WP3: Reduction of uncertainties by integral substructure design
WP3: Reduction of uncertainties by integral substructure design
To make offshore wind cost effective there is a need to reduce the cost of energy. The aim of this work package is to investigate where cost reductions in the support structure are possible while keeping a sound and safe design. Probabilistic design methods (structural reliability methods) are used to study whether there is any conservatism in the design of support structures.
At this moment the design of support structures for offshore wind turbines is done in collaboration between the wind turbine manufacture and a specialized engineering company like Ballast Nedam or Ramboll. The support structure consists of a foundation, a sub-structure and a tower. The tower is designed by the wind turbine manufacture, while the foundation and sub-structure are designed by the engineering company. The structural design is verified by applying partial safety factors as prescribed in the standards of Germanische Lloyd or DNV. Partial safety factors are applied to compensate for uncertainties in the design process, and improved knowledge may help to reduce these factors and hence contribute to reduction in COE without compromising the reliability of the design.
An alternative of applying (possibly conservative) partial safety factors probabilistic design methods will be applied. In the past probabilistic design methods were already used for wind turbines. For instance in the JOULE-III project PRODETO (PRObabilistic DEsign TOol) the PhD-thesis of Veldkamp and the work by Soerensen. The focus in this work package will be on the support structure while taking the complete offshore wind turbine system in to account.
The application of probabilistic design methods should give insight in the following:
- The safety level of an offshore wind turbine based on partial safety factors as prescribed in the standard;
- The safety level in case the life time of 20 years is increased;
- Identification of the importance of the different load parameters to the safety level.
The work is organised along the following tasks:
Task 3.1: Definition data base set up and selection of probabilistic methods and offshore wind turbines
To investigate where cost reductions are possible in the support structure the loads and response (time) histories of an offshore wind turbine are needed. A representative state of the art offshore wind turbine will be selected and the loads and time histories of that turbine will be stored in a data base. The definition of the data base set up should relate to the selected probabilistic method. Before continuing with the next tasks the selected probabilistic method and the data base set up are presented to the advisory board.
Task 3.2: Creation of data base
For the selected offshore wind turbine a data base is created by first modelling the wind turbine in Focus6 and then performing load calculations using the available aeroelastic tool PHATAS. The quality of the data will be checked.
Task 3.3: Implementation of probabilistic methods
Before a probabilistic analysis can be performed the data in the data base needs to be processed by the probabilistic software tool selected.
Task 3.4: Probabilistic analysis
Using the available data base and probabilistic tools the safety level of the offshore wind turbines will be analysed with respect to life time and possibly important cost reducing load parameters for the support structure design will be identified.
Task 3.5: Re-assessment of safety factors
Based on the probabilistic analysis a re-assessment of the safety factors is performed and compared with the factors as prescribed in the standards.
Task 3.6: Assessment of industry questions
During the project questions from industry related to the (probabilistic) design of support structure can come forward. In this task these questions can be assessed.
Software that is used in this work package:
- Offshore wind turbine design tool (FOCUS/PHATAS);
- Probabilistic tool that is to be determined. Choices are Proban, Strurel or open source codes like Ferum.
Return to Work Packages