Aircraft noise source positioning method and device, medium and application

Abstract

The invention relates to an aircraft noise source positioning method and device, a medium and application, and the method comprises the steps: S1, grouping M microphone arrays to form m sub-arrays, M and m being integers; s2, aiming at each sub-array, solving sound source space distribution to obtain a sound source positioning result of each sub-array; and S3, multiplying the m sound source localization results by the m-power root of the product to obtain a result as a final sound source localization result. The method provided by the invention can be suitable for more complex and more practical microphone array distribution, and a sound source positioning result with higher precision can be obtained.

Description

Technical Field

[0001] This invention belongs to the field of acoustic measurement and aeroacoustics technology, specifically relating to a method, device, medium, and application for locating aircraft noise sources. Background Technology

[0002] Aeroacoustic noise testing is an effective means of studying aeroacoustic noise levels and generation mechanisms. In traditional acoustic testing methods, measurement results cannot pinpoint the noise generation area or the amplitude intensity of individual sound sources. The emergence and development of phased microphone array technology in recent years aims to accurately identify the location of noise sources and quantify their intensity, providing a powerful tool for studying noise generation mechanisms and defining noise reduction targets. Beamforming algorithms have been widely used in phased microphone array sound source localization technology. The traditional Delay-Sum (DAS) algorithm obtains the sound source signal by delaying the sound pressure signals received by each array sensor and then summing them. Traditional DAS algorithms have good robustness but suffer from low resolution and high sidelobes. Figure 1 As shown, in 2004, Brooks et al. developed a complete deconvolutional DAMAS algorithm, establishing an inverse problem-solving model containing information about unknown sound sources, which is considered a key breakthrough in microphone array post-processing technology. The DAMAS algorithm establishes a system of linear equations between the sound source intensity at grid points and the DAS algorithm output, and uses the non-negative constraint Gauss-Seidel iterative method to solve this system, improving the sound source localization resolution by more than five times. While the DAMAS algorithm significantly improves the ability of microphone arrays to locate sound sources, it has two main drawbacks: firstly, the non-negative constraint Gauss-Seidel iterative method used in the DAMAS algorithm has a slow convergence speed and long computation time; secondly, the DAMAS algorithm makes the assumption of sound source incoherence. To improve the computational speed of the DAMAS algorithm, Dougherty proposed the DAMAS2 and DAMAS3 algorithms in 2005. In 2012, Huang et al. replaced the DAS algorithm with an adaptive beamforming algorithm, and then used the DAMAS algorithm, finding that it accelerated the convergence speed of the DAMAS algorithm when processing experimental data.

[0003] In recent years, researchers have applied compressed sensing theory to sound source localization. Li et al. developed the SC-DAMAS, CMF, CMF-C, and MACS algorithms. They used interior-point convex optimization algorithms to solve sparse optimization problems, and the SC-DAMAS and CMF algorithms to solve incoherent sound source models. The SC-DAMAS algorithm solves the linear equations established by the DAMAS algorithm, but uses the results of the frequency-domain DAS algorithm to select sound source grid points, thus reducing the size of the linear equations. The CMF algorithm solves the optimization problem established by the cross-spectral matrix (CSM). Numerical experiments show that the SC-DAMAS and CMF algorithms have better robustness than the DAMAS algorithm. Bai and Kuo used compressed sensing convex optimization algorithms to localize sound sources. Zhong et al. proposed a compressed sensing algorithm based on the sampling covariance matrix to solve the two-dimensional sound source localization problem. Ning et al. applied the orthogonal matching pursuit algorithm from the compressed sensing greedy algorithm class to sound source localization. Bai Baohong et al. further improved the accuracy of sound source localization by introducing a non-convex sparse regularization method to solve the acoustic inverse problem. They also located two noise sources at the side edge of the flap for the first time, and experimentally verified the theoretical hypothesis of two sound sources at the side edge of the flap for the first time.

[0004] Furthermore, to achieve better sound source localization, the array configuration needs to be optimized to meet the performance requirements of spatial resolution and dynamic range. Examples include multi-arm spiral array design, iterative optimization design, global optimization methods, and adaptive array optimization design. However, these methods are all for a single array. When the target frequency range of the array is large, a single array configuration can hardly simultaneously meet the array resolution and dynamic range requirements. Simultaneously, the required array channel size increases further, necessitating a nested design of multiple subarrays. Different subarrays are selected for different frequency ranges, and each subarray shares some microphones to improve microphone channel utilization. The most representative work is NASA's nested array design method used in its field noise source localization tests for large passenger aircraft.

[0005] In actual field testing of aircraft noise sources, many practical problems arise. For example, the ground on which the microphone array is placed is not flat, with height differences. Moreover, the microphones are often not placed directly on the ground, but rather in various placement methods. These height differences and different placement methods can cause phase errors in the noise signals measured by the microphones, leading to significant errors in the final sound source localization results and making it impossible to obtain satisfactory sound source localization results.

[0006] For example, microphone height error can significantly affect the sound source localization quality. Because the field test sites for aircraft noise sources are often not flat, the maximum elevation difference within the entire array layout area is close to 1m. Figure 2Taking a multi-arm spiral array as an example, when the simulated sound source height is 120m, the influence of a random error of approximately ±10cm in the microphone array installation height on the sound source localization results is observed. Numerical simulation experiments show that when there is a random error of approximately ±10cm in the radial installation height of the microphone array, the low-frequency 630Hz localization result is relatively clear. Figure 2 (a) shows the positioning results at 630Hz when there is no height difference. Figure 2 (b) shows the localization result at 630Hz when there is a height difference, which can accurately distinguish the three main sound sources; the localization result at 1000Hz (mid-frequency) Figure 2 (c) in the figure represents the positioning result at 1000Hz when there is no height difference. Figure 2 (d) shows the localization result at 1000Hz when there is a height difference. Three main sound sources can be seen, but many false sound sources also appear, and the localization result begins to deteriorate. The high frequencies, approaching 1600Hz, have a very significant impact, affecting the localization effect. Figure 2 (e) in the figure represents the positioning result at 1600Hz when there is no height difference. Figure 2 (f) shows the localization result at 1600Hz when there is a height difference. The result is significantly worse, making it impossible to distinguish the three main sound sources. This indicates that height difference has a significant impact on the localization results of high-frequency sound sources. Therefore, it is necessary to develop sound source localization methods with higher accuracy for microphone arrays with height differences. Summary of the Invention

[0007] In order to overcome the above-mentioned problems in the prior art, the present invention provides an aircraft noise source localization method, device, medium and application to solve the above-mentioned problems in the prior art.

[0008] A method for locating aircraft noise sources, the method comprising: S1. Divide the M microphone arrays into m subarrays, where M and m are both positive integers; S2. For each subarray, solve for the spatial distribution of sound sources to obtain the sound source localization results for each subarray; S3. Multiply the above m sound source localization results by the m-th root of the product and take the result as the final sound source localization result.

[0009] In addition to the aspects described above and any possible implementation, an implementation is further provided in which the same group of microphones in each subarray is mounted in the same manner, and the height difference of the microphones in the same group is less than or equal to 2% of the aperture of the subarray.

[0010] In addition to the aspects described above and any possible implementation, a further implementation is provided in which the largest subarray aperture is less than or equal to twice the smallest subarray aperture among all subarray apertures.

[0011] In addition to the aspects and any possible implementations described above, an implementation is further provided in which S2 includes: S21. Obtain the sound pressure signal in the frequency domain of the sound source received by all microphones in each subarray; S22. Calculate the cross-spectral matrix of each subarray based on the frequency domain sound pressure signal; assuming that the interference noise is incoherent, force the diagonal elements of the cross-spectral matrix to be 0. S23. In the spatial region where the actual sound source is located, divide the sound source scanning surface into a grid, and assume that each scanning surface grid point is a monopole sound source, and construct the guide vector from each scanning surface grid point to the microphone subarray; S24. Estimate the sound source intensity at each grid point in the sound source scanning grid region; S25. Perform the above steps for each subarray and use the resulting sound source intensities as the sound source localization results.

[0012] As described above and in any possible implementation, a further implementation is provided, wherein the sound source frequency domain sound pressure signal It is obtained by adding the product of the sound propagation matrix and the frequency domain amplitude vector of the sound source to the interference noise signal vector received by the microphone.

[0013] In addition to the aspects described above and any possible implementations, a further implementation is provided, wherein the vector calculation formula for the acoustic propagation matrix is: , For the first The sound source scanning grid points are from the first sound source to the second. The distance of one microphone, Here, j is the wave number, and the superscript T is the transpose.

[0014] As described above and in any possible implementation, a further implementation is provided, wherein the cross-spectral matrix in S22 Where H represents the conjugate transpose. It represents the average.

[0015] The present invention also provides an aircraft noise source localization device, the device being used to implement the method, comprising: The grouping module is used to group the M microphone arrays into m subarrays, where M and m are both positive integers; The solution module is used to solve the spatial distribution of sound sources for each subarray and obtain the sound source localization results for each subarray. The calculation module is used to multiply the above m sound source localization results by the m-th root of the product and use the result as the final sound source localization result.

[0016] The present invention also provides a computer storage medium storing a computer program, the computer program being executed by a processor to implement the method described.

[0017] This invention also provides an application of an aircraft noise source localization method on an aircraft. Beneficial effects of the present invention The aircraft noise source localization method of the present invention includes: S1. grouping M microphone arrays into m subarrays; S2. solving the spatial distribution of sound sources for each subarray to obtain the sound source localization result for each subarray; S3. multiplying the above m sound source localization results by the m-th root of the product to obtain the final sound source localization result. The method of the present invention is suitable for more complex and practical microphone array distributions, and can obtain higher accuracy sound source localization results.

[0018] It has the following beneficial effects: 1) This invention is applicable to complex microphone array distributions with ultra-large apertures, and the microphone array grouping does not need to be determined in advance; 2) The subarray division method of the present invention ensures that the height difference of each subarray does not exceed 2% of the aperture, the apertures of all subarrays do not differ too much, and the aperture of the largest subarray does not exceed twice the aperture of the smallest subarray. 3) This invention establishes a novel sound source localization algorithm that combines the sound source localization results of multiple subarrays. Attached Figure Description

[0019] Figure 1 A schematic diagram of sound source localization using a microphone array in the prior art; Figure 2 This is a simulation diagram illustrating the impact of random errors in microphone height on positioning results in existing technologies. Figure 3 This is a flowchart of the sound source localization method of the present invention; Figure 4 For the performance evaluation of the external noise source test array for domestically produced aircraft, (a) and (b) in the figure are the actual array geometric coordinates considering the height difference encountered during actual testing; (c) is the sound source result obtained using the existing technology method, with the sound source height being 80m and the frequency being 400Hz; (d) is the sound source result obtained using the new method. Figure 5 This is a schematic diagram showing the results of locating an external noise source at a frequency of 1000Hz using a single subarray. Figure 6 This is a schematic diagram showing the results of locating an external noise source at a frequency of 1000Hz using two subarrays and a new algorithm. Detailed Implementation

[0020] To better understand the technical solution of this invention, the content of this invention includes, but is not limited to, the specific embodiments described below. Similar technologies and methods should be considered within the scope of protection of this invention. To make the technical problems to be solved, the technical solutions, and advantages of this invention clearer, a detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments.

[0021] It should be understood that the embodiments described in this invention are merely some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0022] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0023] This invention discloses a method for locating aircraft noise sources, the method comprising: S1. Divide the M microphone arrays into m subarrays, where M and m are both positive integers; S2. For each subarray, solve for the spatial distribution of sound sources to obtain the sound source localization results for each subarray; S3. Multiply the above m sound source localization results by the m-th root of the product and take the result as the final sound source localization result.

[0024] The technical problem solved by this invention is that, during actual testing of ultra-large aperture microphone arrays, uneven test surfaces and different microphone installation methods cause phase errors in the noise signals measured by the microphones, ultimately leading to large sound source localization errors. By adopting the above method, this invention improves the sound source localization accuracy and saves testing costs.

[0025] Furthermore, S2 includes: S21. Obtain the sound pressure signal in the frequency domain of the sound source received by all microphones in each subarray; S22. Calculate the cross-spectral matrix of each subarray based on the frequency domain sound pressure signal; assuming that the interference noise is incoherent, force the diagonal elements of the cross-spectral matrix to be 0. S23. In the spatial region where the actual sound source is located, divide the sound source scanning surface into a grid, and assume that each scanning surface grid point is a monopole sound source, and construct the guide vector from each scanning surface grid point to the microphone array; S24. The DAS algorithm is used to estimate the sound source intensity at each grid point in the sound source scanning grid region; S25. Perform the above steps for each subarray and use the resulting sound source intensities as the sound source localization results.

[0026] Specifically, the method for locating aircraft noise sources applicable to complex microphone arrays proposed in this invention has the following specific process: Figure 3 As shown: Step 1: Divide the existing large-aperture microphone array (M microphones) into groups (overlapping is allowed, and some microphones can be reused) to form m subarrays. Each subarray can contain multiple microphones, and a single microphone may appear in multiple subarrays. Grouping ensures that microphones in the same group are installed in the same way within a subarray, and that the height difference between microphones in the same group is small. The subarray division criterion is that the height difference between each subarray does not exceed 2% of the aperture. Height difference refers to the height difference between all microphones in the subarray; ideally, all microphones should be at the same height on the same plane, meaning there is no height difference. Aperture refers to the maximum distance between all microphones in a microphone array. A larger aperture provides better localization of low-frequency sound sources. Each subarray has a corresponding aperture, which is the maximum or farthest distance between any two microphones in that subarray. Setting the above percentage limit aims to minimize the height difference between microphones in each subarray, thus ensuring more accurate sound source localization results from that subarray.

[0027] At the same time, the apertures of all subarrays cannot differ too much; that is, the aperture of the largest subarray cannot exceed twice the aperture of the smallest subarray. As mentioned earlier, each subarray has a unique aperture, which is equal to the maximum distance between any two microphones in that subarray. Therefore, this aperture represents the size of the subarray. The larger the aperture, the larger the subarray, and the better the localization result for low-frequency sound sources. The above convention restricts the size of all subarrays from differing too much. If the apertures differ too much, the localization results of different subarrays will differ too much, resulting in a large error in the final joint localization result.

[0028] Step 2: For each subarray, use existing arbitrary sound source localization methods to solve for the spatial distribution of sound sources and obtain the sound source localization results. ; Step 3: Based on the localization results of these m sound sources, multiply these sound source localization results and take the m-th square root of the result as the final sound source localization result.

[0029] The calculation process for the second step is as follows: First, establish a microphone array signal reception model, assuming the microphone array contains a certain number of microphones. In the spatial region where the actual sound source is located, a sound source scanning area grid is generated, with the number of grid points in the area being [number missing]. , , Assuming that each grid point is a monopole point sound source and is a positive integer, the frequency domain sound pressure signal received by the microphone in free space is: (1) in, The sound pressure level signals received by all the microphones in the array. The frequency domain amplitude vector of the sound source. Let be the vector of the interference noise signal received by the microphone. This interference noise originates from background noise, such as the noise created by wind blowing across the microphone. For the first The sound source scanning grid points are from the first sound source to the second. The distance between the microphones, It is a positive integer. For wave number, Let be the sound propagation matrix, where Let be the column vector of the sound propagation matrix. The calculation formula is as follows: (2) This invention first calculates the cross-spectrum matrix G (CSM) based on the sound pressure signals acquired by the microphone array. The formula for calculating the cross-spectrum matrix G is as follows: (3) Where H represents the conjugate transpose. Represents average, The sound pressure signal is calculated using formula (1).

[0030] Subsequently, this invention uses the DAS algorithm to estimate each grid point in the sound source scanning area. Sound source intensity The coordinates of this grid point are The expression is: (4) Where G is the cross-spectral matrix, It is calculated by formula (3). The guide vector is calculated using the following formula: (5) From the above formula, it can be seen that when the interference noise e is uncorrelated white noise, the steering vector... It is a diagonal matrix, and the elements of each diagonal matrix are... Therefore, it only affects the elements on the diagonal of the cross-spectrum matrix G. In order to reduce the impact of noise on the sound source localization results, this invention adopts the cross-spectrum matrix diagonal restraint technique, that is, the diagonal elements of the cross-spectrum matrix G are forced to be set to 0, and then the sound source is localized using formula (4). Therefore, the area with a larger output value of the DAS algorithm is considered to be the area where the sound source is located. Finally, the sound source localization results of all m sound sources are calculated to obtain , For each sound source localization result, the above formula (4) is used for calculation.

[0031] Finally, based on all the localization results of these m sound sources The final sound source localization result is obtained by multiplying these sound source localization results and taking the m-th root of the result. The formula is as follows: (6).

[0032] As an embodiment of the present invention, the present invention provides an aircraft noise source localization device, the device being used to implement the method, comprising: The grouping module is used to group the M microphone arrays into m subarrays; The solution module is used to solve the spatial distribution of sound sources for each subarray and obtain the sound source localization results for each subarray. The calculation module is used to multiply the above m sound source localization results by the m-th root of the product and use the result as the final sound source localization result.

[0033] As an embodiment of the present invention, the present invention provides a computer storage medium storing a computer program, the computer program being executed by a processor to implement the method described herein.

[0034] As an embodiment of the present invention, the present invention provides an application of an aircraft noise source localization method on an aircraft.

[0035] Figure 4 The performance evaluation of the microphone array for testing external noise sources on domestically produced aircraft is presented. Figure 4 (a) and (b) are the actual array geometric coordinates considering the height difference encountered during actual testing; (c) is the sound source result obtained using the existing old method, with a sound source height of 80m and a frequency of 400Hz. It can be seen that the existing old method produces many side lobes, i.e., pseudo sound sources, and cannot accurately distinguish the real sound source; (d) is the sound source result obtained using the new method. It can be seen that the localization result using this method can accurately distinguish the main sound source, and the side lobe sound sources are significantly reduced.

[0036] To further verify the sound source localization algorithm proposed in this invention, the following verification is performed using field flight test data from a domestically produced aircraft. Figure 5 The results of locating external noise sources on a domestically produced aircraft using all microphones are presented. It can be seen that the location results are very blurry. Figure 6 The results of locating external noise sources of an aircraft using the sound source localization algorithm proposed in this invention are presented. It can be seen that the method significantly improves the accuracy of locating aircraft noise sources, noticeably reduces sidelobe noise sources, and makes the noise sources clearer.

[0037] The foregoing description illustrates and describes several preferred embodiments of the present invention. However, as previously stated, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the inventive concept described herein through the foregoing teachings or techniques or knowledge in related fields. Any modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for locating aircraft noise sources, characterized in that, The method includes: S1. Divide the M microphone arrays into m subarrays, where M and m are both positive integers; S2. For each subarray, solve for the spatial distribution of sound sources to obtain the sound source localization results for each subarray; S3. Multiply the above m sound source localization results by the m-th root of the product and take the result as the final sound source localization result.

2. The method according to claim 1, characterized in that, The microphones in the same group within each subarray are installed in the same way, and the height difference between the microphones in the same group is less than or equal to 2% of the aperture of the subarray.

3. The method according to claim 2, characterized in that, The aperture of the largest subarray is less than or equal to twice the aperture of the smallest subarray.

4. The method according to claim 1, characterized in that, S2 includes: S21. Obtain the sound pressure signal in the frequency domain of the sound source received by all microphones in each subarray; S22. Calculate the cross-spectral matrix of each subarray based on the frequency domain sound pressure signal; assuming that the interference noise is incoherent, force the diagonal elements of the cross-spectral matrix to be 0. S23. In the spatial region where the actual sound source is located, divide the sound source scanning surface into a grid, and assume that each scanning surface grid point is a monopole sound source, and construct the guide vector from each scanning surface grid point to the microphone subarray; S24. Estimate the sound source intensity at each grid point in the sound source scanning grid region; S25. Perform the above steps for each subarray and use the resulting sound source intensities as the sound source localization results.

5. The method according to claim 4, characterized in that, The sound source frequency domain sound pressure signal It is obtained by adding the product of the sound propagation matrix and the frequency domain amplitude vector of the sound source to the interference noise signal vector received by the microphone.

6. The method according to claim 5, characterized in that, The formula for calculating the column vector of the sound propagation matrix is ​​as follows: , For the first The sound source scanning grid points are from the first sound source to the second. The distance of one microphone, Here, j is the wave number, and the superscript T is the transpose.

7. The method according to claim 4, characterized in that, The cross-spectral matrix in S22 Where H represents the conjugate transpose. It represents the average.

8. An aircraft noise source locator, characterized in that, The apparatus is used to implement the method according to any one of claims 1-7, comprising: The grouping module is used to group the M microphone arrays into m subarrays, where M and m are both positive integers; The solution module is used to solve the spatial distribution of sound sources for each subarray and obtain the sound source localization results for each subarray. The calculation module is used to multiply the above m sound source localization results by the m-th root of the product and use the result as the final sound source localization result.

9. A computer storage medium, characterized in that, The medium stores a computer program, which is executed by a processor to implement the method described in any one of claims 1-7.

10. The application of the aircraft noise source localization method according to any one of claims 1-7 on an aircraft.

Images