Coordinateurs du projet
Context
France plans to generate 2,400 MW from offshore wind turbines by 2023.
There is a strong need for tools capable of optimizing the position of turbines in a farm: precise definition of the load to optimize the turbine structure; reduction of maintenance costs; better forecasting of energy production.
In the EOS project, we propose an alternative approach based on the development of a direct CFD model, combined with more simplified formulations to reduce calculation costs.
Analyzing the dynamics around an offshore wind turbine (floating or otherwise) is a challenge that involves studying the interactions between its various components, such as the blades, rotor, float, and control systems. This requires multiphysics modeling including aerodynamics, hydrodynamics, and structural mechanics. In addition, extending these studies to wind farms will also need to take into account the interactions between multiple wind turbines, and in particular wake effects. Changes in velocity (in air and water) and turbulence levels affect the power generated by these farms and can increase mechanical loads on wind turbine structures located downstream.
Today, the design tools used for wind farm design and wind turbine development are imperfect and rely primarily on empirical formulations of turbulence and wake interactions. In this context, the EOS project aims to develop an offshore wind farm simulator based on high-fidelity simulation of aerodynamic and hydrodynamic forces.
Access to high-performance computing will be necessary to accurately simulate the fluid (air)-fluid (water)-structure interactions that occur within an offshore wind farm. This type of simulation will be used to understand the complex physical phenomena involved, to feed or recalibrate simplified models used in engineering, or to calculate certain critical points for dimensioning (under extreme conditions, for example).
Scientific breakthroughs and innovation
Currently, simplified models based on empirical formulations for taking into account wake effects and changes in turbulence levels are used by design engineers. The EOS project aims to achieve the following advances:
- Development of a direct model without simplifying assumptions.
- A massively parallel and highly scalable multiphase CFD code associated with existing computing resources in the region.
- Construction of a simulator that can be edited, distributed, and promoted by a regional SME specializing in software for marine activities.
The main scientific advances concern the implementation of a comprehensive CFD approach: no simplification, no assumptions, and accurate interactions (such as wakes) implicitly taken into account.
In order to reduce computing costs, techniques such as automatic adaptation of anisotropic meshes and massive parallel computing are being implemented. The project also involves tackling difficult subjects such as: the development of a coupled aero-hydro solver in an FSI (Fluid-Structure Interaction: rigid body floating and in motion) context; the study of water/air/wind turbine interactions; the introduction of far-field conditions and boundary conditions in the near-field domain.
In terms of innovation, a massive parallel multiphase CFD code, highly scalable and linked to supercomputer infrastructure in the Pays de la Loire region, is being developed as part of the EOS project.
Expected technical and economic impact
The expected impacts of the EOS project are:
- Advances in understanding the behavior of floating wind turbines within a farm in relation to their environment.
- Developing unique expertise in accurate 3D simulation of floating offshore wind turbines.
Potential value:
- or by a software solutions publisher.
- by engineering firms for design services: for example, extreme calculations for floating wind turbines (whose conditions will have been defined by conventional codes that are faster but less accurate), or to provide input data for these same parametric calculations.
Demonstrator
Simulation platform for offshore wind farm dynamics.
The simulator will be based on the ICI-tech library, developed by the High Performance Computing Institute (ICI-ECN) at École Centrale de Nantes and dedicated to high-performance scientific computing. The implementation of the EOS method for simulating one or more floating wind turbines and the final validation of the simulator, in comparison with experimental results or other models in the literature, but also thanks to participation in the IAEWind project, will be carried out in collaboration with the Marine & Ocean Energy team at the Laboratory of Hydrodynamics, Energy and Atmospheric Environment Laboratory (LHEEA-ECN).
Results
- First CFD simulations with the existing solver.
- Acceleration of phase function calculations to represent different simulation domains.
- Parallel meshing of models with up to 100 wind turbines.
- Development of a wind and FSI simulator. Study of main limitations.
- Development of a wave simulator and HOS coupling. Study of main limitations.
- Start of software prototype development.
First CFD simulations with the existing solver
Acceleration of computation of the phase functions to represent different domains in the simulation
Parallel meshing of models up to 100 wind turbines
Software prototype development
Wave simulator and HOS coupling
Wind simulator and FSI
Perspectives
Simulation of fluid/deformable structures; floating bodies (in progress); more accurate treatment of wake conditions and boundaries.
Use of Artificial Intelligence (AI) and ROM (Reduced Order Modeling) techniques to speed up calculations.
Industrial applications
This EOS simulation platform can:
be edited, distributed, and promoted by a regional SME specializing in marine activities.
also be used by engineering firms for the design of wind turbines and wind farms.