Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
A METHOD AND A SYSTEM FOR DESIGNING A FOUNDATION FOR A WIND TURBINE
Document Type and Number:
WIPO Patent Application WO/2019/201404
Kind Code:
A1
Abstract:
A method, a computer system, and software for designing a foundation for a wind turbine. To facilitate FEED study, the method comprises the use of predefined wind turbines, load parameters, load scenarios, and load models, for establishing load data sets. The load data sets are used for designing foundations for the wind turbines. The method enables a faster design of wind turbine installations including wind turbine and foundation in a process where load data sets are defined or definable for a number of combinations between predefined wind turbines and predefined load scenarios, by use of pre-defined models. Further the method provides a consistent basis for design of wind turbine installations and the method enables several independent entities to design foundations and make evaluation of load conditions in an identical and consistent way.

Inventors:
PERICLEOUS, Alex (150 Station Avenue, Ewell, Surrey KT19 9UG, KT19 9UG, GB)
SATO, Kenji (Thunøgade 14A ST, 8000 Århus C, 8000, DK)
PEDERSEN, Anders Ravn (Frederiks Plads 26, 3.2, 8000 Århus C, 8000, DK)
HALD, Tue (Drejøvej 8, 8370 Hadsten, 8370, DK)
Application Number:
DK2019/050121
Publication Date:
October 24, 2019
Filing Date:
April 16, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
MHI VESTAS OFFSHORE WIND A/S (Dusager 4, 8200 Aarhus N, 8200, DK)
International Classes:
G06F17/50
Other References:
M. MUSKULUS ET AL: "Reliability-based design of wind turbine support structures", SYMPOSIUM ON RELIABILITY OF ENGINEERING SYSTEM, SRES 2015, HANGZHOU, CHINA, 2015, pages 1 - 13, XP055508293
P. PASSON ET AL: "Load calculation methods for offshore wind turbine foundations", SHIPS AND OFFSHORE STRUCTURES, vol. 9, no. 4, 6 August 2013 (2013-08-06), pages 433 - 449, XP055508296, ISSN: 1744-5302, DOI: 10.1080/17445302.2013.820108
S. SCHAFHIRT ET AL: "Decoupled simulations of offshore wind turbines with reduced rotor loads and aerodynamic damping", WIND ENERGY SCIENCE, vol. 3, no. 1, 21 February 2018 (2018-02-21), pages 25 - 41, XP055508302, DOI: 10.5194/wes-3-25-2018
H. ZHANG ET AL: "A model-driven approach to multidisciplinary collaborative simulation for virtual product development", ADVANCED ENGINEERING INFORMATICS, vol. 24, no. 2, April 2010 (2010-04-01), pages 167 - 179, XP027039318, ISSN: 1474-0346, [retrieved on 20090828]
None
Attorney, Agent or Firm:
VESTAS PATENTS DEPARTMENT (Hedeager 42, 8200 Aarhus N, 8200, DK)
Download PDF:
Claims:
CLAIMS

1. A method for designing a foundation (7) for a wind turbine (1) comprising a tower (2) extending between a base and a top, the method comprising the steps of: a) designing at least one wind turbine (1);

b) defining load parameters for each designed wind turbine, the load

parameters defining the influence of external loads on the structure of the wind turbine, and adding the load parameters to a database of a computer system;

c) defining at least one load scenario, each load scenario defining an expected external load on a wind turbine, the load scenario being added to the database of the computer system;

d) providing for each designed wind turbine, a corresponding load model

configured to predict a response of the wind turbine to load, and provided to take as input a load scenario, and a set of load parameter and to provide as output, a load data set for a wind turbine;

e) using the load model for at least one of the wind turbines to provide a load data set for that wind turbine based on one of the load scenarios; and f) using the load data set to design a foundation (7) for the wind turbine. 2. The method according to claim 1, wherein the load data set comprises:

- a damping matrix defining damping with at least six degrees of freedom,

- a stiffness matrix defining stiffness with at least six degrees of freedom,

- a mass matrix defining mass with at least six degrees of freedom, and

- a force-time-series matrix defining force with at least six degrees of freedom . 3. The method according to claim 2, wherein the load model is configured to provide the matrices with values corresponding to a linearization and thus a simplification of a more complex non-linear load on the wind turbine.

4. The method according to claim 3, wherein the load model is configured to provide the matrices with values corresponding to a linearization of a load on a Rotor-Nacelle-Assembly (RNA) carried by the tower (2).

5. The method according to claim 3 or 4, wherein the load model is configured to provide the values of the linearization of a load being exclusively at the top of the tower (2).

6. The method according to any of the preceding claims, wherein a catalogue is defined, the catalogue comprising a plurality of different load data sets, each load data set corresponding to one of the wind turbines and one of the load scenarios.

7. The method according to claim 6, wherein at least a part of the catalogue is transmitted from the computer system to an external computer system, and the external computer system is used for designing a foundation based on at least one of the combinations of wind turbines, load scenarios, and load data sets contained therein.

8. The method according to any of claims 1-6, wherein the computer system comprises a first computer unit and a second computer unit, the first computer unit being configured for access by a first group of users, the second computer unit being configured for access by a second group of users, the second computer unit configured to allow the second group of users to: g) select from the computer system, a wind turbine from the at least one wind turbine in the database of the computer system;

h) select from the computer system, a load scenario from the at least one load scenario in the database of the computer system;

i) use the computer system to define a load data set for the selected wind

turbine; and

j) use the computer system to design a foundation based on the load data set.

9. The method according to claim 8, wherein the first computer unit and second computer unit is at different locations.

10. The method according to any of the preceding claims, wherein the load parameters define at least one of the following :

• a lift and drag of a turbine blade (6) for a given wind speed,

• an angle of attack of the wind on the blade (6),

• information about the wind turbine controller related to:

o pitching of blades,

o rotor speed,

o limit wind speed for shifts from idling to steady power generation, to ramp down, and shut down.

• information from modelling of system losses related to:

o drive train, generator, or converters,

• a pitching angle of a blade for a certain climatic condition. 11. The method according to any of the preceding claims, wherein the design of the foundation includes the step of defining in the computer system, a computer model of the foundation, providing a finite element model of the foundation, applying the load data to the finite element model, and using a simulation process to evaluate load data in the foundation. 12. The method according to any of the preceding claims, wherein each load scenario defines at least one external parameter describing a feature selected from the group consisting of: a wind speed, a wind direction, a wind turbulence, an air density, and a state of the wind turbine.

13. The method according to any of the preceding claims, comprising the step of making the selected wind turbine, making the designed foundation, and setting up the wind turbine on the foundation.

14. A computer system for designing foundations for wind turbines, the computer system being configured to carry out the method according to claims 1-13.

15. The computer system according to claim 14, wherein the computer system comprises a first computer unit and a second computer unit, the first computer unit being configured for the steps a) to e) of the method according to claim 1, and the second computer unit being configured for step f) of the method of claim 1.

16. A computer program configured for the method according to claim 1-13.

17. The computer program according to claim 16, configured to distribute the load data sets between a first computer unit and a second computer unit.

18. A wind turbine installation comprising a wind turbine and a foundation, and where the wind turbine and the foundation are made in a computer system according to claims 14 or 15.

Description:
A METHOD AND A SYSTEM FOR DESIGNING A FOUNDATION FOR A WIND

TURBINE

INTRODUCTION

The present disclosure relates to a method and a system for designing

foundations for wind turbines based on a design of a wind turbine. Particularly, the disclosure relates to early design of foundations.

BACKGROUND

A wind turbine typically generates electrical power from wind energy.

The main components of a traditional horizontal axis wind turbine include a tower and an energy generating unit on top of the tower. The energy generating unit includes a rotor and a nacelle which houses a drive train connected to the rotor. The drive train includes the generator, and the rotor includes blades driven by wind. The tower is fixed to a foundation which, by definition herein, is not part of the wind turbine. By definition herein, the foundation forms part of a complete wind turbine installation including the foundation and the wind turbine.

Often, the wind turbine is designed by a turbine designer and produced in a factory, i.e. away from the site of the wind turbine installation.

The competition in wind auctions applies pressure on prices, delivery schedule, quality, and safety. In practise, all bidders seek to optimize their projects by e.g. carrying out detailed integrated load analysis and design of the complete wind turbine installation. However, due to the very large complexity combined with a tight schedule, optimization of FEED (front end engineering design) studies is difficult and often neglected in favour of providing project proposals on short notice. This may lead to the use of relatively conservative cost and risk

estimates in the bidding phase and hence reduce the likelihood of providing a winning bid. Further, it may lead to a design which is not optimal for the intended location and load conditions.

SUMMARY

Embodiments of the present disclosure seek to address, at least in part, the need to more efficiently provide a condensed version of a model of a full wind turbine installation including the wind turbine and the foundation for use in integrated load/foundation FEED studies.

According to a first aspect of the present disclosure there is provided a method for designing foundations for wind turbines comprising a tower extending between a base and a top, the method comprising the steps of: a) designing at least one wind turbine;

b) defining load parameters for each designed wind turbine, the load

parameters defining the influence of external loads on the structure of the wind turbine, and adding the load parameters to a database of a computer system;

c) defining at least one load scenario, each load scenario defining an expected external load on a wind turbine, the load scenario being added to the database of the computer system;

d) providing for each designed wind turbine, a corresponding load model

configured to predict a response of the wind turbine to load, and provided to take as input a load scenario, and a set of load parameter and to provide as output, a load data set for a wind turbine;

e) using the load model for at least one of the wind turbines to provide a load data set for that wind turbine based on one of the load scenarios; and f) using the load data set to design a foundation for the wind turbine.

Accordingly, the method enables a faster design of wind turbine installations including a wind turbine and a foundation. Particularly, the method facilitates a process where load data sets are defined or definable for a number of combinations between predefined wind turbines and predefined load scenarios, by use of pre-defined models. Further, the method provides a consistent basis for design of wind turbine installations and the method enables several independent entities to design foundations and make evaluation of load conditions in a consistent way.

This disclosure may be particularly relevant for horizontal axis wind turbines of traditional design, i.e. for wind turbines comprising a tower extending between a base and a tower top, and where the base is attached to the foundation and the tower top is attached to a nacelle. However, it may also be applied for other types of wind turbines, including multiple rotor wind turbines.

This disclosure may be particularly relevant for design of complex foundations, e.g. for offshore wind turbine foundations, such as floating and/or non-floating foundations. However, it may also be applied for other types of wind turbines, including wind turbine installations on land.

The computer system could be constituted by one single computer with data storage capacity and computer capability enabling storage of the load

parameters and the load scenarios and execution of the load model. However, the computer system could also be distributed between two or more computer units. It could be distributed between different locations, and it could be set up for controlled access such that specific individuals may access individual parts of the computer system and individual parts of information contained therein.

The database could, by definition herein, be constituted by one single database, or it could be split into a number of separate databases each containing a subset of the data, e.g. one containing the wind turbine design, one containing the load parameter, one containing the load scenarios, one containing the load models, and one containing the load data sets.

The load parameters define how external load, e.g. caused by wind, may influence the structure of the wind turbine. The load parameters may include information about blade shape, such as chord, twist, thickness, and length, aerofoil properties, such as lift and drag, structural properties of the blade(s) and the nacelle, such as stiffness's and mass distribution, wind turbine controller behaviour, including pitching, yawing, and torque, and system losses, including loss in drive train, generator, convertors, etc.

The above examples of load parameters may thereby form the basis for the load model of the at least one wind turbine. The load parameters may be defined by use of a Blade Element Moment (BEM) model, such as FAST. FAST is a

comprehensive aero-elastic simulator using unsteady BEM theory to model a wind turbine as a collection of rigid and flexible bodies in a spatiotemporal field of turbulent flow.

The load model could be a traditional structural analysis model for analysing a structural system and predicting its responses to load, e.g. with the main objective to determine internal forces, stresses, and deformations under various loads. Such tools form part of existing technology, and they come as off-the- shelf products, e.g. from various makers of CAD systems. Generally, for such models to provide a reliable result, the structure should be modelled as exact as possible, and the loads must be applied as exact as possible. According to the method disclosed herein, this task is carried out in separate steps by firstly defining for the designed wind turbines, corresponding load parameters, and secondly by defining load scenarios, and thirdly by allowing the model to operate on the defined wind turbines and load scenarios. By selecting between the predefined wind turbines and multiple load scenarios, a very fast and precise determination of load data can be carried out without the necessity to access the detailed information. This enables the foundation designer to work on the foundation with relatively exact load data without obtaining detailed information about the wind turbine. Particularly in the early stages of a wind turbine project, e.g. in the bidding phase, this is advantageous.

The design of the foundation may include the step of defining in the computer system, a computer model of the foundation, providing a finite element method (FEM) model of the foundation, applying the load data to the finite element model, and using an FEM load analysis process to evaluate load data in the foundation. The use of a computer model of the foundation with the load data in an FEM model may improve FEED by providing a more accurate evaluation of the foundation and thus reduce a risk profile in the project of designing the wind turbine installation.

In one embodiment, the load data is provided in the form of a:

- damping matrix defining damping with at least six degrees of freedom, - stiffness matrix defining stiffness with at least six degrees of freedom,

- mass matrix defining mass with at least six degrees of freedom, and

- force-time-series matrix defining force with at least six degrees of freedom.

Three of the at least six degrees of freedom relates to displacement, three of the at least six degrees of freedom relates to rotation, and remaining degrees of freedom may relate to individual parameters, e.g. internal modes, e.g. an individual blade response etc.

The exchange of linearized matrices including three displacement degrees of freedom and three rotation degrees of freedom provides a simplified but also relatively exact representation of the load and provides a better starting point for foundation design than the traditionally used principle where load is applied as a vector defining only one or a few degrees of freedom.

The load data set could contain simplified data being a result of a set of complex non-linear data representing loads, particularly loads on the rotor-nacelle- assembly (RNA). I.e. even though the load data for the tower top represents a complex non-linear system, they could be simplified to reduce the level of information provided by the wind turbine designer to the foundation designer. The simplified load data may particularly use a linearization of a load part at least partly defined by aerodynamics, wind turbine control, and/or turbulence, or by a combination thereof. The load model may therefore be configured to provide the above mentioned four matrices with numbers corresponding to a linearization and thus a simplification of a more complex non-linear load.

As mentioned in the introduction, in wind turbine design projects, the wind turbine installation is normally split into two main components, one constituting the wind turbine including the tower, and one constituting the foundation. This split, between the entities of the wind turbine installation is applied inter alia since the designer of the wind turbine and the designer of the foundation are typically different entities. In traditional manufacturing, load evaluation is carried out on the full wind turbine installation by use of an FEM model which includes the foundation, the tower, and the energy generating unit. While this may provide an accurate model and a reduced risk profile in the design project, it requires a large amount of data to be compiled, it requires detailed knowledge of the wind turbine and the foundation, and typically, it requires close and confidential interaction between the foundation designer and the wind turbine designer.

The load model may particularly be configured to provide the numbers of the linearization of a load being exclusively at the top of the tower, e.g. to provide linearized values of a complex load on the RNA. Accordingly, the foundation may be designed based exclusively on the load at the tower top.

By providing this new way of splitting the wind turbine installation, the two major components, for the purpose of design and load evaluation, become one component constituting the foundation including the tower, and one component constituting the power generating unit. This new way of splitting may optionally provide one or more of several advantages: a) it separates the load model advantageously into two separate models, one dedicated to a separate part constituted by the rotor and nacelle assembly, and one separate part dedicated to a part of the tower and foundation; b) it separates the design task, and particularly the design load verification task in a particularly advantageous way; and c) it allows exchange of complex construction details in a standardised and confidential way while reducing the amount of data and the complexity of data exchanged between the aforementioned designer of the foundation and designer of the wind turbine. Regarding a)

The load on this part is influenced by aerodynamics including lift proportional to n L 3 (V is the velocity of the blades of the rotor), or including lift proportional to rotor speed L 2. For the complex non-linear part, load could also be influenced by the control of the wind turbine, particularly the pitching of the blades, the yaw angle, and other factors related to various control strategies and modes of operation, such as for example maximising power production, minimising load, start-up, (emergency) shutdown, production, and idling. The control dependency may involve different control strategies at different points in time, and it may include non-linear control terms. For the complex non-linear part, load could also be influenced by wind turbulence.

For the linear part, load is influenced by structural dynamics from the tower and downwards, including the foundation. This part may cover anything that can be modelled as a "spring, mass damper system".

Regarding b) Since the method provides the load on the top of the tower, the foundation designer may, in a simple way, and by use of a relatively uncomplicated model of an essentially linear system, predict the foundation loads and thus provide a FEED with a higher degree of accuracy and/or speed.

Regarding c) To provide load data on the tower top requires detailed knowledge about the geometry and design details of inter alia the rotor, the nacelle, and the wind turbine control. Such information may be considered confidential by the wind turbine designer and exchange of such data may be undesired. By use of a model which provides load data representing load at the tower top, no exchange of confidential information is necessary, and the foundation designer can obtain a detailed FEED based on load data without gaining detailed information about the rotor, nacelle or control of the wind turbine. As a further advantage, the specific split at the tower top may reduce the requirement for computer power since the computation may become less complicated, and it may reduce the amount of data to be exchanged between two computer systems.

The foundation designer needs access to the load data sets such that the foundation can be designed with the correct loads experienced for a specific wind turbine and for a specific load scenario. To increase the ability to exchange data without exchanging confidential information, the disclosure provides two different series of embodiments.

In a first series of embodiments, the foundation designer may receive a complete catalogue of combinations between different wind turbines, load scenarios, and corresponding load data sets. In this embodiment, in the following referred to as the catalogue embodiment, the catalogue may be in the form of a table with wind turbines, load scenarios and corresponding load data, it may be in any other pre-defined format.

In one example, the foundation designer may access the catalogue or selected parts of the catalogue on a computer owned and controlled by the wind turbine designer, and in another embodiment, the foundation designer simply receives the entire catalogue to be uploaded to his own computer for design of the foundation.

In a second series of embodiments, the computer system comprises a first computer unit and a second computer unit, the first computer unit being configured for access by a first group of users, the second computer unit being configured for access by a second group of users.

Access to the wind turbine designs may be restricted to the first group of users. The method may include defining a second group of users, e.g. users being external relative to the wind turbine designer. For this group of users, access to the wind turbine design, and to the model could be restricted such that they are only allowed to execute the model with load parameters of a selected wind turbine and with a selected load scenario. The foundation designer may not even be allowed to see the load model, the load parameters and the load scenarios but may only be allowed to select them.

The second computer unit could be configured to allow the second group of users to: g) select from the computer system, a wind turbine from the at least one wind turbine in the database of the computer system; h) select from the computer system, a load scenario from the at least one load scenario in the database of the computer system; i) use the computer system to define a load data set for the selected wind turbine; and j) use the computer system to design a foundation based on the load data set.

The first computer unit may be owned and controlled by the wind turbine designer and the second computer unit may be owned and controlled by the foundation designer. Accordingly, they could be at different locations.

In one embodiment, steps a)-d) are carried out in the first computer unit and steps e)-f) are carried out in the second computer unit.

In another embodiment, steps a)-e) are carried out in the first computer unit and step f) is carried out in the second computer unit.

The load parameters may define at least one of the following :

• a drag and lift of a turbine blade for a given wind speed,

· an angle of attack of the wind on the blade, • information about the wind turbine controller related to:

o pitching of blades,

o rotor speed,

o limit wind speed for shifts from idling to steady power generation, to ramp down, and shut down.

• information from modelling of system losses related to:

o drive train, generator, or converters,

• pitching angle of a blade for a certain climatic condition etc.

The design of the foundation may include the step of defining a computer model of the foundation, providing a finite element model of the foundation, applying the load data to the finite element model, and using a simulation process to evaluate load data in the foundation. The process of designing the foundation may particularly take place in a computer system, e.g. in the same computer system in which the load data are established by use of the load model. By "same computer system" is herein meant that the computer unit used for designing the foundation is in data communication with the computer unit used for establishing the load data, e.g. in real-time data communication such that data, particularly the load data sets, can be exchanged during the design process.

The computer model of the foundation may include a computer model of the tower. In this way, an FEM model of the foundation including the tower can be defined, and the load data at the top of the tower can be used on the FEM model. In this way, an FEM load analysis process can be applied in the

evaluation of load data in the foundation.

Each load scenario may define at least one external parameter selected from the group consisting of: wind speed, wind direction, wind turbulence, air density, and a state of the wind turbine. These parameters are indicative for the load on the power generating unit above the tower top and thus the load on the tower top. The state of the wind turbine may e.g. determine a criterion for initiating power production with the wind turbine or other process parameters

incorporated in the wind turbine control. If the load data relates to the tower top, other parameters, for example related to wave height etc. could be excluded from the load scenario. This provides an additional advantage and simplifies the process further.

Finally, the method may include the step or steps of making at least a part of the designed foundation, or making the complete foundation and/or the wind turbine and assembling the wind turbine and foundation, e.g. at the site of construction, e.g. offshore. It should be understood, that the turbine and the foundation may also be assembly away from the site and subsequently transported to the site. The term "foundation" may for example cover a monopole, a mono-bucket, a piled jacket, a suction bucket jacket, a tripod, a tri-pile, a gravity based foundation, and floater. Particularly, the disclosure may relate to a non-floating foundation, and other types of foundation may also be considered.

In a second aspect, the disclosure provides a computer system for designing foundations for wind turbines, the computer system being configured to carry out the method according to the first aspect of the invention and may

particularly comprise a first computer unit and a second computer unit, the first computer unit being configured for the steps a) to e) of the method of the first aspect, and the second computer unit being configured for step f) of the method of the first aspect. Alternatively, the first computer unit could be configured for the steps a) to d) of the method of the first aspect, and the second computer unit could be configured for step e) and f) of the method of the first aspect.

In a third aspect, the disclosure provides a computer program for instructing a computer system to perform the method of the first aspect. The program may particularly be configured to distribute the load data sets between a first computer unit and a second computer unit.

In a fourth aspect, the disclosure provides a wind turbine installation comprising a wind turbine and a foundation, and where the wind turbine and the foundation are designed in a computer system according to the second aspect. In a further aspect, this disclosure provides a method for designing foundations for wind turbines comprising a tower extending between a base and a top, and a power generating unit attached at the top, the method comprising the steps of: a) designing at least one wind turbine; b) defining load parameters for each designed wind turbine, and adding the load parameters to a database of a computer system; c) defining at least one load scenario, each load scenario defining an expected external load on a wind turbine, the load scenario being added to the database of the computer system; d) providing for each designed wind turbine, a corresponding load model defining load data based on the load parameters of the wind turbine and a load scenario; wherein the load data provided by the load model represents a load condition at the tower top, particularly, it may represent a load condition exclusively at the tower top.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described in further details with reference to the drawing, in which :

Fig. 1 illustrates a wind turbine, Fig. 2 schematically illustrates a wind turbine installation and a separation between different parts of the wind turbine installation where different models apply,

Fig. 3 illustrates an example of Load Time Series,

Fig. 4 schematically illustrates a distributed computer system, Fig. 5 illustrates a GUI for selection of a wind turbine, and

Fig. 6 illustrates a GUI for selection of a tower design.

DETAILED DESCRIPTION OF THE DRAWINGS

It should be understood that the detailed description and specific examples, while indicating embodiments, are given by way of illustration only, since various changes and modifications within the spirit and scope of the disclosure will become apparent to those skilled in the art from this detailed description.

Fig. 1 illustrates a wind turbine 1. The wind turbine 1 comprises a tower 2 having a number of tower sections and an energy generating unit. The energy generating unit includes a nacelle 3 positioned on top of the tower 2 and a rotor 4 rotatably carried by the nacelle 3. In the following, the nacelle and rotor are referred to as rotor-nacelle-assembly (RNA). The nacelle 3 houses the drive train (not shown) connected to the rotor 4. The drive train includes the generator (also not shown). The tower 2 is fixed to a foundation 7. The nacelle could be on land or off-shore, e.g. on or in the seabed.

The rotor 4 is rotatable with respect to the nacelle 3, and comprises a hub 5 and three blades 6. Wind incident on the blades 6 causes the rotor 4 to rotate with respect to the nacelle 3. The mechanical energy from the rotation of the rotor 4 is converted into electrical energy by the generator in the nacelle 3. The generated electrical energy can be supplied to an electrical grid or to a local community.

Fig. 2 schematically illustrates that the wind turbine installation may be split into different parts which are treated differently. The tower 2 and the foundation 7 may be modelled in a simple way, whereas the RNA may be modelled in a more detailed and complex way. Detailed load parameters may define how external loads on the RNA influence the structure, and detailed load scenarios may define expected external loads on the wind turbine. Correspondingly, a detailed and complex load model may be defined for the RNA. By use of the complex load model, the load parameters and the load scenarios, load data sets can be defined. The load data set is defined as a set of four linearization matrices, each representing at least six degrees of freedom. These four matrices are herein referred to as Mass, Damping, Stiffness, and Load Time Series.

The designer of the wind turbine can distribute the linearization matrices to external foundation designers without disclosing confidential details of the wind turbine design, and the foundation designer can design the foundation based on linearized matrices derived from detailed load parameters, load scenarios, and a complex load model of the RNA. The foundation designer thereby obtains a good basis for correct foundation design without accessing detailed wind turbine design parameters. The spilt between the complex wind turbine model and the simpler tower model and foundation model is illustrated in Fig. 2 by the vertical line 8.

The equation for the loading at the nacelle 3 can be approximated as:

P hase nacelle = P e x ternal + Mass x Acceleration + Stiffness x Displacement

+ Damping x Velocity The 1 st term F externcU represents the loading which would have occurred if the RNA 3 was fixed in space which is relatively invariant with RNA motion. While the following terms represent the loading seen at the base of the RNA due to the motions of the RNA 3, these terms are separated because the motion of the RNA will vary substantially depending on the foundation stiffness and wave loading on the RNA.

The 1 st term representing the external forces is calculate by fixing the RNA in space in the load model, for example by disabling the tower 2 degrees of freedom, then extracting the three moments and forces seen at the base of the RNA 3. An example of a Load Time Series is shown in Fig. 3.

The following terms are then calculated by linearizing the RNA, one simple method of doing this is described below but many other methods exist including : Jacobian linearization, Carleman linearization, Lie Series, iteration methods and feedback linearization. One suggested method includes:

• Apply an external motion to the RNA 3 within load model, with the wind turbine 1 in the target operating state, with constant external wind conditions

· Measure the loading at the base of the RNA 3 which has been caused by this imposed motion

• Fit the terms, M, K and D of the equation :

F = M x Acceleration + K x Displacement + D x Velocity to the measured load levels. As the RNA 3 has at least six degrees of freedom, displacement in three directions and rotations in three directions, these are each a 6 x 6 matrix, this approach can also be extended to include additional internal degrees of freedom of the RNA by perturbing the internal modes and measuring the resulting loading at the base of the RNA, i.e. at the tower top. Example Linearization matrices are shown below, for a 6 degree of freedom set up:

It should be observed that the numbers in these matrices correspond to the linearization of one specific RNA. Fig. 4 illustrates a distributed computer system 9. The computer system comprises a first computer unit 10 and a second computer unit 11. The computer units 10, 11 are located in different locations, for example one at the wind turbine designer, and one at the foundation designer. The computer units communicate over the internet, and thereby constitute networked computers, which share the same purpose of designing a complete wind turbine installation 1 by facilitating the wind turbine design by use of the first computer unit and the foundation design by use of the second computer unit. The computer system 9 may be configured e.g. for "concurrent computing", "parallel computing", and/or "distributed computing". If the computer system 9 is configured for parallel computing, the first computer unit 10 may have a processor having access to computer memory, and the second computer unit 11 may have a processor having access to the same computer memory, i.e. the memory forms a shared memory, and information may be exchanged swiftly and directly via the shared memory. This may allow a higher degree of volatility between the two units, and thereby allow the foundations to be designed based on real time updated computer information from the first computer unit.

If the computer system 9 is configured for distributed computing, the first computer unit 10 and the second computer unit 11 may each have a private memory and share information online or offline, e.g. over the internet. This may allow a higher degree of privacy and independence between the two units, and thereby allow the foundations 7 to be designed more independently on the designer of the wind turbine by exchange of the linearized matrices. In both scenarios, i.e. in distributed or parallel computing, each computer unit may use their own private load model, e.g. in the form of a simulation system, or they may share the same load model.

However, in one aspect of the invention, the load model at least of the wind turbine is not shared with the designer of the foundation. On the contrary, the designer of the foundation can execute the load model of the wind turbine without getting access to the model as such, or the foundation designer may receive a catalogue of pre-defined load data sets.

Subsequently, the foundation designer can use the linearized matrices when simulating loads on the foundation. The arrows 12 illustrate sharing and/or exchange of information.

By way of example, the method for designing a foundation includes a number of steps taking place in the computer system. a) A plurality of different wind turbines 1 is designed and detailed information related to the RNA, e.g. related to the blade shape, such as chord, twist, thickness, and length, aerofoil properties, such as lift and drag, structural properties of the blade(s) 6 and the nacelle 3, such as stiffness's and mass distribution is defined.

In a second step b), a plurality of load parameters for each designed wind turbine 1 is defined and added to a database of a computer system 9. These parameters define which influence load on the wind turbine, particularly on the tower top has on the structure of the wind turbine. An example of such a load parameter could be the drag of a turbine blade 6 for a given wind speed, and angle of attack of the wind on the blade 6. Additionally, the load parameters may include information about the wind turbine controller, such as pitching of blades, particularly at which wind speeds the wind turbine 1 shifts from idling, to steady power generation, to ramp down, and shut down. Additionally, load parameters could be information from modelling of system losses, including drive train, generator, converters. Another example of such load parameter could be the pitching angle of a blade for a certain climatic condition. In a third step c), a plurality of different load scenarios is defined. The load scenarios may e.g. relate to different locations and each load scenario defines a set of external loads which could be expected on the wind turbine 1. The load scenario further defines a scenario specific RNA linearization. An example of such a linearization could be EXPLAIN. In one example, the load scenarios are sorted based on location, and in another example, the load scenarios are sorted based on type of operation situation. By means of an example, the loads could be sorted in the following situations: gust, high wind condition, low wind condition, idling operation of the wind turbine, start-up operation, and shutdown operation of the wind turbine. The parameters could e.g. include wind turbulence intensity which could be defined as a number in percentage points which defines the amount of turbulence in the wind. It is a measure of the variation in the wind speed.

Typical values are between 0 and 1 (where 1 is 100%)

The parameters could also include mean wind speed. This can either be defined as a number, or as a relative amount to the rated wind speed of the turbine. The following are examples of how this can be formatted assuming a rated wind speed and a deviation from the rated wind speed, e.g. if the rated wind speed is 12, then Vr=12. Then the load scenario could be parametrized by specifying a deviation from the rated wind speed, e.g. Vr+5 = 17 or Vr-3 = 9. The parameters could also include the direction the wind is coming from, e.g. expressed as a value in degrees.

The parameters could also include the density of the air in Kg/m 3 , e.g. 1.225 kg/m 3 . Table 1 below illustrates an example of a set of parameters constituting different load scenarios, where DLC is a specific Design Load Case (scenario).

In a subsequent step d) a load model is defined. The load model takes the parameters of the wind turbine and the load scenario as input and provides a set of load data in the form of four linearized matrices as an output. The load model could be implanted e.g. in Blade Element Moment (BEM) model, such as FAST. FAST is a comprehensive aero-elastic simulator using unsteady BEM theory to model a wind turbine as a collection of rigid and flexible bodies in a spatiotemporal field of a turbulent flow. Accordingly, such models are made in accordance with standard procedures existing in the art. In relation to the present disclosure, it is typical to define one model for each of the wind turbines in the library.

The above steps are carried out in the computer system, e.g. in a first computer unit 9 of the system, and typically a computer unit owned and controlled by the wind turbine designer.

Once these steps are carried out, a number of steps can be carried out in the computer of the wind turbine designer or in a computer of the foundation designer.

In accordance with the first series of embodiments, different combinations between wind turbines and load scenarios may be established and by use of the load model, a set of load data could be made for each combination.

Subsequently, the load data sets are stored in a catalogue and provided to the foundation designer, e.g. in a format suitable for being entered directly into an external computer system of the foundation designer for use in further study.

In accordance with the second series of embodiments, the computer system is a distributed system with a first and a second computer unit. The first computer unit is used for that part of the process which takes place at the wind turbine designer, and the second computer unit being used for that part of the process which takes place at the foundation designer. In this system, the foundation designer may select one of the wind turbines 1 from the database in the first computer unit. The user may not gain access to the various values and parameters but merely select one of the turbines. This is illustrated in Figs. 5 and 6, where Fig. 5 illustrates a GUI 13 for selection of a wind turbine 1. The wind turbine 1 is selected in a pull-down menu 14 containing a plurality of wind turbines. Fig. 6 illustrates a GUI 15 for selection of design parameters for the tower 2 of the selected wind turbine 1. In a second step of the second computer unit 10, the user may select one of the load scenarios predefined in the first computer system. Again, the user may or may not gain access to the detailed information in the load scenario. In one embodiment, the user will only know the rough conditions, e.g. a location to which the scenario belongs, and in another embodiment, the user may select detailed elements from the load scenario, e.g. which wind condition the selected turbine 1 is exposed to.

In a third step of the second computer unit, the user may execute the load model, e.g. without getting access to the load model as such. Subsequently, the user may obtain the load data related to the tower top. The load data is received from the first computer unit in the form of four linearized matrices.