Skip to main content

About this book

The book presents novel Computational Fluid Dynamics (CFD) techniques to compute offshore wind and tidal applications. The papers in this volume are based on a mini-symposium held at ECCOMAS 2018. Computational fluid dynamics (CFD) techniques are regarded as the main design tool to explore the new engineering challenges presented by offshore wind and tidal turbines for energy generation.The difficulty and costs of undertaking experimental tests in offshore environments have increased the interest in CFD which is used to design appropriate turbines and blades, understand fluid flow physical phenomena associated with offshore environments, predict power production or characterise offshore environments amongst other topics.

Table of Contents


Simple Models for Cross Flow Turbines

Using a high order discontinuous Galerkin numerical method with sliding meshes, we simulate one, two and three bladed cross-flow turbines to extract statistics of the generated wakes (time averaged velocities and Reynolds stresses). Subsequently, we compare the wakes resulting from simple models (a circular cylinder and an actuator disc) to the time averaged cross-flow turbine wakes. Additionally, we provide results for a reduced order model based on dynamic mode decomposition (Le Clainche and Ferrer, Energies, 11(3), 2018, [1]). Whilst simplified models find difficulties in capturing wake asymmetries characteristic of cross-flow turbines, our proposed reduced order model captures mean values and Reynolds stresses with good accuracy, showing the potential of the last technique to speed up the simulation of cross-flow turbine statistics.
Esteban Ferrer, Soledad Le Clainche

Suppressing Vortex Induced Vibrations of Wind Turbine Blades with Flaps

The present work describes an exploratory work aiming to analyze the impact of trailing edge flaps activation on Vortex Induced Vibrations (VIV) suppression. A computational study of the VIV of the AVATAR rotor blade, a 10 MW design suitable for offshore locations, was performed. A Fluid Structure Interaction (FSI) approach was adopted for the simulations, coupling an Improved Delayed Detached Eddy Simulations (IDDES) flow solver with a beam-based structural model. Initial simulations based on the clean geometry identified significant edgewise VIV for certain free stream velocity and flow inclination angles. Inflow conditions showing the maximum amplitude of blade vibrations were used in order to test several trailing edge flap geometries and operating angles. The best flap configuration found in this parametric study managed to suppress the VIV phenomenon. However, when assessing a wider range of inflow conditions, the amplitudes of vibration of the blade equipped with flaps were found to be equivalent to the ones obtained for its clean counterpart. It is therefore concluded that a re-calibration of the flap operating angle should be required in order to adapt it to the considered wind speed and wind direction.
Sergio González Horcas, Mads Holst Aagaard Madsen, Niels Nørmark Sørensen, Frederik Zahle

Prediction of the Wake Behind a Horizontal Axis Tidal Turbine Using a LES-ALM

A large-eddy simulation-actuator line method (LES-ALM) applied to a single horizontal axis tidal turbine is presented and validated against experimental data. At a reasonable computational cost, the LES-ALM is capable of capturing the complex wake dynamics, such as tip vortices, despite not explicitly resolving the turbine’s geometry. The LES-ALM is employed to replicate the wake behind a laboratory-scale horizontal axis turbine and achieves a reasonably good agreement with measured data in terms of streamwise velocities and turbulence intensity. The turbine is simulated at six tip speed ratios in order to investigate the rate of decay of velocity deficit and turbulent kinetic energy. In the far-wake, these quantities follow a similar decay rate as proposed in the literature with a −3/4 slope. For cases when the turbine spins at or above the optimal tip speed ratio, the levels of turbulent kinetic energy and wake deficit in the far-wake are found to converge to similar values which seem to be linearly correlated. Finally, transverse velocity profiles from the simulations agree well with those from an analytical model suggesting that the LES-ALM is well-suited for the simulation of the wake of tidal stream turbines.
Pablo Ouro, Magnus Harrold, Luis Ramirez, Thorsten Stoesser

Harmonic Balance Navier–Stokes Analysis of Tidal Stream Turbine Wave Loads

ARCTIC, a novel incompressible Reynolds–averaged Navier–Stokes finite volume code for the hydrodynamic analysis of open rotor unsteady loads is presented. One of its unique features is a harmonic balance solver enabling high–fidelity analyses of turbine periodic hydrodynamic loads with runtimes reduced by more than one order of magnitude over conventional time–domain CFD, and with negligible accuracy penalty. The strength of the new technology is demonstrated by analyzing with both harmonic balance and time–domain solvers the load fluctuations of a realistic tidal stream turbine. Such fluctuations are caused by a harmonic perturbation of the freestream velocity similar to that due to surface gravity waves.
A. Cavazzini, M. S. Campobasso, M. Marconcini, R. Pacciani, A. Arnone

Analysis of the Aerodynamic Loads on a Wind Turbine in Off-Design Conditions

In this work, the aerodynamic loads acting on a large horizontal axis wind turbine are analysed in off-design conditions by means of computational fluid dynamics (CFD) simulations. The turbulent wind flow is solved using an unsteady RANS approach and choosing the \(k-\epsilon \) model. Appropriate boundary conditions are used in combination with modified wall functions in order to preserve the atmoshperic boundary layer (ABL) profiles throughout the entire domain. An overset technique is used to handle the rotation of the blades throughout the simulated time. Changing both the pitch angle of the blades and the tip-speed ratio (TSR) of the turbine, several operating points are investigated. The performance and the loads are highly affected by the ABL, whose effect is highlighted. The performance of the wind turbine in each simulated operating point is compared to the nominal operating point (NOP). The aerodynamic loads are monitored, analysed and mutually compared throughout the motion of the rotor, in order to identify the most critical conditions for the blade structures.
G. Santo, M. Peeters, W. Van Paepegem, J. Degroote

An Algorithm for the Generation of Biofouled Surfaces for Applications in Marine Hydrodynamics

The adverse effects of marine biofouling on marine renewable energy devices are well established. In recent fundamental investigations on fluid flow over this type of surface roughness, marine biofouling has mainly been realized as ordered arrangements of roughness elements. These surfaces cannot be compared to realistic biofouled surfaces which show an irregular distribution of roughness features. In this work, a geometric algorithm for generating realistic surface roughness due to barnacle settlement is presented. The algorithm mimics the settlement behaviour of barnacles and allows the generation of a range of fouling states from very sparse rough surfaces to surfaces that are fully covered by barnacle colonies. The generated surfaces can be used in various applications, e.g. in CFD simulations to establish the fluid dynamic roughness effect of different fouling states or as 3D printed surface tiles for use in wind-tunnel and towing tank experiments.
Sotirios Sarakinos, Angela Busse

A Higher-Order Chimera Method Based on Moving Least Squares

The Chimera/overset approach is widely used in the numerical simulation of flows involving moving bodies. In this approach, first used by Steger et al. in 1983, the domain is subdivided into a set of overlapping grids, which provide flexible grid adaptation, the ability to handle complex geometries and the relative motion of bodies in dynamic simulations. However, most of current methods present a second order convergence at most, due to the interpolation between overlapped grids. In this work a higher-order (>2) accurate finite volume method for the resolution of the Euler/Navier–Stokes equations on Chimera grids is presented. The formulation is based on the use of Moving Least Squares (MLS) approximations for transmission of information between the overlapped grids. The accuracy and performance of the proposed method is demonstrated by solving different benchmark problems.
Luis Ramírez, Xesús Nogueira, Pablo Ouro, Fermín Navarrina, Sofiane Khelladi, Ignasi Colominas

A Review on Two Methods to Detect Spatio-Temporal Patterns in Wind Turbines

This Chapter presents a review on two methods for the analysis of flow structures in wind turbines. These methods are higher order dynamic mode decomposition and spatio-temporal Koopman decomposition, which are highly efficient tools suitable for the detection of spatio-temporal patterns in complex flows. These two techniques have been applied to detect the main flow structures in a cross-flow wind turbine in turbulent regime, and in an horizontal wind turbine, which is laminar in the near field but transitioning to turbulence in the far field. Using these methods, a reduced number of traveling waves which are responsible for triggering the flow transition, are able to describe the aforementioned complex flows.
Soledad Le Clainche, José M. Vega, Xuerui Mao, Esteban Ferrer

Towards Numerical Simulation of Offshore Wind Turbines Using Anisotropic Mesh Adaptation

In the context of reducing the cost of floating wind energy, predicting precisely the loads applied on structures and their response is essential. As the simulation of floating wind turbines requires the representation of both complex geometries and phenomena, several techniques have been developed. The wake generated by the aerodynamic loads experienced and the tower can be modeled using methodologies inherited from onshore wind simulation, and coupled with a hydrodynamic codes that were most of the time developed for the oil and gas industry. This work proposes a methodology for the simulation of a single or several turbines with an exact representation of the geometries involved, targeting an accurate evaluation of loads. The software library used is ICI-tech, developed at the High Performance Computing Institute (ICI) of Centrale Nantes. A single computational mesh is used, where every phase is defined through level-set functions. The Navier–Stokes (NS) equations are solved in the Variational MultiScale (VMS) formalism using finite element discretization and a monolithic approach. A coupling with an automatic and anisotropic adaptation procedure guarantees the good representation of the geometries immersed. The adaptation allows the simulation of phenomena with very different orders of magnitude, e.g. aerodynamics around blades and waves propagation. The reduction of the number of points in the mesh and the massive parallelization of the code are also necessary for wind turbine simulation.
L. Douteau, L. Silva, H. Digonnet, T. Coupez, D. Le Touzé, J.-C. Gilloteaux

Numerical Modelling of a Savonius Wind Turbine Using the URANS Turbulence Modelling Approach

This work presents a three-dimensional investigation of the performance prediction of the operation of the vertical axis wind turbine. The analysis was carried out for the micro-turbine equipped with the Savonius rotor. The applied methodology was based on the Computational Fluid Dynamics (CFD) and used the Finite Volume method to solve the unsteady Reynolds Averaged Navier–Stokes equations. We concentrated our investigations on the influence of the turbulence modelling methodology on the simulation results. The analysis considers most of the URANS turbulence models, starting with the commonly used two equation models like \(k-\epsilon \) or \(k-\omega \) SST to the Reynolds Stress Models with quadratic pressure strain modelling. The results show the influence of turbulence models on the results of the predicted flow field and wind turbine performance.
Tomasz Krysinski, Zbigniew Bulinski, Andrzej J. Nowak

The Standard and Counter-Rotating VAWT Performances with LES

Traditionally, the wind turbine performance is defined in terms of power extraction performance (expressed non-dimensionally as power coefficient, CP, with its maximum value \( C_{{P_{B} }} = 16/27 \)) while the turbine ability to start is normally ignored. Nevertheless, if a turbine cannot accelerate through start-up, its power extraction performance is severely limited, especially at low wind speeds. The criterion of starting behavior at relatively low Reynolds numbers, appropriate for the urban application, therefore offers another expectation to improve the overall performance concerning the period that the turbine needs to start might be achieved which might lead to a significant increase in energy turn-out. The work will focus upon vertical-axis machines of Darrieus type using an H-rotor in which the blades are straight and parallel to their axis of rotation. For the small such turbines, i.e. in low Reynolds number flows, some researchers have stated that the Darrieus-type turbine is inherently not self-starting. The concept of a vertical-axis counter-rotating rotor is used to overcome the starting drawback of small Vertical Axis Wind Turbines (hereafter VAWT). For this purpose, we attempt to simulate the flow around a Counter-Rotating VAWT (CR-VAWT) with Large Eddy Simulation (LES) and both starting behavior and power performance is outlined by comparing with an equivalent conventional turbine.
Horia Dumitrescu, Alexandru Dumitrache, Ion Malael, Radu Bogateanu

A High-Order Finite Volume Method for the Simulation of Phase Transition Flows Using the Navier–Stokes–Korteweg Equations

In this work, we employ the Navier–Stokes–Korteweg system of equations for the simulation of phase transition flows. This system belongs to the diffuse interface models, in which both phases are separated by a non-zero thickness interface where the properties vary continuously. The key idea of these methods is the ability to use the same set of equations for the entire computational domain, regardless of the phase of the fluid. However, these methods lead to a system of equations with high-order derivatives, which are difficult to discretize and solve numerically. Here, we propose the use of a high-order Finite Volume method, FV-MLS, for the resolution of the Navier–Stokes–Korteweg equations. The method uses Moving Least Squares approximations for the direct and accurate discretization of higher-order derivatives, which is particularly suitable for simulations on unstructured meshes. In this work, we show two numerical examples in which the interface is set to interact with great changes in the properties, in order to demonstrate the robustness of the method.
Abel Martínez, Luis Ramírez, Xesús Nogueira, Fermín Navarrina, Sofiane Khelladi

An a Posteriori Very Efficient Hybrid Method for Compressible Flows

In this work we present a framework for a high-order hybrid method made up of an explicit finite-difference scheme and a member of the Weighted Essentially Non-Oscillatory (WENO) family. A new a posteriori switching criterion is developed based on the Multidimensional Optimal Order Detection (MOOD) method. The schemes tested here are chosen to illustrate the process, we select non-standard fourth order finite difference and fifth order compact and non-compact WENO, but any other combination of central finite differences and upwind schemes could be used as well. In this work we present a one-dimensional and a two-dimensional case to illustrate the speed, accuracy and shock-capturing properties of the proposed schemes.
Javier Fernández-Fidalgo, Xesús Nogueira, Luis Ramírez, Ignasi Colominas
Additional information