Method and device for identifying circumferential acoustic modes of fan single tone noise of an aero-engine

Abstract

Disclosed is a fan single tone noise circumferential acoustic mode decomposition method based on orthogonal matching pursuit, which comprises the following steps: calculating the dominant circumferential acoustic mode order of fan single tone noise through the number of fan blades at each stage of an aero-engine, determining the maximum number and installation angle of a circumferential microphone array uniformly arranged on a fan casing of the aero-engine, constructing an observation matrix, and performing fast Fourier transform on the sound pressure signals measured by the microphone array; taking an orthogonal Fourier transform basis as a sparse dictionary to construct a sparse representation model of the frequency domain signals of the uniformly arranged microphone array, establishing a sparse representation model of the fan single tone noise signals of the aero-engine based on the sensing matrix and the observation vector; reconstructing the complex amplitude of the sound pressure at the blade passing frequency at the full microphone array through orthogonal matching pursuit; and obtaining the number and amplitude of the fan single tone noise circumferential acoustic modes of the aero-engine through spatial Fourier transform.

Description

Technical Field

[0001] This invention belongs to the technical field of aircraft engine noise testing, and in particular to a method and device for identifying the circumferential acoustic modal of a single-tone noise of an aircraft engine fan based on orthogonal matching pursuit. Background Technology

[0002] With the widespread application of high-bypass turbofan aero-engines, fan noise has gradually become the dominant noise component in aero-engines, attracting widespread social attention. Fan noise can be categorized into single-tone noise at various blade passage frequencies and broadband noise distributed across the entire frequency domain, based on its formation mechanism. Single-tone noise is significantly stronger than other noise components and constitutes the main part of aero-engine fan noise. Refined noise reduction design for aero-engine fans requires a deep understanding of the generation and propagation modes of single-tone noise, and the identification and decomposition of the acoustic modal structure of the duct plays a crucial guiding role in this process.

[0003] Using a ring microphone array to decompose the circumferential acoustic modes of aero-engine fan noise is currently the most common method in the field of aero-engine fan noise mode decomposition. However, classic uniformly arranged microphone arrays suffer from drawbacks such as high testing costs and poor failure resistance, making them difficult to apply to real-world aero-engine duct acoustic mode measurements. To address these issues, this paper constructs an l0-norm optimization problem based on the sparse prior of the single-tone noise circumferential acoustic modes in the wavenumber domain. A compressed sensing model of the circumferential acoustic modes with a small number of measurement points and a non-uniform microphone layout is established. Based on orthogonal matching pursuit, the amplitude of the aero-engine fan single-tone noise circumferential acoustic modes is solved, enabling accurate order identification and amplitude reconstruction of the dominant acoustic modes of aero-engine fan noise using a limited number of microphones.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of the present invention, and therefore may contain information that does not constitute prior art known to those skilled in the art in this country. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a method and device for identifying the circumferential acoustic modes of single-tone noise from an aircraft engine fan based on orthogonal matching pursuit. This invention proposes a sparse decomposition model of the circumferential acoustic modes of pipe noise based on compressed sensing, and reconstructs the circumferential acoustic modes of fan noise through orthogonal matching pursuit, thereby realizing the order identification and amplitude reconstruction of the dominant circumferential acoustic modes. This solves the problems of high testing cost and poor failure resistance of traditional uniformly arranged microphone arrays. The identification of the dominant acoustic modes of aircraft engine fan noise can be completed with far fewer microphone arrays than classical methods.

[0006] The objective of this invention is achieved through the following technical solution: the method for circumferential acoustic modal identification of single-tone noise of aircraft engine fans based on orthogonal matching pursuit includes the following steps:

[0007] In the first step, the order m of the dominant circumferential acoustic mode of the single-tone noise of the aero-engine fan is calculated by the number of rotor blades B and stator blades V. The number of microphones and the installation angle of the full microphone array arranged circumferentially on the aero-engine fan casing are determined by the highest order of the circumferential acoustic mode m of the single-tone noise.

[0008] In the second step, according to the given number of microphones, the microphone installation angles are randomly selected from the microphone installation angles to construct an observation matrix. The sound pressure signal measured by the microphone array is subjected to a fast Fourier transform, and the complex amplitude of the sound pressure signal at the blade passage frequency is taken as the observation vector.

[0009] In the third step, an orthogonal Fourier transform basis is used as a sparse dictionary to construct a sparse representation model of the single-tone noise of the aero-engine fan; based on the observation matrix and the sparse dictionary, a compressed sensing sampling model for sparse decomposition of the circumferential acoustic modes of the single-tone noise of the aero-engine is established.

[0010] In the fourth step, based on the compressed sensing sampling model, the complex amplitude of the sound pressure signal at the blade passage frequency at the full microphone array is reconstructed by orthogonal matching tracking; the circumferential acoustic mode of the single-tone noise of the aero-engine fan is obtained by spatial Fourier transform.

[0011] In the method described above, in the first step, the circumferential dominant acoustic mode order m of the fan single-tone noise is: m = ηB ± ζV, where η represents the pressure pulsation order caused by the unsteady aerodynamic force resulting from the fan's rotation-to-static interference, and ζ represents a non-negative integer. Based on the Nyquist sampling theorem, the number n of microphones and the installation angle θ = [θ1, θ2, ..., θ] in the full microphone array uniformly arranged circumferentially on the aero-engine fan casing are determined. n ] T θ i This represents the installation angle of the i-th sensor.

[0012] In the method described, in the second step, the given number of microphones is k, based on the installation angle θ = [θ1, θ2, ..., θ] of the full microphone array. n ] T Randomly select k positions θ from them k =[θ k1 θ k2 , ..., θ kk ] T Install a microphone array, θ ki Let represent the installation angle of the i-th sensor among k sensors, and construct the observation matrix Ψ:

[0013] right k Fast Fourier Transform of the sound pressure signals measured by a non-uniformly distributed microphone: y f =FFT(y) t ), where y t This represents the time-domain signal of the fan noise sound pressure level measured by the non-uniformly distributed microphone array, where FFT(·) represents the Discrete Fourier Transform, and y f = [P1(ω), P2(ω), ..., P k (ω)] T Let P represent the frequency domain signals from k non-uniformly distributed microphones. i (ω) represents the complex amplitude at the i-th sensor frequency ω. Based on the frequency domain signals of the k non-uniformly distributed microphones, the frequency f through which the fan rotor blades pass is selected. BPF Construct the observation vector y = [P1(f)] using the complex magnitude. PBF ), P2(f BPF ), ..., P k (f BPF )], where f BPF =Bn BPF n BPF Main spindle speed, P k (f BPF ) represents the frequency f of the fan rotor blades passing through the k-th sensor. BPF The complex amplitude below.

[0014] In the method described above, in the third step, an orthogonal Fourier transform basis is constructed as a sparse dictionary Φ. n×n Construct a sparse representation of the frequency domain signals of n uniformly distributed microphones:

[0015] x = Φ T P, where P = [P1(ω), P2(ω), ..., P n [(ω)] represents the complex amplitude of the frequency domain signal measured by the n evenly distributed microphones to be reconstructed at frequency ω, P i (ω) represents the complex amplitude value at the i-th sensor frequency ω, and X represents the sparse representation of the full microphone array frequency domain signal in the wavenumber domain;

[0016] Based on the aforementioned observation matrix Ψ and sparse dictionary Φ, the compressed sensing sampling model for sparse decomposition of the circumferential acoustic modes of single-tone noise is established as follows:

[0017] y = ΨP = ΨΦx = Θ CS x,

[0018] Where, Θ CS This represents the sensing matrix of the compressed sensing sampling model for sparse decomposition of acoustic modes.

[0019] In the method described above, in the fourth step, a sparse decomposition representation model of acoustic modes based on the l0 norm is established based on the compressed sensing sampling model:

[0020]

[0021] Where ||·||0 represents the zero norm, and min represents minimization. The sparse decomposition representation model of the acoustic modes is solved using orthogonal matching pursuit to reconstruct the sparse representation X of the sound field structure measured by n uniformly distributed microphones in the beam domain. Based on the sparse representation X of the full microphone array frequency domain signal in the wavenumber domain, the frequency domain signal P of the n uniformly distributed microphones is solved: P=Ψx. Based on the frequency domain signal P of the full microphone array, the amplitudes P of each boundary circumferential mode on the microphone mounting wall are obtained through spatial Fourier transform. m :

[0022]

[0023] In the method described above, in the first step, the relationship between the number of microphones n and the order m of the dominant circumferential acoustic mode m is: n > 2m.

[0024] In the method described above, the fourth step involves establishing a sparse decomposition representation model of acoustic modes based on the l0 norm and solving it through orthogonal matching pursuit, including the following steps:

[0025] S401. Based on the sparse decomposition model of acoustic modes based on the l0 norm, set the error threshold ∈ 0, the number of iterations k = 0, and the initial solution X. 0 =0, initial residual r 0 =y-Θ CS X 0 = y and the support set of the initial solution

[0026] S402. Increment the iteration count by one at each iteration: k = k + 1;

[0027] S403. Scan all j and select the optimized parameters. Calculation error: Where j represents the perception matrix Θ CS Subscripts in the middle column;

[0028] S404. Determine the point j0 where the error ∈(j) takes its minimum value: ∈(j0)≤∈(j), and update the support set: S k =S k-1 ∪{j0};

[0029] S405, In the support set support{x}=Sk Under the conditions, the calculation makes Minimize the value X k Update residual: r k =y-Θ CS x k , where r k The residual is the number of iterations k.

[0030] S405. Check if the stopping condition has been met: ||r k If ||2≤∈0, and the stopping condition is met, then output the optimized solution x after k iterations. k Otherwise, continue iterating.

[0031] An aero-engine fan single-tone noise circumferential acoustic mode identification device implementing the method includes:

[0032] The fan noise measurement module is used to randomly select microphone positions from the microphone array installation locations and measure the time-domain signal of fan noise.

[0033] The fan single-tone noise circumferential acoustic mode recognition module is used to establish a sparse decomposition compressed sensing sampling model of the circumferential acoustic mode of the aero-engine single-tone noise and to establish an acoustic mode sparse decomposition representation model based on the l0 norm.

[0034] Beneficial effects

[0035] This invention provides a method based on the principle of compressed sensing. By randomly selecting the microphone installation position, an observation matrix is ​​constructed. Based on the sparsity characteristics of the fan noise duct acoustic mode in the wavenumber domain, a sparse dictionary is determined, and a compressed sensing sampling model of single-tone noise from an aero-engine fan is constructed. An optimization problem based on the l0 norm is constructed, and the sound field structure measured by an n uniform microphone array is reconstructed through orthogonal matching pursuit. Based on spatial Fourier decomposition, the identification of the circumferential dominant acoustic mode of single-tone noise is achieved.

[0036] Compared to the traditional method of decomposing the circumferential acoustic modes of fan noise using a full microphone array, the method provided by this invention can achieve accurate identification of the dominant circumferential acoustic mode order of pipe noise and accurate reconstruction of its amplitude, reducing the number of microphones and lowering testing costs.

[0037] The above description is merely an overview of the technical solution of the present invention. In order to make the technical means of the present invention clearer and more understandable, so that those skilled in the art can implement it according to the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more obvious and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0038] Various other advantages and benefits of the present invention will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0039] In the attached diagram:

[0040] Figure 1 This is a flowchart of the method of the present invention;

[0041] Figure 2 This is a flowchart of the orthogonal matching pursuit process of the present invention;

[0042] Figure 3 This is a diagram showing the microphone installation location;

[0043] Figure 4 This is a schematic diagram of the single-tone noise mode of an aero-engine and the amplitude of the noise mode detected by this invention.

[0044] The present invention will be further explained below with reference to the accompanying drawings and embodiments. Detailed Implementation

[0045] The following will refer to the appendix. Figures 1 to 4 Specific embodiments of the invention will be described in more detail below. While specific embodiments of the invention are shown in the accompanying drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0046] It should be noted that certain terms are used in the specification and claims to refer to specific components. Those skilled in the art will understand that different terms may be used to refer to the same component. This specification and claims do not distinguish components based on differences in terminology, but rather on differences in function. The terms "comprising" or "including" used throughout the specification and claims are open-ended and should be interpreted as "comprising but not limited to." The following descriptions are preferred embodiments for carrying out the invention; however, these descriptions are for the purpose of understanding the general principles of the specification and are not intended to limit the scope of the invention. The scope of protection of this invention is determined by the appended claims.

[0047] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. The accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0048] This invention discloses a method for identifying the acoustic mode of single-tone noise from an aero-engine fan based on orthogonal matching pursuit, comprising the following steps:

[0049] In the first step, the order of the dominant circumferential acoustic mode of the single-tone noise of the aero-engine fan is calculated by the number of blades of the rotor and stator of the aero-engine fan, and the number of microphones and the installation angle of the full microphone array arranged circumferentially on the aero-engine fan casing are determined.

[0050] In the second step, according to the given number of microphones, microphone positions are randomly selected from the uniformly arranged microphone positions determined in the first step to construct an observation matrix. The sound pressure signal measured by the microphone array is subjected to a fast Fourier transform, and the complex amplitude of the sound pressure signal at the frequency through which the blade passes is taken as the observation vector.

[0051] In the third step, an orthogonal Fourier transform basis is used as a sparse dictionary to construct a sparse representation model of the single-tone noise of the aero-engine fan; based on the observation matrix and the sparse dictionary, a compressed sensing sampling model for sparse decomposition of the circumferential acoustic modes of the single-tone noise of the aero-engine is established.

[0052] In the fourth step, based on the sparse decomposition representation model of the circumferential acoustic mode of the single-tone noise of the aero-engine, the complex amplitude of the sound pressure signal at the blade passage frequency at the full microphone array is reconstructed by orthogonal matching pursuit; and the circumferential acoustic mode of the single-tone noise of the aero-engine fan is obtained by spatial Fourier transform.

[0053] In the method described above, in the first step, the circumferential dominant acoustic mode order m of the fan single-tone noise is calculated according to the number of rotor blades B and stator blades V of the aero-engine fan: m = ηB ± ζV, where η represents the pressure pulsation order caused by the unsteady aerodynamic forces resulting from the fan rotor-stator interference, and ζ represents a non-negative integer. Based on the Nyquist sampling theorem, the number n of microphones and the installation angle θ = [θ1, θ2, ..., θ] in the full microphone array uniformly arranged circumferentially on the aero-engine fan casing are determined. n ] T θ i The installation angle of the i-th sensor is represented; the relationship between the number of microphones n and the order m of the circumferential acoustic mode dominated by single-tone noise is: n > 2m;

[0054] In the method, in the second step, the given number of microphones is k, based on the installation angle θ = [θ1, θ2, ..., θ] of the full microphone array. n ] TRandomly select k positions θ from them k =[θ k1 θ k2 , ..., θ kk ] T θ ki Let Ψ represent the installation angle of the i-th sensor among k sensors, install the microphone array, and construct the observation matrix Ψ. k×n :

[0055]

[0056] Perform a Fast Fourier Transform on the sound pressure signals measured by the k non-uniformly distributed microphones: y f =FFT(y) t ), where y t This represents the time-domain signal of the fan noise sound pressure level measured by the non-uniformly distributed microphone array, where FFT(·) represents the Discrete Fourier Transform, and y f = [P1(ω), P2(ω), ..., P k (ω)] T P represents the frequency domain signal of k non-uniformly distributed microphones. i (ω) represents the complex amplitude at the i-th sensor frequency ω; based on the frequency domain signals of the k non-uniformly distributed microphones, the frequency f through which the fan rotor blades pass is selected. BPF Construct the observation vector y = [P1(f)] using the complex magnitude. PBF ), P2(f BPF ), ..., P k (f BPF )], where f BPF =B·n BPF n BPF Main spindle speed, P k (f BPF ) represents the frequency f of the fan rotor blades passing through the k-th sensor. BPF Complex amplitudes below;

[0057] In the method described, the third step involves constructing an orthogonal Fourier transform basis as a sparse dictionary Φ. n×n Construct a sparse representation of the frequency domain signals of n uniformly distributed microphones:

[0058] x = Φ T P

[0059] Where, P = [P1(ω), P2(ω), ..., P n [(ω)] represents the complex amplitude of the frequency domain signal measured by the n evenly distributed microphones to be reconstructed at frequency ω, P i(ω) represents the complex amplitude value at the i-th sensor frequency ω, and X represents the sparse representation of the full microphone array frequency domain signal in the wavenumber domain;

[0060] Based on the aforementioned observation matrix Ψ and sparse dictionary Φ, the compressed sensing sampling model for sparse decomposition of the circumferential acoustic modes of single-tone noise is established as follows:

[0061] y = ΨP = ΨΦx = Θ CS x

[0062] Where, Θ CS The sensing matrix represents the compressed sensing sampling model for sparse decomposition of acoustic modes;

[0063] In the method, in the fourth step, based on the circumferential acoustic modal sparse decomposition compressed sensing sampling model of the single-tone noise of the aero-engine, an acoustic modal sparse decomposition representation model based on the l0 norm is established:

[0064]

[0065] Where ||·||0 represents the 0-norm, and min denotes minimization. The sparse representation model of the acoustic modal with respect to the 10-norm is solved using orthogonal matching pursuit to reconstruct the sparse representation X of the sound field structure measured by n uniformly distributed microphones in the beam domain. Based on the sparse representation X of the full microphone array frequency domain signal in the wavenumber domain, the frequency domain signal P of the n uniformly distributed microphones is solved: P = Ψx. Based on the frequency domain signal P of the full microphone array, the circumferential modal amplitudes P on the microphone mounting wall are obtained through spatial Fourier transform. m :

[0066]

[0067] Figure 2 The flowchart for orthogonal matching pursuit is as follows: Figure 2 As shown, in the method, the fourth step, establishing a sparse decomposition representation model of acoustic modes based on the l0 norm, and solving the sparse model through orthogonal matching pursuit, includes the following steps:

[0068] S401, based on l o The norm-based sparse decomposition model of acoustic modes, with an error threshold ∈ 0, iteration number k = 0, and initial solution X. 0 =0, initial residual r 0 =y-Θ CS X 0 = y and the support set of the initial solution

[0069] S402. Increment the iteration count by one at each iteration: k = k + 1;

[0070] S403. Scan all j and select the optimized parameters. Calculation error: Where j represents the perception matrix Θ CS Subscripts in the middle column;

[0071] S404. Determine the point j0 where the error ∈(j) takes its minimum value: ∈(j0)≤∈(j), and update the support set: S k =S k-1 ∪{j0};

[0072] S405, In the support set support{x}=S k Under the conditions, the calculation makes Minimize the value X k Update residual: r k =y-Θ CS X k , where r k The residual is the number of iterations k.

[0073] S405. Check if the stopping condition has been met: ||r k If ||2≤∈0, and the stopping condition is met, then output the optimized solution X after k iterations. k Otherwise, continue iterating.

[0074] The following is in conjunction with the appendix Figures 1 to 4 The invention is further described using a specific simulation example instead of an experimental example. It should be emphasized that the following description is merely exemplary, and the application of the invention is not limited to the following example.

[0075] To better understand, Figure 1 The flowchart shows a method for circumferential acoustic modal identification of single-tone noise from an aircraft engine fan based on orthogonal matching pursuit. Figure 1 As shown, the method for circumferential acoustic modal identification of single-tone noise of aircraft engine fans based on orthogonal matching pursuit includes the following steps:

[0076] 1) Assuming the number of rotor blades in the aero-engine's sound pressure is B = 22 and the number of stator blades is V = 17, according to the formula for calculating the order of the fan single-tone noise-to-stator interference mode, m = ηB ± ζV, usually η = 1 is taken to represent the order of pressure pulsation caused by the unsteady aerodynamic force resulting from the interference of the upstream and downstream blades of the fan, and ζ = 1 is taken to represent the first-order single-tone noise mode. The order of the fan single-tone noise circumferential dominant mode is calculated as m = ηB ± ζV = 22 - 17 = 5. Based on the Nyquist sampling theorem, the number of microphones in the full microphone array uniformly arranged circumferentially on the aero-engine fan casing is determined to be n ≥ 2m. Taking n = 32 satisfies the condition, the installation angle is determined to be θ = [11.25°, 22.5°, ..., 360°]. T

[0077] 2) Assuming a given number of microphones k = 8, the microphone positions θ = [11.25°, 22.5°, ..., 360°] are determined by the uniform arrangement of microphones in the first step. T Randomly select microphone position θ k = [33.75°, 67.5°, 78.75°, 123.75°, 180°, 247.5°, 337.5°, 360°] T ,like Figure 3 The observation matrix Ψ is constructed as shown, and Ψ has a size of 8×32. in, Let represent the element in the i-th row and j-th column of the observation matrix Ψ, with all other positions being 0; construct the time-domain signals of 32 microphones with m=5 in the static-to-static interference mode, and perform a Fast Fourier Transform on the sound pressure signals measured by the 8 non-uniformly distributed microphones, y f =FFT(y) t ), where y t This represents the time-domain signal of the fan noise sound pressure level measured by the non-uniformly distributed microphone array, where FFT(·) represents the Discrete Fourier Transform, and y f =[P1(ω),P2(ω),...,P8(ω)] T This represents the frequency domain signals from 8 non-uniformly distributed microphones, where ω represents frequency; assuming the blade passage frequency is 3000Hz, select the fan rotor blade passage frequency f. BPF The complex magnitudes under the given conditions are used to construct the observation vector y = [P1(3000), P2(3000), ..., P8(3000)];

[0078] 3) Construct an orthogonal Fourier transform basis as a sparse dictionary Φ, and construct a sparse representation of the frequency domain signals of 32 uniformly distributed microphones: x = Φ T P, where P = [P1(ω), P2(ω), ..., P 32[(ω)] represents the complex amplitude of the frequency domain signal measured by the 32 uniformly distributed microphones to be reconstructed at frequency ω, and x represents the sparse representation of the frequency domain signal of the full microphone array in the wavenumber domain; based on the observation matrix Ψ and the sparse dictionary Φ, the compressed sensing sampling model for the sparse decomposition of the circumferential acoustic mode of single-tone noise is established as: v=ΨP=ΨΦx=Θ CS x

[0079] 4) Based on the circumferential acoustic modal sparse decomposition compressed sensing sampling model of the single-tone noise of the aero-engine, establish an acoustic modal sparse decomposition representation model based on the l0 norm. sty = Θ CS x utilizes, for example Figure 2 The orthogonal matching pursuit method is used to solve the sparse representation model of the acoustic modes with respect to the l0 norm, reconstructing the sparse representation x of the sound field structure measured by 32 uniformly distributed microphones in the beam domain; the frequency domain signal P of n uniformly distributed microphones is solved: P = Ψx; based on the frequency domain signal P of the full microphone array, the amplitudes P of each circumferential mode on the microphone mounting wall are obtained by spatial Fourier transform. m : The result is consistent with the calculation in step 1) of m = ηB ± ζV = 22 - 17 = 5, and the output result is as follows: Figure 4 As shown.

[0080] Although embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the specific embodiments and application fields described above. The specific embodiments described above are merely illustrative and instructive, and not restrictive. Those skilled in the art can make many other forms based on the guidance of this specification and without departing from the scope of protection of the claims of the present invention, and all of these are within the scope of protection of the present invention.

Claims

1. A method for circumferential acoustic mode recognition of single-tone noise of aero-engine fans based on orthogonal matching pursuit, comprising the following steps: In the first step (S1), the number of rotor blades of the aircraft engine fan is used to determine the process. With the number of stator blades Calculate the order of the circumferential dominant acoustic mode of single-tone noise from an aero-engine fan. The order of the circumferential acoustic modes of single-tone noise The highest order determines the number of microphones and the installation angle of the full microphone array arranged circumferentially on the upper edge of the aero-engine fan casing; In the second step (S2), according to the given number of microphones, the microphone installation angle is randomly selected from the microphone installation angles to construct an observation matrix. The sound pressure signal measured by the microphone array is subjected to a fast Fourier transform, and the complex amplitude of the sound pressure signal at the blade passage frequency is taken as the observation vector. In the third step (S3), the orthogonal Fourier transform basis is used as a sparse dictionary to construct a sparse representation model of the single-tone noise of the aero-engine fan. Based on the observation matrix and sparse dictionary, a compressed sensing sampling model for sparse decomposition of the circumferential acoustic modes of single-tone noise of aero-engines is established. In the fourth step (S4), based on the compressed sensing sampling model, the complex amplitude of the sound pressure signal at the blade passage frequency at the full microphone array is reconstructed by orthogonal matching pursuit; The circumferential acoustic mode of the single-tone noise of the aero-engine fan is obtained by spatial Fourier transform.

2. The method according to claim 1, wherein, In the first step (S1), the circumferential dominant acoustic mode order of the fan single-tone noise is... : ,in, This indicates the order of pressure pulsations caused by unsteady aerodynamic forces resulting from the interference of the fan's rotation and stationary motion. Representing non-negative integers, the number of microphones in a full microphone array uniformly arranged circumferentially on an aircraft engine fan casing is determined based on the Nyquist sampling theorem. and installation angle , Indicates the first The installation angle of each sensor.

3. The method according to claim 2, wherein, In the second step (S2), the given number of microphones is: Based on the installation angle of the full microphone array Randomly select from them Location Install a microphone array. express The first sensor The installation angle of each sensor was determined, and an observation matrix was constructed. : ,right Fast Fourier Transform of the sound pressure signals measured by a non-uniformly distributed microphone: ,in, This represents the time-domain signal of the fan noise sound pressure level measured by the aforementioned non-uniformly distributed microphone array. Represents the Discrete Fourier Transform. express Frequency domain signal from a non-uniformly distributed microphone Indicates the first Sensor frequency The complex amplitude values ​​below, based on the Frequency domain signal selection from a non-uniformly distributed microphone at the frequency of the fan rotor blades Constructing observation vectors from complex magnitudes ,in, , Main spindle speed Indicates the first The fan rotor blades of the sensor pass through the frequency The complex amplitude below.

4. The method according to claim 3, wherein, In the third step (S3), an orthogonal Fourier transform basis is constructed as a sparse dictionary. ,structure Sparse representation of a uniformly distributed microphone frequency domain signal: ,in, This represents the need for restructuring. The frequency domain signal measured by a uniformly distributed microphone is in the frequency range. The complex amplitude below, Indicates the first Sensor frequency The complex amplitude below, This represents the sparse representation of the frequency domain signal of the full microphone array in the wavenumber domain. Based on the observation matrix With sparse dictionaries The compressed sensing sampling model for sparse decomposition of single-tone noise circumferential acoustic modes is established as follows: ， in, This represents the sensing matrix of the compressed sensing sampling model for sparse decomposition of acoustic modes.

5. The method according to claim 4, wherein, In the fourth step (S4), a model based on the compressed sensing sampling model is established. Norm-based sparse decomposition representation of acoustic modes: ， in, Representing the 0-norm, min denotes minimization. The acoustic modal sparse decomposition representation model is solved using orthogonal matching pursuit to reconstruct... Sparse representation of the sound field structure measured by a uniformly distributed microphone in the beam domain Based on the sparse representation of the full microphone array frequency domain signal in the wavenumber domain. Solve Frequency domain signal of a uniformly distributed microphone : Frequency domain signals based on a full microphone array The amplitudes of the circumferential modes on the microphone mounting wall are obtained by solving the spatial Fourier transform. : 。 6. The method according to claim 2, wherein, In the first step (S1), the number of microphones With single-tone noise dominating the circumferential acoustic mode order The relationship is: .

7. The method according to claim 4, wherein, In the fourth step (S4), establish based on The norm-based sparse decomposition of the acoustic modes represents the model, and the solution obtained by orthogonal matching pursuit includes the following steps: S401, based on Norm-based sparse decomposition model of acoustic modes, with an error threshold set. Number of iterations Initial solution Initial residuals and the support set of the initial solution ; S402. Increment the iteration count by one at the beginning of each iteration: ; S403, For all Perform a scan and select optimized parameters. Calculation error: ,in, Representation of the perception matrix The subscript of the middle column, For the first The installation angle of each sensor; S404, Determine the error The point that takes the minimum value : And update the support set: ; S405, in the support set Under the conditions, the calculation makes Minimize value Update residuals: ,in, For the number of iterations Time residuals; S405. Check if the stopping condition has been met: If the stopping condition is met, the output will be in The optimized solution after the next iteration Otherwise, continue iterating.

8. A device for identifying the circumferential acoustic mode of a single-tone noise of an aircraft engine fan, implementing the method of any one of claims 1-7, comprising: The fan noise measurement module is used to randomly select microphone positions from the microphone array installation locations to measure the time-domain signal of fan noise. A fan single-tone noise circumferential acoustic mode recognition module is used to establish a sparse decomposition compressed sensing sampling model of the circumferential acoustic mode of aero-engine single-tone noise, and to establish a model based on... Norm-based sparse decomposition representation model of acoustic modes.

Images