阅读说明：本技术 一种用于形成翼型的附加部件的方法 (Method for forming additional part of airfoil ) 是由 张子培 菲利普·查尔斯·伍德海德 于 2020-05-14 设计创作，主要内容包括：本发明公开了一种用于形成翼型的附加部件的方法,能够调整翼型的结构,以降低当空气沿从翼型前缘流过翼型后缘的流动方向流动时在频率f-(峰值)处产生的声音的振幅。该方法适用于具有狭缝构造和双根部后缘锯齿(缩写为“DRooTES”)的附加部件。(A method for forming an additional component of an airfoil is disclosed, the configuration of the airfoil being adjustable to reduce the frequency f when air flows in a flow direction from the leading edge of the airfoil over the trailing edge of the airfoil Peak value The amplitude of the sound produced. The method is applicable to add-on components having a slit configuration and dual root trailing edge serrations (abbreviated as "drooeses").)

1. A method for forming an additional component of an airfoil having a leading edge and a trailing edge, the component being formed to reduce the velocity U of the air as it flows in a freestream_∞At a frequency f when flowing in a flow direction from a leading edge of the airfoil over a trailing edge of the component_{Peak value}The amplitude of the generated sound, the method comprising the steps of:

(a) selecting the frequency f of the sound to be reduced_{Peak value}；

(b) Selecting the free flow speed U of the air_∞；

(c) Providing a component having a joining edge for joining to an airfoil trailing edge, and a trailing edge opposite the joining edge;

(d) forming the trailing edge of the component into a plurality of pairs of peaks, each of the pairs having a first valley, a first peak, a second valley, a second peak at a greater distance from the engagement edge than the first peak, and a third valley, wherein the first valley is located on one side of the first peak, the second valley is located between the first peak and the second peak, and the third valley is located on the other side of the second peak opposite the second valley;

(e) wherein the first and third troughs of each pair of peaks lie substantially on a first axis, the second trough of each pair of peaks lies substantially on a second axis, and the second peak of each pair of peaks lies on a third axis;

(f) and wherein the component is formed according to the following formula:

whereinIn the range of 0.4 to 0.8, and

wherein h 'is the shortest distance between the first axis and the second axis, h' is the shortest distance between the second axis and the third axis, f_{Peak value}Is the frequency of the sound to be reduced, and U_∞Is the free stream velocity of air in the flow direction over the trailing edge of the component.

2. The method of claim 1, wherein h ═ h ".

3. The method of claim 1 or 2, wherein for each of the plurality of pairs of peaks, a shortest distance from the first peak to the first axis is the same.

4. The method of any preceding claim, wherein the third valley of a first pair of peaks is the first valley of a second pair of peaks adjacent to the first pair of peaks.

5. A method for reducing the velocity U of air in a free stream_∞At a frequency f when flowing in a flow direction from a leading edge of an airfoil past a trailing edge of said airfoil_{Peak value}A method of processing the amplitude of a generated sound, wherein the method comprises the steps of:

(a) forming an additional component for an airfoil according to the method of any one of claims 1 to 4, and

(b) attaching the component to the airfoil.

6. The method according to claim 5, wherein the method comprises the steps of: forming a plurality of components and attaching the plurality of components to the airfoil.

7. The method of claim 6, wherein three components are formed and attached to the airfoil; wherein the first component is proximate to a tip of the airfoil, the second component is proximate to a middle of the airfoil, and the third component is proximate to another end of the airfoil opposite the tip.

8. The method of claim 6 or 7, wherein the components are identical.

9. The method of claim 6 or 7, wherein the components have different values of h' and/or h ".

10. A method for forming an additional component of an airfoil having a leading edge and a trailing edge, the component being formed to reduce the velocity U of the air as it flows in a freestream_∞At a frequency f when flowing in a flow direction from a leading edge of the airfoil over a trailing edge of the component_{Peak value}The amplitude of the generated sound, the method comprising the steps of:

(a) selecting the frequency f of the sound to be reduced_{Peak value}；

(b) Selecting the free flow speed U of the air_∞；

(c) Providing a component having a joining edge for joining to an airfoil trailing edge, and a trailing edge opposite the joining edge;

(d) forming a trailing edge of the component as a plurality of peaks separated by a plurality of troughs, wherein each peak is connected to each trough on either side of the peak by a wall generally perpendicular to the trailing edge, the plurality of peaks each having a width (a) generally parallel to the trailing edge, the plurality of troughs each having a width (W) generally parallel to the trailing edge, the width (W) also being a gap between adjacent peaks, wherein the plurality of peaks lie generally on a first axis and the plurality of troughs lie generally on a second axis, and wherein the shortest distance from the first axis to the second axis is defined as H;

(e) and wherein the component is formed according to the following formula:

whereinIn the range of 0.4 to 0.8.

11. The method of claim 10, wherein the gaps (W) between adjacent peaks are related to a spanwise correlation length scale (L) of turbulent eddies_y) The ratio of (A) satisfies the expression 0.2 ≤ W/L_y≤0.5。

12. A method for reducing the velocity U of air in a free stream_∞At a frequency f when flowing in a flow direction from a leading edge of an airfoil past a trailing edge of said airfoil_{Peak value}A method of processing the amplitude of a generated sound, the method comprising the steps of:

(a) forming an additional part of an airfoil according to the method of claim 10 or 11, and

(b) attaching the component to the airfoil.

13. The method according to claim 12, wherein the method comprises the steps of: forming a plurality of components and attaching the plurality of components to the airfoil.

14. The method of claim 13, wherein three components are formed and attached to the airfoil, wherein a first component is near a tip of the airfoil, a second component is near a middle of the airfoil, and a third component is near another end of the airfoil opposite the tip.

15. The method of claim 13 or 14, wherein the components are identical.

16. The method of claim 13 or 14, wherein the components have different H values.

Background

The present application relates to a method for forming an additional component of an airfoil, and for reducing the frequency f when air flows in a flow direction from an airfoil leading edge over an airfoil trailing edge_{Peak value}A method of processing the amplitude of the generated sound.

It is well known that harmful noise is essentially a form of health-affecting pollution, especially in the vicinity of airports and wind farms. Research shows that aviation and wind power field noise pollution can generate adverse effects on health, and the adverse effects can cause the harms such as dysphoria, sleep disorder, cognitive disorder and the like. In order to protect the health of the public, strict noise regulations are implemented in aviation and wind energy departments. The european union committee has set forth an aviation goal by the year 2050 that airborne noise emissions will be reduced by 65% over 2000. However, global civil aviation traffic is expected to increase by 4% to 5% annually, with even greater growth in the middle east and asian regions. Additionally, the european wind energy institute issued a report on the projected energy capacity scenario of 2030 in 2014, which indicated that 320 gigawatts of electricity will be produced from wind energy, 78% of which are produced by land wind farms.

The rapid development of the aviation industry and the significant increase in the number of wind farms on land in the European Union (EU) member countries will inevitably lead to more aerodynamic noise pollution generation. Aircraft noise is primarily generated by jet engines and lift devices during takeoff and landing. With the development of jet engine technology, jet noise is reduced, but noise generated by the high bypass ratio fan blades is increased.

Wind turbine mechanical noise is mainly caused by moving parts inside the gearbox and generator.

However, recent efforts at high precision gear tooth profile design and outer shell sound insulation have reduced the source of mechanical noise. In contrast, aerodynamic noise generated by wind turbine blades is more difficult to reduce. To protect local communities, the maximum noise level at 350m from the wind turbine is currently limited to 35-45dB (A). Therefore, wind turbine companies typically reduce the turbine speed in high wind speed scenarios to suppress aerodynamic noise levels from exceeding limits. However, this will substantially reduce the availability of the available wind energy resources. Thus, reducing noise without reducing rotor speed would make wind energy less expensive and a more attractive alternative to fossil energy. Quantitatively, a further 1dB reduction in noise at the current level can result in a significant increase in wind energy production.

Aerodynamic noise may be generated at the leading and trailing edges of an airfoil. Airfoil noise is generated at the leading edge of the blade by interaction with atmospheric turbulence, or turbulence in the boundary layer is formed at the blade surface and scattered as sound, thereby being generated at the trailing edge. Leading edge noise and trailing edge noise are different noise sources that are uncorrelated with each other. Trailing edge noise (also commonly referred to as self-noise) remains one of the most relevant noise sources for the aviation and wind turbine industries.

It is known from the prior art that trailing edge noise reduction can be achieved by using a simple serrated trailing edge. First, the noise reduction achieved by the sawtooth trailing edge is related in part to the destructive interference caused by the phase lag mechanism of turbulent noise scattering on the sloped edge. Second, the root of the sawtooth trailing edge is effective for turbulent noise radiation. Third, two noise sources with a phase angle longitudinally displaced by 180 degrees, such as in the case of leading edge slots, have been shown to minimize the destructive interference of aerodynamic noise.

CN 109292076 Al (harabin industry university) discloses a low self-noise airfoil structure with a multi-wavelength sawtooth trailing edge. The airfoil body is detachably connected with the noise reduction trailing edge plate through a connecting piece.

WO 2017/220594 Al (LM wind power) discloses a wind turbine blade comprising two or more serrations provided along a portion of the trailing edge.

US 2012/027590 Al (bonbon) discloses a rotor blade assembly and a method for reducing noise of a wind turbine rotor blade.

WO 2019/158876 Al (seiko aircraft engine) discloses an air profiled structure comprising a body and a porous sound-absorbing region.

US 2017/0174320 (amazon technologies) discloses a system, method and apparatus for actively adjusting the position of one or more propeller blade handling members of an aircraft propeller blade during operation of the aircraft.

US 2017/0022820 a1 (rales reus corporation) discloses an airfoil for a gas turbine having a series of first and second grooves cut into the leading edge of the airfoil itself. The grooves cut to reduce airfoil leading edge noise are as follows:

f₀＝U/2h_tt

wherein f is₀Is the reduced noise frequency, U is the velocity of the air flowing over the leading edge, and h_ttIs the difference in height between adjacent valleys of the first and second grooves.

Gruber, Azarpeyvand and Joseph compared the trailing edge self-noise reduction measurements obtained using sawtooth and slot profiles on a NACA651210 airfoil in "airfoil trailing edge noise reduction by introducing sawtooth and slot trailing edge geometries" (20 th international acoustic congress, ICA 2010).

Disclosure of Invention

The inventive realization of the present application is that for an incorrect equation disclosed in patent document US 2017/0022820 a1, a more accurate equation as defined in claim 1 is proposed. Furthermore, the solution of constructing a separate part for the airfoil and attaching it to the trailing edge of the airfoil according to the equations of the present invention may obtain improved results compared to the solution of cutting grooves into the airfoil itself as disclosed in US patent document US 2017/0022820 a 1.

Furthermore, equivalent inventive equations may be used to adjust the noise reduction obtainable from airfoils with slit profiles (as disclosed in Gruber, Azarpeyvand and Joseph, above).

A first aspect of the invention provides a method for forming an additional component of an airfoil having a leading edge and a trailing edge, the component being formed to reduce the velocity U of the air as it flows in a freestream_∞At frequency f when flowing in the direction of flow from the leading edge of the airfoil over the trailing edge of the component_{Peak value}The amplitude of the generated sound, the method comprising the steps of:

(a) selecting the frequency f of the sound to be reduced_{Peak value}；

(b) Selecting the free flow speed U of the air_∞；

(c) Providing a component having a joining edge for joining to the airfoil trailing edge, and a trailing edge opposite the joining edge;

(d) forming the trailing edge of the component into a plurality of pairs of peaks, each pair having a first valley, a first peak, a second valley, a second peak spaced a greater distance from the joint edge than the first peak, and a third valley, wherein the first valley is located on one side of the first peak, the second valley is located between the first peak and the second peak, and the third valley is located on the other side of the second peak opposite the second valley;

(e) wherein the first and third troughs of each pair of peaks lie substantially on a first axis, the second trough of each pair of peaks lies substantially on a second axis, and the second peak of each pair of peaks lies on a third axis; and the number of the first and second groups,

(f) wherein the component is formed according to the following formula:

whereinIn the range of 0.4 to 0.8, and

wherein h 'is the shortest distance between the first axis and the second axis, h' is the shortest distance between the second axis and the third axis, f_{Peak value}Is the frequency of the sound to be reduced, and U ∞ is the free flow velocity of the air in the flow direction over the trailing edge of the component.

In one embodiment, h' ═ h ".

For each of the pairs of peaks, a shortest distance of the first peak to the first axis may be the same.

The third valley of the first pair of peaks may be a first valley of the second pair of peaks adjacent to the first pair of peaks.

A method for reducing the velocity U of air in a free stream_∞At a frequency f when flowing in a flow direction from the leading edge of the airfoil over the trailing edge of the airfoil_{Peak value}A method of processing the amplitude of a generated sound, the method comprising the steps of:

(a) forming additional components of an airfoil according to the above method, and

(b) attaching the component to the airfoil.

The method may additionally comprise the steps of: forming a plurality of components and attaching the plurality of components to the airfoil. For example, three components may be formed and attached to the airfoil; wherein the first part is near the tip of the airfoil, the second part is near the middle of the airfoil, and the third part is near the other end of the airfoil opposite the tip.

In one embodiment, the components may be identical. However, in another alternative embodiment, the components may have different values of h' and/or h ".

A second aspect of the invention provides a method for forming an additional component of an airfoil having a leading edge and a trailing edge, the component being formed to reduce the velocity U of the air as a free stream. . At frequency f when flowing in the direction of flow from the leading edge of the airfoil over the trailing edge of the component_{Peak value}The amplitude of the generated sound, the method comprising the steps of:

(a) selecting the frequency f of the sound to be reduced_{Peak value}；

(b) The free flow velocity U of the air is selected. . (ii) a

(c) Providing a component having a joining edge for joining to the airfoil trailing edge, and a trailing edge opposite the joining edge;

(d) forming a trailing edge of the component as a plurality of peaks separated by a plurality of troughs, wherein each peak is connected to each trough on either side of the peak by a wall generally perpendicular to the trailing edge, each peak having a width (a) generally parallel to the trailing edge, each trough having a width (W) generally parallel to the trailing edge, the widths (W) also being the gaps between adjacent peaks, the plurality of peaks being generally located on a first axis, the plurality of troughs being generally located on a second axis, the shortest distance from the first axis to the second axis being defined as H; and

(e) wherein the component is formed according to the following formula:

whereinAt 0.In the range of 4 to 0.8.

In one embodiment, the gap (W) between adjacent peaks is related to the spanwise dependent length dimension (L) of the turbulent eddies_y) The ratio of (A) satisfies the expression 0.2 ≤ W/L_y≤0.5。

A method for reducing the velocity U of air in a free stream_∞At a frequency f when flowing in a flow direction from the leading edge of the airfoil over the trailing edge of the airfoil_{Peak value}A method of processing the amplitude of a generated sound, the method comprising the steps of:

(a) forming additional components of an airfoil according to the above method, and

(b) attaching the component to the airfoil.

The method may additionally comprise the steps of: forming a plurality of components and attaching the plurality of components to the airfoil. For example, three components may be formed and attached to the airfoil; wherein the first part is near the tip of the airfoil, the second part is near the middle of the airfoil, and the third part is near the other end of the airfoil opposite the tip.

In one embodiment, the components may be identical. However, in an alternative embodiment, the components may have different values of H.

In yet another aspect of the invention, a method is provided for forming an additional component of an airfoil having a leading edge and a trailing edge, the component being formed to reduce the frequency f when air flows at a flow velocity U in a flow direction from the leading edge of the airfoil over the trailing edge of the component_{Peak value}The amplitude of the generated sound, the method comprising the steps of:

(a) selecting the frequency f of the sound to be reduced_{Peak value}；

(b) Selecting a flow speed U of air;

(c) providing a component having a joining edge for joining to the airfoil trailing edge, and a trailing edge opposite the joining edge;

(d) forming the trailing edge of the component into a plurality of pairs of peaks, each pair of peaks having a first valley, a first peak, a second valley, a second peak at a greater distance from the joint edge than the first peak, and a third valley, wherein the first valley is located on one side of the first peak, the second valley is located between the first peak and the second peak, and the third valley is located on the other side of the second peak opposite the second valley;

(e) wherein the first and third troughs of each pair of peaks lie substantially on a first axis, the second trough of each pair of peaks lies substantially on a second axis, and the second peak of each pair of peaks lies on a third axis; and the number of the first and second groups,

(f) wherein the component is formed according to the following formula:

h′＝Uκ/4f_{peak value}Wherein

Wherein h 'is the shortest distance between the first axis and the second axis, h' is the shortest distance between the second axis and the third axis, f_{Peak value}Is the frequency of the sound to be reduced and U is the flow velocity of the air in the direction of flow over the trailing edge of the component.

Yet another aspect of the present invention is directed to a method of reducing the frequency f when air flows at a flow velocity U in a flow direction from an airfoil leading edge over an airfoil trailing edge_{Peak value}A method of processing the amplitude of a generated sound, the method comprising the steps of: providing at least one component formed according to the above method and attaching the at least one component to the airfoil.

In yet another aspect, the present invention provides an additional component for an airfoil having a leading edge and a trailing edge, the component being formed to reduce the frequency f when air flows at a flow velocity U in a flow direction from the leading edge of the airfoil over the trailing edge of the component_{Peak value}The amplitude of the generated sound, the component comprising:

a joining edge for joining to an airfoil trailing edge;

a trailing edge opposite the joint edge; and

a surface between the joint edge and the trailing edge;

wherein the trailing edge of the component is formed as a plurality of pairs of peaks, each pair of peaks having a first valley, a first peak, a second valley, a second peak at a greater distance from the joint edge than the first peak, and a third valley, wherein the first valley is located on one side of the first peak, the second valley is located between the first peak and the second peak, and the third valley is located on the other side of the second peak opposite the second valley;

wherein the first and third troughs of each pair of peaks lie substantially on a first axis, the second trough of each pair of peaks lies substantially on a second axis, and the second peak of each pair of peaks lies on a third axis;

wherein h 'is the shortest distance between the first axis and the second axis, and h' is the shortest distance between the second axis and the third axis; and

wherein the maximum thickness of the component in a direction perpendicular to the surface is 0.5mm to 10 mm.

Preferred embodiments of the present invention are described below with reference to the accompanying drawings, in which:

FIG. 1A is a schematic representation of the geometric parameters of SRooTES, DRooTES and slit trailing edge attachments;

FIG. 1B is a schematic illustration of a drooeses and a slit trailing edge attachment for carrying out the method according to the invention;

FIG. 1C is a schematic diagram of an experimental setup for far-field noise measurement as described in the present invention;

FIG. 2 depicts U at 20m/s ≦ U_∞Δ PWL, dB performance comparisons for slot trailing edges with different slot amplitudes H, in the range ≦ 60m/s, where λ is 3 mm;

FIG. 3 depicts U at 20m/s ≦ U_∞In the range of ≦ 60m/s, a comparison of dimensionless frequencies with different slit amplitudes H, wherein λ remains 3mm, W remains 0.3mm, and

FIG. 4 depicts at U_∞30m/s and U_∞Δ PWL for slit trailing edges with different slit wavelengths λ (mm), dB performance comparison, the slit root width W was kept at 0.3mm, and

FIG. 5 depicts U at 20m/s ≦ U_∞Δ PWL, dB performance comparison of baseline and trailing edge of slit (W/λ different but λ the same) in the range ≦ 60m/s, where H is 15mm and λ is 3 mm;

FIG. 6 depicts U at 20m/s ≦ U_∞Δ PWL, dB performance comparison of baseline and droots (with different h') trailing edges in the range ≦ 60m/s, where λ is 3 mm;

FIG. 7 depicts a comparison of dimensionless frequencies of DRooTES with different amplitudes H, where λ is 6mm,

FIG. 8 depicts a comparison of dimensionless frequencies of DRooTES with different root-to-root amplitudes h 'in the range of 20m/s ≦ U ∞ ≦ 60m/s, when h' h ", i.e., k 1, where λ 6mm,

FIG. 9 depicts U at 20m/s ≦ U_∞In the range of less than or equal to 60m/s, comparing APWL and dB of SRooTES, DRooTES and the slit trailing edge, wherein the amplitude H and the wavelength lambda are respectively kept at 15mm and 3 mm; and

FIG. 10 depicts U at 20m/s ≦ U_∞In the range ≦ 60m/s, APWL, dB comparisons for SRooTES, DRooTES and slit trailing edge, where amplitude H and wavelength λ were maintained at 30mm and 3mm, respectively.

In the present invention, the additional part of the airfoil for implementing the method according to the invention, called double root trailing edge serrations (abbreviated as "drooeses"), makes it possible not only to improve the level of reduction of the trailing edge noise, but also to fine-tune the frequencies of interest so as to reduce the noise itself. Drooes takes a nomenclature that is analogous to a standard serration, which is referred to herein as "SRooTES" (single section trailing edge serration).

Naming mode

frequency, hertz (Hz)

H amplitude, meter (m)

λ ═ wavelength, m

h' ═ root to root longitudinal displacement, millimeters (mm)

h ″, root to tip longitudinal displacement, mm

Phi is the angle of the secondary sawtooth tip

λ₀Lateral displacement from root to mid-root, mm

h is half amplitude, m

c₀Length of chord, m

w is the slit width at the root, mm

a is the slit width at the tip, mm

AoA, theta ═ angle of attack

Theta is polar angle, degree

U_∞Speed of free flow, meters per second (m/s)

x-position on the airfoil

n-convection velocity factor

Factor for pressure driven turbulent vortex propagation

PWL ═ sound power level, decibel (dB)

Δ PWL is the difference in acoustic power level, dB

OAPWL ═ total acoustic power level, dB

L_ySpan-wise dependent length scale of turbulent eddies

Theory of science

The invention focuses on the characteristics and mechanism of the drooeses and the acoustic interference generated by the slit trailing edge on the trailing edge noise of the turbulent boundary layer. The basic theory of this work is based on the interference of wave theory. Interference has two forms: constructive and destructive. Theoretically, perfect destructive interference occurs when the acoustic radiation from two acoustic sources (S1 and S2 as shown in fig. 1B) is 180 ° (pi) out of phase. The associated phase difference (or phase angle) may be expressed as n pi, where n is 1, 3, 5, etc. Destructive interference causes the acoustic emissions to cancel each other out. Conversely, when the acoustic radiation between the two sources is in phase (i.e. when n is 2, 4, 6, etc.), perfect constructive interference occurs, resulting in amplification of the acoustic radiation far enoughA field. As a general term, the following equation is defined to determine the phase angle

Where ω is 2 π f equation 1

Where ω is the oscillation frequency, f is the frequency, 1 is the longitudinal displacement between the two sources, and U_∞Is the free stream velocity. Therefore, applying the "out-of-phase" angle (destructive interference) to equation 1, based on the distance between S1 and S2 (scattering sources), defined as H' in the DROOTES case shown in FIG. 1A, and H in the slit trailing edge case, the following equation can be derived:

where St is the strouhal number and represents the corresponding value of dimensionless frequency at which destructive interference occurs at odd numbers and constructive interference occurs at even numbers.

Details of trailing edge parameters

Abbreviations used herein are as follows: slits (Slits) and double-root serrations (drooes).

Experimental setup

NACA65- (12)10 arc airfoil chord length c 0-0.145 m to c 0-0.170 m for the castellated trailing edge case, and (non-castellated) chord length c 0-0.1425 m to c 0-0.155 m for the baseline (B) case, and spanwise length 0.45 m. To ensure similar wetted surface area, different chord lengths are used in the baseline and profile cases. The chord length in the case of the non-serrated trailing edge (baseline) is half the chord length of the serrated trailing edge in the case of Slits and drooes. It should be noted that SRooTES basically represents a simple sawtooth trailing edge. The airfoil is made up of two main parts: a main airfoil body and a detachable flat trailing edge. The main airfoil was made of aluminum alloy, and had surface pressure holes in both the upper and lower surfaces and a 0.8mm slot along the trailing edge. A 0.8mm thick knock-out plate was cut by laser to form different trailing edge shapes.

Error! Are not valid self-referencing bookmarks. A shows the geometrical parameters for the case of the trailing edge slab. These parameters are defined as the serration amplitude (H), the serration wavelength (λ), the longitudinal root 1 to root 2 displacement (H '), the longitudinal root 2 to tip 2 displacement (H'), the angle of the serration tip (φ), and the transverse serration root displacement (λ `)₀). Unless otherwise stated, the longitudinal displacement (h') of the root 1 to the root 2 and the transverse displacement (λ) of the root 1 to the root 2₀) Half the amplitude (H) and half the wavelength (λ), respectively. For the slit case, the geometric parameters are slit amplitude (also defined as H), slit wavelength (also defined as λ), width of the sliding tip (a), and width of the slit gap (W). The present invention investigates four situations: baseline, SRooTES, droootes, and Slit. Fig. 1B compares the drooeses and Slit geometries in more detail and will be referred to below.

A 0.8mm thick flat trailing edge can be inserted into a 0.8mm slot along the trailing end of the main airfoil body. The sawtooth amplitude (H) and the wavelength (lambda) of the Slit, the SRooTES and the DRooTES are respectively the value of every 5mm at intervals within the range of H being more than or equal to 5mm and less than or equal to 30mm, and the value of lambda being more than or equal to 3mm and less than or equal to 35 mm. The transverse displacement from the root 1 to the root 2 of the DRooTES and the angle of the sawtooth tip are respectively 1.5mm and lambda₀The value is taken at intervals of 1.5mm within the range of not more than 4.5mm, and the value is within the range of phi not less than 0 degrees and not more than 84.3 degrees. The baseline trailing edge is half the amplitude in the castellated case to ensure similar wetted surface area as SRooTES and droots. To facilitate comparing SRooTES and droootes to the slip trailing edge, the baseline trailing edge is also assumed to be half the amplitude. Coarse sandpaper was used for long periods of time on the upper and lower surfaces at 30mm from the leading edge of the airfoil to ensure that the boundary layer was fully brought into turbulent conditions. The width of abrasive paper strip is 10mm, and the thickness is 0.95 mm.

Wind tunnel equipment and instrument

Wind tunnel devices and instruments for far-field noise measurements are arranged in the aerodynamic acoustic installation of the university of brunell, which consists of an open jet wind tunnel placed in a sound-damping chamber of 4m × 5m × 3.4 m. The size of the open nozzle was 0.3m × 0.1m (width × height). The airfoil is attached to a side plate flush with the nozzle lip. The operation capacity of the open jet wind tunnel is up to 80m/s, and the typical low turbulence is between 0.1% and 0.2%. The background noise (no airfoil, but side panels) is primarily open jet noise, which is much less than the airfoil self-noise produced at the same flow rate. All far field noise measurements were made at a geometric angle of attack (AoA) θ of 0 °.

Far field array noise measurement

Far-field array noise measurements were made with 8 g.r.a.s. half inch condenser microphones (46AE) mounted at polar angles ranging from Θ to 50 ° to Θ to 120 ° every 10 °. The microphone at 90 ° is located 0.97m above the mid-span and trailing edge of the airfoil. Fig. 1C shows an experimental arrangement of a far field array. Each microphone signal is applied with a gain of ± 20dB by a g.r.a.s.12ax4 channel constant current source (CCP) amplifier. Data was collected using a 16-bit analog-to-digital card manufactured by national instruments ltd. The sampling frequency was 40kHz and the sampling time was 20 seconds. The data is windowed and the Power Spectral Density (PSD) of the 1Hz bandwidth is calculated from the 1024 point Fast Fourier Transform (FFT) and the 50% overlap time. Assuming that the sound wave propagates spherically from the trailing edge, the noise is calculated to obtain the sound power level (PWL). In the range of U-20 m/s to U-60 m/s, the free flow velocity takes a plurality of values at intervals of U-10 m/s to measure the noise value.

A shows the geometrical parameters for the case of the trailing edge slab. For DRooTES, the geometric parameters are defined as the serration amplitude (H), the serration wavelength (λ), the longitudinal displacement (H ') of the root 1 to the root 2, the longitudinal displacement (H') of the root 2 to the tip 2, the angle (φ) of the serration tip, and the transverse displacement (λ) of the serration root₀). Unless otherwise stated, the longitudinal displacement (h') of the root 1 to the root 2 and the transverse displacement (λ) of the root 1 to the root 2₀) Typically half the amplitude (H) and wavelength (λ), respectively. For slitsThe geometric parameters are the slit amplitude (also defined as H), the slit wavelength (also defined as λ), the width of the sliding tip (a) and the width of the slit gap (W).

Experimental setup

NACA65- (12)10 arc airfoil chord length c 0-0.145 m to c 0-0.170 m for the castellated trailing edge case, and (non-castellated) chord length c 0-0.1425 m to c 0-0.155 m for the baseline (B) case, and spanwise length 0.45 m. To ensure similar wetted surface area, different chord lengths are used in the baseline and profile cases. The chord length in the case of the non-serrated trailing edge (baseline) is half the chord length of the serrated trailing edge in the case of Slits and drooes. It should be noted that SRooTES basically represents a simple sawtooth trailing edge. The airfoil is made up of two main parts: a main airfoil body and a detachable flat trailing edge. The main airfoil was made of aluminum alloy, and had surface pressure holes in both the upper and lower surfaces and a 0.8mm slot along the trailing edge. A 0.8mm thick knock-out plate was cut by laser to form different trailing edge shapes.

Error! Are not valid self-referencing bookmarks. A shows the geometrical parameters for the case of the trailing edge slab. These parameters are defined as the serration amplitude (H), the serration wavelength (λ), the longitudinal root 1 to root 2 displacement (H'), the longitudinal root 2 to tip 2 displacement (H "), the angle of the serration tip (φ), and the transverse serration root displacement (λ [)₀). Unless otherwise stated, the longitudinal displacement (h') of the root 1 to the root 2 and the transverse displacement (λ) of the root 1 to the root 2₀) Half the amplitude (H) and half the wavelength (λ), respectively. For the slit case, the geometric parameters are slit amplitude (also defined as H), slit wavelength (also defined as λ), width of the sliding tip (a), and width of the slit gap (W). The present invention investigates four situations: baseline, SRooTES, droootes, and Slit. Fig. 1B compares the drooeses and Slit geometries in more detail and will be referred to below.

A 0.8mm thick flat trailing edge can be inserted into a 0.8mm slot along the trailing end of the main airfoil body. The sawtooth amplitude (H) and the wavelength (lambda) of the Slit, the SRooTES and the DRooTES are respectively the value of every 5mm at intervals within the range of H being more than or equal to 5mm and less than or equal to 30mm, and the value of lambda being more than or equal to 3mm and less than or equal to 35 mm. Root of DRooTES 1 toThe transverse displacement of the root 2 and the angle of the sawtooth tip are respectively more than or equal to lambda and less than or equal to 1.5mm₀The value is taken at intervals of 1.5mm within the range of not more than 4.5mm, and the value is within the range of phi not less than 0 degrees and not more than 84.3 degrees. The baseline trailing edge is half the amplitude in the castellated case to ensure similar wetted surface area as SRooTES and droots. To facilitate comparing SRooTES and droootes to the slip trailing edge, the baseline trailing edge is also assumed to be half the amplitude. Coarse sandpaper was used for long periods of time on the upper and lower surfaces at 30mm from the leading edge of the airfoil to ensure that the boundary layer was fully brought into turbulent conditions. The width of abrasive paper strip is 10mm, and the thickness is 0.95 mm.

Wind tunnel equipment and instrument

Wind tunnel devices and instruments for far-field noise measurements are arranged in the aerodynamic acoustic installation of the university of brunell, which consists of an open jet wind tunnel placed in a sound-damping chamber of 4m × 5m × 3.4 m. The size of the open nozzle was 0.3m × 0.1m (width × height). The airfoil is attached to a side plate flush with the nozzle lip. The operation capacity of the open jet wind tunnel is up to 80m/s, and the typical low turbulence is between 0.1% and 0.2%. The background noise (no airfoil, but side panels) is primarily open jet noise, which is much less than the airfoil self-noise produced at the same flow rate. All far field noise measurements were made at a geometric angle of attack (AoA) θ of 0 °.

Far field array noise measurement

Far-field array noise measurements were made with 8 g.r.a.s. half inch condenser microphones (46AE) mounted at polar angles ranging from Θ to 50 ° to Θ to 120 ° every 10 °. The microphone at 90 ° is located 0.97m above the mid-span and trailing edge of the airfoil. Fig. 1C shows an experimental arrangement of a far field array. Each microphone signal is applied with a gain of ± 20dB by a g.r.a.s.12ax4 channel constant current source (CCP) amplifier. Data was collected using a 16-bit analog-to-digital card manufactured by national instruments ltd. The sampling frequency was 40kHz and the sampling time was 20 seconds. The data is windowed and the Power Spectral Density (PSD) of the 1Hz bandwidth is calculated from the 1024 point Fast Fourier Transform (FFT) and the 50% overlap time. Assuming that the sound wave propagates spherically from the trailing edge, the noise is calculated to obtain the sound power level (PWL). In the range of U-20 m/s to U-60 m/s, the free flow velocity takes a plurality of values at intervals of U-10 m/s to measure the noise value.

Results and discussion

The results will be presented in this section. This section covers when U_∞Noise measurements obtained using drooeses and the trailing edge of the slit were taken at intervals of 10m/s in the range of 20m/s and 60m/s, and the angle of attack θ was 0 °.

The description of the flow rate so far has referred to the free flow value U in the general equations (equations 1 and 2) only_∞. However, in describing the acoustical destruction mechanism of drooeses and slits, the primary parameter is not the free stream velocity U_∞But rather the convective velocity U of the turbulent vortex_c. It should be noted that the convective velocity of turbulent eddies is also affected by the pressure driven secondary flow structure in the direction of the rim. The net convective velocity u of the turbulent vortex has been determined_cBetween 40% and 80% of the value of the free flow velocity, i.e. betweenAndin the meantime. Thus, for the drooeses and Slit trailing edge of the present invention, the general expression in equation 2 can be modified as:

wherein the content of the first and second substances,is the convection velocity factor and has a value between 0.4 and 0.8. The destructive interference mechanism should occur at St 0.5, 1.5, etc. between the roots of droots and between the root and tip of the trailing edge of the Slit, according to the definition of equation 3.

Trailing edge of slit

Slit amplitude (H)

FIG. 2 shows the velocity U of the free stream_∞Comparison of the sensitivity of the trailing edge of the slit with different H when taking different values. Δ PWL is defined as the difference in acoustic power levels of the baseline trailing edge and the slot trailing edge as a function of frequency: Δ PWL ═ PWL_B(f)-PWL_slit(f) In that respect Note that when Δ PWL takes a positive value, it represents a decrease in noise level; when Δ PWL takes a negative value, it represents an increase in noise level. The results clearly show that the trailing edge of the slot is most effective in the low to mid frequency range, up to 7dB of noise reduction can be achieved. The slot trailing edge also results in a significant degradation of performance, up to 5dB, compared to the baseline trailing edge. Furthermore, the unique peak values of the noise reduction Δ PWL achieved by the individual slit trailing edges do not appear to follow the different U_∞Lower slit amplitude H. Maximum Δ PWL observed through individual slot trailing edges exhibits f and U_∞The relationship between H and H is as follows:

1.f∝U_∞

thus, the acoustic frequency at which maximum noise reduction occurs is related to the free stream velocity, and the longitudinal displacement H between the two ends of the slit (root and tip).

Dimensionless frequency-slit amplitude, H

Fig. 3 shows whether the application of the relationship between dimensionless frequency and broadband noise reduction of the slot trailing edge can be generalized at different slot amplitudes (H). Theoretically, perfect destructive interference can occur when the acoustic radiation is 180 ° out of phase between the two sources, resulting in cancellation of the acoustic radiation. Referring to FIG. 1B, an initial assumption of destructive interference between two sources at the root and tip of the trailing edge of a slit is shown. This means that the dimensionless frequency (f.H/uc, where l ═ H when it is the slot trailing edge) defined in equation 3 should be equal to 0.5.

Fig. 4 shows that for different slit amplitudes, the Δ PWL peak does occur around St ≈ 0.5 (except for U ≈ 60m/s, where slight fluctuations in the strouhal number are evident). These results obtained experimentally demonstrate that the destructive interference mechanism associated with 180 ° out of phase cancellation between the two sources (root and tip) minimizes noise.

Furthermore, fig. 3 clearly shows that after all the APWL peaks associated with the slit configuration, there is an accompanying significant drop in Δ PWL (noise increase) at St ≈ 1, which is twice the strouhal value for maximum noise reduction discussed in the previous section. This is another evidence of an acoustic interference mechanism between the sources of the trailing edge of the slot. In summary, the destructive interference mechanism (noise reduction) should occur at St 0.5, 1.5, etc. between the root and tip of the slit trailing edge. This is consistent with the experimentally observed destructive interference at occurrences St ≈ 0.5 and St ≈ 1.5. On the other hand, constructive interference (noise increase) occurs at St ≈ 1.

Dimensionless frequency-wavelength (λ) with constant width (W) of the slot root

Fig. 4 shows a comparison of the sensitivity of dimensionless frequencies when λ of the trailing edge of the slit takes different values, where the slit width W is constantly maintained at 0.3 mm. The results in fig. 4 were observed at all free stream velocities. The results clearly show that λ has no effect on dimensionless frequency, in sharp contrast to the reduction of the peak. Further, for all cases, the Δ PWL peak corresponds to destructive interference at St ≈ 0.5 and St ≈ 1.5, and constructive interference at St ≈ 1. The effect of the acoustic interference mechanism is still observed in the case where λ at the trailing edge of the slit takes the maximum value, but the effect of this mechanism is greatly reduced compared to the case where λ takes the minimum value.

Width (W) of slit root at constant wavelength (lambda)

Fig. 5 shows a comparison of broadband noise emission at the trailing edge of the slot when the aspect ratio W/λ takes different values, where λ is constantly maintained at 3 mm. The results clearly show that at all free stream velocities, the noise performance drops significantly in the mid-low frequency range (Δ PWL is negative) as W increases. As W increases, the slit geometry becomes less and less compact, even eventually exceeding the spanwise integral length scale of the turbulent eddies. At this time, the slits in the root region slowly become a straight trailing edge noise mechanism, which in turn impairs noise reduction capability.

In the experiments, when W/lambda is 0.15, the performance is optimal, and the noise reduction can reach 6dB at most; and W/λ has the worst performance when it takes the maximum value, i.e., 0.5. In the medium-high frequency 600Hz < f <20kHz range, it was observed that only when W/λ is set to a maximum (0.5) does it lead to no noise increase, even a slight reduction of the broadband noise up to 4dB at low speed is possible. On the other hand, all settings in which W/λ takes a small value cause only an increase in noise. Whereas a slot trailing edge with a smaller W/lambda setting results in a larger noise increase. Based on the results presented so far, the following can be summarized:

1) to facilitate wideband noise reduction at medium and low frequencies, a small W/lambda needs to be set. This is to avoid slit widths comparable to the spanwise integral length scale of turbulent eddies, thereby attenuating destructive interference mechanisms.

2) To avoid the increase of high-frequency noise, it is necessary to set a large W/λ. This is to reduce the tendency for cross flow through the slit and to minimize flow leakage.

The reduction of noise is caused by the slit gap (W) and turbulence vortex (L) near the trailing edge_y) Towards the boundary condition of the ratio of the spread-wise correlation length scales. Experiments show that at f_{Peak value}The optimal conditions under which noise reduction occurs are about 0.2W/L or less_y≤0.5。W/L_yThe lower and upper limits of (c) may fluctuate under the influence of flow conditions (e.g., reynolds number, mach number, etc.).

DRooTES

Comparison of noise results for DRooTES with different root-to-root amplitudes (h

Next a new geometry is investigated, also taking advantage of the extra roots of acoustic interference. This new geometry is called drooes. FIG. 6 shows a comparison of the broadband noise reduction (and noise increase) achieved by DRooTES for the baseline trailing edge when the root-to-root amplitude h' (see FIG. 1B) takes different values. In all drooes cases, the maximum amplitude h' is 15mm, which is the best casePerformance of in U_∞The noise reduction is up to 7dB at 60 m/s. The situation when taking the amplitude minimum h' 2.5mm shows noise reduction only at high frequencies for the baseline trailing edge. The general trend is that the noise reduction level decreases with decreasing drooeses amplitude. At high frequencies, droots exhibits an increase in noise up to 2dB with increasing amplitude.

Dimensionless frequency-when root 1 to root 2 (h') of DRooTES is root 2 to tip (h ″)

Assuming that in the case of drooes, there is destructive interference between the two sources as shown in fig. 1B, the distance (h') between them applies to the strouhal number St. For destructive interference of a drooese, the following conditions must be satisfied:

where ω is 2 pi f. U is as previously defined in the trailing edge of the slot_cIs the convective velocity of the turbulent vortex after interaction with the secondary flow driven by the pressure. Experience is due toPossible unequal values between h' and h "in the frequency scaling are taken into account.

FIG. 7 shows dimensionless frequencies f.h'/u_cPlot of the spectrum against Δ PWL. Note that, since h' ═ h ″ i, κ is 1. It can be seen that the maximum noise reduction (max Δ PWL) occurs at St ≈ 0.5, very close to the value observed in the slit trailing edge case. This suggests that the same acoustic destruction mechanism is also applicable to drooes.

Dimensionless frequency-when root 1 to root 2(H ') ≠ root 2 to tip (H') of DRooTES, and H is constant When the amount is 30mm

When h' ≠ h ", then κ is no longer 1. Factor will beSubstituting in frequency scaling, fig. 8 shows the sensitivity of broadband noise radiation at each dimensionless frequency when H 'is different, and then adjusting H' accordingly to produce an overall constant H of 30 mm. The results show a significant drop in each curve. An increase in the value of h' affects the maximum noise reduction peak. The best performance is achieved with a noise reduction of up to 8dB when h 'is 20mm, whereas at the same free flow speed, only a 5dB reduction is achieved when h' is 5 mm. However, the acoustic performance depends on the geometrical parameters and the flow conditions.

Summary of the invention

The following equations are applied to determine the optimal geometry for slits and drooes (H for slits, H' for drooes):

in the case of a slit, the slit,

wherein

In the case of the drooeses,

whereinTo the palace

In general, influenceAnd f_{Peak value}The exact numerical value of (the source of uncertainty) is potentially:

1) turbulent vortices are dispersed into pressure waves near the root rather than just at the root, which affects the "true" longitudinal displacement between the slit root and the tip;

2) a change in a convection velocity factor range of the turbulent eddies;

3) due to the presence of secondary flows (e.g., cross-flow within the slit), there is another factor that affects the turbulent vortex vs. flow velocity, defined as

Comparison of SRooTES, DRooTES and slit trailing edge

Fig. 9 shows a comparison of the trailing edge at different free stream velocities when the amplitude H is 15 mm. With the maximum Δ PWL, the situation with the best performance is the slot trailing edge, which can be reduced by up to 4.8dB compared to the baseline trailing edge. Alternatively, DRooTES produced significant wideband noise reduction up to 4.5dB over a much larger frequency range of 100Hz < f < 4kHz at all free stream velocities than the trailing edge of the slot. In contrast, the free stream velocity based slit achieves a reduction in the small frequency range 100kHz < f < 1.5 kHz. The worst case performance is SRooTES, which produces the least noise reduction in the mid and low frequency range. However, SRooTES produces minimal noise increase in the high frequency range compared to droootes and slit trailing edges. In the high frequency range, the noise increase produced by the trailing edge of the slit is greatest, followed by drooes.

Fig. 10 shows a comparison of the trailing edge at different free stream velocities when the amplitude H is 30 mm. The best performance is drooes, which can achieve up to 7.2dB of noise reduction. In addition, drooes produces minimal noise increase and noise reduction over all frequency ranges compared to baseline. SRooTES produces similar noise reduction effects in the low to mid frequency range, but it produces the greatest increase in noise. The trailing edge of the slit produces a significant drop at a particular frequency in the low to mid frequency range, while producing a drop similar to drooes in the high frequency range.

In summary, the results shown in fig. 9 and 10 clearly demonstrate that:

1. in all trailing edge cases, DRooTES has a significant advantage in improving noise performance. Droots achieved noise reduction of up to 4.5dB and 7.2dB when H15 mm and H30 mm, respectively.

2. For all trailing edge cases, optimizing the geometry can significantly improve noise performance depending on the flow conditions.

All optional, preferred features and modifications of the above described embodiments and dependent claims are applicable to all aspects of the invention taught herein. Furthermore, the individual features of the dependent claims, as well as all optional features, preferred features and modifications of the above-described embodiments, can be combined with and interchanged with one another.

The disclosure of uk patent application No. 1906920.2, to which this application claims priority, and the abstract accompanying this invention are incorporated herein by reference.