Performance enhancement of horizontal axis wind turbine with guide ring

Abstract

This study investigates the impact of integrating a guide ring upstream of the rotor plane for performance increase of a horizontal axis wind turbine rotor. The guide ring is strategically positioned near the rotor hub to reduce the radial component of wind flow and retain more axial velocity through the rotor. The methodology involved physical testing of a rotor retrofitted with guide rings ranging from 15% to 20% of the rotor diameter. Comparative experimental results demonstrate an increase in peak performance for retrofitted rotors, when compared to a standard rotor. The rotor retrofitted with an 18% guide ring emerged as the most beneficial choice, exhibiting a 4.34% peak power output increase and a broader operational range than the standard rotor. Overall, the study contributes valuable insights into post-installation optimisation strategies for horizontal axis wind turbine rotors.

Keywords

Operational range power increase post installation retrofit aerodynamic improvement

Introduction

The design of horizontal axis wind turbines (HAWTs) and optimisation of rotor performance is tailored to the wind conditions of the installation site. Historically, enhancements and optimisations to HAWT rotor performance were conducted before installation, resulting in a fixed performance profile once the turbine was deployed. Modern-day large-scale HAWTs incorporate post-installation optimisation methods that include operational control and maintenance strategies. Table 1 summarises HAWT efficiency improvements for pre- and post-installation requirements.

Table 1.

Summary of HAWT efficiency improvement.

Pre-installation optimisation	Reference
Airfoil Optimisation	Ram et al. (2019)
Shrouded Rotor Design	Katooli and Noorollahi (2022)
Pitch Control Systems	Civelek et al. (2016)
Vortex Generators	Zhao et al. (2022)
Twisted or Variable Geometry Blades	Barlas and van Kuik (2010)
Boundary Layer Suction	Vos (2018)
Post-installation optimisation	Reference
Operational Optimisation	Rodríguez-López et al. (2020)
Maintenance and Inspection	Park and Kim (2016)
Blade Retrofitting	Jiang et al. (2022)
Surface Coatings	Bera et al. (2024)
Wake Steering and Farm Layout Optimisation	Howland et al. (2019)
Dynamic Stall Control	Guoqiang et al. (2019)
Rotor Dynamic Control Strategies	Elkodama et al. (2023)

In the aerodynamics of wind turbine rotors, a portion of the incoming air, initially directed axially as the wind speed, transforms into radial and tangential components as it traverses the plane of rotation of the rotor. This transformation is integral to the energy conversion process and contributes to the power generation mechanism. Specifically, the axial wind flow is partially redirected along the rotor blade, constituting radial flow, and across the blade, manifesting as tangential flow. This experimental study seeks to prevent a portion of the axial wind flow from undergoing conversion to radial flow, by installing a guide ring in front of the rotation plane near the hub region of the rotor. This strategic placement is assumed to retain more axial velocity for the rotor, thus increasing power performance. A secondary assumption is that the supporting struts holding the ring in position will improve power conversion by reducing inlet whirl (or swirl). This would produce a larger change in whirl velocities across the rotor and improved power conversion. Figure 1(a) illustrates a standard rotor, and Figure 1(b) illustrates a rotor retrofitted with a guide ring.

Figure 1.

Standard rotor (a) and rotor retrofitted with guide ring (b).

The rotor without any guide ring was referred to as the standard rotor, and rotors with guide rings were retrofitted rotors for this study.

Rotor efficiency

The efficiency of a HAWT rotor is usually measured using the power coefficient Cp, and can be expressed in various ways, including in terms of the axial induction factor. The axial induction factor α can be defined as the fractional decrease of axial wind velocity between the free stream and the rotor plane (Manwell et al., 2010). It can be expressed as:

α = \frac{V i - Vt}{V i}

(1)

Where $V i$ is the free stream wind velocity (incoming wind velocity far upstream of the rotor) and $Vt$ is the axial wind velocity at the rotor plane. Figure 2 is an illustration of Vi and $Vt$ as axial velocity reaches the rotor plane of rotation.

Figure 2.

Illustration of Vi and Vt (Manwell et al., 2010).

The power coefficient is a dimensionless quantity that relates the power extracted by the rotor to the power available in the wind and it is commonly defined as:

C_{p} = \frac{P_{r}}{0.5 ρ A_{r} V_{i}^{3}}

(2)

Where $P_{r}$ is the power extracted by the rotor, ρ is the water density, and is the rotor-swept area.

The power coefficient can also be expressed in terms of the axial induction factor α as:

C_{p} = 4 α {(1 - α)}^{2}

(3)

This expression relates the power coefficient directly to the axial induction factor, providing insights into how the rotor’s aerodynamic efficiency depends on the degree of air diverted from the rotor (Burton et al., 2011).

Overall, these expressions help characterise the efficiency and performance of HAWT rotors in extracting power from the wind, with the axial induction factor serving as an essential parameter in understanding the aerodynamic behaviour of the rotor.

Another aspect of rotor efficiency is the formation of the near wake, which consists of rotational and axial flow components. The flow rotates (tangentially) in the opposite direction to the rotor. This tangential flow is considered as a loss as it is unavailable for torque extraction on the rotor. The efficiency of the rotor in terms of tangential flow can be quantified using the tangential induction factor.

The tangential induction factor, denoted as a′, represents the ratio of the angular velocity of the rotor to the angular velocity imparted to the wake (Burton et al., 2011). It can be mathematically expressed as

a' = \frac{ω_{w}}{2 ω_{r}}

(4)

where $ω_{w}$ is the angular velocity imparted to the wake, and $ω_{r}$ is the angular velocity of the rotor.

Design methodology

To investigate the impact of integrating a guide ring upstream of the rotor plane, an ideal 280 mm rotor was designed using Blade Element Momentum theory as described by Fawkes (2023a), with an SG6043 aerofoil over the full blade length. To identify the most effective placement for enhancing performance, guide rings were successively installed ranging from 15% to 20% of the rotor diameter at 1% intervals. The design of the guide ring utilised in this investigation is depicted in Figure 3. The guide rings were modelled using SolidWorks software and were designed to integrate seamlessly with the existing rotor.

Figure 3.

Sample of guide ring used for this study.

The guide rings featured NACA 0025 symmetrical aerofoil profiles for the purpose of minimising drag and lift. These rings were affixed to a non-rotating part of the hub, ahead of the rotating rotor. Detailed parameters of both the rotor and guide rings are presented in Table 2.

Table 2.

Rotor and guide ring parameters.

Description	Value
Rotor diameter (m)	0.28
Design tip speed ratio	4.8
Number of blades	3
Hub ratio	0.5
Hub diameter (m)	0.0140
15% guide ring diameter (m)	0.0420
16% guide ring diameter (m)	0.0448
17% guide ring diameter (m)	0.0476
18% guide ring diameter (m)	0.0504
19% guide ring diameter (m)	0.0532
20% guide ring diameter (m)	0.0560

The overall methodology comprised the following steps:

Design of guide rings, using SolidWorks, to integrate seamlessly with the existing rotor design.

Fabrication of the rotor and guide rings using a resin 3D printer.

Physical testing of the rotor without guide ring, to establish a baseline for comparison with retrofitted rotors.

Physical testing of the same rotor retrofitted with various guide rings.

Analysis and comparison of results obtained from experimental testing.

The testing approach in this study was comparative rather than absolute. The tests were carried out using vertical travel test apparatus and water as fluid medium, with a free stream velocity of 0.5 m/s.

Testing methodology

Experimental testing of the rotor and various guide rings utilised a vertical travel test apparatus similar to that used by Fawkes (2023b). This equipment comprises a drop frame positioned atop a water tank measuring 1.4 m in diameter and 1.5 m in height. Within the drop frame, a slider, propelled by a stepper motor, drives the rotor into the water at a constant speed. The slider moves along two parallel 22 mm shafts equipped with four linear bearings. The primary rotor shaft is secured to the slider via two self-aligning angular bearings. The vertical travel test apparatus is shown in Figures 4 –6, with descriptions of all parts.

Figure 4.

Water tank with drop frame.

Figure 5.

(a) Free rotating shaft with rotor attached and (b) slider with free rotating shaft attached.

Figure 6.

Slider with generator and instrumentation.

The stepper motor responsible for driving the slider is managed by an Arduino controller utilising the G-code reference block library (grbl) Computer Numerical Control (CNC) software. A unipolar 3.8 Nm 1.8° stepper motor provided the slider motion, and was connected via a GT2 toothed belt and 60-tooth pulley on the driving stepper motor.

The equipment mounted on the slider included a 0.28 Nm 1.8° stepper motor repurposed as a generator. A single phase of this generator provided suitable torque resistance. This ‘stepper-generator’ was rotated by the main rotor shaft via a GT2 9 mm belt, with both the shaft and the ‘stepper-generator’ equipped with 60-tooth pulleys to maintain a 1:1 rotation ratio. A single-phase rectifier converted the alternating current (AC) output to direct current (DC). The DC output was then directed to a resistor bank - the resistance of which could be adjusted. This resistor bank allowed for rotation of the ‘stepper-generator’ under varying loads and therefore at different rotation speeds. The loads utilised during testing were power resistors with resistances of 50, 20, 10, and 5 Ω, and 2 Ω.

An eight-slot encoder disc was mounted at the top of the primary shaft, with rotation speed monitored using an infrared optical sensor interfaced with an Arduino controller board. Data transmission from the Arduino controller was facilitated through serial communication with a laptop. Water temperature readings were obtained using a digital LCD thermometer and recorded before each test run.

Before testing, calibration of the stepper generator was conducted by correlating rotation speed with DC voltage output across the different resistance loads employed during testing. This calibration procedure occurred on a workbench, where the ‘stepper-generator’ was driven by a DC motor via a GT2-toothed belt, ensuring a constant rotation speed. The rotation was monitored using a handheld tachometer, while the voltage output was measured with a multimeter. Figure 7 illustrates the calibration graph for the ‘stepper-generator’ across all resistive loads utilised during testing.

Figure 7.

Generator calibration for all resistive loads.

A large diameter water tank was utilised to minimise blockage effects during testing. Blockage ratio is the area of the rotor divided by the test section area (tank cross sectional area). This test equipment resulted in a blockage ratio of 4%. Studies conducted by Chen and Liou (2011) indicated that blockage ratios below 10% in wind tunnel testing yielded less than 5% blockage errors. Previous research by Fawkes (2023b) utilising the same rotor and tank size as in this study, demonstrated that with a blockage ratio of 4%, the blockage error remained under 1%, making it acceptable for research purposes.

Additionally, considerations were made regarding the length of travel of the rotor within the water tank. It was imperative to ensure that the near wake was fully developed during testing and that the stopping position of the rotor was sufficiently elevated from the tank floor to avoid flow interference. A travel distance of 1,140 mm, equivalent to over four rotor diameters, was utilised, with the stopping position set at 560 mm above the tank floor. These constraints on travel length and stopping position ensured full development of the near wake and prevented flow interference from proximity to the tank floor. Good coverage of the expected length of a turbine rotor near-wake is provided by Bastankhah and Porté-Agel (2017) and the rationale for the 560 mm stopping distance is discussed by Fawkes (2023b).

To summarise, the vertical travel test apparatus lowered the rotor at a consistent velocity of 0.5 m/s over a distance of 1,140 mm, exceeding four rotor diameters. The rotation of the rotor was monitored using an infrared optical sensor interfaced with an Arduino controller board. The ‘stepper-generator’ was connected to various resistances to provide load braking during testing.

Testing operation

Seven distinct rotors underwent testing, comprising one standard rotor, serving as a baseline for comparison, and six rotors retrofitted with guide rings. The guide rings were positioned at intervals of 15%, 16%, 17%, 18%, 19%, and 20% of the rotor diameter. Each rotor underwent five test runs per resistive load, with rotation speed and elapsed time recorded through serial communication from the Arduino board. In total, 175 test runs were conducted for the complete set of rotors.

Data handling

The raw test data included temperature readings for each run, the rotor rotation speed, and the elapsed time, corresponding to a relative fluid velocity of 0.5 m/s. The average rotor rotation speed in rpm versus elapsed time for five runs per resistive load was plotted for analysis. Afterwards, an exponential decay function (5) was fitted to the data using Excel Solver.

Y = Y_{0} + A e^{- k t}

(5)

Where Y is the rpm curve value at any time t, and Y₀ is the asymptote value. The amplitude of the curve is described by constant A, and the decay rate constant is k. Excel solver optimises Y₀, A, and k by minimising the sum of δ² values of the raw data. The asymptote of the exponential decay function provided the final rotation speed of the rotor.

Throughout the test runs, the rotation speed starts at a high level and gradually decelerates as the near wake forms. A notable indication, that the near wake was fully developed within the 4-diameter travel distance, was observed in the exponential decay curves of the test runs, which approached the asymptote towards the end of the elapsed time. As an example, a sample plot of the 17% guide ring rotor with a 5 Ω resistive load is shown in Figure 8.

Figure 8.

Exponential decay curve and asymptote for rotation speed of 17% guide ring rotor.

Power generated by the rotors P_r, was calculated using the final rotation speed obtained from the previously determined asymptote and the ‘stepper-generator’ calibration curve to derive the corresponding DC voltage. The power calculation was performed using

P_{r} = \frac{v^{2}}{R}

(6)

where $v$ represents the DC voltage and $R$ denotes the resistive load.

Power versus rotation speed curves were generated for each rotor by determining the asymptote of the rotation speed data for each test run. The rotation speed asymptote was recorded as well as the ‘stepper-generator’ DC output voltage and calculated power. This procedure was repeated for each set of resistive loads, including 50, 20, 10, and 5 Ω, and 2 Ω.

Temperature data from the test runs was recorded to assess its impact on water density and influence on the performance of the rotors. This approach ensured that the results were compared fairly, accounting for variations in power P_w available from the water. Available power from the water is

P_{w} = 0.5 ρ A_{r} V_{i}^{3}

(7)

where $ρ$ is the water density influenced by water temperature, A_r is the rotor swept area and V_i is the relative water velocity.

Results

The experimental test results of power versus rotation speed, for all rotors, at a relative fluid velocity of 0.5 m/s is shown in Figure 9. The solid line, representing the standard rotor, serves as the baseline for further results discussion.

Figure 9.

Power-rotation speed graphs for all rotors.

Figure 9 shows that all rotors retrofitted with guide rings demonstrated a notable increase in peak performance, a shift of the power peak towards higher velocities, and in the 17%, 18%, and 19% guide rings, a broadening of the operational range of angular velocity. At higher rotational speed the retrofitted rotors have a steeper power drop-off compared to the standard rotor. An explanation for this is that the standard rotor makes less use of the near-hub region for power conversion – particularly at high rotation speeds when radial flow over the blades is exacerbated. In contrast, the retrofitted rotors rely on the near-hub region to contribute significantly to the rotor power conversion – particularly at higher rotation speeds. However, near the hub, angle of attack and blade efficiency are more negatively affected by change in rotation speed than at the tip of the blade. The broadening of the operational range is the result of more successful flow control through the rotor (less power lost to radial flow), whereas the more rapid drop-off near the upper limit of rotational speed is to do with the efficiency of the near-hub blade profiles operating at an off-design angle of attack. At lower rotational speed the standard rotor had less radial flow along the blade, so the retrofitted guide rings provide little net benefit, and in some cases a net deficit of power. Losses associated with guide rings include skin friction drag, form drag (aerofoil profile and angle of attack) and the impact of a wind velocity profile transition inboard and outboard of the guide ring at the rotor plane.

The peak performance increase of retrofitted rotors generally fell within a similar range of 0.09 W, though with variations in rotation speeds where the peaks occurred. When selecting the most suitable guide ring for enhancement, the highest peak performance and the expanded operational width was considered. This entails assessing retrofitted rotors that outperform the standard rotor across a broader range of rotation speeds.

Concerns regarding underperformance at high and low rotation speeds are mitigated by recognising that HAWT rotors are not designed to operate at these extremes. Considering both criteria of peak performance increase and operational range enhancement, the rotor retrofitted with the 18% guide ring emerged as the most beneficial choice.

To facilitate clearer comparison, power curves for the standard rotor and best-performing retrofitted rotor with an 18% guide ring are presented in Figure 10.

Figure 10.

Power-rotation speed graphs for standard and 18% guide ring rotors.

Based on the observations from Figure 10, it is apparent that the 18% retrofitted rotor demonstrated increased performance in peak power output and operational range compared to the standard rotor. When focussing solely on peak performance, the standard rotor exhibited a peak power output of 0.0874 W. In contrast, the retrofitted rotor achieved a higher peak power output of 0.0912 W. This represents a 4.34% improvement in peak performance compared to the standard rotor.

The retrofitted rotor had a broader operational range with consistently higher performance when compared to the standard rotor. As shown in Figure 10, the 18% retrofitted rotor outperformed the standard rotor across a range of approximately 150 rpm to about 173 rpm. This broader operational range with increased performance makes the retrofitted rotor a more favourable choice for practical applications where a variable wind speed might be expected.

Temperature recorded during test runs fell within a range of 24°C–25.7°C. The impact on density for the range of temperature had a 0.043% effect on power available from the water. Therefore, no correction to power for the tests was deemed necessary.

Conclusion

This experimental study showed improved peak performance when a 280 mm rotor, tested in water at a free stream velocity of 0.5 m/s, was retrofitted with guide rings. The 18% guide ring rotor was shown to have the best peak performance increase (4.35%) compared to the standard rotor and produced improved performance across a wider range of rotational speeds. Near the upper limit of rotation speed, the decline in performance of the retrofitted rotors was more rapid than for the standard rotor due to an unfavourable angle of attack in the near-hub region of the rotor – a region that is utilised more by guide ring retrofitted rotors. At lower rotational speeds the standard rotor would experience less radial flow along the blade, so the benefit of retrofitted guide rings is reduced, and in some cases results in a net deficit of power. Losses associated with the presence of guide rings include skin friction drag, form drag (aerofoil profile and angle of attack) and the impact of the wind velocity profile transition inboard and outboard of the guide ring at the rotor plane.

While the physical testing equipment did not allow for visualisation or measurement of flow vectors through the rotor, this study provides evidence suggesting that retrofitted rotors with guide rings could potentially serve as a solution for enhancing performance of existing HAWT installations through retrofit.

It is important to acknowledge that the results of this study are specific to a rotor with a 5% hub ratio, utilising water as the working fluid. These findings serve as a foundation for further studies, as discussed in the recommendations below.

Recommendations

This study of retrofitting a 280 mm rotor with a guide ring needs to be simulated using Computational Fluid Dynamics (CFD) software. The simulation will allow visualisation of the effect that the guide rings have on the near wake, and on the radial and tangential flow at the rotor plane. Flow visualisation will allow for better comparison between the various rotors. CFD simulation will also allow parameters such as radial and tangential flow values to be extracted - which was not possible during physical testing. Case studies with CFD for larger retrofitted HAWTs and with higher wind speeds is recommended to verify the performance increase and the position of the guide ring for best performance relative to the diameter of the rotor.

To assess whether guide rings can enhance the performance of rotors experiencing higher radial and tangential flows on the blades, it is advisable to simulate the rotors in this study with higher relative fluid velocities beyond the capabilities of the water drop test equipment. Additionally, further investigation into optimising the aerofoil profiles of the guide rings and adjusting the aerofoil angle relative to the axial fluid velocity could potentially lead to further performance improvements.

Footnotes

Appendix

Acknowledgements

The author acknowledges the support of the Cape Peninsula University of Technology.

Authors contribution

Daniel Barnard: Conceptualisation, Investigation, Methodology, Data curation, Visualisation, Writing – original draft. Howard Fawkes: Writing – review & editing, Supervision, Validation. Graeme Oliver: Project administration, Software, Supervision

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 iDs

Daniel R Barnard

Howard T Fawkes

Research data

Research data is available on request from the author and via .

References

Barlas

van Kuik

GAM

(2010) Review of state of the art in smart rotor control research for wind turbines. Progress in Aerospace Sciences 46(1): 1–27.

Bastankhah

Porté-Agel

(2017) A new miniature wind turbine for wind tunnel experiments. Part I: Design and performance. Energies 10(7): 908.

Bera

Lakshmi

Pathak

, et al. (2024) Recent progress in the development and evaluation of rain and solid particle erosion resistant coatings for leading edge protection of wind turbine blades. Polymer Reviews 64(2): 639–689.

Burton

Jenkins

Sharpe

, et al. (2011) Wind Energy Handbook. John Wiley & Sons, Vol. 3, p. 45.

Chen

Liou

(2011) Blockage corrections in wind tunnel tests of small horizontal-axis wind turbines. Experimental Thermal and Fluid Science 35(3): 565–569.

Civelek

Lüy

Çam

, et al. (2016) Control of pitch angle of wind turbine by fuzzy pid controller. Intelligent Automation & Soft Computing 22(3): 463–471.

Elkodama

Ismaiel

Abdellatif

, et al. (2023) Control methods for horizontal axis wind turbines (HAWT): state-of-the-art review. Energies 16(17): 6394.

Fawkes

(2023a) A BEM theory adaption for inclusion of hub ratio effects in HAWT rotor design and analysis. Wind Engineering 47(4): 867–882.

Fawkes

(2023b) Hub ratio of horizontal axis wind turbine rotors for optimal performance. Available at: https://etd.cput.ac.za/handle/20.500.11838/3960 (accessed 2024).

10.

Guoqiang

Weiguo

Yubiao

, et al. (2019) Experimental investigation of dynamic stall flow control for wind turbine airfoils using a plasma actuator. Energy 185: 90–101.

11.

Howland

Lele

Dabiri

(2019) Wind farm power optimization through wake steering. Proceedings of the National Academy of Sciences 116(29): 14495–14500.

12.

Jiang

Finnegan

Flanagan

, et al. (2022) Optimisation of highly efficient composite blades for retrofitting existing wind turbines. Energies 16(1): 102.

13.

Katooli

Noorollahi

(2022) Shrouded wind turbines: A critical review on research and development. International Journal of Ambient Energy 43(1): 8775–8791.

14.

Manwell

McGowan

Rogers

(2010) Wind Energy Explained: Theory, Design and Application. John Wiley & Sons Ltd, Vol. 3, p. 94.

15.

Park

Kim

(2016) Advanced wind turbine blade inspection and evaluation techniques. In Renewable and Sustainable Energy Reviews (Vol. 60). Available at: https://research.tue.nl/en/studentTheses/improvement-of-aerodynamic-performance-of-vertical-axis-wind-turb

16.

Ram

Lal

Ahmed

(2019) Design and optimization of airfoils and a 20kW wind turbine using multi-objective genetic algorithm and HARP_Opt code. Renewable Energy 144: 56–67.

17.

Rodríguez-López

Del Rio

Cerdá

(2020) Modeling wind-turbine power curves: Effects of environmental temperature on wind energy generation. Clean Technologies and Environmental Policy 13(2): 4941.

18.

Vos

(2018) Improvement of aerodynamic performance of vertical axis wind turbines using boundary layer suction. Available at: https://research.tue.nl/en/studentTheses/improvement-of-aerodynamic-performance-of-vertical-axis-wind-turb

19.

Zhao

Jiang

Feng

, et al. (2022) Researches on vortex generators applied to wind turbines: A review. Ocean Engineering 253: 111266.