Noise source positioning method and system based on multi-integral-surface acoustic monitoring

Abstract

The invention discloses a noise source positioning method and system based on multi-integral-surface acoustic monitoring. The method comprises the following steps: arranging a microphone array on an outermost integral surface, converting a sound pressure signal into a time-frequency domain through short-time Fourier transform, and reconstructing a whole sound field at one time by using a regularization equivalent source method; and then the active sound intensity on the multi-layer nested integral surface is calculated in sequence, and the net sound power is solved, so that the normalized sound power divergence of the adjacent shell layers is obtained. And performing time-frequency accumulation and peak value detection on the divergence spectrum to quickly lock a shell layer where a sound source is located, and determining a three-dimensional coordinate in the shell layer according to an equivalent source energy density extreme value. The device comprises an array module, an acquisition module, a time-frequency conversion module, a reconstruction module, a divergence calculation module and a peak value detection and positioning module, and can complete noise source visualization in real time. The method has the advantages of high spatial resolution, strong noise immunity and millisecond processing speed, and is suitable for online noise source identification and fault diagnosis of vehicles, aerospace and industrial equipment.

Description

Technical Field

[0001] This invention relates to the field of acoustic signal processing and noise source localization technology, and in particular to a noise source localization method and system based on multi-integral surface acoustic monitoring. Background Technology

[0002] Existing noise source localization technologies mainly rely on methods such as acoustic holography, beamforming, or acoustic power methods, but these have many limitations. For example, traditional acoustic holography is easily affected by reflection and reverberation in complex sound fields, limiting its localization accuracy; beamforming methods are sensitive to the direction of the sound source and have difficulty handling multi-source or strong reflection scenarios; while acoustic power methods can reflect the energy distribution of the sound source, they lack precise mapping of spatial location. In addition, existing methods are prone to localization errors in low signal-to-noise ratio or complex sound fields, making it difficult to meet the high-precision requirements for industrial noise source identification.

[0003] In existing technologies, some studies have attempted to improve positioning accuracy through multi-layer integral surfaces or acoustic power difference methods, but their core problems are: 1) Sound field reconstruction methods (such as equivalent source methods) are sensitive to noise, and it is difficult to balance reconstruction accuracy and computational complexity; 2) Acoustic power difference methods do not adequately describe the physical correlation of the spatial distribution of sound sources, making it difficult to directly locate the sound source position; 3) The lack of dynamic modeling of the energy flow of sound sources in complex sound fields makes the positioning results susceptible to environmental interference.

[0004] While existing literature involves multi-layer integral surface or acoustic power analysis, it does not combine acoustic power divergence with optimization algorithms for dynamic localization, nor does it address the multi-source localization problem in complex sound fields. Furthermore, existing methods still need improvement in terms of real-time performance, anti-interference capability, and multi-source separation ability. Summary of the Invention

[0005] To address the shortcomings of the existing technology, this invention provides a noise source localization method based on multi-integral surface acoustic monitoring, comprising the following steps: Sound pressure was measured by arranging an array on the outermost layer. Convert the signal to the time-frequency domain; The entire sound field is reconstructed using the regularized equivalent source method; The sound pressure, particle velocity, and active sound intensity on the multi-layer integral surface are calculated sequentially from the reconstruction results. The net acoustic power was obtained for each area integral; The normalized acoustic power divergence of adjacent shell layers is calculated and accumulated over time and frequency. Peak detection is performed on the accumulated divergence spectrum to pinpoint the shell layer containing the sound source, and the three-dimensional coordinates within that layer are obtained using the extreme value of the equivalent source energy density. The peak shell layer is used to locate the layer where the sound source is located, and the three-dimensional coordinates are obtained within that layer using the extreme value of the equivalent source energy density.

[0006] The shell refers to a three-dimensional spatial region surrounded by two adjacent, virtual nested integral surfaces.

[0007] The acoustic power divergence is a physical quantity used to characterize the net generation rate of acoustic energy within a certain shell. It is calculated by dividing the difference in net acoustic power flowing out of the inner and outer surfaces of the shell by the volume of the shell.

[0008] The integral surface is a series of virtual, nested concentric spheres, ellipsoids, or curved surfaces conforming to the device housing; the microphone array is arranged on the surface of the outermost integral surface or in the measurement area immediately adjacent to it, so as to reduce the condition number of the reconstructed transfer function and improve the positioning stability under large curved surfaces or conformal surfaces.

[0009] The regularization parameter is adaptively updated in each time-frequency unit through the L-curve or GCV. Meanwhile, the equivalent source is arranged in a non-uniform grid with a denser a priori source region and a sparser non-source region, taking into account both the stability and spatial resolution of the solution.

[0010] Among them, the particle velocity is obtained by directly weighting and summing the gradient of the equivalent source Green's function in the frequency domain using the Euler equation, avoiding numerical difference of the measured sound pressure, suppressing high-frequency noise and improving the accuracy of velocity estimation.

[0011] Prior to the step of 'obtaining net acoustic power from each area integral', a background noise suppression step is included: estimating the background sound intensity threshold based on the signal acquired during the passive period, and setting the portion of the active sound intensity vector whose amplitude is lower than the threshold to zero, in order to reduce spurious peaks caused by reverberation and measurement noise.

[0012] In this process, a nonlinear power transform is applied to the time-frequency averaged acoustic power divergence spectrum, and the peak-to-valley ratio is used as the confidence level. Only the high-confidence peaks are retained to enhance the contrast between strong and weak sources and reduce misjudgments.

[0013] When multiple peaks are detected, energy density extremum search is performed independently on each peak shell to achieve multi-source parallel localization, and a list of source intensities and confidence levels is output.

[0014] The initial positioning result is used as a new priori equivalent source arrangement, and the reconstruction-divergence-peak process is executed again to form an iterative closed loop, so that the source coordinates gradually converge to sub-grid accuracy.

[0015] Among them, overlapping summation STFT, pre-decomposed transfer matrix and GPU parallel solution of linear equations are used to achieve millisecond-level frame processing, which meets the real-time positioning requirements of vehicle passing or transient noise of aircraft.

[0016] This invention also proposes a noise source localization system based on multi-integral surface acoustic monitoring, comprising: a microphone array, a data acquisition module, a time-frequency conversion module, an adaptive equivalent source reconstruction module, a multi-layer acoustic power divergence calculation module, a peak detection and three-dimensional localization module, and a terminal for real-time display and storage; the system executes the above method to complete the online localization and visualization output of the noise source.

[0017] Compared with the prior art, the present invention has the following advantages: Multi-layer integral surface and acoustic power divergence analysis. By virtually setting up multi-layer nested integral surfaces and combining them with acoustic power divergence analysis, this method quantifies the distribution differences of sound source energy in different shell regions, significantly improving the ability to perceive the spatial location of sound sources. This method reveals the physical laws of sound source energy flow through the dynamic changes in acoustic power divergence, providing a more direct physical basis for localization. Its core physical principle lies in the fact that acoustic power divergence quantitatively describes the "source" and "sink" of sound energy in space. Within the actual shell where the sound source is located, sound energy is generated internally and radiates outwards, resulting in the shell having the largest positive divergence value; while in passive shells or strong reverberation regions, the sound energy flow tends to be balanced or chaotic, with its divergence value approaching zero or exhibiting noise characteristics. Therefore, by detecting the peak value of the multi-layer shell divergence spectrum rather than the absolute magnitude of sound pressure or sound intensity, it is possible to fundamentally distinguish the radiated energy of the sound source from the interference energy reflected and scattered by the environment, thereby achieving robust localization in complex sound fields.

[0018] Equivalent source reconstruction and optimization algorithm. The sound field is reconstructed using a regularized equivalent source method, combined with optimization algorithms (such as the L-curve method) to balance reconstruction accuracy and computational efficiency, solve noise interference problems in complex sound fields, and improve positioning accuracy.

[0019] Multi-source localization and dynamic optimization. Through collaborative analysis of multi-layer integral surfaces, the separation and localization of multiple sound sources are achieved. The localization results are then combined with feedback correction and iterative optimization to adapt to dynamic changes in complex sound fields.

[0020] Real-time performance and anti-interference capability. Through GPU / FPGA accelerated computing and real-time processing modules, the system meets the real-time requirements of industrial scenarios and enhances anti-interference capabilities.

[0021] The technical effects of this invention are reflected in: 1) positioning accuracy is improved by more than 30%, which is suitable for complex sound fields and multi-source scenarios; 2) anti-interference ability is significantly enhanced, which is suitable for low signal-to-noise ratio environments; 3) multi-source positioning capability is improved, which supports industrial noise source identification and positioning. Attached Figure Description

[0022] The above and other objects, features, and advantages of exemplary embodiments of the present disclosure will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the present disclosure are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein: Figure 1 This is a flowchart illustrating a noise source localization method based on multi-integral surface acoustic monitoring according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating a noise source localization system based on multi-integral surface acoustic monitoring according to an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0024] 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, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0025] It should be understood that although the terms first, second, third, etc., may be used to describe... in the embodiments of the present invention, these... should not be limited to these terms. These terms are only used to distinguish... For example, first... may also be referred to as second... without departing from the scope of the embodiments of the present invention, and similarly, second... may also be referred to as first...

[0026] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0027] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0028] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0029] The optional embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] Example 1 like Figure 1 As shown, this invention discloses a noise source localization method based on multi-integral surface acoustic monitoring, comprising the following steps: Sound pressure was measured by arranging an array on the outermost layer. Convert the signal to the time-frequency domain; The entire sound field is reconstructed using the regularized equivalent source method; The sound pressure, particle velocity, and active sound intensity on the multi-layer integral surface are calculated sequentially from the reconstruction results. The net acoustic power was obtained for each area integral; The normalized acoustic power divergence of adjacent shell layers is calculated and accumulated in time and frequency. Peak detection is performed on the accumulated divergence spectrum to locate the sound source layer corresponding to the peak value. The three-dimensional coordinates are obtained in the layer with the extreme value of the equivalent source energy density. The peak shell layer is used to locate the layer where the sound source is located, and the three-dimensional coordinates are obtained within that layer using the extreme value of the equivalent source energy density.

[0031] The equivalent source energy density refers to the sound energy density generated by the equivalent source at a point r in space after the sound field is reconstructed using the equivalent source method. Its calculation formula is: , where q s For the equivalent source intensity, G s (r) is the transfer function, ρ0 is the density of the medium, and c is the speed of sound.

[0032] The integral surface is a series of virtual, nested concentric spheres, ellipsoids, or curved surfaces conforming to the device housing; the microphone array is arranged on the surface of the outermost integral surface or in the measurement area immediately adjacent to it, so as to reduce the condition number of the reconstructed transfer function and improve the positioning stability under large curved surfaces or conformal surfaces.

[0033] The regularization parameter is adaptively updated in each time-frequency unit through the L-curve or GCV. Meanwhile, the equivalent source is arranged in a non-uniform grid with a denser a priori source region and a sparser non-source region, taking into account both the stability and spatial resolution of the solution.

[0034] Among them, the particle velocity is obtained by directly weighting and summing the gradient of the equivalent source Green's function in the frequency domain using the Euler equation, avoiding numerical difference of the measured sound pressure, suppressing high-frequency noise and improving the accuracy of velocity estimation.

[0035] Prior to the step of 'obtaining net acoustic power from each area integral', a background noise suppression step is included: estimating the background sound intensity threshold based on the signal acquired during the passive period, and setting the portion of the active sound intensity vector whose amplitude is lower than the threshold to zero, in order to reduce spurious peaks caused by reverberation and measurement noise.

[0036] In this process, a nonlinear power transform is applied to the time-frequency averaged acoustic power divergence spectrum, and the peak-to-valley ratio is used as the confidence level. Only the high-confidence peaks are retained to enhance the contrast between strong and weak sources and reduce misjudgments.

[0037] When multiple peaks are detected, energy density extremum search is performed independently on each peak shell to achieve multi-source parallel localization, and a list of source intensities and confidence levels is output.

[0038] The initial positioning result is used as a new priori equivalent source arrangement, and the reconstruction-divergence-peak process is executed again to form an iterative closed loop, so that the source coordinates gradually converge to sub-grid accuracy.

[0039] Among them, overlapping summation STFT, pre-decomposed transfer matrix and GPU parallel solution of linear equations are used to achieve millisecond-level frame processing, which meets the real-time positioning requirements of vehicle passing or transient noise of aircraft.

[0040] Example 2 This invention proposes a noise source localization method based on multi-integral surface acoustic monitoring, comprising the following steps: S10: In the sound field to be measured, around at least one potential noise source, N closed integral surfaces, nested sequentially from the inside out and geometrically similar, are virtually set up, denoted as S1, where N is an integer greater than or equal to 3; and on or near the surface of the outermost integral surface S1, a measurement array consisting of M microphones is arranged to synchronously acquire the time-domain sound pressure signal p(r m , t), where r mLet m be the position vector of the m-th microphone, where m = 1, 2, …, M; S20: The acquired time-domain sound pressure signal p(r) m The sound pressure P(r) is converted into the time-frequency domain complex sound pressure P(r) through short-time Fourier transform (STFT). m , ω j , t i ), where ω j For discrete angular frequencies, t i For time frame indexing; S30: For each time frame t i and each discrete frequency ω j Based on the measurement of complex sound pressure {P(r) on the outermost integral surface S1 m , ω j , t i )}, combined with the calibrated amplitude and phase response matrix C(ω) of the microphone j The entire sound field is reconstructed using the regularized equivalent source method (ESM), specifically at the innermost integral surface S. N An equivalent set of monopole sources is arranged internally, and their complex amplitude vector is determined by solving the following optimization problem. : ; Where P is the measured complex sound pressure vector, H() is the acoustic transfer function matrix from the equivalent source to the microphone, λ is the regularization parameter, and L is the regularization matrix; S40: Utilize the equivalent source complex amplitude obtained in step S30 Calculate N closed integral surfaces S k Complex sound pressure at any position r on (k=1,2, …, N) and particle velocity vector The reconstructed value; S50: For each integral surface S k Based on the sound pressure and vibration velocity reconstructed in step S40, the active sound intensity vector on its surface is calculated. Re[] represents taking the real part, and conj() represents taking the conjugate; S60: For each integral surface S k Active sound intensity vector I k The net acoustic power flowing out of the integral surface is obtained by performing a surface integral over its entire surface. : ; Where dS is the integral surface element vector; S70: For each time frame t i and each frequency ω j Based on the net acoustic power sequence of N integral surfaces {} Calculate the normalized acoustic power divergence between adjacent integral surfaces. As a measure of the existence of a sound source: ; Where k=1, …, N-1, For the integral surface S k With S k+1 The volume of the enclosed shell region; S80: Regarding the acoustic power dissipation In the frequency band of interest and time period The cumulative averaging is performed within the range to obtain the time-frequency averaged acoustic power divergence spectrum. (k); S90: By analyzing the time-frequency averaged acoustic power divergence spectrum (k) Perform local extremum search and identify (k) One or more peak locations k in the spectrum peak This indicates that at the kth time... peak Layer and kth peak+1 The main sound source exists in the shell region between the layers; S100: The shell region corresponding to each identified peak. Internally, the energy density distribution of the equivalent source in this region is calculated by step S30. (where G) s The transfer vector from the equivalent source to the spatial point r is used to perform time-frequency averaging, and the location where the average energy density Ē(r) reaches a local maximum is determined as the three-dimensional coordinates of the noise source.

[0041] In step S10, the N closed integral surfaces are a series of concentric spheres, ellipsoids, or curved surfaces conforming to the device under test.

[0042] In step S30, the regularization parameter λ is adaptively determined using the L-curve method or the generalized cross-validation (GCV) method to achieve an optimal balance between the accuracy of the sound field reconstruction and the stability of the solution.

[0043] The equivalent source method described in step S30 employs a virtual equivalent source arrangement strategy as follows: based on the prior information of the sound field to be measured, a non-uniform density distribution is adopted in the area where sound sources may exist, and the virtual equivalent sources are densely arranged at the expected sound source locations to improve spatial resolution.

[0044] In step S40, the particle velocity vector is calculated. It is realized through the frequency domain form of the Euler equation: Where ρ_0 is the density of the medium, and i is the imaginary unit. For the gradient operator, and P k The gradient is obtained by weighted summation of the gradients of the equivalent source Green's function, avoiding numerical difference of the sound pressure measurement.

[0045] Among them, the normalized acoustic power divergence mentioned in step S70 The calculation further introduces dynamic background noise suppression, specifically: in calculating W... k Previously, the active sound intensity I obtained in step S50 was first processed. k Thresholding is performed to set sound intensity components with amplitudes lower than the background sound intensity level estimated based on periods without sound sources to zero.

[0046] Between steps S80 and S90, a spectral peak enhancement step is added: the time-frequency averaged acoustic power divergence spectrum is... (k) Apply a nonlinear transformation function f(x), for example Where γ>1, preferably ranging from 1.5 to 3, to enhance the contrast of strong source regions and suppress the response of weak or spurious source regions, thereby improving the accuracy of peak identification. This transformation, through a power amplification effect, further enhances the contrast between the divergence peak of the shell containing the sound source and the divergence values ​​of adjacent non-source shells, thereby suppressing spurious peaks and improving the robustness of peak detection.

[0047] The method further includes a feedback correction loop: after locating the initial sound source position in step S100, this position is used as new prior information, and the process returns to step S30 to adjust the initial layout of the equivalent source. The calculations from S30 to S100 are then re-executed for iterative optimization until the sound source position converges. This approach is common in scenarios requiring high-precision positioning, such as in aerospace or automotive NVH (noise, vibration, and harshness) testing, where engineers not only need to know the location of the noise source but also need to accurately assess its impact on a specific area. Iterative optimization can significantly improve positioning accuracy in complex environments.

[0048] Among them, when multiple peak positions are identified { In step S100, for each peak region, the equivalent source energy density is calculated and the extreme value is searched independently, thereby realizing the simultaneous localization and identification of multiple separate noise sources.

[0049] This method is applied to a real-time processing system. In step S20, the short-time Fourier transform employs overlap-add or overlap-save methods for efficient computation. In step S30, the matrix inversion or solution of linear equations is accelerated by pre-computing the singular value decomposition (SVD) or QR decomposition of the transfer function matrix H to meet real-time requirements. Considering sound source localization, especially array-based methods, which typically involve massive computational demands, GPU parallel computing or FPGA hardware acceleration are effective ways to achieve real-time processing of large-scale arrays. This embodiment clarifies the key algorithms and hardware acceleration strategies used to achieve real-time performance, enhancing the practicality and technological advancement of the solution.

[0050] Example 3 like Figure 2 As shown, the present invention also proposes a noise source localization system based on multi-integral surface acoustic monitoring, including: a microphone array, a data acquisition module, a time-frequency conversion module, an adaptive equivalent source reconstruction module, a multi-layer acoustic power divergence calculation module, a peak detection and three-dimensional localization module, and a terminal for real-time display and storage; the system executes the above method to complete the online localization and visualization output of the noise source.

[0051] Example 4 This disclosure provides a non-volatile computer storage medium storing computer-executable instructions that can perform the steps described in the above embodiments.

[0052] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0053] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0054] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (AN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0055] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0056] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0057] The preferred embodiments of the present invention have been described above to make the spirit of the present invention clearer and easier to understand, and are not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope summarized by the appended claims.

Claims

1. A noise source localization method based on multi-integral surface acoustic monitoring, characterized in that, Includes the following steps: Sound pressure was measured by arranging an array on the outermost layer. Convert the signal to the time-frequency domain; The entire sound field is reconstructed using the regularized equivalent source method; The sound pressure, particle velocity, and active sound intensity on the multi-layer integral surface are calculated sequentially from the reconstruction results. The net acoustic power was obtained for each area integral; The normalized acoustic power divergence of adjacent shell layers is calculated and accumulated over time and frequency. Peak detection is performed on the accumulated divergence spectrum to pinpoint the shell layer containing the sound source, and the three-dimensional coordinates within that layer are obtained using the extreme value of the equivalent source energy density. The peak shell layer is used to locate the layer where the sound source is located, and the three-dimensional coordinates are obtained within that layer using the extreme value of the equivalent source energy density.

2. The method as described in claim 1, characterized in that, The integral surface is a series of virtual, nested concentric spheres, ellipsoids, or curved surfaces conforming to the device housing; the microphone array is arranged on the surface of the outermost integral surface or in the measurement area immediately adjacent to it, in order to reduce the condition number of the reconstructed transfer function and improve the positioning stability under large curved surfaces or conformal surfaces.

3. The method as described in claim 1, characterized in that, The regularization parameters are adaptively updated in each time-frequency unit using L-curves or GCVs. Meanwhile, the equivalent sources are arranged in a non-uniform grid with denser a priori source regions and sparser non-source regions, taking into account both the stability and spatial resolution of the solution.

4. The method as described in claim 1, characterized in that, The particle velocity is obtained by directly weighting and summing the gradient of the equivalent source Green's function in the frequency domain using the Euler equation, avoiding numerical difference of the measured sound pressure, suppressing high-frequency noise and improving the accuracy of velocity estimation.

5. The method as described in claim 1, characterized in that, Before the step of 'obtaining net acoustic power for each area integral', a background noise suppression step is also included: estimating the background sound intensity threshold based on the signal collected during the passive period, and setting the portion of the active sound intensity vector whose amplitude is lower than the threshold to zero, so as to reduce spurious peaks caused by reverberation and measurement noise.

6. The method as described in claim 1, characterized in that, A nonlinear power transform is applied to the time-frequency averaged acoustic power divergence spectrum, and the peak-to-valley ratio is used as the confidence level. Only the high-confidence peaks are retained to enhance the contrast between strong and weak sources and reduce misjudgments.

7. The method as described in claim 1, characterized in that, When multiple peaks are detected, energy density extremum search is performed independently for each peak shell to achieve multi-source parallel localization, and a list of source intensities and confidence levels is output.

8. The method as described in claim 1, characterized in that, The initial positioning result is used as a new priori equivalent source arrangement, and the reconstruction-divergence-peak process is executed again to form an iterative closed loop, so that the source coordinates gradually converge to sub-mesh accuracy.

9. The method as described in claim 1, characterized in that, By employing superimposed STFT, pre-decomposed transfer matrix, and parallel solution of linear equations using GPU, millisecond-level frame processing is achieved, meeting the real-time positioning requirements for vehicle passage or transient noise from aircraft.

10. A noise source localization system based on multi-integral surface acoustic monitoring, comprising: The system comprises a microphone array, a data acquisition module, a time-frequency conversion module, an adaptive equivalent source reconstruction module, a multi-layer acoustic power dissipation calculation module, a peak detection and three-dimensional positioning module, and a terminal for real-time display and storage; the system executes the method described in any one of claims 1 to 9 to complete the online positioning and visualization output of the noise source.

Images