Elsevier

Applied Energy

Volume 185, Part 2, 1 January 2017, Pages 1100-1108
Applied Energy

Computational assessment of the DeepWind aerodynamic performance with different blade and airfoil configurations

https://doi.org/10.1016/j.apenergy.2015.10.038Get rights and content

Highlights

  • A BEM model for Darrieus wind turbines is developed and validated.

  • The DeepWind performances in the baseline configuration are estimated.

  • The impact on loads is estimated considering different number of blades.

  • A comparison based on performance is performed considering different airfoils.

  • An optimization algorithm is adopted to find the best performing chord distribution.

Abstract

An aerodynamic improvement of the DeepWind rotor is conducted adopting different rotor geometries and solutions with respect to the original configuration while keeping the comparison as fair as possible. The objective of this work is to find the most suitable configuration in order to maximize the power production and minimize the blade stress and the cost of energy. Different parameters are considered for the study. The DeepWind blade is characterized by a shape similar to the Troposkien geometry but asymmetric between the top and bottom parts. The blade shape is considered as a fixed parameter in the optimization process and, because of different blade element radii, it will experience different tip speed ratios in the same operational condition. This leads to a complex optimization problem, which must be carefully analyzed in order to find the most suitable parameter set. The number of blades in the analysis is varied from 1 to 4. In order to keep the comparison fair among the different configurations, the solidity is kept constant and, therefore, the chord length reduced. A second comparison is conducted by considering different blade profiles belonging to the symmetric NACA airfoil family. Finally, a chord optimization along the blade span is conducted, in order to find the optimal chord distribution to maximize the power production.

Introduction

MW class horizontal-axis wind turbine (HAWT) installations experienced an exponential growth in the last decades [1], due to the increasing awareness of an upcoming necessity of renewable energy production. The rotor size has been experiencing a constant increase in order to minimize the cost of energy, but recently an upper boundary has been reached due to the difficulty to engineer huge components and guarantee a smooth operational life. In this context, vertical axis wind turbines (VAWT) can play a crucial role: in fact, their inherent characteristics allow big size rotors affected by lower operational loads and simpler maintenance policies. Among VAWTs, the Darrieus architecture provides the best performance [2] and is therefore considered the most promising rotor configuration. Nevertheless, the vertical-axis concept does not have a well-consolidated design methodology and, differently from the horizontal-axis architecture, the optimal shape for the Darrieus wind turbine is still object of debate [3]. The main rotor configurations considered in literature are the straight blade (H-rotor) and the Troposkien shape, often transformed in the more practical Straight–Circular–Straight (SCS) shape [4]. The Troposkien geometry has the structural advantage of being subjected only to normal stress, whereas the straight blade shape experiences also bending loads. On the other hand, considering the same rotor occupancy, the straight bladed turbine is characterized by a wider swept area than Troposkien shape.

Seventh Framework Programme (FP7) DeepWind is the first project aiming to design a big-scale offshore floating vertical axis wind turbine with a nominal power generation of 5 MW [5], [6], [7]. Given the considerable rotor size needed to achieve this power production, the Troposkien shape is chosen in order to reduce the blade stresses. The rotor shape was further analyzed and optimized from the baseline to account for the gravity effect, leading to a new blade geometry designed for the target operative conditions [8]. Estimations of the rotor loads obtained with the aeroelastic code HAWC2 are provided in the final report from DTU Wind Energy [9] The purpose of this work is to conduct an aerodynamic improvement with different blade configurations, considering different numbers of blades, profiles and chord distributions.

Two types of simulation approaches are usually adopted in order to predict wind turbine performance: Computational Fluid Dynamic (CFD) codes, based on the numerical resolution of the Navier–Stokes equations, and semi-empirical codes as Blade Element-Momentum (BEM) and Vortex based models, obtained by combining physical formulations and experimental data. The DeepWind concept was usually analyzed considering the Actuator Cylinder model implemented in HAWC2 [10], which can be classified in the second category. Whereas the first models allow a deeper analysis of the rotor aerodynamics, the second ones require a much lower computational effort, providing a quick result that can be easily adopted for rotor design. Among the authors who supported the first approach, Carrigan et al. [11] applied a differential evolutionary algorithm combined with a CFD simulation tool, registering an improvement in the overall rotor performance of about 6% with respect to the baseline geometry, achieved with a considerable variation in both solidity and profile thickness. Also Bourguet et al. [12] adopted a multi-criteria optimization algorithm, aiming at both reducing blade weight and increasing the power production by means of a full campaign of CFD simulations each one requiring 7 h, converging to an optimal airfoil shape very close to NACA 0025 airfoil. Ferreira et al. [13] coupled a genetic optimization algorithm with a potential flow solver using the polars generated by RFOIL [14], [15]. The optimized airfoils show an increased thickness which range from 15% to 35%.

The BEM approach represents the most diffused semi-empirical algorithm for vertical axis wind turbine simulations. Several authors adopted algorithms based on the BEM theory, originally developed by Glauert for helicopter simulations [16] and by Templin and Rangi [17] for vertical axis wind turbine simulations and successively improved by Strickland [18], [19] and Paraschivoiu [20], [21], respectively introducing the multiple streamtube and double disk theories. This tool was successfully adopted for evaluating the optimal blade configuration for Darrieus wind turbines by Bedon et al. [22], showing that, depending on the target operative conditions considered, different blade designs should be adopted to maximize the power coefficient.

In the present work, a BEM code based on the Paraschivoiu model [20], [21] is developed and validated against experimental results. The improvement of the original configuration is conducted by considering different numbers of blades and airfoil sections and optimizing the blade chord distributions, to maximize the energy production and to minimize the blade stress and the cost of energy.

Section snippets

Simulation algorithm

The BEM algorithm considered in the present work is the Double Disk Multiple Streamtube [20], [21], an improvement of the Single Disk Multiple Streamtube model originally proposed by Strickland [18]. The method relies on an iterative procedure to obtain the air induction factor, defined as:a=1-ViVwhere Vi is the velocity at the upwind or downwind blade and V is the free-stream velocity, which allows to estimate the energy extracted from the air by the rotor. The iterative procedure is conducted

Case study

In the present work, the rotor architecture from the DeepWind project is considered. The rotor shape was designed considering a maximum allowable strain [8] and is kept constant in the aerodynamic optimization. The blade shape is shown in Fig. 2.

The baseline rotor configuration, developed in the DeepWind project and characterized by the parameters reported in Table 1, is simulated using the BEM algorithm described in the previous section. A uniform wind speed ranging from 4 m/s to 25 m/s is

Aerodynamic optimization

This section presents the results obtained from the analysis of different blade configurations in order to maximize the power production and minimize the blade loads and manufacturing cost. Two different campaigns are conducted:

  • 1.

    variation of the number of blades keeping the solidity constant;

  • 2.

    variation of the profile thickness.

Finally, an optimization of the chord distribution along the blade span is conducted.

Conclusion

The present work considered the FP7 DeepWind rotor for an aerodynamic analysis of different blade parameters. In the baseline design, a two bladed rotor with NACA 0018 profiles was considered. The comparison shows the aerodynamic effect of a change in the number of blades. A blade number equal to two or three is equally a good choice with respect to the rotor loads. Further reduction in the loads can be obtained by adding one additional blade, however this would increase substantially the rotor

References (37)

  • Reis GE, Blackwell BF. Practical approximations to a Troposkien by straight-line and circular-arc segments. Tech rep....
  • Paulsen US, Madsen HA, Kragh KA, Nielsen PH, Baran I, Ritchie E, et al. The 5 MW DeepWind floating offshore vertical...
  • Verelst DRS, Madsen HA, Kragh KA, Belloni F. Detailed load analysis of the baseline 5MW DeepWind concept. Tech rep....
  • Madsen HA, Paulsen US, Vita L. Analysis of VAWT aerodynamics and design using the actuator cylinder flow model. In:...
  • T.J. Carrigan et al.

    Aerodynamic shape optimization of a vertical-axis wind turbine using differential evolution

    ISRN Renew Energy

    (2012)
  • R. Bourguet et al.

    Aerodynamic multi-criteria shape optimization of VAWT blade profile by viscous approach

    Wind Energy

    (2007)
  • Ferreira CS, Geurts B. Aerofoil optimization for vertical-axis wind turbines. Wind Energy....
  • van Rooij R. Modification of the boundary layer calculation in RFOIL for improved airfoil stall prediction. Tech rep....
  • Cited by (42)

    • Towards a standard approach for future Vertical Axis Wind Turbine aerodynamics research and development

      2021, Renewable and Sustainable Energy Reviews
      Citation Excerpt :

      While some research uses original data from the older Sandia reference turbines [1], this design is not compatible with some significant advances in VAWT research such as pitch control. Some articles reference newer H-turbine based results however these use small scale turbines which may not be applicable for utility scale [2]. The methodology is also variable, with wind tunnel and field testing unavailable to many researchers, rapid numerical analysis methods lacking accuracy under many conditions, and Computational Fluid Dynamics (CFD) methods showing contrasting validations.

    View all citing articles on Scopus

    This paper was presented at the 7th International Conference on Applied Energy (ICAE2015), March 28–31, 2015, Abu Dhabi, UAE (Original paper title: “Aerodynamic Benchmarking of the Deepwind Design” and Paper No.: 674).

    View full text