shown in Fig. 3a, those after downscaling in Fig. 3b. An improved match is seen throughout the inner simulation domain, demonstrating the value of dynamical downscaling in resource and site assessment.

page 30

Coarse-to-fine model calibration with Artificial Neural Network (ANN) There is always a conflict between accuracy and runtime in deriving a site climatology from a mesoscale model. The greater the resolution of small-scale processes sought, the greater the computational cost, and the less practical simulation over climatologically robust periods becomes. The use of

FIG. 4. Scatter plot of predicted westerly component of 10 m winds at the location of the FINO1 met mast, as derived from output over one year from a WRF simulation: predictions from results at 9 km horizontal resolution are plotted against results at 3 km. Red crosses show WRF values at 9 km, green crosses the values after ANN processing.

www.norcowe.no


an ANN to calibrate model results at a site at different resolutions, and thus allow extension of high-resolution results to long time scales, is being investigated. One year’s worth of data, from July 1st 1999 to June 30th 2000, at the location of the FINO1 met mast in the German Bight was selected from the decadal WRF simulation. Outputs from the outer 9 km domain, comprising the wind vector at 10 m and the friction velocity, have been fed to the network, and used to predict values of these variables within the inner 3 km domain. (As the domains are two-way nested, a smaller difference between results at the two resolutions is to be expected than in the one-way case.) A number of network architectures have been examined, with no account thus far taken of the order of data in the input time series. A training subsample extracted from the input data comprises a randomly-selected time series one quarter the size (these data are thereafter excluded from the calibration). In Fig. 4, the westerly component of 10 m winds predicted at 3 km horizontal model resolution using ANN is plotted against the value actually obtained at this resolution from the mesoscale simulation. Also plotted are the corresponding simulation values at 9 km, which comprise part of the input to the network. A perfect predictor of the 3 km values would produce points lying exactly on a straight line of 1:1 mapping. The fit obtained by ANN is significantly tighter with respect to the mesoscale simulation (correlation coefficient of 0.9993 vs. 0.9937; r.m.s. error of 0.24 m s−1 vs. 0.72 m s−1). Thus it is clear that a usefully accurate relation between inputs and outputs of the mesoscale model may be established on a wholly empirical basis, without knowledge of the dynamical modelling applied in the downscaling.

(a)

(b)

Downscaled wave climatology Surface-wave climatologies over the north-west European shelf have been obtained using the WAM model developed by the international Wave Modelling Group. A simulation extending over 2008 has been completed, forced by the wind field from WRF and ERA-Interim wind and wave boundary conditions. The coupling with WRF is one-way, the wave field thus not feeding back into WRF (the wave field in the standalone or one-way-coupled WRF is assumed fully developed). Such a feedback may be assumed not to impact on the wave field significantly. Histograms of spectral characteristics are straightforward to obtain, and to validate (a focus of future work). WAM is, however, a spectral wave model, the directional wave spectrum thus being modelled at each grid point in space and time, and so it is also possible to build up the directional wave spectrum pertaining over a climatologically-significant period at a site (another future focus). In Fig. 5, histograms of simulated spectral characteristics at the location of the Ekofisk oil platform in the central North Sea are presented. The plots are consistent with waves being more likely to originate from the northern open boundary than anywhere else, in which case swell dominates (Fig. 5c); but with a wind-driven sea nonetheless being much more likely than a swell-dominated one.

(c)

FIG. 5. Histograms of predicted spectral characteristics of the surface wave field at Ekofisk obtaining over 2008, a) significant height, b) mean period (weighted on energy per unit sea surface area and angular and frequency increment) c) originating directions of surface wind (blue) and net wave-energy flux (green).

www.norcowe.no

page 31


The Norwegian Meteoro A new partner in NORCOWE The group of oceanography and marine meteorology (OM) in the Research and Development department of The Norwegian Meteorological Institute (MET Norway) in Bergen, has been involved in NORCOWE since the start-up of the centre. Through a subcontract with Uni Research, the PhD work of Ole Johan Aarnes on extreme value statistics of waves was partly funded, and scientists from OM has been active in the internal scientific committee of NORCOWE and the arrangement of the summer schools. In 2013, the Geophysical Institute at UiB and MET Norway carried out a 5-month comparison experiment with LiDAR and Radiosonde at Sola airport. Through further involvement in NORCOWE, MET Norway sees a potential for more of this kind of interesting scientific collaboration with partners in the centre. About MET Norway MET Norway was founded in 1866 and has today approximately 430 employees. The main office is in Oslo and regional offices are in Tromsø (Forecasting Division for Northern Norway) and Bergen (Forecasting Division for Western Norway).

The main duties of MET Norway • • • •

• • •

To issue weather forecasts To study the national climate conditions and produce climate reports To provide meteorological observations from Norway, adjacent sea areas, and from the Svalbard area To carry out research and development in support of the institute’s operational functions to ensure that the services are of the highest possible standard To make available the results of our work To provide special services for the public and private interests on a commercial basis To participate in international meteorological co-operation

Vervarslinga på Vestlandet and the Geophysical Institute, Bergen.

page 32

www.norcowe.no


ological Institute that they carry information about the uncertainty of the forecasted values. These techniques are being further improved and tested during the Research Council of Norway project “Decision support for offshore wind farms”, lead by CMR.

Figure 1: Satellite image of wind map.

Met Norway provides meteorological services for both civil and military purposes, for protection of life and property, and for protection of the environment. Users include authorities, commerce and industry, institutions and the general public. Vervarslinga på Vestlandet Weather services for western Norway and ocean areas up to mid-Norway are performed by the marine forecasters in Bergen, at the Forecasting Division for Western Norway (Vervarslinga på Vestlandet, VV). VV is organised as part of Meteorological department of MET Norway. Research and development department In total about 70 scientists work in the Research and development department in Oslo and Bergen. The research, development and forecasting at MET Norway which is relevant for offshore wind energy is carried out mainly in the Numerical modelling division in Oslo and the Oceanography and marine meteorology group in Bergen. The Numerical modelling group is responsible for atmosphere model development, ensemble model systems and data assimilation. This group also carry out important work on validation methods and statistical calibration techniques that are used for e.g. forecasting power output from wind farms. An advantage of these methods is

www.norcowe.no

The scientists in Bergen are working with wave observations, modelling and forecasting, and drift models. i.e. models for calculating the location of a man-over-board, a ship or the drift and extent of an oil spill. Through downscaling to 10km grid size of the ERA-40 data the Norwegian Reanalysis Archive of wind and waves was produced and is now widely used for offshore wind and wave statistics. NORA10 covers Norwegian ocean areas, including the North Sea, Northern North Atlantic and up to Svalbard and Novaya Zemlya. The first part of the archive is a downscaling of the ERA-40 data set over the original model period, starting in September 1957 and ending in August 2002. Using analyses from the ECMWF forecast model, NORA10 is being extended further on a semiannual basis. A new, shorter archive is presently under way. This will be a downscaling of the ERA-interim to the same 10km grid as NORA10. The group also runs projects related to marine meteorology and satellite remote sensing. Forecasting for Hywind and analysis of the wave and meteorological conditions at the planned Havsul wind park are some of the projects. High resolution satellite wind retrieval is added to the processing of satellite Synthetic Aperture Radar (SAR) data. These wind maps are useful for studying topographic effects and are also interesting for wind resource mapping. An example of a wind speed map from Radarsat2 ScanSAR on the 30th of January is shown in Figure 1 (wind speed in m/s). From 2015, much more SAR data will be available through the launch of the European Space Agency satellite Sentinel1. The processing from satellite image to the wind map shown in Figure 1 is based on new open-source software being developed at Nansen Environmental and Remote Sensing Center and MET Norway. It is anticipated that easier access to SAR wind maps will increase the use and benefit of this type of satellite data. This is one of the topics in the NORCOWE work plan for 2014. Testing of new wave measuring techniques using autonomous vessels is another activity included.

page 33


NORCOWE strengthens condition monitoring Jasna Bogunović Jakobsen, Professor, University of Stavanger

One of the research topics addressed by NORCOWE is the influence of atmospheric stability on wind power production and wind turbine structural response. In addition to numerical analysis, it is planned to investigate this matter based on full-scale observations. For a thorough assessment of turbulent flow characteristics, NORCOWE has purchased three WindMaster Pro ultrasonic anemometers 1561-PK-020, by Gill Instruments. In a preliminary campaign prior to deployment in wind energy, the sensors are installed on Lysefjord bridge, where characterization of turbulent flow field across the fjord is essential for validation of computational models

Bridge location at the inlet of Lysefjord (norgeskart.no)

for wind-induced bridge vibrations. The wind records from the NORCOWE sensors will increase the data base on spatial flow characteristics along the bridge, in parallel with measurements of the bridge girder accelerations. The measurement campaign is managed by the University of Stavanger, with Professor Jasna Bogunoviċ Jakobsen and Professor Jonas Thor Snæbjörnsson (Reykjavik University) as principle investigators. The work is supported by the Norwegian Public Road Administration (NPRA), in Norway known as Statens Vegvesen.

Jarle Berge (NPRA), Jasna Bogunović Jakobsen and Jonas T. Snæbjörnsson

page 34

The measurement data collected since November 2013 are currently being analysed by Etienne Cheynet, a PhD student at the University of Stavanger, and three master students. The impact of the topography on wind conditions on the bridge is investigated with reference to the data from the weather station at Sola airport, some 30 km to the West, in a flat coastal area. Depending on the prevailing global wind direction, the wind speed, wind direction and the structure of turbulence observed on the bridge are seen to be affected by the fjord topography to a variable degree. Details of the wind flow on the bridge and the bridge vibration data are being

www.norcowe.no


s its capacity for wind Sonic anenometers, first developed in the 1950s, use ultrasonic sound waves to measure wind velocity. They measure wind speed based on the time of flight of sonic pulses between pairs of transducers. Their temporal resolution is very fine, which makes them well suited for turbulence measurements. Since the speed of sound varies with temperature and is virtually stable with pressure change, sonic anemometers are also used as thermometers. Source: Wikipedia - Anenometer

PhD student Etienne Cheynet with a data acquisition unit.

further investigated. The NPRA has also approved funding for a pilot measurement campaign in 2014, in which a LiDAR will be introduced for a simultaneous monitoring of the wind field at the bridge location. The LiDAR measurements will be undertaken by Professor Joachim Reuder’s research group from the University of Bergen. The pilot test campaign will spur off further research on the applicability of LiDARs for turbulence monitoring and for wind surveying for fjord crossings planned as part of Coastal Highway E39 project.

www.norcowe.no

page 35


Measurement campaign

Birgitte R. Furevik, Norwegian Meteorlogical Institute, Valerie Kumer and Joachim Reuder, the Geophysical Institute at the University of B

The new generation of scanning LiDARs enable not only a visualization of the ambient flow field by scanning two dimensional vertical or horizontal cross sections, but also show a higher range. Scanning in 2D makes new measurement strategies possible and opens a new way of analyzing boundary layer structures as well as interacting processes like wind turbine wakes. However, the wind measurements require new validation methods since the measurement range exceeds the one of meteorological masts. The University of Bergen conducted therefore in 2013 the LiDAR Measurement Campaign Sola (LIMECS) in collaboration with NORCOWE and the Norwegian Meteorological Institute (MET), to investigate coastal boundary layer processes and perform a comparison of scanning LiDAR and Radiosonde wind profiles.

internationally through the global telecommunication system (GTS) for assimilation into weather forecast models. However, the radiosonde records every 2 seconds, which with a rising speed of about 5m/s gives a vertical resolution of around 10m. Wind speed is measured by recording the GPS position as the balloon drifts with the wind while rising. The 2-second wind speed from radiosonde was validated against in situ measurements in a previous work which gave confidence enough to use them for comparison to LiDAR winds. Also the simultaneous measurements of the atmospheric stability are very valuable for the experiment.

Figure 1: Land breeze observed by LiDAR on March 12, 2013. Note the reversal of flow. It occurs here at about 360 m above the ground.

A Radiosonde is a rising weather balloon equipped with a GPS antenna and sensors for temperature, humidity and pressure measurements. Radiosondes have been used since the 1920s to measure profiles throughout the whole atmosphere and are operated daily by Meteorological Institutes all over the world. Radiosonde data from certain levels are distributed

page 36

Operating a scanning LiDAR at the Norwegian coast allowed us to additionally investigate boundary layer transition processes from sea to land. A common process is the sea breeze. It is generated by stronger solar heating of the land areas than the sea, leading to rising air over land. As air from the sea flows toward land, it is turned to the right due to

www.norcowe.no


ns at Sola and ECN

Bergen

Coriolis and a northerly wind is set up on the west coast. This usually happens in the spring and early summer when the sea is cold. In the example shown in figure 1, the opposite, less pronounced phenomenon of a land breeze is observed in the morning on March 13 2013. In the case of the land breeze the sea is warmer than the land, due to night-time cooling. The cooled air flows downhill towards the sea. Due to pressure differences a compensating air flow aloft is directed towards land, where it may sink and thus lead to a closed circulation. The land breeze, with velocities of 1-2 m/s and a depth of around 300 m, is not as strong as the sea breeze with speeds of > 5 m/s and a depth of 1-2 km. Both processes are of great importance for offshore wind energy as they influence the energy production of wind parks in coastal areas.

LiDAR – Light Detection and Ranging The LiDAR uses laser to measure the wind speed and direction. The laser is reflected by aerosols in the atmosphere, and the resulting frequency shift (Doppler shift) measured in the returning wave is used to establish the velocity with which the aerosols are moving. NORCOWE owns several LiDARs from the producer Leosphere, which became a NORCOWE partner from January 2014.

The article continues on the next page.

PhD student Valerie Kumer at the ECN test site in Wieringermeer, the Netherlands.

www.norcowe.no

page 37


Figure 2: Map of the LIMECS site.

LIMECS was set up at two sites at the airport of Stavanger in Sola and lasted from March 1st, 2013 until August 24th, 2013. The scanning WindCube (WLS100S-8) and a WindCube v1 (WLS7-67) measured wind fields and profiles from above the rooftop of the fire brigade building at Stavanger airport (site 1 in the map in Figure 2). The fire brigade building is located 1.7 km from the coast line. With that, the measurements were in general inside the transition zone between the maritime boundary layer and the one over land. Further inland and 2.3 km southeast of site 1 the second WindCube v1 (WLS7-65) measured wind profiles next to the autosonde from the MET (site 2 in Figure 2). Preliminary results of the LIMECS campaign were presented at the 11th Deep Sea Offshore Wind R&D Conference in January 2014 in Trondheim. Based on the results and experiences of LIMECS a new international measurement campaign has been set up in cooperation with the Energy Center of the Netherlands (ECN). This

experiment started in the beginning of November 2013 and is intended to last until the end of April 2014. A total of 4 static LiDAR wind profilers WindCube v1 and v2 from Leosphere and one scanning WindCube WS100S are located in the vicinity of the 5 test turbines at the ECN test site Wieringermeer. In addition one of the turbines is equipped with two nacelle mounted LiDAR systems: one AVENT Wind Iris looking upstream into the incoming flow and one horizontally looking ZephIR 300 pointing downstream into the turbine wake. The instrumental set-up is sketched in Figure 3. The main goal of WINTWEX-W (Wind Turbine Wake Experiment – Wieringermeer) is the measurement of single wind turbine wakes under a wide range of atmospheric conditions. With the collected data we are aiming for a qualitative and quantitative description of single wind turbine wake evolution, propagation and persistency, as well as to improve the full scale data basis for the initialization, test and validation of corresponding CFD models.

Figure 3: Instrumental set-up during the WINTWEX-W campaign at the ECN test-site.

page 38

www.norcowe.no


Valerie Kumer is doing her PhD at the University of Bergen with Professor Joachim Reuder as main supervisor. What is your scientific background? I got my diploma in meteorology at the University of Vienna in 2012, where I also worked as a research assistant in the theoretical meteorology group. What topic is addressed in your PhD? As being part of the experimental meteorology group at the University of Bergen I am working with Doppler LiDAR wind measurements. During my PhD I am testing static and scanning LiDARs for their application in wind energy, with the focus on wind turbine wakes. What are your main results? Preliminary results of comparison and performance tests of the scanning LiDAR to radiosonde wind profiles during the LiDAR Measurement Campaign Sola (LIMECS) lead to the conclusion that the WindCube 100S wind speeds and directions correlate well with wind measurements of radiosondes. Additionally, the WindCube 100S was able to catch coastal processes, such as the land-sea breeze circulation, making it a useful tool for boundary layer studies. While there were some technical and software related issues, the data availability was linked to the planetary boundary layer depth.

www.norcowe.no

Are there any international or industrial collaborations you would like to highlight? After an upgrade of the scanning LiDAR WindCube 100S at the manufacturer Leosphere, I visited them for some expert training outside of Paris. I enjoyed the expert training very much, as it answered some of my questions about the device and I learned new manual settings on distance calibration and focusing height. The visit additionally strengthened the scientific collaboration with Leosphere, as the device was prepared for the Wind Turbine Wake Experiment - Wieringermeer (WINTWEX-W) and scanning strategies were discussed. At the moment our LiDARs are collecting data within the scope of a collaborative measurement campaign with the Energy Centre of the Netherlands (ECN). The WINTWEX-W campaign is conducted at the wind turbine test site of ECN. The aim of the campaign is to measure wind fields and profiles of single wind turbine wakes, in order to later analyse their structure and behaviour under different weather conditions. Additionally, the data will be used for wake model validations.

Valerie Kumer also contributes to the blog Climatesnack (www.climatesnack.com). In May 2013 she wrote “Earth Science goes Star Wars – how lasers can catch the wind”

page 39


Leosphere A new partner in NORCOWE Founded in 2004 by Alexandre and Laurent Sauvage in France, Leosphere is today a recognized global leader in the laser radar (LiDAR) remote sensors market, present at the crossroads of atmospheric environmental applications such as Air Quality and risk, Weather and Climate, Aviation Weather and Wind Power. The flagship product of Leosphere in the Wind Power market is the Windcube v2, a vertical ground based wind profiler based on a unique Doppler pulsed LiDAR technology, allowing direct measurement of the wind and flow inclination as well as the vertical dimension of the wind. Why did you join NORCOWE? Being part of this consortium is a great opportunity for Leosphere to promote our LiDAR technology in various applications, mature or emerging. With some major resources invested in science and R&D, being part of this consortium is part of our strategy of developing partnerships for scientific development.

What is your key activity within wind energy? Leader in the LiDAR technology, Leosphere positions itself as a unique one-stop-shop. We provide a solution to optimize the value of our customers’ projects with best-in-class LiDAR wind measurements. What have been your previous contacts with NORCOWE? The first contacts with the NORCOWE consortium took place in 2012 through the University of Bergen that acquired a scanning Windcube 100S LiDAR for some specific studies on the wind flow transition from offshore to onshore and for some wake studies. One of the PhD students came for a few weeks training at Leosphere on the LiDAR technology. How do you see the future within offshore wind? With the current highest capacity factors due to stable and regular winds, offshore wind has some good potential for further development in the coming years. One of the major drawbacks of offshore wind power is the current develop-

NORCOWE and ECN personnel setting up a Leosphere LiDAR.

page 40

www.norcowe.no


ment and construction costs, for which we think the LiDAR can bring a solution for e.g. the wind resource assessment and the wake study. About the company : In 2009, Leosphere created the spin off Avent LiDAR Technology specialized in Turbine Nacelle mounted LiDARs with the Wind Iris. This product completes the portfolio of LiDARs with one being able to measure the upstream or downstream wind from the nacelle of a turbine, for applications such as yaw alignment, power curve or performance optimisation. In addition, Leosphere offers the Windcube v2 with offshore and buoy versions, and the scanning Windcube for long range wind measurements. In 2013, 70% of the sales of the group Leosphere were done in the wind power market and 85% were non-domestic. In order to maintain its technical leader position, Leosphere spends more than 20% of its revenue in R&D. At the present date, more than 400 systems have been deployed in over 25 countries. LiDARs’ use in the wind power market 1. Wind Resource assessment: The profitability of a wind farm directly depends on uncertainty reduction during the wind resource assessment and wind turbine optimization phases. Projects risks vary from site to site, depending upon wind farm type (onshore/offshore), size and site complexity. www.norcowe.no

2. Power curve measurement: In order to assess the performance of a wind turbine, power curve measurements are usually performed. It allows measuring the correlation between the wind speed and the output power of the turbine and to compare this measurement to the contractual power curve. 3. Performance optimization: For existing wind farms, power performance optimization is a possibility. Seeing the wind exactly as the turbine sees it is essential for meaningful performance analysis. Positioned on top of the nacelle, a LiDAR like the Wind Iris characterizes the approaching wind at hub height. 4. New Applications: Thanks to the versatility and polyvalence of LiDARs, new applications are now emerging in the wind research community, such as wake studies and wind farm forecasting optimization for example. Power forecasting is the ability to predict the energy that will be produced by a wind farm in a mid to longterm time-line. A scanning Windcube can measure the wind remotely of a wind farm in order to estimate the upcoming wind. Mid to long term forecasting is also key in terms of data assimilation that would help to improve forecasting models.

page 41


Mostafa B. Paskyabi is doing his PhD at the University of Bergen with Professor Joachim Reuder as main supervisor. What is your scientific background? I have a degree in applied mathematics with focus on partial differential equations and stochastic differential equations for solving fluid mechanic problems. Whilst still at university, I got the chance to work in a research institute as an engineer in signal processing, computer programing for real time data acquisition, and numerical modelling of ocean wave propagation. I stayed there for about 8 years, keeping close contact with academia also through teaching and co-supervising a master thesis. I became interested in problems related to offshore hydrodynamics and air-sea interaction, and decided to return to academia to do a PhD. What topic is addressed in your PhD? I am presently studying turbulence in the upper ocean and the marine atmospheric boundary layer in the presence of wavy and windy air-sea interface. In this study, a dedicated autonomous subsurface moored platform (called MATS) was developed to measure turbulence variability near the sea surface when waves and winds impose a substantial amount of momentum and energy into both the water column and the atmosphere. We used MATS to do a couple of deployments mainly in a measurement site near the Havsul窶的 area off the west coast of Norway, the first site in Norway with a concession for an offshore wind park, alongside with a tethered free-fall microstructure profiler deployed from the ship during the cruises. We have also developed a wave toolbox in the General Ocean Turbulence Model (GOTM) incorporating page 42

wave forcing in the upper ocean mixing models. Furthermore, a signal processing toolbox was developed to decontaminate the wave-induced motion from our acquired microstructure signals measured by MATS. We checked its reliability by applying it for high resolution wind measurements from a sonic anemometer mounted on a discus buoy with sensors located at the wave-affected boundary layer. The results were very encouraging and successful. In addition to the motion correction toolbox, we could estimate the surface gravity waves using data acquired from a motion-corrected pressure sensor mounted on the MATS. What are your main results? Surface gravity waves are one of the main sources of vorticity, momentum and energy in the upper ocean mixed layer in the way that they can substantially affect the whole mixed layer. A large wind farm in the presence of surface gravity waves enhances the variability of the upper ocean by organizing upwelling and downwelling regions around the wind farm. Wave directional spreading together with wave breaking parameterizations are important factors in estimating the drag coefficients in both the marine atmospheric boundary layer and the upper ocean mixed layer. Motion correction algorithms should be equipped with flow distortion and correlated wave-platform dynamics. These result in a new correction algorithm which can cover successfully expected behaviors of obtained solutions. www.norcowe.no


Martin Flßgge is finishing his PhD at the University of Bergen with Professor Joachim Reuder as main supervisor. You are close to finishing your PhD. Congratulations! What are your main findings? During my PhD, I worked with the application of the NORCOWE Direct Covariance Flux System which has been developed in collaboration with the turbulence group at the University of Ireland. The system is utilized for offshore applications and can easily be mounted on ships and buoys. Measurements of turbulent air-sea exchange processes from floating platforms are contaminated by platform motions. A correction algorithm has been successfully used for measurements taken from ships in the late 1990’s. Based on this algorithm, we have found that turbulence measurements of heat and momentum taken from buoys can be corrected to give accurate results. The usage of buoys for these kinds of offshore measurements makes the site specific investigation of the vertical turbulence structure cheaper, as buoys have lower operation costs and can easily be deployed within a wind farms. Will your research be continued within NORCOWE? The flux system will definitively be used for further investigations of atmospheric turbulence structures within NORCOWE. The collected turbulence data are used by other NORCOWE researchers in wind farm- and marine boundary layer modelling. Furthermore, the motion correction algorithms can

www.norcowe.no

easily be adapted to remotely piloted aircraft systems which are able to investigate the turbulence structure within wind farms. What will you be doing now? At the moment, I am finalizing my PhD thesis which I want to deliver this spring. I would like to continue my work within the OBLO project (described below). Such a position would give me the opportunity to use the research results from my PhD thesis within NORCOWE for measurements and mapping of wind resources. The Offshore Boundary Layer Observatory (OBLO) is one part of the NFR funded joint NORCOWE-NOWITECH infrastructure project NOWERI (Norwegian Offshore Wind Energy Research Infrastructure). OBLO is hosted at the Geophysical Institute of the University of Bergen and will be realized in close collaboration with CMR. The main aim is the provision and operation of state-of-theart instrumentation and measurement capabilities for a wide range of atmospheric and oceanic parameters relevant for offshore wind energy applications. OBLO started up in 2013 and is at the moment in the middle of the purchase process of advanced meteorological and oceanographic instrumentation, like microwave temperature profilers, sodar/RASS wind and temperature profilers, scanning wind LiDAR systems and underwater current and turbulence measurement systems.

page 43


The Gwind story Arnfinn Nergaard, University of Stavanger The arrival of Hywind in Åmøyfjorden in the spring of 2009 triggered a process of considering alternative designs for floating wind turbines at the University of Stavanger (UiS). Studying the difference between horizontal axis wind turbines (HAWTs) and vertical axis wind turbines (VAWTs), we concluded that the VAWTS might have certain merits superior to HAWTS when installed on a floater, the two key elements being 1) the advantage of moving heavy equipment down and into the hull and 2) the potential advantage of gyroscopic suppression of pitch and roll motion. Based on a patent application filed late 2009 by the TTO office Prekubator, a patent for “Gyrostabilized Floating Wind Turbine” was granted in January 2011. The concept was thereafter named Gwind indicating relation to the gyroscopic effect. The WAVT concept applied on a floater was identified as one potential new concept under WP 2, Innovative Concepts, in NORCOWE during the 2010 program. This was followed by a separate two-year NORCOWE project focusing on floating WAVT technology in general. WP 2.2, Development and qualification of floating VAWTS, was initiated early 2011 and concluded at the end of 2012. The deliverables were mainly

page 44

- so far

global designs, energy capture potential, mooring design and global motion analysis. A benchmark with a Hywind type concept concluded that there might be a potential of reducing the overall size and weight of the floater by 25 %. The work was performed by CMR Prototech and UiS. The two-year project was planned to be Phase I in a 7-year development program taking the concept through the necessary phases of design, testing and qualification to commercialization. However, due to budget constraints the NOCOWE VAWT project was discontinued at the end of 2012. The key element of the Gwind patent is the potential merit of gyroscopic forces suppressing roll and pitch motion and thereby reducing overall global motion of the floating VAWT. Prekubator, being a Technology Transfer Office owned by UiS and IRIS, initiated a project in parallel with the NORCOWE WP 2 in 2012. The project focused on verifying the gyroscopic effect and preparing strategies for commercialization of the Gwind concept. The project has been granted about 3.2 mill. NOK from the Research Council of Norway’s FORNY program under the project title: “Development, qualification and commercialization of floating VAWT - Proof of concept.” The scope of work had four elements: 1) Model design and construction, 2) Model tests, 3) Commercial proofing and 4) Commercial development. The project activities started in January 2012 and have continued through December 2013. The project has been supported with additional financial contribution from VRI (another RCN program) with 0.2 mill. NOK and from Innova-

www.norcowe.no


tion Norway with 0.3 mill. NOK. The work has been performed in close cooperation with CMR Prototech. A cooperation agreement with NORCOWE has secured access to the results obtained in NORCOWE as the Gwind activities is partly a spinoff of the initial NORCOWE program. The following milestones have been achieved: • Design and construction of laboratory gyro-model Q2, 2012 (CMR) • Testing for gyro-effect at Bergen University College, October 2012 • Further testing at Stadt Towing Tank, March 2013 • Preparation and presentation of commercial models, 2013 • Design and construction of a fjord scale model Q2, 2013 • Launching and Christening of the two ton “Spinwind I” fjord model, August 2013 • Initial testing of Spinwind I, September 2013 through January 2014

The business model is based on the following three-step strategy: 1. Constructing prototype 250 kW for in fjord fish farm 2. Constructing first single-unit offshore floating VAWT for powering platform 3. Full multi-unit wind farm development, see illustration below For step 1 Gwind has developed strong relations with Grieg Seafood, Blue Planet and Arena Ocean of Opportunities, the business model being replacement of diesel generator systems on fish farms in Norwegian fjords. As a part of the commercialization activities company Gwind AS was established December, 2012. From the start the company is owned 1/3 each by Prekubator TTO, UiS and Arnfinn Nergaard, initiator of the Gwind patent application.

The laboratory tank testing has concluded that the gyroscopic effect might eliminate peak excessive motion at natural pitch and roll periods when applying powerful gyros. The testing has verified theoretical models that enable assessment of magnitudes of motion suppression given gyro power and size and the geometry of the total system. The fjord testing of Spinwind I has not been concluded yet as data collection has suffered from downtime on the recording system. Video recordings are being analyzed during spring of 2014, while tests are planned to be resumed in April 2014. A successful unplanned survival test took place early December as Spinwind I was hit by a severe storm. A video recording of this episode can be found on NORCOWE’s home page.

www.norcowe.no

page 45


National and internatio NORCOWE has become more visible during 2013. In particular the measurement campaigns and the new motion laboratory have drawn a lot of attention. Topics addressed by NORCOWE NORCOWE addresses topics like meteorology, oceanography, maintenance and operation, wind farm control, marine operations and wind farm layout. The new partners (Acona Flow Technology, Aquiloz, Leosphere, Norwegian Meteorological Institute and StormGeo) have strengthened NORCOWE with respect to meteorology and oceanography, LiDAR measurement, CDF modeling and operation of wind farms. NORCOWE and NOWITECH There are some joint projects, like the “NOWITECH-NORCOWE Workshop on Wind and Wake Modelling”. This is an annual event, alternating between Trondheim and Bergen. Another joint project in 2013 was “Laboratory test of wind turbine foundations (Jackets) against breaking wave loads”.

The tests were performed in the Large Wave Flume (GWK) in Hannover. There is also cooperation on other topics, like wind farm operation. NORCOWE is in many ways complementary to NOWITCH when it comes to the topics addressed by the two centres. Joint scientific projects There is a focus on strengthening the cooperation internally in NORCOWE and with national and international partners by setting up more joint projects. An example is the measurement campaign at Wieringermeer which is a joint project between NORCOWE partners and ECN. NORCOWE Reference Wind Farm (NRWF) is being defined in order to strengthen the cooperation and to help quantifying the cost reduction by applying tools and methods from NORCOWE. NRWF will help to tie together work done in NORCOWE and it may also be a basis for joint projects outside NORCOWE. NORCOWE has also worked closely together with the industry network Arena NOW.

Geir Pedersen (CMR), Benny Svardal (CMR), Peter Eecen (ECN), Valerie Kumer (UiB) and Jan Willem Wagenaar (ECN).

page 46

www.norcowe.no


onal cooperation Project team for new wind turbine standard IEC 61400-6: Wind Turbines: Tower and foundation design NORCOWE partners participate in three IEA wind tasks: IEA Wind Task 31- WakeBench: Benchmarking of wind farm flow models IEA Wind Task 32 - LiDAR: Wind LiDAR Systems for Wind Energy Deployment IEA Wind Task 33 - Reliability Data - Standardization of data collection for wind turbine reliability and operation & maintenance analyses NORCOWE partners are active in other international bodies like EERA Wind, EWEA and in EU project. Kristin Guldbrandsen Frøysa, NORCOWE and John Olav Tande and Hans Christian Bolstad, NOWITECH.

MoU NORCOWE has now MoU with DTU Wind, Fraunhofer IWES, The National Renewable Energy Lab (NREL) in USA, ECN (The Netherlands) and Arena NOW (Norway). The agreement (MoU) with ECN was signed in 2013. International Scientific cooperation The international focus has been strengthened in NORCOWE during 2013. NORCOWE partners have scientific cooperation with a wide range of international research institutions, e.g. University of Strathclyde, DTU Wind, Fraunhofer IWES, NREL, Georgia Institute of Technology (USA), German Wind Energy Institute (DEWI), Universitat Politecnica de Catalunya (UPC) Barcelona, University of Bremen, TU Delft, Lund University, National Centre for Atmospheric Research (NCAR), Hannover Universität, Military University of Technology (Poland), Lodz University of Technology, ECN (The Netherlands), Harbin Institute of Technology (China), Nelson Mandela University (NMU), Port Elizabeth (South Africa), DHI (Denmark), Swinburne University of Technology (Australia) and University of Cincinnati. NORCOWE partners have projects together with DONG, Vattenfall and Vestas. Standardization committees and international bodies Aalborg University and the University of Stavanger take part in three standardization groups: Maintenance group for revision of IEC 61400-1: 2005: Wind turbines - Part 1: Design requirements Maintenance group for revision of IEC 61400-3: 2009: Design requirements for offshore wind turbines www.norcowe.no

Public outreach NORCOWE has two annual scientific meetings. The 2013 spring meeting was held at University of Stavanger May 21st-22nd. NORCOWE Day took place in Bergen September 11th. NORCOWE also organizes conferences which are open to non-NORCOWE members, like Science Meets Industry. Arena NOW and NORCOWE organized Science Meets Industry in Stavanger April 17th and in Bergen September 10th 2013 There will be two Science Meets Industry in 2014: April 2nd in Stavanger and September 9th in Bergen. NORCOWE was well represented at EWEA Offshore in Frankfurt and at the DeepWind conference in Trondheim. There were five issues of the NORCOWE newsletter in 2013. Please send an email to post@norcowe.no to sign up for the newsletter.

Linn Olsen (StormGeo) and Bjarte Sandal (Arena NOW) at the NORCOWE/Arena NOW stand, EWEA Offshore 2013.

page 47


Centre Management Gro Angus Graham Angus Graham leads the Environmental Flow Group at Uni Research in Bergen. He has a degree in physics from the University of Sussex, UK, and a doctorate in applied mathematics from the University of Southampton. He has ten years’ research experience in physical oceanography, gained at Southampton Oceanography Centre, and ten years of experience in meteorology and air quality, gained at Manchester Metropolitan University. At Uni Research he applies his expertise to geophysical boundary layers; air-sea interaction and surface waves; wakes, jets and plumes; remote measurement techniques and image processing; and stochastic simulation. There are two main issues within offshore wind energy that he addresses in his work: The need for necessary and sufficient wind climatology statistics for accurate calculation of energy yields and loads over the lifetime of a prospective wind farm; and the need for identification of optimal site- and routespecific weather windows, weighted on uncertainty levels, for the installation and maintenance of offshore wind turbines. Angus is leading the WP2 on Wind energy estimation.

Joachim Reuder Joachim Reuder is professor in experimental meteorology and Deputy Head of Department at the Geophysical Institute at the University of Bergen. He has more than 20 years of experience in atmospheric boundary layer research, including within the fields of turbulence and turbulent exchange processes, air-sea interaction, atmospheric radiation, and the effects of orography on the wind field and precipitation. In the last years he has been intensively working on the marine

page 48

atmospheric boundary layer from the aspect of offshore wind energy installations. The main focus area in this field is the experimental characterization of boundary layer turbulence in general and its interaction with wind turbines in particular. Both in-situ measurements, e.g. by the buoy mounted and motion corrected NORCOWE eddy covariance system, and ground based remote sensing, e.g. by static and scanning wind LiDAR systems are used. The latest instrumental development is the preparation and characterization of the Remotely Piloted Aircraft System) SUMO (Small Unmanned Meteorological Observer) for direct turbulence measurements in the vicinity of wind turbines. The SUMO can be glimpsed in the foreground of the picture of Joachim. Joachim is leading the WP1 Met-Ocean measurements and database in NORCOWE, managing the national Norwegian offshore wind energy related infrastructure projects EFOWI and OBLO and is also involved internationally in EERA and IEA.

Stian Anfinsen Stian Anfinsen works at Christian Michelsen Research (CMR), at the department CMR Computing. He graduated from the University of Bergen in 1997 with a M.Sc in cryptography, and has worked at CMR since his graduation except for a period of two years. During this period he worked as a project manager for a Norwegian software consultancy firm. At CMR Computing he is department manager, with main responsibility for human resources, quality assurance and the activities relating to Search and Rescue and maritime applications. Within offshore wind his work has been focused on decision support and the development of distributed software solutions. In NORCOWE he has been responsible for the activities relating to setting up and maintaining an online storage solution for NORCOWE’s scientific data, called Metawind (http://.metawind.met.no).

www.norcowe.no


oup Geir Hovland Geir Hovland is a professor in mechatronics at the University of Agder. He has been an active participant in the NORCOWE centre since the beginning when the application to the Research Council of Norway was written and submitted in 2008. He is still the general NORCOWE representative at UiA, being part of the Center Management Group. Geir has been the prime driver behind the establishment of the Norwegian Motion Laboratory, where he has been the technical coordinator since the start in 2010. The first project was started in September 2013, while the official opening took place on the 27th of November the same year (see page 24). His scientific work within NORCOWE has been focused on two topics: condition-based maintenance and marine operations. He will be supervising a new PhD candidate within the area of marine operations for offshore wind energy, with startup in 2014.

Birgitte Rugaard Furevik Birgitte R. Furevik works at the Norwegian Meteorological Institute (Met Norway). She has a MSc in Physical Oceanography from Copenhagen University. She has a PhD from the Nansen Environmental and Remote Sensing Center (NERSC) and the University of Bergen, on Synthetic Aperture Radar for ocean wind applications in the marginal ice zone and for wind energy purposes. She worked at NERSC until 2005 when she started at MET Norway in Bergen.

Geophysical Institute, University of Bergen. Areas of interest are related to wave modelling, –forecasting and –energy, offshore wind energy and applications of radar remote sensing in meteorology. She has been heavily involved in NORCOWE’s summer schools and has been a member of NORCOWE’s scientific committee for many years.

Jasna Bogunović Jakobsen Jasna Bogunović Jakobsen is professor in structural engineering at the University of Stavanger. She holds a PhD in structural engineering from the Norwegian Institute of Technology (now NTNU) and a M.Sc. in structural engineering from the University of Novi Sad, Serbia. Her main research area is structural dynamics, with emphasis on wind loads and wind-induced vibrations of slender structures. The research topics include generation of buffeting loads on bridge box-girders typically used in long-span bridges and methods to capture the wind-structure interaction responsible for bridge flutter, from ambient vibration data. At present, she is involved in studies of aerodynamics of bridge stay cables in the critical Reynolds number range, in full-scale wind and bridge response measurements, and in the evaluation of flow conditions on wind turbine rotor blades, when moving in the along-wind direction. She is principal investigator for the measurement campaign at the Lysefjord bridge (see page 34). Jasna has been heavily involved in the planning and teaching of NORCOWE’s summer school, and is supervisor of PhD student Lene Eliassen on the dynamic analysis of offshore wind turbines.

Birgitte has also a 20% position as associate professor in meteorology and oceanography for renewable energy at the www.norcowe.no

page 49


Finn Gunnar Nielsen Finn Gunnar Nielsen holds a position as senior advisor in Statoil. He has a background in engineering from NTNU, with a PhD within marine hydrodynamics. For more than 30 years has he worked within R&D related to dynamics of offshore structures. He has also had a position as adjunct professor at NTNU, teaching marine operations. At Statoil his main activities are related to R&D within offshore wind. In particular, he has headed the R&D project that lead to the Hywind floating offshore wind concept. In addition to his position at Statoil, Finn Gunnar is adjunct professor at the University of Bergen (UiB) where a master program in energy has been initiated. Within NORCOWE he has supervised master students at UiB, as well as PhD students at the University of Stavanger. He is head of NORCOWE’s scientific committee and has been heavily involved in the planning and teaching of NORCOWE’s summer schools. Finn Gunnar has also participated in several national and international committees related to offshore wind and marine renewable energy more generally.

Thomas Bak Thomas Bak is professor in automatic control at Aalborg University, Denmark. He graduated in 1993 with a M.Sc. in control from Aalborg University, and a PhD in 1998. He has worked at Aalborg University ever since, except for a 3-year period when he worked at a government research laboratory, and a one-year visit to UC Berkeley in 2003.

increase turbulence levels resulting in loads. Initially in NORCOWE the focus was on floating wind turbines. The floating structure, and the hydrodynamics change the behavior of the system and conventional pitch control introduces a negative damping term resulting in larger loads if not handled by the control system design. Thomas Bak has been involved in the NORCOWE summer schools, and is now work package leader of WP3 – Design, installation and operation of offshore wind turbines.

Marwan Khalil Marwan Khalil has been working with computational fluid dynamics and heat transfer as a research engineer for more than 10 years. He is currently a senior research engineer at GexCon AS, Bergen, Norway. GexCon is a worldleading company in the field of safety and risk management and advanced dispersion, explosion and fire modelling. GexCon develops, maintains and uses the industry standard CFD software for modelling gas explosions; FLACS. CMR-Wind is a research version built on top on FLACS that is focused on CFD modelling of wind farms. Marwan has a master’s degree in computational and experimental turbulence from Chalmers University of Technology, Goteborg, Sweden. His research is focused on computational fluid dynamics and heat transfer, turbulence modeling (DNS, LES, RANS), atmospheric flows, wind turbine aerodynamics, and wind farm modeling.

His research is mainly dedicated to improve the use of control methodology within relevant application domains. One important domain is offshore wind energy, with focus on wind farm control. The problem is, however, quite complex. Within large wind farms, turbines interact through their wakes, and page 50

www.norcowe.no


Kristin Guldbrandsen Frøysa Centre Director

Background Kristin Guldbrandsen Frøysa holds a dr. scient. in applied mathematics from the University of Bergen. Her first job was at Norsk Hydro, modeling flow in porous media. Later she worked with fish stock assessment and development of software for fish stock assessment at the Institute of Marine Research and with marketing of instrumentation for current measurement and environmental monitoring. Her last position before joining Christian Michelsen Research (CMR) and NORCOWE was as department manager at Reinertsen. Having worked both as scientist and as a manager in the private sector is very useful in my work as director of NORCOWE, she says. Working at NORCOWE She started as NORCOWE director in January 2010. At that time there was a great optimism with respect to offshore wind energy as a new industrial opportunity for the Norwegian industry. Maritime companies were eager to learn more about offshore wind.

Then the oil prices started to increase and the interest in getting into the offshore wind business decreased. This has also been reflected in NORCOWE, where many of the original user partners have left the consortium. Today she feels there is a good spirit in NORCOWE. Five new partners have recently joined and NORCOWE is internationally recognized as an important player in the offshore wind energy community. Management philosophy My role is to make sure that the NORCOWE partners get added value by being a member. It is important to make sure that we both keep an eye on the long term goals and that we deliver according to the shorter-term plans and time schedules. I think the position as centre director has some similarities with the orchestra conductor, Kristin says. A director doesn’t achieve much alone, but a good director may help the staff to work towards the goals in a coordinated way and make sure that the resources are used in a good way. I am pleased that the Centre Management Group, the Board, the scientific committee and the scientists in NORCOWE all play an active role in development of NORCOWE, Kristin says.

Annette Fagerhaug Stephansen Centre Coordinator

Background Annette F. Stephansen works at CMR as centre coordinator of NORCOWE and as a researcher. She has a MSc in aerospace engineering from PoliMi in Italy, where she specialized in aerodynamics. She initially moved to Italy to attend the United World College of the Adriatic, obtaining an IB, but decided to prolong her stay. She then left for France where she obtained a PhD in applied mathematics from École des Ponts ParisTech. Before coming to CMR, Annette worked at Uni Research, Uni CIPR, as a researcher, and she has also taught mathematics at the University of Bergen. Her current research is on model reduction, and she will be supervising a master student on this topic during the first semester of 2014. You may read more about model reduction on page 18.

www.norcowe.no

Working at NORCOWE Being a Centre Coordinator is an interesting complement to that of the researcher, as one is continuously forced to see the bigger picture, says Annette. I keep in close touch with researchers in neighbouring fields, the industry, the Research Council and organizations like INTPOW (Norwegian Renewable Energy Partners). We also want to inform the public and are for instance involved in a collaboration with the University of Bergen regarding their Centre for Science Education. Here we give a yearly talk informing teachers about the status of offshore wind technology and research. Annette finds that the work as a coordinator is eased by the close collaboration with the centre administration of FME SUCCESS, a FME centre on CO2-storage, which is also hosted by CMR. It is of course a challenge to coordinate so many different partners spread both inside and outside of Norway’s borders, but it is also a great pleasure to meet so many inspiring people engaged and interested in all the different aspects of offshore wind, says Annette.

page 51


Summer school

page 52

www.norcowe.no


NORCOWE’s summer school has now been arranged a total of four times. It is a five-days in depth workshop on offshore wind, open for both industry employees and PhD-students. Non-NORCOWE members are welcome, and all courses are in English, opening up for international participation. In 2013 the official host was the University of Stavanger (UiS), and the summer school took place at Preikestolen Fjellstue, close to the famous Pulpit Rock (Preikestolen in Norwegian).

“A five-days in depth workshop on offshore wind, open for both industry employees and PhD-students“

The aim of the summer school was to give an overview over some of the main challenges connected with planning a wind park. Lecturers from the UiS, the University of Bergen, the Norwegian Meteorological Institute, Statoil and Kjeller Vindteknikk shared their expertise on the following topics: • The offshore wind field • Wind turbines - how they work and important challenges • Offshore wind turbine foundations • Analysis of time series • Planning of wind parks • Response statistics The participants were also given a 2 hours course on how to analyze wind data using Matlab, with an instructor from MathWorks. The airport of Stavanger is Sola, where a NORCOWE measurement campaign took place in 2013 (see page 36). One of the LiDARs from the campaign was actually brought from the airport to Preikestolen Fjellstue and set up just outside the lecture room, so that the students could inspect it and learn more about how it works. A substantial part of the summer school is devoted to group work. This year the exercises were split in two, with one part accompanying each lecture and a second part consisting in using the information to present a convincing case for establishing a wind park. The students could thus get an understanding of the information needed for making the various decisions involved in establishing a wind park. The members of the winning group (elected through voting) were each awarded a Pulpit Rock t-shirt. Half a day was also devoted to the hike up the Pulpit Rock, a cliff of 604 metres which derives its name from having a large plateau at the top that is almost flat. The cliff overlooks Lysefjorden, the fjord where the NORCOWE bridge measurement campaign is taking place (see page 34). NORCOWE’s summer school 2013 counted 27 participants including 8 lecturers. 11 nationalities were present, with the participant from Japan having the longest trip. NORCOWE’s summer school 2014 will be hosted by the University of Agder, and will take place at Strand Hotel Fevik at the south coast of Norway in mid-August. The chosen topic for the workshop is Innovative Methods and Concepts in Offshore Wind Energy. www.norcowe.no

page 53


page 54

www.norcowe.no


Organization Scientific Committee Lead: Finn Gunnar Nielsen, University of Bergen / Statoil

as of January 1st, 2014

General assembly All partners and EB chairman

International Scientific Advisory Committee Lead: Finn Gunnar Nielsen, University of Bergen / Statoil

Executive Board (EB) 9 representatives Chair: Bjørn H. Hjertager, University of Stavanger

Centre Administration Hosted by Christian Michelsen Research AS Centre Director Kristin Guldbrandsen Frøysa Centre Coordinator Annette Fagerhaug Stephansen

Work Package 1 Met-ocean data Lead: Joachim Reuder, University of Bergen Work Package 2 Wind energy estimation Lead: Angus Graham, Uni Research Work Package 3 Design, installation and operation of offshore wind turbines Lead: Thomas Bak, Aalborg University

www.norcowe.no

page 55


Contact info Kristin Guldbrandsen Frøysa, Centre Director Annette F. Stephansen, Centre Coordinator Postal Address FME-NORCOWE Christian Michelsen Research AS P.O. Box 6031 NO-5892 Bergen, Norway Visiting Address Christian Michelsen Research AS Fantoftvegen 38 Bergen, Norway post@norcowe.no Kristin@cmr.no web: www.norcowe.no

page 56

www.norcowe.no

Annual Report NORCOWE 2013  

Annual report, NORCOWE, Offshore wind

Read more
Read more
Similar to
Popular now
Just for you