Elsevier

Energy

Volume 191, 15 January 2020, 116433
Energy

HYBRI – A combined Savonius-Darrieus wind turbine: Performances and flow fields

https://doi.org/10.1016/j.energy.2019.116433Get rights and content

Highlights

  • Wind tunnel measurements on a novel hybrid Savonius-Darrieus wind turbine

  • Low tip speed ratio with careful design of shape, size and position of blades

  • Performances, efficiency and working conditions with electrical measurements

  • Velocity fields and blade interactions by means of PIV

  • Power coefficient around 0.2 for tip speed ratios between 0.5 and 4

Abstract

In this paper, wind tunnel measurements on a model of a vertical axis wind turbine (VAWT) are reported. The turbine is a novel hybrid Savonius-Darrieus combined rotor which aims optimizing performances in medium-low wind regimes, by using a careful design of the shape, size and relative positions of Savonius and Darrieus blades. To this end, a dynamically scaled turbine model is tested in wind tunnel to derive instantaneous and averaged velocity fields by means of Particle Image Velocimetry (PIV), which allows deriving wakes and specific fluid flow phenomena on each single configuration (Savonius or Darrieus) and interactions on the combined geometry. These results are coupled with electrical measurements to determine global performances, efficiency and best working conditions for each separate turbine and for the combined turbine. Data are also compared with results obtained by other authors in previously reported combined hybrid configurations. The proposed system is able to work with good performances (power coefficient equal or slightly lower than 0.2), on an extended range of operative conditions, covering those of each component alone, i.e. for tip speed ratios between 0.5 and 4, in comparison to the ranges 0.5 ÷ 1 and 1.7 ÷ 4 of the used Savonius and Darrieus components. Motivations for the increased performances and working ranges reached by the proposed combined turbine, especially in the low tip speed ratio regime, are given as derived from detailed PIV velocity measurements.

Introduction

Vertical Axis Wind Turbines (VAWT) represent one of the most promising solutions for wind energy production at small scales (with characteristic size from 1 m to 10 m and relatively small wind velocities, i.e. from 3 m/s to 20 m/s) (Rogers et al. 2009 [1], Bhutta et al. 2011 [2]). The major advantages of VAWT turbines over the usually employed Horizontal Axis Wind Turbine (HAWT) are ([1,2], Wood 2011 [3]):

  • -

    Operative conditions almost independent on wind direction;

  • -

    Capability of working even in presence of turbulent streams;

  • -

    Good operative conditions also with vertical wind components;

  • -

    Low starting torque allowing working with small wind intensity and/or turbine size;

  • -

    Power generation system mounted close to the turbine basis, so far reducing operative costs in comparison to HAWT;

  • -

    Reduced noise emission in comparison to HAWT of similar size and power (Mollerstrom et al. 2016 [4], Weber et al. 2015 [5]).

As a consequence, wind turbines based on VAWT solutions are very useful for small systems to be used in highly turbulent regions as for example in urban areas (van Bussel et al. 2003 [6]).

On the other hand, HAWT still retain advantages, especially in terms of power coefficient, Cp, defined as the ratio among the power extracted by the wind turbine (which is proportional to the pressure drop across the wind turbine) and the total power of the wind streamCp=P12ρAU3(where ρ is the density of the fluid, U the reference upstream velocity and A the cross-section of the turbine), and in terms of the tip speed ratio, λ,λ=ωRU,

(ω is the turbine rotational velocity and R its radius). Specifically, HAWTs allow obtaining the following ([1,3], Adaramola et al. 2011 [7], Dou et al. 2019 [8], Abkar 2019 [9], Bastankhah et al. 2017 [10], Ryan et al. 2016 [11]):

  • -

    High power conversion efficiency, being the maximum power coefficient Cp ≈ 0.4 in comparison to smaller values obtained with VAWTs (Cp < 0.2)

  • -

    More extended working conditions for tip speed ratio, typically λ ≈ (1–10), in comparison to VAWTs (λ ≈ 1)

  • -

    Constant interaction among turbine wings and wind during turbine rotation

  • -

    Reduced wake in comparison to VAWT and possibility to act on pitch and yaw angles for turbine control and optimization purposes.

So far, there are several of the previous points which are strictly related to the design and aerodynamics of each element (blade or equivalent) and of the whole rotor. Specifically, considering the relevant objective of reducing the wake extension of VAWT, it would be very important to move from a condition comparable to a bluff body, as for the wakes of simply drag-type rotors closely resembling the cylinder wake, to more clever solutions. A similar goal is aimed when concerning the objective of extending the working conditions of tip speed ratios for VAWTs.

To this end, among possible models of VAWT, those which reached the largest diffusion were the Savonius (Chauvin et al. 1989 [12], Hara et al. 2004 [13], Ikeda et al. 2008 [14]) and the Darrieus rotors (Shibuya et al. 2001 [15], Paraschivoiu 2002 [16], Gorelov 2010 [17]). The geometry of the first, which is a typical example of drag-type rotor, simply consists of two semi-cylinders (in the simplest version) mounted on the vertical axis one facing the other forming an “S" shape in a horizontal cross section. Under an orthogonal wind stream, one of the two semi-cylinder experiences drag, whereas the other produces a thrust which is slightly larger than the former due to the curved surfaces. This difference allows the turbine to rotate and to convert energy, even if with low efficiency and in a range of very small tip speed ratio, typically λ ≈ (0.5–1.5), if compared to HAWT (Wenehennubun et al. 2015 [18], Roy et al. 2015 [19]). This is schematically represented in Fig. 1, taken from Ref. [20], where it is shown that the Savonius turbine is able to achieve a relatively small maximum power coefficient (Cp < 0.15), with rotational speed of the same order of the wind speed (λ ≈ 1). Even with small wind speeds, it is able to start rotating and preserving rotational speed, due to the large wetted surface and the related large torque. Therefore, its performances are very similar to those of the classical American farm windmill, which however is strongly wind direction dependent. Recent modifications of the basic configuration in terms of blade shape, number and separation led to even improved performances (Roy and Saha 2013 [21], Alom and Saha 2018 [22]).

On the other hand, the Darrieus rotor is geometrically more complicated than the Savonius one, so far consisting of several slots, with airfoil cross-section, mounted on a rotating shaft. Thus, the lift exerted by the slots is responsible for the rotation and due to the small airfoil surfaces (even if partially compensated by the large radius of rotation) the torque is not so large, thus requiring an independent system to start rotation at low wind speeds. In this case, as also reported in Fig. 1, typical working conditions are for λ > 4, i.e. for large rotational speed in comparison to the wind velocity. The power coefficient attains a value as high as Cp ≈ 0.4, which is comparable to that of a HAWT (Hashem et al. 2018 [23], Tjiu et al. 2015 [24]). From the point of view of the wake generation and extension, the Darrieus turbine clearly attains smaller wakes in comparison to the Savonius and to HAWTs, but performances are highly dependent on the incoming aerodynamic wake ([15], Kyozuka 2008 [25]). The research was recently focused onto the profile of the Darrieus blade, on the number of blades and on the specific arrangement (straight or helicoidal) of the blades (Bhutta et al. [26], Li et al. 2015 [27], Jafari et al. 2018 [28]).

From the analysis of Fig. 1, it is interesting to consider that the working ranges of Savonius and Darrieus VAWT rotors are usually not overlapping. Therefore, the innovative hybrid concept of a combined Savonius-Darrieus turbine, mounted on the same shaft, should highlight advantages of each configuration and at least partially compensate disadvantages ([25], Gupta et al. 2008 [29], Miller et al. 2018 [30]). Specifically, the low starting torque, typical of the Darrieus turbine, is balanced by the Savonius auto-start counterpart, whereas the small power coefficients and low tip speed ratio regimes of the latter are improved by the former. The main drawback of this hybrid configuration is the increased geometrical complexity, the way in which the two single counterparts are assembled being important in order to maximize the performances of the combined rotor, in comparison to those of each single component (Muhamad 2013 [31], Lap et al. [32]). Indeed, this also affects the request for an increased working range in terms of tip speed ratios, i.e. from λ ≈ 0.5 to λ ≈ 5, in order to cover the gap between that typical of Savonius or Darrieus turbine alone. In this sense, an effective verification of aerodynamic interactions among the two turbine components, on the global performances and efficiency should be carried out for any combined turbine. In addition, the investigation of the wake generation mechanism in comparison to those of non-hybrid configuration should be performed. Previous investigations on hybrid Savonius-Darrieus configuration, certify that among the others the most promising ones are the one with the Savonius part at the top and Darrieus at the bottom and the one with Savonius rotor at the center, surrounded by the Darrieus blades (Abid et al. 2015 [33], Srenivasan et al. 2017 [34]). The former allows a slight improvement in tip speed ratio regimes and reduction of cut-in velocity, with the advantage of no aerodynamic interaction between the Savonius and Darrieus counterparts, at the price of a reduced increase in performances in comparison to the Darrieus configuration alone. The latter gets a relevant improvement of performances at low tip speed ratio regimes, but the specific design and arrangement of the Darrieus blades must be investigated in detail in order to minimize aerodynamic interactions.

In this paper, a combined Savonius-Darrieus wind turbine, identified and patented as HYBRI (EUIPO, Community design No 004035269-0001, 2017) (Savino et al., 2012 [35]) and based on the second arrangement previously described is presented and tested, both from the point of view of power coefficient and rotational performances and from detailed measurements of the aerodynamic generation of the wake, using Particle Image Velocimetry (PIV). The design of the scaled prototype of the turbine is described in detail, by focusing on the geometrical and dynamical parameters useful to extend information on the full scale turbine. The investigation is performed on the basis of the comparisons of obtained results for the hybrid rotor with those on the two separate configurations, Savonius and Darrieus alone. In addition, results are compared also with those on recently proposed hybrid configurations ([31,32]), in order to highlight performances, advantages, limitations and working regime ranges. The attention is also focused onto the link between the global performances of the rotor and the detailed aerodynamic interactions among the single counterparts.

Section snippets

Materials and methodology

A scaled-prototype was designed, assembled and tested in a low speed wind tunnel (open test section with diameter equal to about 1 m, maximum velocity equal to 40 m/s), at Reynolds numbers Re ≈ 105 (using the wind tunnel velocity, U, from 10 m/s to 20 m/s, and the radius of the turbine equal, R = 0.2 m), defined asRe=UR/v,where ν is the kinematic viscosity of air (equal to 1.5 × 10−5 m2/s at ambient temperature).

The assembled combined wind turbine is shown in Fig. 2, where it is also possible

Results

The results are presented separately for the electrical tests, aiming to demonstrate the performances of the proposed hybrid turbine, and the velocity measurements, used to deepen the motivations of the observed behaviors and to reveal some peculiarities of the developing near wake.

Discussion of results and remarks

To give a more strict relation between velocity measurements made by PIV and power coefficients attained in section 3.1, the wake defect is investigated. Indeed, this defect can be computed by considering the mean velocity profiles, from the velocity fields presented in previous figures, specifically by the amount of reduction of velocity in the wake of the model and by the extension of the region in which such a reduction takes place. It is also directly related to the drag exerted by the

Conclusions

Wind turbine model investigation in large wind tunnel allows performing combined electrical and fluid flow measurements to understand some of the most important aerodynamics aspects related to geometrically complex vertical axis wind turbines. In particular, Savonius and Darrieus configurations are investigated in this work, both alone and in a novel proposed combined hybrid configuration, using geometrical and dynamical similarities. The main results are hereafter summarized:

  • -

    the comparison

Patent

The combined Savonius-Darrieus wind turbine is patented with the name HYBRI (EUIPO, Community design No 004035269-0001, 2017) by Cloudwise srl.

Author contributions

Conceptualization, A.P. and G.P.R.; Data curation, D.P. and G.P.R.; Methodology and Formal analysis, A.P. and G.P.R.; Investigation, D.P. and A.P.; Resources, G.P.R.; Writing—original draft, G.P.R.; Writing—review and editing, D.P. A.P. and G.P.R.; Supervision, G.P.R.

Funding source

This research received no external funding.

Acknowledgments

The authors acknowledge the technical contribution by Mr. Alberto Savino and the support from the company Cloudwise srl.

References (38)

  • R. Gupta et al.

    Comparative study of a three-bucket Savonius rotor with a combined three-bucket Savonius–three-bladed Darrieus rotor

    Renew Energy

    (2008)
  • P. Lap-Arparat et al.

    Real-time maximized power generation of vertical axis wind turbines based on characteristic curves of power coefficients via fuzzy pulse width modulation load regulation

    Energy

    (2019)
  • R. Howell et al.

    Wind tunnel and numerical study of a small vertical axis wind turbine

    Renew Energy

    (2010)
  • A.L. Rogers et al.
  • M.M.A. Bhutta et al.

    Vertical axis wind turbine – a review of various configurations and design techniques

    Renew Sustain Energy Rev

    (2012)
  • D. Wood
  • E. Möllerström et al.

    Noise emission of a 200 kW vertical Axis wind turbine

    Energies

    (2016)
  • J. Weber et al.

    Aeroacoustics of Darrieus wind turbine

    Aeroacoustics

    (2015)
  • G. van Bussel et al.

    Performance of an h-Darrieus in the skewed flow on a roof

    J Sol Energy Eng

    (2003)
  • Cited by (33)

    • Wind energy system for buildings in an urban environment

      2023, Journal of Wind Engineering and Industrial Aerodynamics
    • Investigation on harvesting characteristics of convective wind energy from vehicle driving on multi-lane highway

      2023, Energy
      Citation Excerpt :

      The efficiency is improved by blade height [27–29], with an average increase of 18.6%. The efficiency is improved by combining the advantages and disadvantages of two rotors to complement each other [30–32], with an average increase of 12.5%. The efficiency is improved by studying the fan flow field [33–35], with an average increase of 22.3%.

    View all citing articles on Scopus
    View full text