A method of blast noise localization imaging

Abstract

The application relates to a kind of air blower noise positioning imaging method, by the noise data of the surrounding environment of air blower when not running and running is collected, the noise data of the surrounding environment of air blower when running is screened, the region where the decibel value is highest is taken as non-synchronous measurement area, non-synchronous measurement is carried out to the air blower in operation, the non-synchronous measurement data of air blower in multiple positions is obtained, the non-synchronous measurement data of air blower in multiple positions obtained is preprocessed, multiple non-synchronous measurement data of air blower after preprocessing is obtained, the energy spectrum matrix construction is carried out to the non-synchronous measurement data of air blower after preprocessing obtained, multiple energy spectrum matrices are obtained, then based on the space position of each energy spectrum matrix, diagonalization arrangement is carried out, and the energy matrix to be filled is obtained.

Description

Technical Field

[0001] This application relates to the field of noise localization technology, and in particular to a method for noise localization imaging of a blower. Background Technology

[0002] Non-synchronous measurement (NSM) is an effective method for achieving low-frequency acoustic localization performance through sequential scanning of the sound field. It achieves larger and denser aperture arrays by moving a single planar array. By performing non-synchronous measurements on a moving microphone array, the accuracy of sound source localization in a two-dimensional plane can be improved.

[0003] However, existing methods can only perform sound source imaging at frequencies above 2000 Hz. In noise source identification and localization, sound source imaging studies generally use planar arrays to image in a two-dimensional plane, with asynchronous measurements primarily involving planar movement. The asynchronous measurement method for two-dimensional planar arrays involves continuously moving the planar array to scan spatially distributed sound sources, approximating measurements of large-aperture and high-density microphone arrays, and then using traditional beamforming algorithms for imaging to locate the sound source.

[0004] Blowers are a crucial component in industrial products for eliminating exhaust emissions, ventilating air, compressing air, and driving air conditioning systems. These blowers typically operate continuously for extended periods, generating considerable noise pollution during this time. Unusual noise levels are often associated with blower malfunctions and structural design flaws, including unbalanced rotors, loose fasteners, and blade scratches. Blower performance and safety are affected by these malfunctions and structural defects. The characteristic frequencies of blowers under normal and abnormal conditions provide effective acoustic characteristics for type identification and operational status monitoring. Therefore, source localization at the blower's characteristic frequencies can improve the imaging of sound source distribution. If the characteristic frequencies of the sound source are known, the spatial resolution and power accuracy of the imaging method are significantly improved. This enhances the estimation of the target sound source, resulting in higher accuracy and resolution of the imaging method, leading to better identification and localization of noise sources.

[0005] The analysis of the noise mechanism of the blower was improved by locating the noise source, which also helps in the diagnosis of mechanical product problems and facilitates the design of noise reduction solutions.

[0006] Although asynchronous measurement methods improve the spatial resolution of acoustic imaging at lower frequencies, their application in real-world scenarios has not yet been realized due to the complexity of the algorithms and the difficulty in implementing them effectively. Summary of the Invention

[0007] Based on this, it is necessary to address the problem that traditional asynchronous measurement methods have improved the spatial resolution of acoustic imaging at lower frequencies, but their application in practical scenarios has not yet been realized due to the complexity of the algorithm and the difficulty in effective implementation. Therefore, a blower noise localization imaging method is proposed.

[0008] This application provides a method for locating and imaging noise in a blower, including:

[0009] Acquire ambient noise data when the blower is off and ambient noise data when the blower is on.

[0010] Asynchronous measurements were performed on the area with the highest noise decibel value in the ambient noise data when the blower was turned on, and asynchronous measurement data of blower noise at multiple locations were obtained.

[0011] Based on the ambient noise data when the blower is off, the asynchronous measurement data of the blower noise at each location is preprocessed to obtain the preprocessed asynchronous measurement data of the blower noise at each location.

[0012] Based on the preprocessed asynchronous measurement data of blower noise at each location, energy spectrum matrices are constructed to obtain multiple energy spectrum matrices.

[0013] Based on the different positions obtained from each energy spectrum matrix, the matrix is ​​diagonally arranged to obtain the energy matrix to be filled.

[0014] The matrix filling algorithm is used to fill the energy matrix to be filled, and the filled energy matrix is ​​obtained.

[0015] The sound source localization imaging of the blower is obtained by performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm.

[0016] Based on the obtained blower sound source localization image and blower body image, visible light fusion is performed to obtain a visible light fused image of the blower body.

[0017] Furthermore, the asynchronous measurement of the area with the highest noise decibel value in the ambient noise data when the blower is turned on is performed to obtain asynchronous measurement data of blower noise at multiple locations, including:

[0018] Acquire noise data around the blower;

[0019] Analyze the noise data around the blower to obtain multiple noise data points in the circumferential direction of the blower;

[0020] The region with the highest noise decibel value among multiple noise data points in the circumferential direction of the blower is selected as the target region.

[0021] Asynchronous measurements were performed on the target area to obtain asynchronous measurement data for the blower.

[0022] Furthermore, the asynchronous measurement data of blower noise at each location based on ambient noise data when the blower is off is preprocessed to obtain preprocessed asynchronous measurement data of blower noise at each location, including:

[0023] Select asynchronous measurement data of blower noise at one location;

[0024] The asynchronous measurement data of the blower noise at this location is analyzed to obtain the environmental noise data and the measured data of the blower noise in the asynchronous measurement data of the blower noise at this location;

[0025] Based on the environmental noise data, the measured data of the blower noise are preprocessed to obtain the asynchronous measurement data of the blower noise at this location.

[0026] Return to the asynchronous measurement data of the blower noise at the selected location, until the asynchronous measurement data of the blower noise at each location has been selected once.

[0027] Furthermore, the preprocessing of the measured blower noise data based on environmental noise data to obtain preprocessed asynchronous measurement data of the blower noise at that location, further includes:

[0028] Spectral analysis was performed on the preprocessed asynchronous measurement data of blower noise to obtain the spectrum diagram of the preprocessed asynchronous measurement of blower noise;

[0029] The operating parameters of the blower are obtained by analyzing the spectrum of the preprocessed blower noise from asynchronous measurements.

[0030] Furthermore, based on the preprocessed asynchronous measurement data of the blower noise at each location, energy spectrum matrices are constructed to obtain multiple energy spectrum matrices, including:

[0031] Select preprocessed asynchronous measurement data of blower noise at one location;

[0032] The asynchronous measurement data of preprocessed blower noise at this location is analyzed to obtain multiple preprocessed acoustic signal data in the asynchronous measurement data of preprocessed blower noise at this location. The multiple preprocessed acoustic signal data are blower noise data detected at different locations.

[0033] Generate the corresponding energy transfer matrix based on multiple preprocessed acoustic signal data;

[0034] Return to the preprocessed asynchronous measurement data of the blower noise at the selected location, until the preprocessed asynchronous measurement data of the blower noise at each location has been selected once.

[0035] Furthermore, the step of diagonalizing the energy matrix obtained from the different positions of each energy spectrum matrix to obtain the energy matrix to be filled includes:

[0036] Based on the position of each energy transfer matrix, spatial positions are arranged to obtain a spatial order of multiple energy transfer matrices.

[0037] Create an empty matrix;

[0038] The two energy transfer matrices located at the beginning and end of the spatial order of multiple energy transfer matrices are filled into the two diagonals of the blank matrix to obtain the preliminary filled energy matrix.

[0039] Based on the spatial ordering of multiple energy transfer matrices, the remaining energy transfer matrices are filled into the initial filling matrix to obtain the energy matrix to be filled.

[0040] Furthermore, the sound source localization imaging based on the joint maximum a posteriori distribution algorithm on the filled energy matrix to obtain the blower sound source localization imaging includes:

[0041] The formula for localizing and imaging the sound source of the blower is shown in Formula 1.

[0042]

[0043] Where P represents the collected noise data of the blower; ω is the vector form of the filled energy matrix; ω is the compensation value for the time delay and amplitude attenuation of sound propagation.

[0044] Furthermore, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes:

[0045] The imaging results are discretized to obtain results that are discrete into multiple grid points;

[0046] A linear energy propagation model is constructed based on the results obtained from multiple grid points;

[0047] The formula for the linear energy propagation model is shown in Formula 2;

[0048]

[0049] Where P represents the collected noise data of the blower; G is the energy transfer matrix; S is the high-resolution sound source energy distribution of the blower; ∈ represents the model error; g n1，n2For elements in the energy transfer matrix; g n1 g is the guide vector from the n1th scan grid point to the planar array; n2 This is the guide vector from the n2th scan grid point to the planar array.

[0050] Furthermore, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes:

[0051] The high-resolution sound source energy distribution of the blower is obtained by solving the linear energy propagation model based on the joint maximum a posteriori distribution algorithm.

[0052] Furthermore, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes:

[0053] The high-resolution sound source energy distribution of the blower is imaged onto a schematic diagram of the blower to obtain an example image;

[0054] Obtain a simulated image of the blower;

[0055] Calculate the error between the simulated imaging and the example imaging to obtain the error value;

[0056] Determine whether the error value is less than the error threshold;

[0057] If the error value is less than the error threshold, the high-resolution sound source energy distribution of the blower is defined as qualified.

[0058] This application relates to a method for noise localization imaging of a blower. The method involves collecting noise data from the surrounding environment when the blower is not running and when it is running. The noise data from the running environment is then filtered, and the area with the highest decibel value is designated as the asynchronous measurement area. Asynchronous measurements are then performed on the running blower to obtain asynchronous measurement data at multiple locations. This data is then preprocessed to obtain multiple preprocessed asynchronous measurement data sets. Energy spectrum matrices are constructed from these preprocessed data sets, resulting in multiple energy spectrum matrices. These matrices are then diagonally arranged based on their spatial positions to obtain an energy matrix to be filled. A matrix filling algorithm is used to fill the energy matrix to obtain a filled energy matrix. Finally, a joint maximum a posteriori distribution algorithm is used to perform sound source localization imaging on the filled energy matrix to obtain the sound source localization image of the blower. The sound source localization image of the blower is then fused with a visible light image of the blower itself to obtain a noise distribution map of the blower. Attached Figure Description

[0059] Figure 1 This is a schematic flowchart of a blower noise localization imaging method provided in an embodiment of this application.

[0060] Figure 2 This is a noise radar image of a blower during stable operation in a blower noise localization imaging method provided in an embodiment of this application.

[0061] Figure 3 This is a schematic diagram of a computer-simulated blower imaging method for locating blower noise, provided in an embodiment of this application.

[0062] Figure 4 This is a schematic diagram of an example of blower imaging in a blower noise localization imaging method provided in an embodiment of this application.

[0063] Figure 5 This is a schematic diagram of a two-dimensional planar measurement matrix in a blower noise localization imaging method provided in an embodiment of this application.

[0064] Figure 6 This is a schematic diagram of the movement path of the two-dimensional plane measurement matrix in a blower noise localization imaging method provided in an embodiment of this application.

[0065] Figure 7 This is a schematic diagram of the two-dimensional planar measurement matrix for blower noise measurement in a blower noise localization imaging method provided in an embodiment of this application. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0067] like Figure 1 As shown, in one embodiment of this application, the blower noise localization imaging method includes the following steps S100 to S800:

[0068] S100 acquires ambient noise data when the blower is off and ambient noise data when the blower is on.

[0069] Specifically, the ambient noise data when the blower is off and the ambient noise data when the blower is on are both collected circumferentially from the blower.

[0070] S200 performs asynchronous measurements on the area with the highest noise decibel value in the ambient noise data when the blower is turned on, and obtains asynchronous measurement data of blower noise at multiple locations.

[0071] Specifically, such as Figure 5 and Figure 7 As shown, asynchronous measurement refers to arranging multiple microphones for sound measurement in a matrix on a two-dimensional plane to form a two-dimensional plane measurement matrix. This two-dimensional plane measurement matrix is ​​then placed over the target measurement area of ​​the blower. This target measurement area refers to the region with the highest decibel value in the ambient noise data when the blower is running. Then, measurements are taken at multiple locations by moving the entire two-dimensional plane measurement matrix. These measurements can be taken along a straight line or along a curve, measuring multiple nodes, such as... Figure 6 As shown, in this embodiment, the measurement is performed by translating along a horizontal straight line, and the blower operates stably throughout the measurement process.

[0072] S300 preprocesses the asynchronous measurement data of blower noise at each location based on the ambient noise data when the blower is off, and obtains the preprocessed asynchronous measurement data of blower noise at each location.

[0073] S400, based on the asynchronous measurement data of the preprocessed blower noise at each location, constructs an energy spectrum matrix to obtain multiple energy spectrum matrices.

[0074] S500, based on the different positions obtained from each energy spectrum matrix, is diagonally arranged to obtain the energy matrix to be filled.

[0075] S600 uses a matrix filling algorithm to fill the energy matrix to be filled, thus obtaining the filled energy matrix.

[0076] S700 uses the joint maximum a posteriori distribution algorithm to perform sound source localization imaging on the filled energy matrix to obtain the sound source localization imaging of the blower.

[0077] S800 performs visible light fusion based on the obtained blower sound source localization image and blower body image to obtain a visible light fused image of the blower body.

[0078] In this embodiment, noise data of the surrounding environment when the blower is not running and when it is running are collected. The noise data of the surrounding environment when the blower is running is filtered, and the area with the highest decibel value is taken as the asynchronous measurement area. Asynchronous measurement is performed on the running blower to obtain asynchronous measurement data of the blower at multiple locations. The obtained asynchronous measurement data of the blower at multiple locations is preprocessed to obtain multiple preprocessed asynchronous measurement data of the blower. Energy spectrum matrices are constructed for each of the multiple preprocessed asynchronous measurement data of the blower to obtain multiple energy spectrum matrices. Then, based on the spatial position of each energy spectrum matrix, it is diagonally arranged to obtain the energy matrix to be filled. The matrix filling algorithm is used to fill the energy matrix to be filled to obtain the filled energy matrix. Then, the sound source localization imaging is performed on the filled energy matrix using the joint maximum a posteriori distribution algorithm to obtain the sound source localization imaging of the blower. The sound source localization imaging of the blower is fused with the visible light image of the blower body to obtain the noise distribution map of the blower.

[0079] like Figure 2 As shown, in one embodiment of this application, the asynchronous measurement of the area with the highest noise decibel value in the ambient noise data when the blower is turned on, to obtain asynchronous measurement data of blower noise at multiple locations, includes the following steps S201 to S204:

[0080] S201, acquire noise data around the blower.

[0081] S202, Analyze the noise data around the blower to obtain multiple noise data points around the blower.

[0082] S203, select the area with the highest noise decibel value among multiple noise data in the circumferential direction of the blower as the target area.

[0083] S204, perform asynchronous measurements based on the target area to obtain asynchronous measurement data of the blower.

[0084] In this embodiment, the area with the highest noise decibel value within a certain distance range during the operation of the blower is selected for asynchronous measurement, so that the detected noise data represents the larger noise area during the operation of the blower.

[0085] In one embodiment of this application, the asynchronous measurement data of blower noise at each location based on ambient noise data when the blower is off is preprocessed to obtain preprocessed asynchronous measurement data of blower noise at each location, including the following steps S301 to S304:

[0086] S301, Select asynchronous measurement data of blower noise at a location.

[0087] S302, analyze the asynchronous measurement data of the blower noise at this location to obtain the environmental noise data and the measured data of the blower noise in the asynchronous measurement data of the blower noise at this location.

[0088] S303, based on environmental noise data, preprocesses the measured data of blower noise to obtain the preprocessed asynchronous measurement data of blower noise at this location.

[0089] S304, return the asynchronous measurement data of the blower noise at the selected location, until the asynchronous measurement data of the blower noise at each location has been selected once.

[0090] In this embodiment, since the multiple noise data of asynchronous measurement include noise data from the surrounding environment, in order to improve the accuracy of the multiple noise data of asynchronous measurement, it is necessary to cancel the side lobes of the environmental noise data of the asynchronous measurement noise data at each location, so as to obtain the preprocessed asynchronous measurement noise data at each location.

[0091] Furthermore, the asynchronous measurement noise data at each location is subjected to sidelobe cancellation of environmental noise data, including precise sidelobe cancellation of environmental noise data detected by a single microphone at the same location and noise data detected during blower operation.

[0092] In one embodiment of this application, the preprocessing of the measured blower noise data based on environmental noise data to obtain preprocessed asynchronous measurement data of the blower noise at that location is further included in steps S305 to S306:

[0093] S305. Based on the preprocessed asynchronous measurement data of blower noise, a spectrum analysis is performed to obtain the spectrum diagram of the preprocessed asynchronous measurement of blower noise.

[0094] S306, the preprocessed spectrum of the blower noise obtained from asynchronous measurement is analyzed to obtain the operating parameters of the blower.

[0095] In this embodiment, the preprocessed asynchronous blower noise measurement data at each location is subjected to spectral analysis to obtain a spectrum diagram of the preprocessed asynchronous blower noise measurement data at each location. Then, the spectrum diagrams of the preprocessed asynchronous blower noise measurement data at each location are analyzed to obtain the blower operating parameters detected at different measurement locations. Furthermore, the blower operating frequency is statistically analyzed based on the blower operating parameters detected at multiple locations, and the average value of the blower operating frequency at multiple locations is taken as the blower operating frequency.

[0096] In one embodiment of this application, the asynchronous measurement data of preprocessed blower noise at each location is used to construct an energy spectrum matrix, resulting in multiple energy spectrum matrices, including the following steps S401 to S404:

[0097] S401, Select pre-processed asynchronous measurement data of blower noise at a location.

[0098] S402, parse the asynchronous measurement data of the preprocessed blower noise at this location to obtain multiple preprocessed acoustic signal data in the asynchronous measurement data of the preprocessed blower noise at this location, wherein the multiple preprocessed acoustic signal data are blower noise data detected at different locations.

[0099] S403 generates the corresponding energy transfer matrix based on multiple preprocessed acoustic signal data.

[0100] Specifically, the energy transfer matrix expression is as follows: Then the measured signal data vector P at position l l The energy transfer matrix is ​​represented as

[0101] in, It is the energy transfer matrix of unrelated sources. S represents th The power of the source, () H This represents the conjugate transpose of a matrix.

[0102] S404, return the preprocessed asynchronous measurement data of the blower noise at the selected location, until the preprocessed asynchronous measurement data of the blower noise at each location has been selected once.

[0103] In this embodiment, an energy spectrum matrix is ​​constructed by preprocessing the asynchronous measurement data of the blower noise at each location. Specifically, the parameters of each microphone used to detect noise are obtained on the two-dimensional plane matrix, and the energy spectrum matrix at the asynchronous measurement location is constructed based on the position of each microphone on the two-dimensional plane.

[0104] In one embodiment of this application, the step of diagonalizing the energy matrix to be filled based on the different positions of each energy spectrum matrix includes the following steps S501 to S504:

[0105] S501, based on the position of each energy transfer matrix, spatial positions are arranged to obtain a spatial position sort of multiple energy transfer matrices.

[0106] S502, Create a blank matrix.

[0107] S503, fill the two energy transfer matrices located at the beginning and end of the spatial order of multiple energy transfer matrices into the two diagonals of the blank matrix to obtain the preliminary filled energy matrix.

[0108] S504, based on the spatial position sorting of multiple energy transfer matrices, fill the remaining energy transfer matrices into the preliminary filling matrix to obtain the energy matrix to be filled.

[0109] In this embodiment, the asynchronous measurement method is to translate the two-dimensional plane matrix used for measurement by a horizontal straight line. Therefore, based on the spatial position, there are two positions, the first and the last, among the multiple positions of asynchronous measurement. Then, a blank matrix is ​​created, and the two energy spectrum matrices located at the first and the last positions of asynchronous measurement are filled into the two diagonals of the blank matrix to obtain a preliminary filling matrix. Based on the spatial position, the remaining energy spectrum matrices located between the first and the last energy spectrum matrices are filled into the preliminary filling matrix to obtain the energy matrix to be filled.

[0110] In one embodiment of this application, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain blower sound source localization imaging includes the following S701:

[0111] S701, the formula for blower sound source localization imaging is shown in Formula 1;

[0112]

[0113] Where P represents the collected noise data of the blower; ω is the vector form of the filled energy matrix; ω is the compensation value for the time delay and amplitude attenuation of sound propagation.

[0114] Specifically, It is calculated from the measured sound pressure level, which is derived from the actual position of the microphone array at each location. ω is used to compensate for the time delay and amplitude attenuation during forward propagation, and ||g|| represents the norm of g.

[0115] In one embodiment of this application, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain blower sound source localization imaging further includes the following steps S702 to S703:

[0116] S702 discretizes the imaging results, resulting in a result that is discrete into multiple grid points.

[0117] S703, a linear energy propagation model is constructed based on the results obtained from multiple grid points;

[0118] The formula for the linear energy propagation model is shown in Formula 2;

[0119]

[0120] Where P represents the collected noise data of the blower; G is the energy transfer matrix; S is the high-resolution sound source energy distribution of the blower; ∈ represents the model error; g n1，n2 For elements in the energy transfer matrix; g n1 g is the guide vector from the n1th scan grid point to the planar array; n2 This is the guide vector from the n2th scan grid point to the planar array.

[0121] In one embodiment of this application, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain blower sound source localization imaging further includes the following S704:

[0122] S704 uses the joint maximum a posteriori distribution algorithm to solve the linear energy propagation model and obtains the high-resolution sound source energy distribution of the blower.

[0123] Specifically, the student t-distribution is assigned as the prior distribution to the target variable S (i.e., the high-resolution sound source energy distribution of the blower under test) and the error variable ∈.

[0124]

[0125] in, For probability density function (PDF), Let be the conditional probability, representing the probability given the parameter α. S In the case of S, the probability of S occurring is given by St(·), which is a student t-distribution, and α is given by α. S For hidden variables express The product of n from 1 to N. Due to the characteristics of the Student's t-distribution as an infinite Gaussian-scale mixture (IGSM), it can be represented by a hierarchical Gaussian-gamma distribution, i.e.:

[0126]

[0127] Where I is the identity matrix, Represented by a Gaussian distribution, with a mean of 0 and a variance of α. S I. It has an inverse gamma distribution. and Let represent the hyperparameters. Similarly, the prior distribution of can also be expressed as:

[0128]

[0129] Where, α ∈ For the variance of random noise, and Represents the hyperparameters. From P = GS + ∈, its inverse likelihood model can be expressed as:

[0130]

[0131] in, This indicates that the relationship between two quantities is positively correlated. This indicates that the mean is GS and the variance is α. ∈ I follows a normal distribution.

[0132] In the Bayesian framework, variables S and α S α ∈ The joint posterior PDF can be represented as

[0133]

[0134] In this embodiment, the energy matrix to be filled is filled using a matrix filling algorithm to obtain the filled energy matrix. Sound source localization imaging is then performed on the filled energy matrix to obtain the blower source localization imaging result. Next, the obtained blower source localization imaging result is discretized to obtain multiple discrete point results. Based on the multiple discrete point results, a linear energy propagation model is created. Then, the joint maximum a posteriori distribution algorithm is used to solve the linear energy propagation model to obtain the high-resolution source energy distribution of the blower.

[0135] like Figures 3 to 4 As shown, in one embodiment of this application, the step of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain blower sound source localization imaging further includes the following steps S705 to S709:

[0136] S705, the high-resolution sound source energy distribution of the blower is imaged onto the blower schematic diagram to obtain an example image.

[0137] S706, acquires simulated images of the blower.

[0138] S707 calculates the error between the simulated image and the example image to obtain the error value.

[0139] S708, determine whether the error value is less than the error threshold.

[0140] S709, if the error value is less than the error threshold, the high-resolution sound source energy distribution of the blower is defined as qualified.

[0141] Specifically, if the error value is greater than the error threshold, the asynchronous measurement is optimized, such as by adjusting the matrix arrangement of the asynchronous measurement or the movement path during the asynchronous measurement.

[0142] The error between the two was compared by imaging a schematic diagram of the blower based on actual measurements and a schematic diagram of the blower based on simulation.

[0143] The error in sound source imaging is calculated using the 2-norm:

[0144]

[0145] Where L is the value of the error; ||·||2 represents the 2-norm, and P 测试 The results of computer simulation imaging of the blower's sound source localization; P 真实位置 This is the result of actual blower sound source localization imaging through asynchronous measurement.

[0146] In this embodiment, the blower parameters are set in a computer simulation model, and the parameters of the blower in the computer simulation model are consistent with the actual blower being tested. Then, the operating noise of the blower in the computer simulation model is simulated and detected, thereby obtaining a simulated image. The error between the obtained simulated image and the actual measured image is calculated to obtain the error between the two. When the error value is less than the error threshold, the high-resolution sound source energy distribution of the actual measured blower is defined as qualified.

[0147] The technical features of the above embodiments can be combined arbitrarily, and the execution order of the method steps is not restricted. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0148] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for locating and imaging noise in a blower, characterized in that, The blower noise localization imaging method includes: Acquire ambient noise data when the blower is off and ambient noise data when the blower is on. Asynchronous measurements were performed on the area with the highest noise decibel value in the ambient noise data when the blower was turned on, and asynchronous measurement data of blower noise at multiple locations were obtained. Based on the ambient noise data when the blower is off, the asynchronous measurement data of the blower noise at each location is preprocessed to obtain the preprocessed asynchronous measurement data of the blower noise at each location. Based on the preprocessed asynchronous measurement data of blower noise at each location, energy spectrum matrices are constructed to obtain multiple energy spectrum matrices. Based on the different positions obtained from each energy spectrum matrix, the matrix is ​​diagonally arranged to obtain the energy matrix to be filled. The matrix filling algorithm is used to fill the energy matrix to be filled, and the filled energy matrix is ​​obtained. The sound source localization imaging of the blower is obtained by performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm. Based on the obtained blower sound source localization imaging and blower body image, visible light fusion is performed to obtain a visible light fused image of the blower body. The process of diagonalizing and arranging the energy spectral matrix based on its different positions to obtain the energy matrix to be filled includes: Based on the position of each energy transfer matrix, spatial positions are arranged to obtain a spatial order of multiple energy transfer matrices. Create an empty matrix; The first and last two energy transfer matrices in the spatial order of multiple energy transfer matrices are filled into the two diagonals of the blank matrix to obtain the preliminary filled energy matrix. Based on the spatial ordering of multiple energy transfer matrices, the remaining energy transfer matrices are filled into the initial filling matrix to obtain the energy matrix to be filled.

2. The blower noise localization imaging method according to claim 1, characterized in that, The method involves asynchronously measuring the area with the highest noise decibel value in the ambient noise data when the blower is turned on, obtaining asynchronous measurement data of blower noise at multiple locations, including: Acquire noise data around the blower; Analyze the noise data around the blower to obtain multiple noise data points in the circumferential direction of the blower; The region with the highest noise decibel value among multiple noise data points in the circumferential direction of the blower is selected as the target region. Asynchronous measurements were performed on the target area to obtain asynchronous measurement data for the blower.

3. The blower noise localization imaging method according to claim 2, characterized in that, The asynchronous measurement data of blower noise at each location, based on ambient noise data when the blower is off, are preprocessed to obtain preprocessed asynchronous measurement data of blower noise at each location, including: Select asynchronous measurement data of blower noise at one location; The asynchronous measurement data of the blower noise at this location is analyzed to obtain the environmental noise data and the measured data of the blower noise in the asynchronous measurement data of the blower noise at this location; Based on the environmental noise data, the measured data of the blower noise are preprocessed to obtain the asynchronous measurement data of the blower noise at this location. Return to the asynchronous measurement data of the blower noise at the selected location, until the asynchronous measurement data of the blower noise at each location has been selected once.

4. The blower noise localization imaging method according to claim 3, characterized in that, The process of preprocessing the measured blower noise data based on environmental noise data to obtain preprocessed asynchronous measurement data of the blower noise at that location, followed by: Spectral analysis was performed on the preprocessed asynchronous measurement data of blower noise to obtain the spectrum diagram of the preprocessed asynchronous measurement of blower noise; The operating parameters of the blower are obtained by analyzing the spectrum of the preprocessed blower noise from asynchronous measurements.

5. The blower noise localization imaging method according to claim 4, characterized in that, The asynchronous measurement data of preprocessed blower noise at each location are used to construct energy spectrum matrices, resulting in multiple energy spectrum matrices, including: Select preprocessed asynchronous measurement data of blower noise at one location; The asynchronous measurement data of preprocessed blower noise at this location is analyzed to obtain multiple preprocessed acoustic signal data in the asynchronous measurement data of preprocessed blower noise at this location. The multiple preprocessed acoustic signal data are blower noise data detected at different locations. Generate the corresponding energy transfer matrix based on multiple preprocessed acoustic signal data; Return to the preprocessed asynchronous measurement data of the blower noise at the selected location, until the preprocessed asynchronous measurement data of the blower noise at each location has been selected once.

6. The blower noise localization imaging method according to claim 5, characterized in that, The sound source localization imaging based on the joint maximum a posteriori distribution algorithm on the filled energy matrix to obtain the blower sound source localization imaging includes: The formula for localizing and imaging the sound source of the blower is shown in Formula 1. Official 1; in, The noise data collected from the blower; This is the vector form of the filled energy matrix; This is the compensation value for the time delay and amplitude attenuation of sound propagation.

7. The blower noise localization imaging method according to claim 6, characterized in that, The method of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes: The imaging results are discretized to obtain results that are discrete into multiple grid points; A linear energy propagation model is constructed based on the results obtained from multiple grid points; The formula for the linear energy propagation model is shown in Formula 2; Official 2; in, The noise data collected from the blower; For energy transfer matrix; High-resolution sound source energy distribution for the blower; This refers to model error; For elements in the energy transfer matrix; This is the guide vector from the n1th scan grid point to the planar array; This is the guide vector from the n2th scan grid point to the planar array.

8. The blower noise localization imaging method according to claim 7, characterized in that, The method of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes: The high-resolution sound source energy distribution of the blower is obtained by solving the linear energy propagation model based on the joint maximum a posteriori distribution algorithm.

9. The blower noise localization imaging method according to claim 8, characterized in that, The method of performing sound source localization imaging on the filled energy matrix based on the joint maximum a posteriori distribution algorithm to obtain the blower sound source localization imaging also includes: The high-resolution sound source energy distribution of the blower is imaged onto a schematic diagram of the blower to obtain an example image; Obtain a simulated image of the blower; Calculate the error between the simulated imaging and the example imaging to obtain the error value; Determine whether the error value is less than the error threshold; If the error value is less than the error threshold, the high-resolution sound source energy distribution of the blower is defined as qualified.

Images