Numerical and Experimental Investigations of Axial Flow Fan with Bionic Forked Trailing Edge
Numerical and Experimental Investigations of Axial Flow Fan with Bionic Forked Trailing Edge
Liang, Zhong;Wang, Jun;Wang, Wei;Jiang, Boyan;Ding, Yanyan;Qin, Wanxiang
2023-01-23 00:00:00
machines Article Numerical and Experimental Investigations of Axial Flow Fan with Bionic Forked Trailing Edge 1 1 , 1 1 1 2 Zhong Liang , Jun Wang * , Wei Wang , Boyan Jiang , Yanyan Ding and Wanxiang Qin School of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan 430074, China Guangdong Sunwill Precising Plastic Co., Ltd., Foshan 528305, China * Correspondence: wangjhust@hust.edu.cn; Tel.: +86-27-8754-2517-85 Abstract: To improve the performance of the aerodynamic properties and reduce the aerodynamic noise of an axial flow fan in the outdoor unit of an air conditioner, this study proposed a bionic forked trailing-edge structure inspired by the forked fish caudal fin and implemented by modifying the trailing edge of the prototype fan. The effect of the bionic forked trailing edge on the aerodynamic and aeroacoustic performance was investigated experimentally, and detailed analyses of the blade load and internal vortex structures were performed based on large-eddy simulations (LES). It is shown that the bionic forked trailing edge could effectively adjust the blade load distribution, reduce the pressure difference between the pressure side and suction side near the trailing edge of the blade tip region, and weaken the intensity and influence range of the inlet vortex (IV) and the tip leakage vortex (TLV). The discrete noise caused by the vortices in the rotor tip area was also reduced, particularly at the blade passing frequency (BPF) and its harmonic frequency. The experimental results confirmed the existence of an optimal bionic forked trailing-edge structure, resulting in the maximum power-saving rate
of 7.5% and the reduction of 0.3 ~ 0.8 dB of aerodynamic noise, with an included angle of 13.5 . The detailed analysis of the internal vortex structures provides a good reference for the efficiency improvement and noise reduction of axial flow fans. Keywords: axial flow fan; bionic forked trailing edge; tip leakage vortex; inlet vortex; aerodynamics; aeroacoustics Citation: Liang, Z.; Wang, J.; Wang, W.; Jiang, B.; Ding, Y.; Qin, W. Numerical and Experimental Investigations of Axial Flow Fan with 1. Introduction Bionic Forked Trailing Edge. Machines 2023, 11, 155. As air conditioners have become one of the essential electrical appliances in homes, https://doi.org/10.3390/ office buildings, and public places, they are also one of the major sources of energy con- machines11020155 sumption and noise pollution in people’s daily lives [1]. The axial flow fan is the principal working component and noise source of the outdoor unit of the split air conditioner. This Academic Editor: Davide Astolfi axial flow fan is also called a semi-open axial flow fan [2] because the shroud only covers Received: 24 November 2022 the rear area of its rotor tip. The efficiency improvement and noise reduction of axial flow Revised: 12 January 2023 fans are crucial for protecting the environment and improving people’s quality of life. Accepted: 18 January 2023 The flow loss model and noise characteristics of axial flow fans have constantly been Published: 23 January 2023 popular research topics in turbomachinery. Denton [3] proposed a tip leakage loss model for ducted axial flow fans and determined that approximately one-third of the aerodynamic loss occurs in the tip area. Jang et al. [4,5] conducted an experimental analysis using laser Doppler velocimetry (LDV) measurements and a numerical analysis using a large-eddy Copyright: © 2023 by the authors. simulation (LES). They found that the flow field in the rotor tip of a propeller fan formed Licensee MDPI, Basel, Switzerland. three vortex structures: a tip vortex (TV), a leading-edge separation vortex (LSV), and a tip This article is an open access article leakage vortex (TLV). Large eddy simulation (LES) captures the unsteady flow features and distributed under the terms and provide relatively accurate solutions [6]. Jung and Joo [7,8] inferred the loss caused by blade conditions of the Creative Commons profile, tip, and hub according to the distribution of stagnation pressure loss coefficient at Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ the outlet surface of the axial flow fan in the outdoor unit of the split air conditioner. The 4.0/). loss proportions were 11.8% in the wake area, 69.7% in the tip area, and 18.5% in the hub Machines 2023, 11, 155. https://doi.org/10.3390/machines11020155 https://www.mdpi.com/journal/machines Machines 2023, 11, 155 2 of 18 area. Furthermore, the effects of entrance hub geometry, tip clearance, winglets, and shroud height on axial flow fan aerodynamic performance were explored. The aerodynamic loss of the axial flow fan in the outdoor unit of the split air conditioner is more concentrated at the blade tip. The three-dimensional structure of the flow field within the outdoor unit of the split air conditioner was determined using a laser particle image velocimetry (PIV) [9]. The results showed a complex vortex flow field at the blade tip which began from the suction side of the blade tip and decreased in magnitude on the pressure side. The noise of the outdoor unit of the split air conditioner is composed of broadband and discrete frequency noise [10], in which broadband noise accounts for the main component. According to the vortex sound theory, the broadband noise is mainly caused by the vortices shedding at the blade’s trailing edge, while the discrete noise is caused by the tip vortices. Scholars have proposed various control technologies based on the understanding of the flow field characteristics and the aeroacoustic performance of the axial flow fans. Some scholars [11,12] installed a winglet on the axial flow fans, effectively reducing the overall noise level and changing the noise spectrum in the low-frequency range. Zhou et al. [13] designed a local convexity-preserving structure at the leading edge (LE), which reduced the noise by 1.1 dB while maintaining the performance of the axial flow fan. Park et al. [14] attached a fence on the shroud near the trailing edge of an axial flow fan, which blocked the reverse leakage flow near the shroud and weakened the movement of the tip leakage vortex (TLV) in the azimuth direction; this improved the fan’s efficiency. With the maturation of bionics, scholars increasingly applied the biological charac- teristics developed over millions of years of biological evolution to blade modification to obtain better aerodynamic and aeroacoustic performance. For instance, the bionic air- foil [15,16], serrated trailing edge [17,18], and blade tip winglet [19,20] were designed to imitate birds’ wings. Simultaneously, research on fish structures developed progressively, and the fish caudal fin is one of the hotspots. Fish caudal fins have evolved into different shapes, most of which are forked [21]. Other shapes include lunate, indented, rounded, and eel-like. Lighthill [22,23] developed a mathematical model and applied the theory of aerodynamics to study fish swimming. The quantitative analysis showed that the caudal fin with high aspect ratio can improve the propulsion efficiency. The results of Buren’s study [24] demonstrated that the concave trailing edge of a rigid pitching panel could delay the natural vortex bending and compression of the wake. Li et al. [25] used the multi-block Lattice Boltzmann Method (LBM) and Immersed Boundary (IB) method to investigate the locomotion performance of flapping plates with typical fish-like tail shapes. Numerical results showed that fish-like forked configurations had more thrust and better efficiency than unforked plates. Matta [26] used a biomimetic robotic tuna to investigate the influence of different caudal fin plane shapes on the thrust production and flow structures during swimming. In addition to generating the maximum thrust, the swept fin best stabilizes the leading-edge vortex that formed in the stroke’s second half. Low [27] investigated the influence of caudal fins with various design parameters on the thrust capability. The higher thrust generated at small sweep angle is attributed to the variation in spanwise flow and leading-edge vortex dynamics [28]. The trailing-edge modification was also applied to the axial flow fans by imitating the fish caudal fin. The noise reduction produced by the concaved trailing edge of the axial flow fan was 1.3 dB, measured in the anechoic chamber [29,30]. The optimized trailing-edge structure had minor pressure variation and pressure difference at the trailing edge, and the low-frequency sound pressure level was reduced [31]. The aerodynamic noise was the main focus of earlier studies on modifying the trailing edge of the axial flow fan. The present study introduces the bionic forked trailing edge into the axial flow fan. Its impact on the internal vortex structures, aerodynamic and aeroacoustic performance of the axial flow fan is thoroughly examined through numerical simulation and experimental testing. Machines 2023, 10, x FOR PEER REVIEW 3 of 18 into the axial flow fan. Its impact on the internal vortex structures, aerodynamic and aer- oacoustic performance of the axial flow fan is thoroughly examined through numerical simulation and experimental testing. Machines 2023, 11, 155 3 of 18 2. Geometric Model and Experimental Test 2.1. Axial Flow Fan Model The object of this study was an axial flow fan of the outdoor unit of the split air con- 2. Geometric Model and Experimental Test ditioner. Figure 1a shows that the outdoor unit contains an axial flow fan, rectification 2.1. Axial Flow Fan Model compressor, finned-tube heat exchanger, motor, motor support, partition board, shrouded The object of this study was an axial flow fan of the outdoor unit of the split air and conditioner air outlet .lo Figur uver, e 1 wi a shows th ovethat rall the dim outdoor ensions unit of 7 contains 00 × 280 an × 5 axial 10 m flow m. The fan, m rectification ain parameters compressor, finned-tube heat exchanger, motor, motor support, partition board, shrouded of the axial flow fan are shown in Figure 1b, and their corresponding values are provided and air outlet louver, with overall dimensions of 700 280 510 mm. The main parameters in Table 1. There are three end-bend blades on the axial flow fan. The hub and the blade of the axial flow fan are shown in Figure 1b, and their corresponding values are provided tip radius are 45mm and 211mm, separately. There is a non-uniform gap between the in Table 1. There are three end-bend blades on the axial flow fan. The hub and the blade tip shroud and the blade tip, with a minimum value of 7mm. Since the shroud only covers radius are 45mm and 211mm, separately. There is a non-uniform gap between the shroud the rear area of the rotor tip (70%~110% axial chord length), the air flows into the rotor and the blade tip, with a minimum value of 7mm. Since the shroud only covers the rear from axial and radial directions, respectively. The flow rate is 2336 m /h under the rated area of the rotor tip (70%~110% axial chord length), the air flows into the rotor from axial and radial directions, respectively. The flow rate is 2336 m /h under the rated operating operating condition, and the Reynolds number based on the blade tip radius Rtip is Re = condition, and the Reynolds number based on the blade tip radius R is Re = 6.7 10 . 6.7 × 10 . tip Figure 1. Geometric model of the entire machine: (a) Outdoor unit of the air conditioner; (b) An Figure 1. Geometric model of the entire machine: (a) Outdoor unit of the air conditioner; (b) An axial axial flow fan. flow fan. Table 1. Main parameters of the axial flow fan. Table 1. Main parameters of the axial flow fan. Parameter Value Parameter Value Number of blades 3 Number of blades 3 Blade tip radius, R 211mm Blade tip radius, Rtip 211mm tip Rated operating condition, Q 2336 m /h v,r Rated operating condition, Qv,r 2336 m /h Axial tip chord length, C /R 0.590 x tip Axial tip chord length, Cx/Rtip 0.590 Tip chord length, C /R 1.521 tip Hub chord length, C /R 0.394 Tip chord le hngt tip h, Ct/Rtip 1.521 Minimum tip celerance, t /R 0.033 gap tip Hub chord length, Ch/Rtip 0.394 Hub-to-tip ratio, R /R 0.213 hub tip Minimum tip celerance, tgap/Rtip 0.033 Reynolds number, Re 6.7 10 Hub-to-tip ratio, Rhub/Rtip 0.213 2.2. Bionic Forked T Rrailing eynolds Edge number, Re 6.7 × 10 Most fish species have developed forked caudal fins, which ameliorate their swimming efficiency and direction control, as shown in Figure 2a. Four control points (1, 2, 3, and 2.2. Bionic Forked Trailing Edge 4) are selected to define the bionic forked trailing edge to apply the forked fish caudal fin Most fish species have developed forked caudal fins, which ameliorate their swim- structure to the modification of the trailing edge of the axial flow fan. Figure 2b illustrates ming efficiency and direction control, as shown in Figure 2a. Four control points (1, 2, 3, the bionic forked trailing edge of the axial flow fan, where the black dotted line represents and 4) are selected to define the bionic forked trailing edge to apply the forked fish caudal the original trailing edge (OTE), and the solid blue line represents the modified bionic fin structure to the modification of the trailing edge of the axial flow fan. Figure 2b illus- forked trailing edge (BFT). Control points 1, 2, 3, and 4 are located at radii of 0.92R , tip 0.61R , 0.28R , and 0.61R , respectively, and control points 1, 4, and 3 are on the original trates the bionic forked trailing edge of the axial flow fan, where the black dotted line tip tip tip trailing edge. The shape of the bionic forked trailing edge is determined by the angle represents the original trailing edge (OTE), and the solid blue line represents the modified between control points 2 and 4, and the control points 1, 2, and 3 are connected through a Machines 2023, 10, x FOR PEER REVIEW 4 of 18 bionic forked trailing edge (BFT). Control points 1, 2, 3, and 4 are located at radii of 0.92Rtip, 0.61Rtip, 0.28Rtip, and 0.61Rtip, respectively, and control points 1, 4, and 3 are on the original Machines 2023, 11, 155 4 of 18 trailing edge. The shape of the bionic forked trailing edge is determined by the angle θt between control points 2 and 4, and the control points 1, 2, and 3 are connected through a smooth curve to obtain the modified bionic forked trailing edge. From this point onwards, smooth curve to obtain the modified bionic forked trailing edge. From this point onwards, we refer to the fans with and without bionic forked trailing edges as BFT and prototype we refer to the fans with and without bionic forked trailing edges as BFT and prototype fans, fans r,espectively respectively. . Figure 2. Schematic diagram of the bionic forked trailing edge: (a) The forked fish caudal fin; (b) Figure 2. Schematic diagram of the bionic forked trailing edge: (a) The forked fish caudal fin; (b) The The blade with the bionic forked trailing edge. blade with the bionic forked trailing edge. 2.3. Experimental Setup 2.3. Experimental Setup The aerodynamic test of the fan is executed according to the international standard The aerodynamic test of the fan is executed according to the international standard ISO 5801-2017. The performance of the fans is tested using standardized airways. The fan aerodynamic test rig is composed of a chamber, manometer, multiple nozzles, flow ISO 5801-2017. The performance of the fans is tested using standardized airways. The fan setting grids, and an auxiliary fan, as shown in Figure 3a. The air conditioner outdoor aerodynamic test rig is composed of a chamber, manometer, multiple nozzles, flow setting unit with an axial fan is installed on the wall of the test rig to simulate the inlet and outlet grids, and an auxiliary fan, as shown in Figure 3a. The air conditioner outdoor unit with boundary conditions during its actual operation. The motor ’s rotational speed determines Machines 2023, 10, x FOR PEER REVIEW 5 of 18 an axial fan is installed on the wall of the test rig to simulate the inlet and outlet boundary the working state of the fan. The static pressure is sensed by four manometers evenly conditions during its actual operation. The motor’s rotational speed determines the work- spaced along the wall of the wind chamber at the fan’s outlet. ing state of the fan. The static pressure is sensed by four manometers evenly spaced along the wall of the wind chamber at the fan’s outlet. The volume flow rate Qv and shaft power W0 of the test fan are calculated based on the experimental data by the following equations: 2ΔP 2 m Q=α d v m m 4 ρ (1) W = UIη 0 motor where the parameters αm, dm, ∆Pm, ρ, U, I, and ηmotor are the coefficients of the multiple nozzles, the diameter of the multiple nozzles, the static pressure difference between two sides of multi-nozzles, air density, input voltage, current, and efficiency of the motor, re- spectively. Figure 3. Experimental setup for aerodynamic performance: (a) Schematic diagram of the fan aero- Figure 3. Experimental setup for aerodynamic performance: (a) Schematic diagram of the fan dynamic test rig; (b) Photograph of the experimental facility. aerodynamic test rig; (b) Photograph of the experimental facility. The international standard ISO 3745-2012 entitled “Acoustics—Determination of sound power levels and sound energy levels of noise sources using sound pressure—Pre- cision methods for anechoic rooms and hemi-anechoic rooms“ constitutes the standard for measuring the aeroacoustics of the test fan. The aeroacoustic performance is measured in a semi-anechoic chamber of dimensions 5.6 × 4.8 × 3.6 m. As shown in Figure 4, the microphone is placed at the test point (MP), one meter away from the floor and the axial flow fan outlet (FO), respectively. Concurrently, the angle between the line connecting the microphone to the outlet of the axial flow fan (FO) and the fan rotation axis is 45°. Figure 4. Experimental setup for the aeroacoustic performance measurement in a semi-anechoic chamber. 3. Numerical Method 3.1. Governing Equations and Turbulence Modeling The highly reliable and robust commercial solver ANSYS FLUENT is selected as the numerical simulation tool. The large-eddy simulation (LES) models the turbulent flow. LES is a turbulence simulation method intermediate between Direct Numerical Simula- tion (DNS) and Reynolds-Averaged Navier–Stokes (RANS) turbulence modeling. The basic idea of LES is to filter the small turbulent scales in the flow field and directly solve the large-scale vortices within the Kolmogorov spectrum (“−5/3 law”) range of the energy Machines 2023, 10, x FOR PEER REVIEW 5 of 18 Machines 2023, 11, 155 5 of 18 The volume flow rate Q and shaft power W of the test fan are calculated based on v 0 the experimental data by the following equations: ( q p 2 2DP Q = a d v m (1) W = UI motor where the parameters , d , DP , , U, I, and are the coefficients of the mul- m m m motor tiple nozzles, the diameter of the multiple nozzles, the static pressure difference be- Figure 3. Experimental setup for aerodynamic performance: (a) Schematic diagram of the fan aero- tween two sides of multi-nozzles, air density, input voltage, current, and efficiency of dynamic test rig; (b) Photograph of the experimental facility. the motor, respectively. The international standard ISO 3745-2012 entitled “Acoustics—Determination of The international standard ISO 3745-2012 entitled “Acoustics—Determination of sound power levels and sound energy levels of noise sources using sound pressure— sound power levels and sound energy levels of noise sources using sound pressure—Pre- Precision methods for anechoic rooms and hemi-anechoic rooms“ constitutes the standard cision methods for anechoic rooms and hemi-anechoic rooms“ constitutes the standard for measuring the aeroacoustics of the test fan. The aeroacoustic performance is measured for measuring the aeroacoustics of the test fan. The aeroacoustic performance is measured in a semi-anechoic chamber of dimensions 5.6 4.8 3.6 m. As shown in Figure 4, the in a semi-anechoic chamber of dimensions 5.6 × 4.8 × 3.6 m. As shown in Figure 4, the microphone is placed at the test point (MP), one meter away from the floor and the axial microphone is placed at the test point (MP), one meter away from the floor and the axial flow fan outlet (FO), respectively. Concurrently, the angle between the line connecting the flow fan outlet (FO), respectively. Concurrently, the angle between the line connecting the microphone to the outlet of the axial flow fan (FO) and the fan rotation axis is 45 . microphone to the outlet of the axial flow fan (FO) and the fan rotation axis is 45°. Figure 4. Experimental setup for the aeroacoustic performance measurement in a semi-anechoic Figure 4. Experimental setup for the aeroacoustic performance measurement in a semi-anechoic chamber. chamber. 3. 3.Numerical NumericalMethod Method 3.1. Governing Equations and Turbulence Modeling 3.1. Governing Equations and Turbulence Modeling The highly reliable and robust commercial solver ANSYS FLUENT is selected as the The highly reliable and robust commercial solver ANSYS FLUENT is selected as the numerical simulation tool. The large-eddy simulation (LES) models the turbulent flow. numerical simulation tool. The large-eddy simulation (LES) models the turbulent flow. LES is a turbulence simulation method intermediate between Direct Numerical Simulation LES is a turbulence simulation method intermediate between Direct Numerical Simula- (DNS) and Reynolds-Averaged Navier–Stokes (RANS) turbulence modeling. The basic idea tion (DNS) and Reynolds-Averaged Navier–Stokes (RANS) turbulence modeling. The of LES is to filter the small turbulent scales in the flow field and directly solve the large-scale basic idea of LES is to filter the small turbulent scales in the flow field and directly solve vortices within the Kolmogorov spectrum (“