Rotorcraft coaxial contra-rotating sound source positioning method and system

Abstract

This invention relates to a method and system for locating coaxial counter-rotating sound sources in rotorcraft, belonging to the technical field of sound source identification methods. It includes: processing the sound pressure signal of the moving sound source using continuous wavelet transform to obtain time-frequency domain variation results; performing Doppler effect correction based on experimental conditions; estimating the spatial location and relative intensity of the noise source using a cross-spectral matrix and weight vector; and performing phase-locked analysis of the rotor noise source intensity using a phase-locked averaging method. This invention differs from traditional time-domain delay superposition algorithms and frequency-domain cross-correlation spectrum methods. It can image unsteady sound sources using time-frequency domain characteristics, solving the acoustic imaging problem of counter-rotating propeller noise and possessing potential applications in the rotorcraft industry.

Description

Technical Field

[0001] This invention relates to the field of sound source identification technology, specifically to a method and system for locating coaxial rotating sound sources in rotorcraft. Background Technology

[0002] The coaxial counter-rotating propeller is an old propulsion concept, dating back to Lanchester's patent in 1907. However, this concept suffers from problems such as mechanical complexity, manufacturing cost, increased weight, vibration, propeller oscillation, and noise. Among these, the noise problem of counter-rotating propellers remains an unsolved challenge. For stationary sound sources, beamforming is a measurement technique used to estimate the location of noise sources and has been widely applied in aerospace acoustics research to identify major aircraft noise sources. Most classical beamforming methods, including delay summation and conventional beamforming, are effective in both the time and frequency domains and can successfully handle both stationary sound sources and noise sources from moving aircraft flying over them. To address specific aerospace applications, several advanced acoustic imaging methods have been proposed based on these classical beamforming methods. However, existing acoustic imaging methods cannot accurately locate rotating target sound sources.

[0003] To address the problem of rotating noise sources, Sijtsma first developed the so-called Rotating Source Identifier (ROSI), extending the delay summation method to locate noise sources in a constantly rotating reference frame. In the ROSI method, Sijtsma employed a "de-Dopplerization" technique in the time domain to account for the rotational effect (Doppler effect). Later, Doughery and Walker, along with Feng et al., proposed using a virtual rotating array to account for the rotating source. They employed a circular phased array coaxial with the propeller and virtually set the rotation by processing signals recorded by a microphone array. Classical beamforming methods were then applied to the processed signal. Furthermore, Pannert and Maier utilized a circular phased array and proposed a mode decomposition-based method. Guérin and Weckmüller proposed a beamforming method based on short-time Fourier transform, which utilizes Fourier transforms based on time delay and the Doppler effect to achieve "de-Dopplerization." These methods can be applied to steady-state rotation problems, but they encounter difficulties with transient phenomena, such as those described by Fleury and Chélius in their work using these methods on a counter-rotating device. They found that the noise source was located at positions before and after the plane of rotation, positions where no propeller noise was generated.

[0004] Continuous wavelet transform (CWT) methods can analyze sound pressure signals in the time-frequency domain. Combined with acoustic imaging methods, this can be used to trace the sound sources generated by counter-rotating propellers and analyze noise generation mechanisms. This time-frequency domain-based beamforming method can be applied to analyze stator-rotor interactions, shear layer-propeller interactions, and blade vortex interactions. However, unlike the interaction problems discussed in previous studies, the noise sources in counter-rotating propellers have different rotational directions. The coherence between noise sources on different propeller blades leads to low spatial resolution in acoustic imaging and poses a challenge to wavelet-based beamforming methods. Summary of the Invention

[0005] To address the problem of ambiguous sound source localization caused by the noise coherence and Doppler frequency shift of coaxial counter-rotating propellers, this invention provides a method for localizing coaxial counter-rotating sound sources in rotorcraft, comprising:

[0006] S1: Perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal;

[0007] S2: Calculate the displacement vector based on the scanning point and sensor coordinates, calculate the velocity vector of the point on the blade based on time and blade rotation speed, and correct the sound pressure signal based on the Doppler effect to obtain the corrected time-frequency domain signal;

[0008] S3: Construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients;

[0009] S4: Through phase-locked averaging, the noise sources of opposite rotation in the coaxial counter-rotating propeller are separated, and the spatiotemporal distribution results of noise within the rotation period are generated.

[0010] Furthermore, in S1, the time-frequency domain sound pressure signal passes through:

[0011]

[0012] Obtain, among which, It is a time-frequency domain sound pressure signal. This is the microphone sensor serial number. The time of signal acquisition. To calculate the frequency for continuous wavelet transform, It's a Morse wavelet. The scaling function, For sensors In time The original sound pressure signal;

[0013] Morse wavelet The scaling function is obtained through:

[0014]

[0015] Obtain, among which, Morse wavelet The center frequency is 1Hz. The sampling frequency.

[0016] Furthermore, in S2, the corrected time-frequency domain signal is processed through:

[0017]

[0018] Obtain, among which, This is the corrected time-frequency domain signal. The propagation time from the scanning surface to the sensor. This represents the actual frequency of the generated noise.

[0019] Furthermore, in S3, the cross-spectral matrix is ​​obtained through:

[0020]

[0021] Obtain, among which, Indicates function conjugation;

[0022] Imaging results passed:

[0023]

[0024] get, For the imaging results, For the frequency before correction for the Doppler effect, These are the weighting coefficients.

[0025] Furthermore, in S4, the spatiotemporal distribution of noise during the rotation period is obtained through:

[0026]

[0027] Obtain, among which, The results show the spatiotemporal distribution of noise during the rotation period. To calculate the time, The average number of periods, For period counting index, The time it takes to complete one rotation.

[0028] It also provides a method and system for locating coaxial counter-rotating sound sources in rotorcraft, including:

[0029] The transformation module is used to perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal;

[0030] The correction module is used to calculate the displacement vector based on the scanning point and sensor coordinates, calculate the velocity vector of the point on the blade based on time and blade rotation speed, and correct the sound pressure signal based on the Doppler effect to obtain the corrected time-frequency domain signal.

[0031] The imaging module is used to construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients.

[0032] The separation module is used to separate the noise sources of the opposite rotation in the coaxial counter-rotating propeller through phase-locked averaging processing, and generate the spatiotemporal distribution results of the noise within the rotation period.

[0033] The beneficial effects of this invention are:

[0034] This invention proposes a method for identifying counter-rotating sound sources in rotor experiments. It employs continuous wavelet transform to process the sound pressure signal of the moving sound source, obtaining time-frequency domain variation results, thus solving the problem that traditional sound source identification methods cannot identify moving sound sources. Doppler effect correction is applied according to the test conditions to reduce the impact of Doppler frequency shift caused by blade rotation on the positioning results. A wavelet transform-based sound source imaging method is used to calculate the instantaneous change process of the counter-rotating rotor noise. A phase-locked averaging method is employed to average the noise source in different rotational directions, reducing background noise interference and highlighting the main noise from the front and rear blades, thereby improving the accuracy of the acoustic imaging results. Attached Figure Description

[0035] Figure 1 This is a flowchart of the method of the present invention;

[0036] Figure 2 A schematic diagram for setting up the experiment;

[0037] Figure 3 Spatial distribution diagram of 7kHz noise sources in a counter-rotating propeller;

[0038] Figure 4 This is an average acoustic imaging result of the noise of a counter-rotating propeller at 4800 rpm. Detailed Implementation

[0039] The technical solution of the present invention will be further described below with reference to embodiments, but it is not limited thereto. Any modifications or equivalent substitutions to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be covered within the protection scope of the present invention. In the following embodiments, the process equipment or apparatus not specifically specified are all conventional equipment or apparatus in the art. Unless otherwise specified, the raw materials used in the embodiments of the present invention are all commercially available; unless otherwise specified, the technical means used in the embodiments of the present invention are all conventional means well known to those skilled in the art.

[0040] Example 1, combined with Figure 1 This embodiment describes a method for locating a coaxial counter-rotating sound source in a rotorcraft, including:

[0041] S1: Perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal;

[0042] S2: Calculate the displacement vector based on the scanning point and sensor coordinates, calculate the velocity vector of the point on the blade based on time and blade rotation speed, and correct the sound pressure signal based on the Doppler effect to obtain the corrected time-frequency domain signal;

[0043] S3: Construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients;

[0044] S4: Through phase-locked averaging, the noise sources of opposite rotation in the coaxial counter-rotating propeller are separated, and the spatiotemporal distribution results of noise within the rotation period are generated.

[0045] Specifically, the placement of the testing instruments is as follows: Figure 2 As shown, the test was conducted in an anechoic chamber. This invention employed two counter-rotating propellers, each 0.22 meters in diameter and 0.02 meters axially separated. The front and rear propellers rotated clockwise and counter-clockwise at the same speed, periodically overlapping on the x and y axes. Here, the origin of the rotation angle was defined as the position where the front and rear propellers overlapped on the x-axis, and the encoder generated a pulse signal at this position. A 56-channel microphone array was used to measure the noise signal generated by the propellers. This array was parallel to the plane of rotation and 0.8 meters from the center of the propellers. The propeller rotation angle and the corresponding acoustic signal were simultaneously acquired using a National Instruments acquisition system at a sampling frequency of 50 kHz. Sound pressure level (SPL) was calculated for 50 blocks (based on the acquired pulse signals), each block containing 1 × 10,000 samples, to obtain reliable statistical results.

[0046] This invention uses continuous wavelet transform to process the sound pressure signal of a moving sound source to obtain the time-frequency domain variation results; it performs Doppler effect correction based on the experimental conditions; it uses cross-spectral matrix and weight vector to estimate the spatial location and relative intensity of the noise source; and it uses phase-locked averaging to perform phase-locked analysis on the rotor noise source intensity, ultimately obtaining the accurate location of the noise generated by the rotor.

[0047] In S1, the time-frequency domain sound pressure signal passes through:

[0048]

[0049] Obtain, among which, It is a time-frequency domain sound pressure signal. This is the microphone sensor serial number. The time of signal acquisition. To calculate the frequency for continuous wavelet transform, It's a Morse wavelet. The scaling function, For sensors In time The original sound pressure signal;

[0050] Morse wavelet The scaling function is obtained through:

[0051]

[0052] Obtain, among which, Morse wavelet The center frequency is 1Hz. The sampling frequency.

[0053] Specifically, this step uses continuous wavelet transform to process the sound pressure signal of the moving sound source, obtains the time-frequency domain change results, and solves the problem that traditional sound source identification methods cannot identify moving sound sources.

[0054] In S2, the corrected time-frequency domain signal passes through:

[0055]

[0056] Obtain, among which, This is the corrected time-frequency domain signal. The propagation time from the scanning surface to the sensor. This represents the actual frequency of the generated noise.

[0057] Specifically, this step involves correcting for the Doppler effect based on the test conditions to reduce the impact of the Doppler frequency shift caused by the blade rotation on the positioning results.

[0058] The actual noise frequency is obtained through:

[0059]

[0060] Obtain, among which, For the frequency before correction for the Doppler effect, This is the Doppler frequency shift coefficient. This represents the Doppler frequency shift.

[0061] Doppler frequency shift coefficients are obtained through:

[0062]

[0063] Obtain, among which, The rotor rotational velocity vector, For the propagation speed of the medium wave, The angle between the rotor speed direction and the sound wave propagation direction.

[0064] The Doppler frequency shift satisfies:

[0065]

[0066] The angle between the rotor velocity direction and the sound wave propagation direction satisfies the following relationship with the rotor rotation velocity vector:

[0067]

[0068] in, The displacement vector originates from a point on the scanning surface and terminates at the sensor.

[0069] The propagation time from the scanning surface to the sensor is as follows:

[0070]

[0071] get.

[0072] In S3, the cross-spectral matrix is ​​obtained through:

[0073]

[0074] Obtain, among which, Indicates function conjugation;

[0075] Imaging results passed:

[0076]

[0077] get, For the imaging results, For the frequency before correction for the Doppler effect, These are the weighting coefficients.

[0078] Specifically, this step utilizes a wavelet transform-based sound source imaging method to calculate the instantaneous change process of the rotor noise.

[0079] Weighting coefficients passed:

[0080]

[0081] Obtain, among which, Represents norm calculation, This is the Doppler-corrected steering vector.

[0082] The Doppler-corrected steering vector passes through:

[0083]

[0084] get.

[0085] Figure 3 The acoustic imaging results of the propeller at different rotational speeds at 7 kHz are presented.

[0086] In S4, the spatiotemporal distribution of noise during the rotation period is obtained through:

[0087]

[0088] Obtain, among which, The results show the spatiotemporal distribution of noise during the rotation period. To calculate the time, The average number of periods, For period counting index, The time it takes to complete one rotation.

[0089] Specifically, this step uses a phase-locked averaging method to average the noise sources in different rotation directions, reduce background noise interference, highlight the main noise from the front and rear blades, and improve the accuracy of acoustic imaging results.

[0090] Over a long period of observation of a specified propeller, noise sources with different rotation directions and speeds will become background noise through long-term averaging. The experimental processing results are as follows: Figure 4 As shown, the average acoustic imaging results of the clockwise and counterclockwise rotating propellers at 4800 rpm are presented. The noise sources are averaged in different rotation directions to highlight the main noise from the front and rear propellers.

[0091] It also provides a method and system for locating coaxial counter-rotating sound sources in rotorcraft, including:

[0092] The transformation module is used to perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal;

[0093] The correction module is used to calculate the displacement vector based on the scanning point and sensor coordinates, calculate the velocity vector of the point on the blade based on time and blade rotation speed, and correct the sound pressure signal based on the Doppler effect to obtain the corrected time-frequency domain signal.

[0094] The imaging module is used to construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients.

[0095] The separation module is used to separate the noise sources of the opposite rotation in the coaxial counter-rotating propeller through phase-locked averaging processing, and generate the spatiotemporal distribution results of the noise within the rotation period.

Claims

1. A method for locating a coaxial counter-rotating sound source in a rotorcraft, characterized in that, include: S1: Perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal; S2: Calculate the displacement vector based on the scanning point and sensor coordinates, and correct the sound pressure signal based on the Doppler effect according to the rotor rotation speed vector to obtain the corrected time-frequency domain signal; S3: Construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients; S4: Through phase-locked averaging, the noise sources of opposite rotation in the coaxial counter-rotating propeller are separated, and the spatiotemporal distribution results of noise within the rotation period are generated. In S1, the time-frequency domain sound pressure signal passes through: ； Obtain, among which, It is a time-frequency domain sound pressure signal. This is the microphone sensor serial number. The time of signal acquisition. To calculate the frequency for continuous wavelet transform, It's a Morse wavelet. The scaling function, For sensors In time The original sound pressure signal; Morse wavelet The scaling function is obtained through: ； Obtain, among which, Morse wavelet The center frequency is 1Hz. The sampling frequency; In S2, the corrected time-frequency domain signal passes through: ； Obtain, among which, This is the corrected time-frequency domain signal. The propagation time from the scanning surface to the sensor. This refers to the actual frequency of the generated noise. The actual noise frequency is obtained through: ； Obtain, among which, For the frequency before correction for the Doppler effect, This is the Doppler frequency shift coefficient. This is the Doppler frequency shift; Doppler frequency shift coefficients are obtained through: ； Obtain, among which, The rotor rotational velocity vector, For the propagation speed of the medium wave, The angle between the rotor speed direction and the sound wave propagation direction; The Doppler frequency shift satisfies: ； The angle between the rotor velocity direction and the sound wave propagation direction satisfies the following relationship with the rotor rotation velocity vector: ； in, The displacement vector originating from a point on the scanning surface and ending at the sensor. The propagation time from the scanning surface to the sensor is as follows: ； get.

2. The method for locating the coaxial counter-rotating sound source of a rotorcraft according to claim 1, characterized in that, In S3, the cross-spectral matrix is ​​obtained through: ； Obtain, among which, Indicates function conjugation; Imaging results passed: ； get, For the imaging results, For the frequency before correction for the Doppler effect, These are the weighting coefficients.

3. The method for locating the coaxial counter-rotating sound source of a rotorcraft according to claim 2, characterized in that, In S4, the spatiotemporal distribution of noise during the rotation period is obtained through: ； Obtain, among which, The results show the spatiotemporal distribution of noise during the rotation period. To calculate the time, The average number of periods, For period counting index, The time it takes to complete one rotation.

4. A method and system for locating the coaxial counter-rotating sound source of a rotorcraft, characterized in that, include: The transformation module is used to perform continuous wavelet transform on the sound pressure signal acquired by the microphone array to obtain the time-frequency domain sound pressure signal; The correction module is used to calculate the displacement vector based on the scanning point and sensor coordinates, and correct the sound pressure signal based on the Doppler effect according to the rotor rotation speed vector to obtain the corrected time-frequency domain signal. The imaging module is used to construct a cross-spectral matrix using the corrected time-frequency domain signal, and calculate the sound source imaging results by combining the steering vector and weighting coefficients. The separation module is used to separate the noise sources of counter-rotating coaxial counter-rotating propellers through phase-locked averaging processing, and generate the spatiotemporal distribution results of noise within the rotation period. In the transformation module, the time-frequency domain sound pressure signal is processed through: ； Obtain, among which, It is a time-frequency domain sound pressure signal. This is the microphone sensor serial number. The time of signal acquisition. To calculate the frequency for continuous wavelet transform, It's a Morse wavelet. The scaling function, For sensors In time The original sound pressure signal; Morse wavelet The scaling function is obtained through: ； Obtain, among which, Morse wavelet The center frequency is 1Hz. The sampling frequency; In the correction module, the corrected time-frequency domain signal is processed through: ； Obtain, among which, This is the corrected time-frequency domain signal. The propagation time from the scanning surface to the sensor. This refers to the actual frequency of the generated noise. The actual noise frequency is obtained through: ； Obtain, among which, For the frequency before correction for the Doppler effect, This is the Doppler frequency shift coefficient. This is the Doppler frequency shift; Doppler frequency shift coefficients are obtained through: ； Obtain, among which, The rotor rotational velocity vector, For the propagation speed of the medium wave, The angle between the rotor speed direction and the sound wave propagation direction; The Doppler frequency shift satisfies: ； The angle between the rotor velocity direction and the sound wave propagation direction satisfies the following relationship with the rotor rotation velocity vector: ； in, The displacement vector originating from a point on the scanning surface and ending at the sensor. The propagation time from the scanning surface to the sensor is as follows: ； get.

Images