Aerodynamic and aeroacoustic performance investigations on modified H-rotor Darrieus wind turbine

Abstract

The present paper investigates the aerodynamic and aeroacoustic characteristics of the H-rotor Darrieus vertical axis wind turbine (VAWT) combined with very promising energy conversion and steering technology; a fixed guide-vanes. The main scope of the current work is to enhance the aerodynamic performance and assess the noise production accomplished with such enhancement. The studies are carried out in two phases; the first phase is a parametric 2D CFD simulation employing the unsteady Reynolds-averaged Navier-Stokes (URANS) approach to optimize the design parameters of the guide-vanes. The second phase is a 3D CFD simulation of the full turbine using a higher-order numerical scheme and a hybrid RANS/LES (DDES) method. The guide-vanes show a superior power augmentation, about 42% increase in the power coefficient at λ = 2.75, with a slightly noisy operation and completely change the signal directivity. A remarkable difference in power coefficient is observed between 2D and 3D models at the high-speed ratios stems from the 3D effect. As a result, a 3D simulation of the capped Darrieus turbine is carried out, and then a noise assessment of such configuration is assessed. The results show a 20% increase in power coefficient by using the cap, without significant change in the noise signal.

Keywords

Aerodynamic aeroacoustic capped turbine Darrieus turbine guide-vanes hybrid RANS/LES (DDES)

Introduction and purpose of the present work

Two technologies convert wind energy based on the alignment of the rotational axis. The two types of wind turbines are distinguished: Horizontal Axis Wind Turbine (HAWT) with a rotational axis enduring parallel to the ground and the mainstream flow, and Vertical Axis Wind Turbine (VAWT) with rotational axis standing vertically perpendicular to the mainstream. Besides their different appearances, the underlying working principle is the same. Wind turbine extracts kinetic energy from the wind, converting it into mechanical energy in the form of rotational motion; through a built-in electricity generator, the mechanical energy is converted into electric power.

Although considerable progress has already been achieved, the available technical design is not yet adequate to develop reliable wind energy converters for conditions corresponding to low wind speeds and urban areas. Darrieus VAWT turbine appears to be particularly promising for such conditions but suffers from poor performance compared to HAWT. However, it still the nearest performance to HAWT from all other VAWTs configurations (Wakui et al., 2005). VAWTs were initially being considered very promising, before being subrogated by the modern, horizontal axis turbines. There is now a resurgence of interests in VAWTs, in particular, Darrieus turbines. The French engineer George Jeans Mary Darrieus presented the Darrieus VAWT. His patent in 1931 (Marie, 1931) was included both the Curved Bladed and Straight-bladed (H-rotor) VAWTs. The Darrieus VAWTs are lift-type wind turbines. The H-rotor turbine consists of two or more airfoil-shaped blades which are combined with a rotating vertical shaft, as shown in Figure 1. The wind flows over the airfoil profiles of the blade and generates an aerodynamic lift and draws the blades along.

Figure 1.

Three-bladed H-rotor Darrieus turbine (Dessoky et al., 2019).

VAWTs presents many advantages, as an omnidirectional machine, they do not need an extra device to adapt the turbine with respect to the wind direction to produce power, eliminates the need for a pitching system hence reduces mechanical complexity (Islam et al., 2008). The mechanical power generation equipment can be installed on the ground level, which makes it easy for maintenance and the ground-based facilities (e.g. transmission and electrical generation, etc.) performs VAWTs lighter (Dabiri, 2011). Additionally, there is a possibility for a large power generation with VAWTs because their size can be significantly augmented. In offshore applications, a low center of gravity allows improved floating stability and a reduced gravitational load. The feasibility of underwater electricity generation further decreases the cost and size of the floating support construction (He, 2013). VAWTs have considerable potentials in wind farm operations. As the output power is in reliance on the turbine projected area, the growing demand for power output has driven the size of HAWTs to restrict. More massive wind turbine size results in tremendous centrifugal force and bending moment. Longer blade claims larger turbine spacing hence lower wind farm density (Peace, 2004). On the other hand, the size of VAWT can be prolonged vertically without a meaningful increase in the occupied area. The wake recovery of a VAWT is faster, allowing for clustered array and increases wind farm power output (Dabiri, 2011; Kinzel et al., 2012). However, VAWT has its problems, besides the well-known self-starting difficulty, the existence of dynamics stall at low speed ratio also reduces the turbine growth. The inherent unsteadiness of VAWTs leads to a complicated rotor loading and flow phenomena, which challenge blade designs and flow investigations (Brahimi et al., 1995). Furthermore, they have lower aerodynamic performance compared to horizontal axis wind turbines (HAWT). Therefore, the performance enhancement of the current H-rotor could help in wind energy utilization in the urban and low wind speed region.

Dessoky et al. (2019) proved that the accelerated approaching wind highly enhances the aerodynamic performance, by using the shrouded wind-lens. As the results of the attractive characteristic of shrouded wind-lens, which drives the author to extend this published work by using the guide-vanes instead of the shrouded wind-lens, that has the same concept of operation.

Chong et al. (2013) used Omni-direction-guide-vane, which enhances the starting capabilities, global performance, and decreases the negative visual impact of the turbine. Here the guide vanes are inclined at a prescribed angle. However, there is no evidence that the introduced design parameters achieved the best performance. Furthermore, despite installing such a system in the urban region, there is no information about the noise accomplished by applying such technology to assess public acceptance.

The enhancement achieved by using such guide-vanes drove Cho et al. (2017) to use the guide-vanes as well, but here the regular nozzle guide-vanes were used. The author used ANSYS CFX to simulate a 2.5D model to assess the impact of each single design parameter on the global performance at a single speed ratio. Although the Darrieus turbine with guide-vanes showed a performance enhancement, it is not sufficient to assess the aerodynamic performance using only a single-speed ratio. It could be an indication but never emphasize the performance enhancement, which is evaluated by the maximum power coefficient and the speed ratio operating range.

Therefore, this work aims to optimize the design parameters of fixed guide-vanes, considering the 3D effect and the noise level caused by such a system. Furthermore, Dessoky et al. (2019) introduced the 2D modeling that shows a higher difference in output power compared to experimental results that were argued due to the 3D effect. Therefore, a turbine with a cap is investigated as well, considering two different cap thicknesses to eliminate the tip loss. Then an aeroacoustic noise assessment for the new configuration is carried out.

Fixed guide-vanes

Wind power generation is proportional to the wind speed cubed. Therefore, a significant increase in output is brought about if it is possible to create even a slight increase in the velocity of the approaching wind. Furthermore, the flow blade incidence angle strongly affects the produced lift to drag ratio, affecting the produced turbine power. The proposed technology could perform both as velocity increasing and flow guidance device. However, the configuration of guide-vanes is a critical aspect, as the flow could be blocked by the guide-vanes, causing performance deterioration. Therefore, the effect of every single parameter will be investigated through 2D simulations, and for a selected configuration a 3D simulation at a single speed ratio will be carried out to assess the aerodynamic and aeroacoustic performance of the technology as shown in Figure 2.

Figure 2.

Schematic of: (a) VAWT installed inside the guide-vanes and (b) design parameters within the guide-vanes.

Blade cap

The aerodynamic characteristics of the finite span blade (3D model) are quite different from the airfoil section (2D model), as the finite span blade is a three-dimensional body. Therefore, there is a flow component in the spanwise direction. Basically, the flow over the airfoil creates sides of high pressure (pressure side) and low pressure (suction side), where the lift is generated as a result of the unbalance created by the two sides. However, as shown in Figure 3(a) the flow near the tip tends to curl around the tip from the pressure side toward the suction side, the generated circular motion near the tip is called the tip vortex, and it moves in the blade downstream. These tip vortices create a small downward flow component that is called downwash. The downwash over the finite blade tends to decrease the angle of attack and creates an excessive drag that is called induced drag, which leads to a lower lift to drag ratio compared to the infinite blade that is more obvious at high flow velocity (Anderson, 2010). Consequently, by installing a cap near the blade tip as in Figure 3(b), it avoids the flow circulation near the tip, and the finite span blade behaves as the infinite one. As a result, higher power at the high-speed ratios is produced.

Figure 3.

Schematic of finite span blade: (a) without cap and (b) with cap.

Numerical setup

The assessments are based on the CFD simulations that were carried out using a block-structured solver FLOWer from the German Aerospace Center (DLR) (Kroll et al., 2000). Institute of Aerodynamics and Gas Dynamics at the University of Stuttgart extended the code over the years for wind turbine applications. The time integration was carried out by an explicit hybrid 5-Stage Runge-Kutta scheme. The higher-order WENO (Weighted Essentially Non-Oscillatory) scheme for the reconstruction of the convective flux at cell boundaries together with an HLLC Riemann solver on block-structured grids of the background was applied to improve conservation of vortical structures (Kowarsch et al., 2013, 2015; Liu et al., 1994). The multigrid level 3 and implicit residual smoothing with variable coefficients were applied to accelerate the convergence. The time discretization is achieved by applying dual time-stepping according to Jameson (1991). Delayed Detached-Eddy Simulation (DDES) (Shur et al., 2008; Weihing et al., 2016) was applied among unsteady Reynolds-averaged Navier-Stokes (URANS) approach employing the Shear-Stress-Transport (SST) $k - ω$ turbulence model according to Menter (1994). During the 2D modeling, the corresponding URANS model as the full DES model, namely the SST $k - ω$ model, is applied. The flux discretization uses the central Jameson-Schmidt-Turkel (JST) scheme (Jameson et al., 1981) for all 2D simulations. The physical time step size used is equivalent to 0.5° of blade revolution (Castelli et al., 2013; Li et al., 2013; McLaren, 2011) with 100 sub-iterations per time step. Convergence behavior for NavierStokes calculation with SST $k - ω$ turbulence model was presented by Kroll et al. (2000). The noise radiation toward an investigation point in the far-field is estimated based on the in-house Ffowcs Williams-Hawkings (FW-H) equation code ACCO (Dessoky et al., 2019; Kessler and Wagner, 2004).

The turbine model is based on the published experimental data by Li et al. (2016). The turbine model is a two-bladed rotor with a NACA0021 airfoil section. It has a radius of 1 $m$ , chord length of 0.265 $m$ , the pitch angle of 6°, 0.216 $m$ central shaft diameter, 1.2 $m$ blade height, and 8 $m / s$ uniform wind velocity.

The present study employed the fully structured mesh method. The turbine mesh initially consists of three grid components, namely, background, rotor blades, and the central shaft ( see Figure 4(a) and (b)). For evaluating the impact of the guide vanes, another component was added, as shown in Figure 4(c). The grid overlapping (Chimera) method was employed, allowing high quality mesh components to be created independently. The total number of cells in all zones based on the published procedure and mesh independence study is about 35 million, with the background dimensions shown in Figure 5 (Dessoky et al., 2019).

Figure 4.

Computational mesh used in the simulations: Background (a), rotor with the shaft (b), and fixed guide-vanes (c).

Figure 5.

Sectional 3D computational mesh used in the simulations (Dessoky et al., 2019).

Fixed guide-vanes 2D parametric study

The guide-vanes are surrounding the rotor; therefore, the flow velocity could be increased by the contraction through the nozzle shape. Furthermore, the guide-vanes could act as a wind direction modulator, for better flow blade incidence angle. The generated power strongly depends on the location of guide-vanes with respect to the flow. As a result, this section investigates the impact of the guide-vanes design parameter through 2D modeling to achieve the best configuration.

Guide-vanes angle ( $β_{v}$ )

In this section, a general consideration of the diffuser main design parameters, inner radius ratio ( $R_{i}$ / $R$ ) and outer radius ratio ( $R_{o}$ / $R$ ) are equal to 1.1 and 3.5 respectively. The 2D simulations are assessing the the turbine performance at different values of $β_{v}$ parameter as shown in Table 1.

Table 1.

Investigated guide-vanes angle values.

β_{v}

0°

15°

30°

45°

Figure 6(a) shows that at $β_{v}$ = 0° the flow enters the turbine smoothly from the flow areas 1 and 2, unblocked, and the flow is passing through the whole area, increasing the flow velocity at the guide-vanes outlet. Increasing $β_{v}$ to 15° as in Figure 6(b), the flow area 1 is partially blocked, resulting in lower velocity at the nozzle 1 outlet. On the other hand, the flow through area 2 performs a little better and the blocking is less, showing a slight increase in velocity at the nozzle 2 outlet. Further guide-vanes angle increase to $β_{v}$ = 30° as in Figure 6(c) the flow through area 1 is totally blocked and the flow rate through area 2 is decreased by means of nozzle inlet edges and separated flow, resulting in less velocity and worse flow orientation at the turbine inlet. At $β_{v}$ = 45° as in Figure 6(d) the flow in area 1 is still blocked, however, more mass flow rate interact with the blade at the effective power generation azimuthal position causing better turbine performance compared to $β_{v}$ = 30°. As a result, the guide-vanes angle $β_{v}$ = 0° shows the better orientation that performs the highest mass flow rate through the turbine. Consequently, Figure 7 shows the impact of such flow behavior on the turbine performance. At guide-vanes angle $β_{v}$ =0°, it shows the highest maximum power coefficient, wider operating range, and the power coefficient is the maximum over the whole speed ratio operating range. Increasing the guide-vanes angle to $β_{v}$ = 15° and $β_{v}$ = 45° shows a higher power coefficient at high-speed ratio range and wider operating range compared to the turbine without guide-vanes. Increasing the angle to $β_{v}$ = 30° causing a deterioration in the turbine performance, and is considered as the worst behavior.

Figure 6.

Velocity contours at $λ$ = 3.25 for: (a) $β_{v}$ = 0°, (b) $β_{v}$ = 15°, (c) $β_{v}$ = 30°, and (d) $β_{v}$ = 45° at $θ$ = 0° (2D model).

Figure 7.

Comparison of power coefficient $C_{P}$ curve between Darrieus turbine without the guide-vanes and a Darrieus turbine with the guide-vanes at different angles (2D model).

Guide-vanes outer radius design parameter

The guide-vanes design parameters, inner radius ratio ( $R_{i}$ / $R$ ) and guide-vanes angle ( $β_{v}$ ), equal to 1.3 and 0°, respectively. The impact of the outer radius is investigated for different values, as shown in Table 2. Such values are chosen to be in a reasonable range with respect to a small turbine for energy production in urban regions. As illustrated in Figure 8, the larger the outer radius, the higher the blockage by the frontal guide-vanes. As a result, the high-pressure region near the guide-vanes is extended to affect the inlet flow. However, the larger the outer radius increases the frontal flow area that allows more mass flow rate through the guide-vanes. Which effect is dominant, leads to the higher or lower mass flow rate through the turbine. In the selected range of outer radius, it is observed that at the outer radius ratio of 4.5 as in Figure 8 achieves the highest turbine inlet velocity at $λ$ = 3.25. That means the effect of the increase in the frontal area is dominant. On the other hand, the effect of the increase in static pressure at the inlet could be dominant at other range of outer radius, which might be not adequate for the urban region applications or at other speed ratio values that also affect the flow topology inside the turbine and the inlet condition (Cho et al., 2017). As a result, Figure 9 emphasizes that the highest maximum power coefficient and the wider operating range are at $R_{i}$ /R = 4.5, where the highest wind speed at the turbine inlet is achieved.

Table 2.

Investigated guide-vanes outer raduis values.

R_{o}

R

3.5

4.5

Figure 8.

Velocity contours at $λ$ = 3.25 for: (a) $R_{o}$ / $R$ = 3, (b) $R_{o}$ / $R$ = 3.5, (c) $R_{o}$ / $R$ = 4, and (d) $R_{o}$ / $R$ = 4.5 at $θ$ = 0° (2D model).

Figure 9.

Comparison of power coefficient $C_{P}$ curve between Darrieus turbine without the guide-vanes and a Darrieus turbine with different outer raduis ratio guide-vanes (2D model).

Guide-vanes inner radius design parameter

The inner radius of the guide-vanes can affect the wind direction and velocity. Therefore, the performance of the VAWT was computed with various inner radius ratios as shown in Table 3, with fixed outer radius ratio and guide-vanes angle of $R_{o}$ /R = 3.5 and $β_{v}$ = 0°, respectively. Figure 10 represents the velocity contours for different inner radius ratios at $λ$ = 3.25. It is observed that at a smaller ratio, the flow is directed toward the lower blade side, and the flow velocity at the upper side is decreased, producing more pressure difference. On the other hand, Figure 11 demonstrates that at the smaller ratio, there is more chance for the blade to interact with the vortices generated by the edge of the guide-vanes, and this chance is diminished as the edge is further away from the blade. Therefore, the inner radius should be small enough to best guide the flow toward the blade without dispersion and with minimal blade-vortex interaction. Figure 12 shows that the best flow condition, where the highest maximum power is achieved with the wider operating range, is at an inner radius ratio of 1.2.

Table 3.

Investigated guide-vanes inner-raduis ratio values.

R_{i}

R

1.1

1.2

1.3

1.5

Figure 10.

Velocity contours at $λ$ = 3.25 for: (a) $R_{i}$ /R = 1.1, (b) $R_{i}$ / $R$ = 1.2, (c) $R_{i}$ / $R$ = 1.3, and (d) $R_{i}$ / $R$ = 1.5 at $θ$ = 0° (2D model).

Figure 11.

Vorticity contours at $λ$ = 3.25 for: (a) $R_{i}$ / $R$ = 1.1, (b) $R_{i}$ / $R$ = 1.2, (c) $R_{i}$ / $R$ = 1.3, and (d) $R_{i}$ / $R$ = 1.5 at $θ$ = 0° (2D model).

Figure 12.

Comparison of power coefficient $C_{P}$ curve between Darrieus turbine without the guide-vanes with different inner raduis ratio guide-vanes, and a Darrieus rotor (2D model).

3D modeling investigations

Darrieus VAWT with fixed guide-vanes

Aerodynamic assessment

The previous section illustrated that the Darrieus turbine with a certain configuration of the fixed guide-vanes demonstrates better performance compared to the conventional turbine, as the results of accelerating and guiding the flow toward the rotor blades. For better describing how the power augmentation takes place by using the guide-vanes, a full model simulation is carried out with the design parameter that achieves the best performance during the 2D parametric study (see Table 4), the deviation of the 2D from the full model is discussed in details by Dessoky et al. (2019). Figure 13 shows that, by using the guide-vanes, the power coefficient increases by approximately 42% at $λ$ = 2.75. This outcome also verifies that the 2D results overestimate the performance improvement, but on the other hand, also confirms that the performance indeed increases due to the guide-vanes.

Table 4.

Selected guide-vanes design parameters.

$β_{v}$	$R_{o}$ / $R$	$R_{i}$ / $R$
$0 °$	3.5	1.2

Figure 13.

Comparison of power coefficient $C_{P}$ between Darrieus turbine without the guide-vanes and a Darrieus turbine with guide-vanes.

Figures 14 and 15 show that at $θ$ = 0° and $θ$ = 270° the guide-vanes increase the angle of attack compared to the blade without guide-vanes, produces a higher pressure at the pressure side and lower pressure in the suction side. As a result, an increase in the generated torque at such azimuthal positions for the turbine with guide-vanes as in Figure 18. Figures 16 and 17 illustrate that at $θ$ = 90° the guide-vanes produce a near blade wake and the velocity deficits; therefore, the produced torque deteriorates. At $θ$ = 180°, Figures 16 and 17 show that the guide-vanes cause a flow velocity deficiency as a result of the flow blocking that leads to a lower torque coefficient (see Figure 18). Therefore, it is concluded that the guide-vanes produce azimuthal positions of a higher torque due to the increase in the angle of attack and other positions of lower torque due to wake generated and flow blocked by the guide-vanes, compared to the turbine without the guide-vanes. However, the average generated torque over the complete revolution is higher than the conventional turbine and a 42% increase in the power coefficient at $λ$ = 2.75 is achieved.

Figure 14.

Pressure ratio ( $P$ / $P_{\infty}$ ) contours and streamlines of the turbine with guide-vanes at: (a) $θ$ = 0° and (b) $θ$ = 270°.

Figure 15.

Pressure ratio ( $P$ / $P_{\infty}$ ) contours and streamlines of the turbine without guide-vanes at: (a) $θ$ = 0° and (b) $θ$ = 270°.

Figure 16.

Velocity contours in m/sec of the turbine with guide-vanes at: (a) $θ$ = 90° and (b) $θ$ = 180°.

Figure 17.

Velocity contours in $m / seconds$ of the turbine without guide-vanes at: (a) $θ$ = 90° and (b) $θ$ = 180°.

Figure 18.

Comparison of torque coefficient $C_{T}$ curve between Darrieus turbine without the guide-vanes and a Darrieus turbine with guide-vanes.

Aeroacoustic assessment

The noise transmitted from the rotor is depending on the flow displaced by the blades during the rotation (thickness noise or monopole noise), and the blades load variation (loading noise or dipole noise). The monopole noise is eternally associated with the dipole noise but not vice versa (Dessoky et al., 2019).

According to He (2013), the rotor is split into four distinct regimes, namely, upwind, leeward, downwind, and windward, as shown in Figure 19, with distinguished physical characteristics of each of them (Dessoky et al., 2019). Therefore, the observer locations are spread at the beginning and end of each region. Figure 20(a) describes the sound pressure level spectra at $λ$ = 2.75 for the open turbine. The sound pressure level (SPL) at the blade passing frequency (BPF 7 Hz) is higher at observer positions $M_{1}$ and $M_{3}$ . Therefore, the blade passing frequency is resulting from the variations of the higher values of the angle of attack, the relative velocity between the incoming flow and blade, and BVI produces a higher load fluctuation. Dynamic stall begins while the blade is near the upwind section, and the massively separated flow occurs. The harmonics (<60 Hz), describe the noise stems from the rise in the angle of attack, the relative velocity variations, and BVI, as the SPL at the observer positions $M_{1}$ and $M_{4}$ are the highest over most of the frequency range, while the higher harmonics (>60 Hz) describe the noise from the separated flow as $M_{2}$ shows a higher SPL (Dessoky et al., 2019). Figure 21 shows the sound pressure level footprints surrounding the turbine based on the BPF for the open turbine and the turbine with fixed guide-vanes. The open turbine shows a higher noise level from the rotor blade positions indicating the highest fluid blade relative speed fluctuations, the angle of attack variations, and the BVI. The turbine with fixed guide-vanes does not show a higher noise level compared to the open turbine. However, the noise propagation direction is in the flow direction, while in the other direction, a blockage experienced by the guide-vanes results in low blade load fluctuations and fluid displaced by the blades.

Figure 19.

Schematic of VAWT rotor division and observer positions $M_{1}$ , $M_{2}$ , $M_{3}$ , and $M_{4}$ .

Figure 20.

Sound pressure level spectra at the prescribed observer positions for : (a) open turbine and (b) turbine with fixed guide-vanes.

Figure 21.

Aeroacoustic noise footprint on a carpet 50 rotor radius at the BPF for: (a) open turbine and (b) turbine with fixed guide-vanes.

Capped Darrieus VAWT

Aerodynamic assessment

The induced three-dimensional flow by the blade tip vortices alters the pressure distributions on the blade that produce an imbalance in the wind direction, causing excessive induced drag from a pressure drag type and reduce the angle of attack as well (Anderson, 2010). The induced drag has an impact on the Darrieus turbine global power; such an impact could differ according to the speed ratio range. A cap is fixed at the tip of the Darrieus turbine blades, as shown in Figure 22. Therefore, it could avoid or partially avoid the three-dimensional flow. Two capes with different $ϵ$ values ( $ϵ$ = 0.4 and $ϵ$ = 0.8) are investigated, where $ϵ$ is defined as the cap thickness to radius ratio as the following:

ϵ = \frac{{R_{c}}_{o} - {R_{c}}_{i}}{R}

(1)

Where ${R_{c}}_{o}$ and ${R_{c}}_{i}$ are the outer cap and inner cap radius respectively as in Figure 22(b).

Figure 22.

Schematic of Darriues turbine with cap (a) 3D view and (b) design parameters.

The two caps configurations are examined at $λ$ = 2.75. Figure 23 illustrates that the turbine with $ϵ$ = 0.4 cap shows a higher power coefficient that is increased by about 20% and very close to the 2D results where the tip vortex does not take into account. On the other hand, the turbine with $ϵ$ = 0.8 cap shows less power coefficient not only compared to $ϵ$ = 0.4 cap but also compared to the turbine without a cap; it decreases by about 2%. Therefore, a drawback of the turbine cap might appear by installing the larger cap.

Figure 23.

Comparison of power coefficient $C_{P}$ of Darrieus turbine with different configurations.

According to Simpson (2001) as in Figure 24 an induced unsteadiness near the junction occurs. A large horseshoe vortex exists in front of the blade, produced by the wind flow in the junction, and impinging on the blade creating a primary vortex near the blade. A secondary vortex and tertiary vortices are formed by wall shear. These vortices merge, creating a stronger large horseshoe vortex. After a while, the flow becomes completely unstable, and the developed strong vortex moves to the blade downstream. Therefore, the horseshoe vortices are observed upstream, and downstream of the blade and the incoming velocity is decelerating toward the leading edge (Bangga et al., 2017). Figure 25 shows that the vortex strength accompanied with $ϵ$ = 0.8 cap is higher compared to $ϵ$ = 0.4 cap in blade upstream and the downstream region.

Figure 24.

Schematic of vortex evolution near cap junction.

Figure 25.

Wake of the Darruies VAWT with cap at $λ_{2}$ = $- 0.006$ iso-surface, $λ$ = 2.75 and for (a) $ε$ = 0.4 and (b) $ε$ = 0.8.

Figure 26 shows that the pressure coefficient difference is higher with $ϵ$ = 0.8 cap, and it has the smallest minimum pressure coefficient as well over the whole azimuthal range except at $θ$ = 90° as in Figure 27. Furthermore, Figure 28 illustrates that the turbine cap always shows a far stagnation point from the leading edge compared to the turbine without the cap. Therefore, a higher angle of attack by using the cap, and it is quite clear that the effect of tip vortex is to reduce the angle of attack. With $ϵ$ = 0.8 cap, the angle of attack at some azimuthal positions is higher and at others is lower. However, at $θ$ = 90° that is considered as the angle of maximum torque, the angle of attack is approximately equal to the angle of the turbine without cap configuration. Figure 29 illustrates that the turbine with $ϵ$ = 0.4 cap always produced the highest torque, even at the azimuthal position the produced a higher lift by using $ϵ$ = 0.8 cap. As the resultant torque is the reflection of the lift to drag ratio, therefore use of $ϵ$ = 0.4 cap shows the minimum drag by avoiding the tip vortex effect with the minimum impact from the blade-cape junction horseshoe.

Figure 26.

Temporal variation of the pressure distribution at different azimuth angles near to the blade tip.

Figure 27.

Minimum pressure coefficient for different values of $ε$ at a near tip spanwise section.

Figure 28.

Staganation point location with respect to chord lengh for diffrent values of $ε$ at a near tip spanwise section.

Figure 29.

Turbine generated torque for Darrieus turbine with different configurations.

Aeroacoustic assessment

Figure 30 shows the sound pressure level footprints surrounding the turbine, based on the BPF for different turbine configurations. Cap results in a horseshoe from junction effect and reducing the effective angle of attack. However, it does not approximately affect the fluid displaced by the blade. Therefore, it is quite evident that the cap does not show any change in signal directivity and produces a little increase in the sound pressure level change the noise produced by the turbine, due to the junction effect. Furthermore, it is clear that the tip vortex has no significant contributions to the noise generated by the Darrieus VAWT. By analyzing the spectra as in Figure 31, it could be concluded that that the capped turbine shows a higher sound pressure level at higher harmonics as a reflection of the horseshoe effect on the cap surface and the blade tip. However, at the observer positions, $M_{1}$ , $M_{2}$ , and $M_{3}$ , the sound pressure level at higher harmonics ( $approx . > 60 Hz$ ) is lower for the capped turbine by means of the lower angle of attack that is dominant over the junction effect.

Figure 30.

Aeroacoustic noise footprint on a carpet 50 rotor radius at the BPF for: (a) turbine without cap and (b) capped turbine ( $ε$ = 0.4).

Figure 31.

Sound pressure level spectra at observer positions: (a) $M_{1}$ , (b) $M_{2}$ , (c) $M_{3}$ , and (d) $M_{4}$ .

Conclusions

Darrieus VAWT has the merit to work under low wind speed and in urban area conditions, but it suffers from poor aerodynamic performance compared to horizontal axis wind turbines. The results presented in this paper show the impact of using guide-vanes incorporated in Darrieus VAWT and assess the aerodynamic and aeroacoustic performance. Therefore, 2D modeling using URANS simulation is taken on to study the effect of each design parameter of the guide-vanes to reach the best configuration. Then the full 3D model of the best configuration using high order numerical scheme with $6^{th}$ order WENO method and a hybrid RANS/LES (DDES) approach for assessing the real aerodynamic performance is conducted. Furthermore, during the validation of the aerodynamic model, differences in power coefficient values between 2D and 3D models are observed. This gap stems from the tip loss in the 3D model. As a result, a capped Darrieus turbine is used to avoid the 3D effect and enhance turbine performance. Finally, the solution of the CFD flow field is forwarded to the CAA simulation to investigate the noise radiation. Accordingly, the following conclusions are taken:

The optimal design parameters of the guide-vanes and the impact of every single parameter on the global aerodynamic performance are introduced through 2D modeling.

The 3D modeling shows that the guide-vanes increase the generated torque around the central shaft at some azimuthal position and reduce it at some other positions. However, the average torque over a complete revolution is increased, achieving an approximately 45% increase in power coefficient at $λ$ = 2.75.

By using the guide-vanes, the directivity of the noise signal is changed, and the total noise value increases when it is predicted at prescribed observer positions.

The capped Darrieus turbine shows about a 20% increase in power coefficient by using the cap with 0.4 thickness to radius ratio while using 0.8 thickness to radius ratio causes a 2% reduction in the power coefficient at $λ$ = 2.75 due to the impact of the junction effect. Moreover, the capped turbine does not show any change in noise signal directivity; only a slight increase in the noise level. As a result, it is obvious that the tip vortex has a minimal contribution to the noise radiated by the H-rotor turbine.

Footnotes

Appendix

Acknowledgements

The authors gratefully acknowledge the German Academic Exchange Service (DAAD), the Ministry of Higher Education and Scientific Research of the Arab Republic of Egypt (MoHE) for the funding through the GERLS scholarship program, the High-Performance Computing Center Stuttgart (HLRS), and Leibniz supercomputing center (SuperMUC) for the computational resources.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Amgad Dessoky

References

Anderson

Jr (2010) Fundamentals of Aerodynamics. Pennsylvania Plaza New York City, USA: Tata McGraw-Hill Education.

Bangga

Ashfahani

Sugianto

, et al. (2017) Three-dimensional flow in the vicinity of a circular cylinder mounted to a flat plate at high reynolds number. In: AIP Conference Proceedings, vol. 1788, p. 030012. Melville, NY, USA: AIP Publishing.

Brahimi

Allet

Paraschivoiu

(1995) Aerodynamic analysis models for vertical-axis wind turbines. International Journal of Rotating Machinery 2(1): 15–21.

Castelli

Dal Monte

Quaresimin

, et al. (2013) Numerical evaluation of aerodynamic and inertial contributions to darrieus wind turbine blade deformation. Renewable Energy 51: 101–112.

Cho

S-Y

Choi

S-K

Kim

J-G

, et al. (2017) Numerical study to investigate the design parameters of a wind tower to improve the performance of a vertical-axis wind turbine. Advances in Mechanical Engineering 9(12): 1687814017744474.

Chong

Fazlizan

Poh

, et al. (2013) The design, simulation and testing of an urban vertical axis wind turbine with the omni-direction-guide-vane. Applied Energy 112: 601–609.

Dabiri

(2011) Potential order-of-magnitude enhancement of wind farm power density via counter-rotating vertical-axis wind turbine arrays. Journal of Renewable and Sustainable Energy 3(4): 043104.

Dessoky

Bangga

Lutz

, et al. (2019) Aerodynamic and aeroacoustic performance assessment of h-rotor darrieus vawt equipped with wind-lens technology. Energy 175: 76–97.

(2013) Wake dynamics study of an h-type vertical axis wind turbine. Master’s Thesis. Delft University of Technology, Netherlands: Technische Universiteit Delft.

10.

Islam

Ting

DS-K

Fartaj

(2008) Aerodynamic models for darrieus-type straight-bladed vertical axis wind turbines. Renewable and Sustainable Energy Reviews 12(4): 1087–1109.

11.

Jameson

(1991) Time dependent calculations using multigrid, with applications to unsteady flows past airfoils and wings. AIAA Paper 1596: 1991.

12.

Jameson

Schmidt

Turkel

(1981) Numerical solution of the euler equations by finite volume methods using runge kutta time stepping schemes. In: 14th Fluid and Plasma Dynamics Conference, p. 1259. Palo Alto, USA: AIAA.

13.

Kessler

Wagner

(2004) Source-time dominant aeroacoustics. Computers & Fluids 33(5–6): 791–800.

14.

Kinzel

Mulligan

Dabiri

(2012) Energy exchange in an array of vertical-axis wind turbines. Journal of Turbulence 13(1): N38.

15.

Kowarsch

Keßler

Krämer

(2013) High order cfd-simulation of the rotor-fuselage interaction. European Rotorcraft Forum (ERF) Document Repository. http://hdl.handle.net/20.500.11881/605

16.

Kowarsch

Lippert

Schneider

, et al. (2015) Aeroacoustic simulation of an ec145t2 rotor in descent flight. In: 71th American Helicopter Society Annual Forum, Virginia. American Helicopter Society.

17.

Kroll

Rossow

C-C

Becker

, et al. (2000) The megaflow project. Aerospace Science and Technology 4(4): 223–237.

18.

Maeda

Kamada

, et al. (2016) Study on power performance for straight-bladed vertical axis wind turbine by field and wind tunnel test. Renewable Energy 90: 291–300.

19.

Zhu

Y-I

, et al. (2013) 2.5 D large eddy simulation of vertical axis wind turbine in consideration of high angle of attack flow. Renewable Energy 51: 317–330.

20.

Liu

X-D

Osher

Chan

(1994) Weighted essentially non-oscillatory schemes. Journal of Computational Physics 115(1): 200–212.

21.

Marie

DGJ

(1931) Turbine having its rotating shaft transverse to the flow of the current. US Patent 1,835,018.

22.

McLaren

(2011) A numerical and experimental study of unsteady loading of high solidity vertical axis wind turbines., PhD Thesis, McMaster University, Hamilton, Ontario, Canada.

23.

Menter

(1994) Two-equation eddy-viscosity turbulence models for engineering applications. AIAA Journal 32(8): 1598–1605.

24.

Peace

(2004) Another approach to wind. Mechanical Engineering Magazine Select Articles 126(06): 28–31.

25.

Shur

Spalart

Strelets

, et al. (2008) A hybrid rans-les approach with delayed-des and wall-modelled les capabilities. International Journal of Heat and Fluid Flow 29(6): 1638–1649.

26.

Simpson

(2001) Junction flows. Annual Review of Fluid Mechanics 33(1): 415–443.

27.

Wakui

Tanzawa

Hashizume

, et al. (2005) Hybrid configuration of darrieus and savonius rotors for stand-alone wind turbine-generator systems. Electrical Engineering in Japan 150(4): 13–22.

28.

Weihing

Letzgus

Bangga

, et al. (2016) Hybrid rans/les capabilities of the flow solver flowerapplication to flow around wind turbines. In: Symposium on Hybrid RANS-LES Methods, Springer, pp.369–380. New York City, USA: Springer.

Aerodynamic and aeroacoustic performance investigations on modified H-rotor Darrieus wind turbine

Abstract

Keywords

Introduction and purpose of the present work

Fixed guide-vanes

Blade cap

Numerical setup

Fixed guide-vanes 2D parametric study

Guide-vanes angle ( β v )

Guide-vanes outer radius design parameter

Guide-vanes inner radius design parameter

3D modeling investigations

Darrieus VAWT with fixed guide-vanes

Aerodynamic assessment

Aeroacoustic assessment

Capped Darrieus VAWT

Aerodynamic assessment

Aeroacoustic assessment

Conclusions

Footnotes

Appendix

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iD

References

Guide-vanes angle ( $β_{v}$ )