Method for positioning low-frequency noise source of transformer substation

Abstract

The invention relates to the technical field of power systems, in particular to a transformer substation low-frequency noise source positioning method. The method comprises the following steps: obtaining a three-dimensional environment model of each structure in a transformer substation, and obtaining a geometric body topological relation matrix in the three-dimensional environment model; performing field measurement on the transformer substation to obtain noise measurement data; and performing outer convolution fusion on noise images between adjacent units according to the geometric topological relation matrix and the local noise image data, and iteratively spreading the noise images to the whole transformer substation area to generate initial sound source distribution estimation data of the whole transformer substation. The sound energy flow is utilized to analyze and identify the source position, the static source and the moving source are separated and tracked in combination with direction estimation and trajectory reconstruction, the static source can be accurately positioned, meanwhile, the motion trajectory of the moving source is obtained, and the separation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for locating a low-frequency noise source in a transformer substation. Background Art

[0002] Substation is an important part of the power system, used to convert high voltage electric energy into low voltage electric energy or vice versa to meet the needs of power transmission, distribution and power supply. It is a key link in the power system, used to achieve the transmission, distribution and conversion of electric energy. In the power system, electric energy needs to be transmitted from the power plant to the user through transmission lines. Due to the long transmission distance of electric energy, in order to reduce energy loss and improve the efficiency of electric energy transmission, electric energy is usually boosted to a higher voltage for transmission during the transmission process. This requires substations to perform voltage and current conversion operations of electric energy.

[0003] The low-frequency noise of a substation refers to the low-frequency noise signal generated during the operation of the substation. Compared with high-frequency noise, the frequency range of low-frequency noise is lower, usually below 20Hz. This noise has a longer wavelength and stronger propagation ability, and can penetrate buildings and obstacles. The low-frequency noise of a substation may have an impact on the surrounding environment and personnel, such as causing human discomfort and interfering with the lives and work of surrounding residents. The location of the low-frequency noise source of a substation refers to the process of determining the specific location of the low-frequency noise generated inside or around the substation. However, there are many different types of power equipment in the substation, and the noise sources are distributed in different locations. There are complex situations with multiple dispersed sound sources, which often cause the source location to have the problem of sound source dispersion. Secondly, the structures in the substation area are intricate, which will produce serious echoes and multipath effects, which are easily confused with direct sound signals, resulting in the location of noise sources often having serious echo and interference problems. In addition, some equipment such as circuit breakers will generate mobile sound sources during operation, and their transient and ringing characteristics make it difficult for noise source location methods to capture and track. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a method for locating a low-frequency noise source in a substation to solve at least one of the above technical problems.

[0005] To achieve the above object, a method for locating a low-frequency noise source in a substation comprises the following steps:

[0006] Acquire a three-dimensional environmental model of each structure in the substation, and obtain a topological relationship matrix of geometric bodies in the three-dimensional environmental model; perform on-site measurement of the substation to obtain noise measurement data; correspond the noise measurement data to the three-dimensional environmental model in blocks based on the measurement position, and perform registration to obtain local noise image data;

[0007] According to the geometric body topological relationship matrix and the local noise image data, the noise images between adjacent units are fused by outer convolution, and iteratively propagated to the entire substation area to generate initial sound source distribution estimation data of the entire substation;

[0008] Separating the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data;

[0009] The echo decoupling data is dimensional compressed and the eigenmode of the low-order scattering field is extracted to obtain the eigenmode set of the scattering field; the parameters of the field mode of the eigenmode set of the scattering field are estimated to obtain the eigenscattering cloud map data;

[0010] The acoustic energy flow around the center of the intrinsic scattering cloud image data is identified, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; the position information of the stationary source and the mobile source is distinguished and integrated according to the noise source motion trajectory data and the potential noise source position data to obtain the moving source path data;

[0011] The modal contribution analysis of the dynamic source path data is performed according to the vibration modes of the substation equipment, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution map.

[0012] The present invention obtains a three-dimensional environmental model of each structure in a substation, which can be performed by using laser scanning technology or photogrammetry technology. These technologies can obtain the geometric shape and structural information of the building and construct an accurate three-dimensional model. Then, the topological relationship and boundary conditions of the geometric body are extracted from the three-dimensional environmental model, which helps to establish the connection relationship and boundary constraints between the structures. The noise images between adjacent units are fused by the outer product convolution fusion method using the geometric body topological relationship matrix and the local noise image data. This process can be regarded as propagating and integrating local noise information, and finally obtaining the initial sound source distribution estimation data of the entire substation area. The result of this estimation can be used as the starting point for subsequent noise source positioning and separation. Using the generalized coordinated filtering algorithm, the direct sound signal and the multipath echo components are separated according to the initial sound source distribution estimation data. At the same time, the influence of multipath noise is removed by the manifold learning method to generate echo decoupling data. The purpose of this step is to reduce the interference of multipath echoes on noise source positioning and separation, and improve the accuracy of positioning and separation. The echo decoupling data is dimensionally compressed, and dimensionality reduction techniques such as principal component analysis (PCA) can be used. Then, the eigenmodes of the low-order scattering field are extracted from the compressed data. These eigenmodes reflect the characteristics of the scattering field. The parameters of the eigenmode set of the scattering field are estimated by the nonlinear regression model to obtain the eigenscattering cloud map data. The purpose of this step is to extract the characteristic information of the scattering field and provide a more accurate data basis for the location and separation of the noise source. According to the eigenscattering cloud map data, the center of the acoustic energy flow is identified, and the position of the noise source is matched with the position of the potential noise source. According to the potential noise source position data, the instantaneous movement direction of the source point is locked, and the motion trajectory of the source point is reconstructed to obtain the motion trajectory data of the noise source. At the same time, according to the noise source motion trajectory data and the potential noise source position data, the position information of the stationary source and the mobile source is distinguished, and the dynamic source path data is integrated. The purpose of this step is to accurately locate and track the position and motion trajectory of the noise source and provide detailed dynamic information. According to the vibration mode analysis of the substation equipment, the modal contribution analysis of the dynamic source path data is performed. By analyzing the contribution of the dynamic source path data under different vibration modes, the radiation mode of the stationary source and the mobile source can be separated. Then, these modes are projected into the scalar sound pressure field to obtain the distribution map of the separated sources. The purpose of this step is to distinguish the radiation modes of stationary sources and moving sources according to the characteristics of the vibration modes, and present their distribution in space, providing intuitive noise source positioning and separation results. In summary, the detailed explanation of these steps includes obtaining a three-dimensional environmental model, establishing geometric topological relationships, fusing local noise images, separating direct sound and echo components, removing multipath noise effects, dimensional compression and eigenmode extraction, noise source location identification and motion trajectory reconstruction, dynamic source path distinction, modal contribution analysis, and noise source distribution map after separation.The combined application of these steps can achieve the precise positioning and effective separation of noise sources, providing useful information and decision support for noise control and management.

[0013] The present invention also provides a low-frequency noise source positioning device, comprising:

[0014] The three-dimensional environment modeling module is used to obtain the three-dimensional environment model of each structure in the substation, obtain the geometric body topological relationship matrix in the three-dimensional environment model, perform on-site measurement of the substation, and obtain noise measurement data; based on the measurement position, the noise measurement data is divided into blocks corresponding to the three-dimensional environment model, and registered to obtain local noise image data;

[0015] A noise measurement module is used to perform outer convolution fusion on the noise images between adjacent units according to the geometric body topological relationship matrix and the local noise image data, and iteratively propagate them to the entire substation area to generate initial sound source distribution estimation data for the entire substation;

[0016] A noise distribution estimation module is used to separate the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data;

[0017] The echo decoupling module is used to perform dimension compression on the echo decoupling data and extract the eigenmodes of the low-order scattering field to obtain the eigenmode set of the scattering field; perform parameter estimation on the eigenmode set of the scattering field to obtain the eigenscattering cloud map data;

[0018] The dynamic source path analysis module is used to identify the center of the acoustic energy flow around the intrinsic scattering cloud map data, and correspond the noise source position to obtain the potential noise source position data; according to the potential noise source position data, the instantaneous motion direction of the source point is locked, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; according to the noise source motion trajectory data and the potential noise source position data, the position information of the stationary source and the mobile source is distinguished and integrated to obtain the dynamic source path data;

[0019] The modal contribution analysis module is used to perform modal contribution analysis on the dynamic source path data according to the vibration modes of the substation equipment, separate the radiation modes of the stationary and moving noise sources, project them into the scalar sound pressure field, and obtain the separated noise source distribution map. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0021] Figure 1 It is a schematic diagram of the steps of the method for locating the low-frequency noise source of a substation according to the present invention;

[0022] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0023] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0024] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0025] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0026] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0027] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for locating a low-frequency noise source in a substation, the method comprising the following steps:

[0028] Acquire a three-dimensional environmental model of each structure in the substation, and obtain a topological relationship matrix of geometric bodies in the three-dimensional environmental model; perform on-site measurement of the substation to obtain noise measurement data; correspond the noise measurement data to the three-dimensional environmental model in blocks based on the measurement position, and perform registration to obtain local noise image data;

[0029] According to the geometric body topological relationship matrix and the local noise image data, the noise images between adjacent units are fused by outer convolution, and iteratively propagated to the entire substation area to generate initial sound source distribution estimation data of the entire substation;

[0030] Separating the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data;

[0031] The echo decoupling data is dimensional compressed and the eigenmode of the low-order scattering field is extracted to obtain the eigenmode set of the scattering field; the parameters of the field mode of the eigenmode set of the scattering field are estimated to obtain the eigenscattering cloud map data;

[0032] The acoustic energy flow around the center of the intrinsic scattering cloud image data is identified, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; the position information of the stationary source and the mobile source is distinguished and integrated according to the noise source motion trajectory data and the potential noise source position data to obtain the moving source path data;

[0033] The modal contribution analysis of the dynamic source path data is performed according to the vibration modes of the substation equipment, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution map.

[0034] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for locating a low-frequency noise source in a substation according to the present invention. In this example, the method for locating a low-frequency noise source in a substation includes the following steps:

[0035] Acquire a three-dimensional environmental model of each structure in the substation, obtain the three-dimensional environmental model and obtain a geometric topological relationship matrix in the model; perform on-site measurement of the substation to obtain noise measurement data; correspond the noise measurement data to the three-dimensional environmental model in blocks based on the measurement position, and perform registration to obtain local noise image data;

[0036] The embodiment of the present invention can use tools such as laser scanners, photogrammetry technology or CAD software to obtain three-dimensional model data of various structures in the substation. The three-dimensional environmental model is processed to extract the geometric topological relationship and boundary condition information therein, such as the connection relationship and spatial position relationship between the structures. Noise measurement is performed in the actual scene of the substation to obtain noise data. The measurement position is aligned with the three-dimensional environmental model, and the noise data is associated with the model of the corresponding position to obtain local noise image data.

[0037] According to the geometric body topological relationship matrix and the local noise image data, the noise images between adjacent units are fused by outer convolution, and iteratively propagated to the entire substation area to generate initial sound source distribution estimation data of the entire substation;

[0038] The embodiment of the present invention uses the geometric topological relationship matrix to perform outer convolution on the local noise image between adjacent units, fuses the noise data of different units, and obtains the initial sound source distribution estimation data. Through the iterative propagation algorithm, the initial sound source distribution estimation data is propagated in the entire substation area, and the sound source estimation value of each unit is updated until convergence, and the sound source distribution estimation data of the entire substation is obtained.

[0039] Separating the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data;

[0040] The embodiment of the present invention uses a generalized coordinated filtering algorithm to process the initial sound source distribution estimation data to separate the direct sound signal and the multipath echo component. The manifold learning method is applied to model and learn the multipath echo component, remove the influence of multipath noise, and obtain echo decoupling data.

[0041] The echo decoupling data is dimensional compressed and the eigenmode of the low-order scattering field is extracted to obtain the eigenmode set of the scattering field; the parameters of the field mode of the eigenmode set of the scattering field are estimated to obtain the eigenscattering cloud map data;

[0042] The embodiment of the present invention compresses the dimension of the echo decoupling data, reduces the dimension of the data, and reduces redundant information. An appropriate method is applied to extract the eigenmode of the low-order scattering field to obtain the eigenmode set of the scattering field. The nonlinear regression model is used to estimate the parameters of the eigenmode set of the scattering field to obtain the parameters of the field mode; according to the parameter estimation results, the eigenscattering cloud map data is generated for subsequent analysis and processing.

[0043] The acoustic energy flow around the center of the intrinsic scattering cloud image data is identified, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; the position information of the stationary source and the mobile source is distinguished and integrated according to the noise source motion trajectory data and the potential noise source position data to obtain the moving source path data;

[0044] The embodiment of the present invention identifies the central area surrounded by the acoustic energy flow by analyzing the intrinsic scattering cloud map data, where there may be a potential noise source. The potential noise source position is matched with the center of the acoustic energy flow, and the position data of the potential noise source is determined. According to the potential noise source position data, the instantaneous motion direction of the noise source is inferred. Based on the instantaneous motion direction and time information of the source point, the motion trajectory data of the noise source is reconstructed. According to the noise source motion trajectory data and the potential noise source position data, the position information of the stationary source and the mobile source is distinguished and integrated to obtain the moving source path data.

[0045] The modal contribution analysis of the dynamic source path data is performed according to the vibration modes of the substation equipment, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution map.

[0046] The embodiment of the present invention analyzes the vibration modes of the substation equipment and determines the contribution of each mode. The dynamic source path data and the contribution of the vibration mode are analyzed correspondingly to determine the contribution of the dynamic source path in the vibration mode. According to the dynamic source path data and the modal contribution, the radiation modes of the stationary noise source and the mobile noise source are separated. The separated noise source radiation mode is projected onto the scalar sound pressure field to obtain a source distribution map of the noise source.

[0047] The present invention obtains a three-dimensional environmental model of each structure in a substation, which can be performed by using laser scanning technology or photogrammetry technology. These technologies can obtain the geometric shape and structural information of the building and construct an accurate three-dimensional model. Then, the topological relationship and boundary conditions of the geometric body are extracted from the three-dimensional environmental model, which helps to establish the connection relationship and boundary constraints between the structures. The noise images between adjacent units are fused by the outer product convolution fusion method using the geometric body topological relationship matrix and the local noise image data. This process can be regarded as propagating and integrating local noise information, and finally obtaining the initial sound source distribution estimation data of the entire substation area. The result of this estimation can be used as the starting point for subsequent noise source positioning and separation. Using the generalized coordinated filtering algorithm, the direct sound signal and the multipath echo components are separated according to the initial sound source distribution estimation data. At the same time, the influence of multipath noise is removed by the manifold learning method to generate echo decoupling data. The purpose of this step is to reduce the interference of multipath echoes on noise source positioning and separation, and improve the accuracy of positioning and separation. The echo decoupling data is dimensionally compressed, and dimensionality reduction techniques such as principal component analysis (PCA) can be used. Then, the eigenmodes of the low-order scattering field are extracted from the compressed data. These eigenmodes reflect the characteristics of the scattering field. The parameters of the eigenmode set of the scattering field are estimated by the nonlinear regression model to obtain the eigenscattering cloud map data. The purpose of this step is to extract the characteristic information of the scattering field and provide a more accurate data basis for the location and separation of the noise source. According to the eigenscattering cloud map data, the center of the acoustic energy flow is identified, and the position of the noise source is matched with the position of the potential noise source. According to the potential noise source position data, the instantaneous movement direction of the source point is locked, and the motion trajectory of the source point is reconstructed to obtain the motion trajectory data of the noise source. At the same time, according to the noise source motion trajectory data and the potential noise source position data, the position information of the stationary source and the mobile source is distinguished, and the dynamic source path data is integrated. The purpose of this step is to accurately locate and track the position and motion trajectory of the noise source and provide detailed dynamic information. According to the vibration mode analysis of the substation equipment, the modal contribution analysis of the dynamic source path data is performed. By analyzing the contribution of the dynamic source path data under different vibration modes, the radiation mode of the stationary source and the mobile source can be separated. Then, these modes are projected into the scalar sound pressure field to obtain the distribution map of the separated sources. The purpose of this step is to distinguish the radiation modes of stationary sources and moving sources according to the characteristics of the vibration modes, and present their distribution in space, providing intuitive noise source positioning and separation results. In summary, the detailed explanation of these steps includes obtaining a three-dimensional environmental model, establishing geometric topological relationships, fusing local noise images, separating direct sound and echo components, removing multipath noise effects, dimensional compression and eigenmode extraction, noise source location identification and motion trajectory reconstruction, dynamic source path distinction, modal contribution analysis, and noise source distribution map after separation.The combined application of these steps can achieve the precise positioning and effective separation of noise sources, providing useful information and decision support for noise control and management.

[0048] Preferably, a three-dimensional environment model of each structure in the substation is obtained, and a topological relationship matrix of geometric bodies in the three-dimensional environment model is obtained; on-site measurement of the substation is performed to obtain noise measurement data; based on the measurement position, the noise measurement data is divided into blocks corresponding to the three-dimensional environment model, and registration is performed to obtain local noise image data, including:

[0049] Obtain the 3D environmental model data of each structure in the substation;

[0050] Extract and represent the geometrical body of the three-dimensional environment model, and the geometrical body represents the model data;

[0051] Extract the topological relationship and boundary conditions of the geometric bodies in the model according to the three-dimensional environment model data and the geometric body representation model data, and obtain the geometric body topological relationship matrix;

[0052] Conduct on-site measurements of substations to obtain noise measurement data, including noise decibel data and measurement point location data;

[0053] Perform wavelet transform filtering on the noise decibel data to obtain noise image data;

[0054] According to the measurement point position data, the noise image data is divided into blocks and matched to the three-dimensional environment model to obtain the block noise data;

[0055] Image registration is performed based on the block noise data and the model units in the three-dimensional environment model to obtain local noise image data.

[0056] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0057] Obtain the 3D environmental model data of each structure in the substation;

[0058] The embodiment of the present invention uses a laser scanner to scan the substation to obtain point cloud data. A drone or other photogrammetry equipment is used to take aerial or ground photos to obtain image data of the substation. A three-dimensional model is constructed using computer-aided design software based on the design drawings and measurement data of the substation.

[0059] Extract and represent the geometrical body of the three-dimensional environment model, and the geometrical body represents the model data;

[0060] The embodiment of the present invention processes the point cloud data acquired by the laser scanner to extract geometric information therein, such as walls, pillars, equipment, etc. The image data acquired by photogrammetry is processed, and the geometric information in the image is extracted using a computer vision algorithm; the three-dimensional model data is imported through CAD software to extract the geometric information therein.

[0061] Extract the topological relationship and boundary conditions of the geometric bodies in the model according to the three-dimensional environment model data and the geometric body representation model data, and obtain the geometric body topological relationship matrix;

[0062] The embodiment of the present invention represents model data based on geometric bodies, segments geometric bodies in the model, and divides them into different units or components; extracts topological relationships between geometric bodies, such as adjacency relationships, connection relationships, etc., based on the relative positions and connection relationships of the geometric bodies; extracts boundary condition information, such as sound-absorbing materials on the wall, sound insulation pads on the floor, etc., based on the boundary features of the geometric bodies. Based on the extracted topological relationships and boundary conditions, a geometric body topological relationship matrix is ​​constructed for subsequent noise image data processing.

[0063] Conduct on-site measurements of substations to obtain noise measurement data, including noise decibel data and measurement point location data;

[0064] The embodiment of the present invention uses a noise measuring instrument to measure noise decibels at different locations of the substation and record the noise level at the measuring point. A total station, GPS or other measuring equipment is used to measure the position coordinates of each measuring point and record the spatial position information of the measuring point.

[0065] Perform wavelet transform filtering on the noise decibel data to obtain noise image data;

[0066] The embodiment of the present invention applies wavelet transform to the noise decibel data to convert the time domain signal into a frequency domain signal. The frequency domain signal after wavelet transform is filtered to remove unnecessary frequency components and retain the frequency information related to the noise. The filtered frequency domain signal is inversely transformed to convert the frequency domain signal back to the time domain signal to obtain the noise image data.

[0067] According to the measurement point position data, the noise image data is divided into blocks and matched to the three-dimensional environment model to obtain the block noise data;

[0068] The embodiment of the present invention corresponds the position data of the measuring point to the geometric body in the three-dimensional environment model. According to the spatial position information of each measuring point, the geometric body or model unit corresponding to it in the three-dimensional environment model is determined. The noise image data is divided into blocks according to the corresponding relationship of the measuring point position. According to the geometric body or model unit where the measuring point is located, the noise image data is divided into corresponding blocks, and each block corresponds to a geometric body or model unit.

[0069] Image registration is performed based on the block noise data and the model units in the three-dimensional environment model to obtain local noise image data.

[0070] The embodiment of the present invention registers the block noise data with the model unit in the three-dimensional environment model so that they are aligned in the same coordinate system. Image processing algorithms such as feature point matching, transformation model and other methods can be used for registration. Through the registration operation, the block noise data is aligned with the corresponding model unit to obtain local noise image data. Each local noise image data corresponds to a model unit, reflecting the noise level in the model unit.

[0071] The present invention can accurately represent the geometric shape and structural information of the substation by acquiring the three-dimensional environmental model data of the substation. This is very important for the subsequent location and separation of noise sources, because the propagation and reflection of noise will be affected by buildings. By processing the three-dimensional environmental model, the geometric bodies therein are extracted and represented. The advantage of this is that a complex three-dimensional environmental model can be converted into a group of geometric bodies, simplifying the complexity of subsequent processing. According to the three-dimensional environmental model data and the geometric body representation model data, the topological relationship and boundary conditions between the geometric bodies are extracted. This information is very important for establishing the connection relationship and boundary constraints between the geometric bodies, which is helpful for the subsequent location and separation of noise sources. By performing on-site measurement at the substation, actual noise measurement data can be obtained. These data include the intensity of the noise (in decibels) and the location information of the measurement point. These data can be used for subsequent analysis and processing. The noise decibel data is subjected to wavelet transform filtering. Wavelet transform filtering can smooth and denoise the noise data to obtain clearer and more accurate noise image data. These noise image data can be used for the subsequent generation of local noise image data. Using the measurement point location data, the noise image data is divided into blocks, and each block is matched to the corresponding position in the three-dimensional environment model. This can match the noise data with the actual geometric structure of the substation, providing an accurate data basis for the subsequent location and separation of noise sources. Image registration is performed using the block noise data and the model unit in the three-dimensional environment model. Image registration can align the block noise data with the corresponding model unit to obtain local noise image data on each model unit. These local noise image data can be used for subsequent noise source location and separation to provide more accurate local noise information. In summary, the three-dimensional environment model data of each structure in a substation can be obtained, and the geometric body can be extracted and represented. Then, by extracting the topological relationship and boundary conditions of the geometric body, the geometric body topological relationship matrix is ​​obtained. At the same time, the substation is measured on site to obtain noise measurement data, including noise decibel data and measurement point location data. Noise image data is obtained by wavelet transform filtering the noise decibel data. Then, the noise image data is divided into blocks and matched to the three-dimensional environment model, and image registration is performed according to the model unit to obtain local noise image data. The above steps build an accurate 3D environment model, extract geometric topological relationships, obtain actual noise measurement data, denoise and smooth noise data, and obtain accurate local noise image data. These effects provide the basis and support for tasks such as noise analysis, noise source location and separation.

[0072] Preferably, the topological relationship and boundary conditions of the geometric bodies in the model are extracted according to the three-dimensional environment model data and the geometric body representation model data to obtain a geometric body topological relationship matrix, including:

[0073] Performing basic shape type recognition on the geometric body representation model data to obtain geometric body type data;

[0074] The embodiment of the present invention uses geometric analysis software, such as CAD software or a special geometric processing tool, to import geometric representation model data. Apply a geometric type recognition algorithm to classify the geometric body into basic shape types, such as cubes, cylinders, spheres, etc., by analyzing the shape, boundary features, curvature and other properties of the geometric body. Use an automatic geometric recognition tool to automatically recognize and classify the geometric representation model data through pattern matching, feature extraction and other technologies.

[0075] Extract the acoustic impedance and absorption coefficient based on the material information according to the three-dimensional environment model data and the geometric representation model data to obtain the geometric boundary condition data;

[0076] The embodiment of the present invention obtains the material information of each geometric body according to the three-dimensional environment model data, including the surface material, density, roughness, etc. The acoustic impedance of each geometric body is calculated according to the material information of the geometric body using the acoustic engineering calculation method, that is, the ability of the geometric body interface to reflect sound waves is calculated. The absorption coefficient of each geometric body is calculated according to the material information of the geometric body using the acoustic engineering calculation method, that is, the ability of the geometric body to absorb sound waves is calculated.

[0077] Performing topological relationship analysis of adjacency, inclusion, separation and intersection between geometric bodies according to geometric body representation model data and geometric body type data to obtain geometric body topological relationship data;

[0078] The embodiment of the present invention extracts spatial relationships between geometric bodies, such as distance, relative position, etc., based on geometric body representation model data. The type of each geometric body is determined based on geometric body type data. Topological relationship analysis algorithms, such as Euclidean space analysis and bounding box analysis, are applied to determine the adjacent, contained, separated, and intersecting relationships between geometric bodies.

[0079] Based on the geometric body topological relationship data and the geometric body boundary condition data, the topological relationship and boundary conditions are integrated into a matrix form to obtain a geometric body topological relationship matrix.

[0080] The embodiment of the present invention constructs a matrix to represent the topological relationship between geometric bodies based on geometric body topological relationship data. The rows and columns of the matrix correspond to each geometric body, and the matrix elements represent the topological relationship between the corresponding geometric bodies, such as adjacent, contained, separated, intersecting, etc. According to the geometric body boundary condition data, a matrix is ​​constructed to represent the boundary conditions of the geometric body. The rows of the matrix correspond to each geometric body, the columns correspond to each boundary condition, and the matrix elements represent the properties of the corresponding geometric body under the boundary condition, such as acoustic impedance, absorption coefficient, etc. By integrating the topological relationship matrix and the boundary condition matrix, the two matrices can be merged or combined according to certain rules to form a comprehensive geometric body topological relationship matrix, which contains the topological relationship between the geometric bodies and the boundary condition information of each geometric body.

[0081] The embodiment of the present invention can determine the type of each geometric body, such as a sphere, a cube, a cylinder, etc., by performing basic shape type recognition on the geometric body representation model data. These geometric body type data can be used for analysis and processing in subsequent steps. Based on the recognition and classification of the geometric body representation model data, the basic features and properties of the geometric body can be extracted, providing a basis for further acoustic, topological relationship and boundary condition analysis. By extracting the acoustic impedance and absorption coefficient based on material information, the boundary conditions of each geometric body, that is, the acoustic characteristics between it and the surrounding environment, can be determined. These boundary condition data are very important for acoustic simulation and acoustic analysis. By extracting the acoustic characteristics of the geometric body, such as reflection, absorption and scattering, the behavior and influence of the geometric body in the acoustic environment can be evaluated. This helps to design and optimize acoustic systems, improve sound propagation and reduce noise. By performing topological relationship analysis on the geometric body representation model data, the adjacent, contained, separated and intersecting relationships between the geometric bodies can be determined. These topological relationship data are very important for structural analysis, collision detection and object interaction of the model. By analyzing the topological relationship between the geometric bodies, their organizational structure and connection mode can be understood, which helps to understand the shape, layout and function of the object. By integrating the topological relationships and boundary conditions of geometric bodies into a matrix form, the relationships between various geometric bodies can be represented in a structured manner. This matrix representation facilitates further analysis and processing, such as topological optimization, graph analysis, and network analysis. Integrating topological relationships and boundary conditions into a matrix helps to organize and manage a large amount of geometric body-related data and improve the efficiency of data processing and calculation. At the same time, the matrix representation can also support the application of algorithms and technologies based on matrix operations. In summary, through the above steps, the identification and classification of geometric body representation model data, acoustic property extraction, topological relationship analysis, and data integration and structuring can be achieved. These can provide detailed information about the type of geometry, boundary conditions, acoustic properties, and topological relationships, providing a basis and support for subsequent tasks such as geometry analysis, acoustic simulation, structural analysis, and optimization.

[0082] Preferably, the noise images between adjacent units are fused by outer convolution according to the geometric topological relationship matrix and the local noise image data, and iteratively propagated to the entire substation area to generate initial sound source distribution estimation data of the entire substation, including:

[0083] According to the geometric topological relationship matrix, the substation space is discretized into small space unit grids to obtain space grid unit data;

[0084] Assigning local noise image data to the spatial unit of spatial grid unit data according to the measurement point position data to obtain initial grid noise distribution data;

[0085] The initial grid noise distribution data and the geometric body topological relationship matrix perform outer convolution on the noise of adjacent grid cells based on boundary conditions to obtain the fusion result data;

[0086] The fusion result data is used as the initial value, and the outer convolution operation based on the boundary conditions is repeated until it propagates to the entire substation area to obtain the grid noise distribution data;

[0087] The grid noise distribution data is interpolated and denoised to obtain the initial sound source distribution estimation data.

[0088] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0089] According to the geometric topological relationship matrix, the substation space is discretized into small space unit grids to obtain space grid unit data;

[0090] The embodiment of the present invention divides the space of the substation into small discrete unit grids according to the geometric body topological relationship matrix. A grid generation algorithm, such as a finite element method or a voxel grid method, can be used to divide the space into discrete grid units. For each grid unit, relevant data such as its position, size, and boundary information are recorded. These data can be used for subsequent calculations and analysis.

[0091] Assigning local noise image data to the spatial unit of spatial grid unit data according to the measurement point position data to obtain initial grid noise distribution data;

[0092] The embodiment of the present invention obtains the position data of the measuring points, which are located inside or around the substation, for measuring the noise level. According to the position data of the measuring points, the local noise image data of the corresponding positions are obtained. Noise collection can be performed using a sensor or a noise measuring device, or the noise data can be obtained using an analog or numerical simulation method. The local noise image data corresponding to the position data of the measuring point is assigned to the corresponding spatial grid unit. An interpolation method, such as nearest neighbor interpolation or bilinear interpolation, can be used to assign the local noise value to the grid unit closest to the measuring point.

[0093] The initial grid noise distribution data and the geometric body topological relationship matrix perform outer convolution on the noise of adjacent grid cells based on boundary conditions to obtain the fusion result data;

[0094] The embodiment of the present invention is based on the initial grid noise distribution data obtained in the previous step, and each grid unit contains local noise information. According to the geometric body topological relationship matrix, an outer product convolution operation based on boundary conditions is performed between adjacent grid units. Outer product convolution can be used to simulate the propagation and interaction process of sound waves. According to the geometric body topological relationship matrix and boundary condition data, the boundary conditions between adjacent grid units, such as acoustic impedance, absorption coefficient, etc., are determined. The results of the outer product convolution are merged with the boundary conditions to obtain new grid noise distribution data. These data reflect the sound wave propagation and interaction effects between adjacent grid units.

[0095] The fusion result data is used as the initial value, and the outer convolution operation based on the boundary conditions is repeated until it propagates to the entire substation area to obtain the grid noise distribution data;

[0096] The embodiment of the present invention uses the above-obtained fusion result data as the initial value; repeats the above outer product convolution operation, and uses the fusion result data as the new initial value. The number of iterations and the convergence condition can be set according to the specific situation. The outer product convolution operation is continuously iterated until the fusion result data is propagated to the entire substation area. This ensures that the propagation and interaction effects of noise are considered in the entire area, and more accurate grid noise distribution data is obtained.

[0097] The grid noise distribution data is interpolated and denoised to obtain the initial sound source distribution estimation data.

[0098] The embodiment of the present invention performs interpolation processing on the discretized grid noise distribution data to convert it into continuous sound source distribution data. Common interpolation methods include linear interpolation, spline interpolation, etc., and a suitable interpolation method is selected according to the needs. The sound source distribution data obtained by interpolation is denoised to eliminate the influence of noise or outliers. The denoising processing method may include filters, wavelet transforms, statistical methods, etc., and an appropriate denoising method is selected according to the characteristics of the noise.

[0099] The present invention can divide the substation space into small spatial unit grids by discretizing it according to the geometric body topological relationship matrix. These spatial grid unit data provide a structured description of the substation area, which provides a basis for subsequent noise distribution analysis and processing. By gridding the substation space, the complex geometric structure can be converted into a regular grid structure, simplifying the processing and calculation of the problem. Gridding also provides convenience for subsequent operations such as interpolation, noise propagation and sound source distribution estimation. By assigning the measurement point position data and the local noise image data to the spatial unit of the spatial grid unit data, the initial grid noise distribution data can be obtained. These data describe the noise level in different spatial units of the substation, providing an initial state for subsequent noise propagation and analysis. By associating the measurement point position data with the spatial grid unit data, the actual measured noise data can be matched with the spatial grid unit, realizing the spatial representation of the noise data, and facilitating the analysis and processing of the noise. By performing outer product convolution based on boundary conditions on the initial grid noise distribution data, the noise between adjacent grid units can be fused. In this way, the mutual influence and propagation law between the grid units can be taken into account, and more accurate noise distribution data can be obtained. Through the outer product convolution based on boundary conditions, the propagation process of noise between adjacent grid cells can be simulated. This helps to understand the diffusion law of noise in the substation space and provide a basis for subsequent noise control and optimization. By repeatedly performing outer product convolution based on boundary conditions, the noise can be gradually propagated to the grid cells of the entire substation area, and more accurate grid noise distribution data can be obtained. These data provide detailed noise distribution information and can be used for noise assessment and the formulation of control measures. By repeatedly performing outer product convolution operations based on boundary conditions and taking the fused result data as the initial value of the next iteration, the precision and accuracy of the noise distribution data can be gradually improved. Such an iterative process can better simulate the propagation and distribution of noise and improve the reliability of the model. By interpolating and denoising the grid noise distribution data, the initial sound source distribution estimation data can be obtained. These data describe the location and intensity distribution of sound sources in the substation, providing a basis for sound source control and optimization. Through interpolation and denoising, the gaps and noise interference in the grid noise distribution data can be filled to improve the integrity and accuracy of the data. This helps to better understand and analyze the distribution of sound sources in the substation. In summary, the above steps include providing spatial grid unit data, realizing data association, simulating noise propagation and distribution, optimizing the precision and accuracy of noise distribution data, and obtaining initial sound source distribution estimation data. These effects help to deeply understand the noise characteristics in substations and provide a scientific basis for noise control and optimization.

[0100] Preferably, separating the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data includes:

[0101] The generalized coordinated filtering algorithm is used to perform parameterized modeling on the initial sound source distribution estimation data, and the model is divided into two parts: the direct sound component and the multipath sound component, thus obtaining the direct sound model and the multipath sound model.

[0102] The generalized coordinated filtering of the embodiment of the present invention is a statistical model estimation method for separating signal components. By performing parameterized modeling on the sound source distribution data, the sound source model is decomposed into a direct sound component and a multipath sound component. According to the initial sound source distribution estimation data, the sound source is parameterized modeled using an appropriate mathematical model. Commonly used methods include Gaussian mixture models, sparse representation models, etc. Parametric modeling can decompose the sound source into different components, such as direct sound and multipath sound. Through parameterized modeling, the model of the direct sound component is extracted. The direct sound model describes the sound wave component that is directly propagated from the sound source to the receiving point. Through parameterized modeling, the model of the multipath sound component is extracted. The multipath sound model describes the sound wave component that is propagated from the sound source to the receiving point through multiple paths such as reflection and diffraction.

[0103] A joint optimization objective function is constructed based on the direct sound model and the multipath sound model, the two models are weighed by the coordination factor, and the objective function is solved using the generalized least squares method to obtain the separated direct sound component data and multipath sound component data;

[0104] The embodiment of the present invention constructs a joint optimization objective function based on the direct sound model and the multipath sound model, which includes the separation constraints of the direct sound and the multipath sound. The specific form of the objective function can be defined according to the specific sound source model and the separation requirements. A coordination factor is introduced to weigh the importance between the direct sound and the multipath sound. The coordination factor can be a parameter, and the contribution of the direct sound and the multipath sound in the separation result can be controlled by adjusting its value. The objective function is solved by the generalized least squares method to minimize the objective function and obtain a solution that optimizes the separation effect of the direct sound and the multipath sound. The generalized least squares method is a commonly used parameter estimation method, which can be solved by an optimization algorithm, such as the gradient descent method or the Newton method.

[0105] A random state space model of the multipath echo path is constructed based on the multipath acoustic component data, and each path is filtered and tracked based on the Kalman particle filter framework. The path delay, attenuation and direction are statistically analyzed to obtain preliminary multipath path parameter data.

[0106] The embodiment of the present invention models the multipath acoustic component data as a random state space model, in which the state represents the parameters of the multipath echo path, such as the delay, attenuation and direction. According to the acoustic characteristics and signal propagation theory, a suitable state space model can be selected, such as a linear dynamic system model. The multipath echo path is tracked and estimated using the Kalman particle filter framework. The Kalman particle filter is a recursive filtering algorithm that can estimate the state variables of a dynamic system. By continuously iterating and updating the weights of the particles, an estimate of the multipath path parameters can be obtained. Based on the tracking results of the Kalman particle filter, the parameters of the multipath path, such as the delay, attenuation and direction, are statistically analyzed to obtain preliminary multipath path parameter data.

[0107] The multipath acoustic component data is sparsely represented on the wavelet basis to obtain a noise sparse representation model;

[0108] The embodiment of the present invention performs wavelet transform on the multipath acoustic component data, converting it from the time domain to the wavelet domain. Wavelet transform can represent the signal as a linear combination of wavelet basis functions of different scales and frequencies. In the wavelet domain, the multipath acoustic component data is represented by a sparse representation method. The goal of sparse representation is to approximate the original signal by a small number of wavelet basis function coefficients, so that most of the coefficients are zero or close to zero. The noise sparse representation model of the multipath acoustic component data is obtained by using the sparse representation. The noise sparse representation model describes the sparsity characteristics of the multipath acoustic component data on the wavelet basis.

[0109] Adaptively model the multipath noise according to the noise sparse representation model and preliminary multipath path parameter data, and subtract the multipath noise component from the direct sound component data to obtain the multipath noise-removed field data;

[0110] The embodiment of the present invention designs an iterative algorithm to decompose and optimize the multipath acoustic component data based on the sparse representation model and the multipath path parameter data. The goal of the iterative algorithm is to separate the multipath echo path data and the background noise data, and continuously update the optimization result. Through the iterative algorithm, the multipath acoustic component data is decomposed into multipath echo path data and background noise data. The decomposition process can utilize the sparsity characteristics of the sparse representation to represent the multipath echo path data as a small number of non-zero wavelet basis function coefficients, and represent the background noise data as a sparse vector with most coefficients close to zero. In the iterative process, the optimization result is updated according to the decomposition result and the multipath path parameter data, so that the separated multipath echo path data and background noise data are more accurate and reliable.

[0111] The residual noise manifold is constructed for the residual noise of the multipath denoised field data using manifold learning technology, and the multipath denoised field data is mapped onto the noise manifold to obtain the reconstructed residual multipath noise component data;

[0112] The embodiment of the present invention preprocesses the multipath denoised field data, such as removing the direct sound component and the multipath noise component, and retaining only the residual noise. Use a manifold learning tool (such as t-SNE or Isomap) to perform manifold learning on the residual noise data to construct a residual noise manifold. Select appropriate parameters according to the specific situation, such as neighborhood size, dimensionality reduction, etc. Preprocess the multipath denoised field data to remove the direct sound component and retain only the multipath denoised component. Use a manifold learning mapping method (such as LLE, Isomap, etc.) to map the multipath denoised field data to the constructed noise manifold. This will obtain the corresponding position of the multipath denoised field data on the noise manifold.

[0113] The reconstructed residual multipath noise component data is subtracted from the multipath de-noised field data to obtain the echo decoupling data.

[0114] The embodiment of the present invention applies wavelet transform to the multipath de-noised field data to convert it into wavelet coefficients. Appropriate wavelet basis functions and scales are selected to adapt to the characteristics of the multipath de-noised field data. The wavelet coefficients of the multipath de-noised field data are sparsely represented using a sparse representation method. An L1 norm minimization algorithm (such as the Lasso algorithm) or a dictionary learning algorithm can be used to find as few non-zero coefficients as possible to represent the multipath de-noised field data. Based on the result of the sparse representation, a noise sparse representation model of the multipath de-noised field data is obtained, which contains fewer non-zero coefficients to represent the noise component.

[0115] The present invention performs parameterized modeling on the initial sound source distribution estimation data through a generalized coordinated filtering algorithm. This represents the sound source distribution data as a combination of two models, a direct sound component and a multipath sound component, which better describes the propagation and reflection characteristics of sound in a substation. The sound source is divided into two parts, a direct sound component and a multipath sound component, and corresponding models are established respectively. The direct sound model describes the path of sound propagating directly from the source to the receiving point, while the multipath sound model describes the path of sound reaching the receiving point after multiple reflections, diffractions and diffractions. Such decomposition helps to better understand the propagation mechanism of sound. By combining the direct sound model and the multipath sound model to construct a joint optimization objective function, the trade-off relationship between the two can be balanced. This helps to accurately separate the direct sound component and the multipath sound component from the mixed sound source distribution data, and improve the interpretability and accuracy of the data. By applying the generalized least squares method to solve the joint optimization objective function, the optimized direct sound component data and the multipath sound component data can be obtained. Such a solution method can effectively handle the parameter estimation problem and obtain a more accurate separation result. By constructing a random state space model of the multipath echo path based on the multipath acoustic component data, the propagation path of the multipath sound can be modeled. This helps to understand the multiple reflections and propagation characteristics of sound in the substation and provide a parameterized description of the path. The multipath echo path is tracked and estimated using the Kalman particle filter framework. This can extract statistical information such as the delay, attenuation and direction of the path from the observed data, providing a basis for subsequent multipath denoising and separation. By sparsely representing the multipath acoustic component data on a wavelet basis, the noise component can be separated from the signal and a sparse representation model of the noise can be extracted. This helps to further analyze and process the noise component and provides a basis for subsequent multipath denoising and noise separation. Based on the noise sparse representation model and preliminary multipath path parameter data, the multipath noise is adaptively modeled. This allows for more accurate modeling and estimation of multipath noise, helping to more effectively reduce the impact of multipath noise. Multipath denoised field data is obtained by subtracting the multipath noise component from the direct acoustic component data. This can remove the impact of multipath noise, making the sound signal cleaner and clearer, and improving the accuracy of subsequent processing tasks. Through manifold learning technology, the residual noise in the multipath denoised field data can be modeled and analyzed. This helps to better understand the inherent structure and characteristics of the residual noise and provide more accurate information for subsequent processing. By mapping the multipath denoised field data onto the noise manifold, the reconstructed residual multipath noise component data can be obtained. In this way, the information of the noise component can be extracted, the interference of multipath noise can be further reduced, and the quality of the signal can be improved. By subtracting the reconstructed residual multipath noise component data from the multipath denoised field data, the echo decoupling data can be obtained. In this way, the influence of the echo can be effectively eliminated, so that only the components related to the direct sound are retained in the signal, improving the clarity and comprehensibility of the signal.By removing the reconstructed residual multipath noise component, the interference of multipath noise on the signal can be effectively reduced, the quality and accuracy of the signal can be improved, and the subsequent processing tasks can be made more reliable and stable.

[0116] Preferably, the echo decoupling data is dimensional compressed, and the eigenmodes of the low-order scattering field are extracted to obtain the eigenmode set of the scattering field; the parameters of the field mode are estimated for the eigenmode set of the scattering field to obtain the eigenscattering cloud map data, including:

[0117] The echo decoupling data is discretely sampled in the time and space domains and arranged in a nuclear tensor form to obtain nuclear tensor representation data;

[0118] The embodiment of the present invention performs discrete sampling on the echo decoupling data in the spatiotemporal domain, and selects a suitable sampling method and sampling rate to obtain discrete data points. The data obtained by discrete sampling is arranged in a certain order in the form of a kernel tensor. According to the sampling dimension and the number of sampling points, a corresponding kernel tensor is constructed. For example, for three-dimensional data, a three-dimensional kernel tensor can be constructed.

[0119] The supersymmetric matrix decomposition technology is used to perform low-rank decomposition approximation on the core tensor representation data to obtain low-dimensional embedded vector data;

[0120] The embodiment of the present invention uses supersymmetric matrix decomposition technology to perform low-rank decomposition on the core tensor representation data. An appropriate supersymmetric matrix decomposition method is selected, and parameters are set according to the nature and dimension of the data. After decomposition, low-dimensional embedded vector data is obtained.

[0121] Mapping the low-dimensional embedded vector data onto the differential manifold and reconstructing the eigenmode of the scattering field to obtain the eigenmode set of the scattering field;

[0122] The embodiment of the present invention uses a manifold learning tool to map low-dimensional embedded vector data to a differential manifold to obtain a representation in the manifold space. Select a suitable manifold learning algorithm and parameter settings. Use an eigenmode reconstruction method to process the mapped data, for example, apply PCA or ICA and other techniques to extract the eigenmodes of the scattered field. Eigenmodes are characteristic patterns with statistical significance that can reflect important structures in the data.

[0123] The parameters of the field pattern of the eigenmode set of the scattering field are estimated through a nonlinear regression model to obtain the intrinsic scattering cloud map data.

[0124] The embodiment of the present invention uses the eigenmode set of the scattered field as input data and prepares corresponding target data, such as real field mode parameters. The selected nonlinear regression model is trained using the prepared input data and target data. The parameters of the model are iteratively optimized so that the model can estimate the parameters of the field mode for the input data; the trained nonlinear regression model is used to estimate the parameters of the field mode for the eigenmode set of the scattered field. Each eigenmode is used as input, and the model outputs the corresponding field mode parameter value. The field mode parameters are combined with the eigenmode to generate eigenscattering cloud map data. According to specific application requirements, the parameter values ​​can be used to reconstruct the scattering field or perform other subsequent processing.

[0125] The present invention can convert continuous signal data into discrete sampling points by discretely sampling echo decoupling data in the time and space domain, which is convenient for subsequent processing and calculation. The data obtained by discrete sampling is arranged in the form of a core tensor, and the multidimensional data structure can be converted into a higher-order tensor representation. In this way, the multidimensional characteristics and structural information of the data can be better captured, and a richer data representation form can be provided for subsequent processing. The core tensor representation data is approximated by low-rank decomposition through supersymmetric matrix decomposition technology. Low-rank decomposition can reduce high-dimensional data to a lower dimension, reduce redundant information of the data, and extract the main features of the data. The low-dimensional embedded vector data obtained by low-rank decomposition can map the original data to a lower-dimensional space. In this way, the storage and calculation costs of the data can be reduced, and the main features of the data are retained, which is helpful for subsequent processing and analysis. The low-dimensional embedded vector data is mapped to a differential manifold, and the data can be represented as a point on the manifold. In this way, the structure and geometric properties of the manifold can be used to analyze the data and extract the implicit features of the data. By mapping the low-dimensional embedded vector data to the differential manifold, the eigenmode of the scattered field can be reconstructed. The eigenmode of the scattering field is a characteristic mode that describes the scattering field. By reconstructing the eigenmode set of the scattering field, the main modes and characteristics of the scattering field can be extracted. By applying the nonlinear regression model, the parameters of the eigenmode set of the scattering field are estimated. The nonlinear regression model can obtain the parameter estimation results of the model by fitting the relationship between the actual data and the model. By estimating the parameters of the eigenmode set of the scattering field through the nonlinear regression model, the field pattern parameters of the scattering field can be obtained. These parameters can reflect the characteristics and properties of the scattering field and provide an in-depth understanding and description of the scattering field. Through the field pattern parameter estimation, the intrinsic scattering cloud map data can be obtained.

[0126] Preferably, the parameters of the field mode are estimated for the eigenmode set of the scattering field by using a nonlinear regression model to obtain the intrinsic scattering cloud map data, including:

[0127] The nonlinear scattering model of eigenmode theory is established according to the eigenmode set of the scattered field and the topological relationship matrix of the geometric body;

[0128] The embodiment of the present invention constructs a geometric model according to the shape and position of the actual scatterer, and establishes a geometric topological relationship matrix. A nonlinear scattering model is established using the eigenmode set of the scattering field and the geometric topological relationship matrix. Numerical methods, physical models or statistical methods can be used to select a suitable modeling method according to the actual problem.

[0129] The free parameters of wave number, attenuation rate and phase in the nonlinear scattering model are identified to obtain a free parameter set;

[0130] The embodiment of the present invention selects a suitable parameter identification method according to the specific nonlinear scattering model and data characteristics. The selected parameter identification method is used to identify free parameters such as wave number, attenuation rate and phase in the nonlinear scattering model. The estimated value of the free parameter is obtained by the degree of fit with the observed data.

[0131] According to the preset measured pattern data, the scattering field intrinsic mode set and the nonlinear scattering model difference are constructed as an optimization objective function to obtain a nonlinear regression objective function;

[0132] The embodiment of the present invention collects or generates a set of measured scattering pattern data as preset target data. By comparing the difference between the scattering field intrinsic mode set and the nonlinear scattering model on the measured pattern data, an optimization objective function for measuring the difference is constructed. The specific difference measurement method can be the mean square error, correlation coefficient, etc.

[0133] Solve the model parameters of the nonlinear regression objective function according to the free parameter set to obtain the free parameter estimation value data;

[0134] The embodiment of the present invention sets the initial parameter value of the nonlinear regression method according to the characteristics of the problem, and uses the selected nonlinear regression method to solve the nonlinear regression objective function to obtain the optimal free parameter estimation value data.

[0135] Substituting the free parameter estimation data into the nonlinear scattering model, the intrinsic scattering cloud map data is generated.

[0136] The embodiment of the present invention substitutes the free parameter estimation value data obtained in step S444 into the nonlinear scattering model to replace the original free parameters. The nonlinear scattering model after substitution is used to calculate and generate intrinsic scattering cloud map data, which reflects the characteristics of the scattering field.

[0137] The present invention establishes a nonlinear scattering model based on the eigenmode theory, which can more accurately describe the physical process and phenomenon of the scattering field. By considering the eigenmode set of the scattering field and the topological relationship of the geometric body, the model can capture more scattering phenomena and provide a more accurate description. The nonlinear scattering model can provide the ability to explain various physical phenomena in the scattering field. By establishing the model, the wave behavior in the scattering process, the interaction between scatterers, etc. can be understood, and the characteristics of the scattering field can be deeply studied. By identifying parameters such as wave number, decay rate and phase in the nonlinear scattering model, the free parameters in the model can be determined. These parameters control the behavior and characteristics of the model. Identifying them can better adjust and optimize the model to make it more consistent with the actual observation data. After obtaining the free parameter set, it can be used as a subsequent parameter estimation and optimization process. These parameter sets provide a preliminary estimate of the model parameters and provide a basis for further analysis and research. By constructing the difference between the eigenmode set of the scattering field and the nonlinear scattering model as an optimization objective function, the gap between the model and the actual observation data can be quantified. This objective function can be used as a basis for nonlinear regression optimization to help determine the optimal model parameters. By optimizing the objective function, the nonlinear regression model can be made to fit the measured data better. This can improve the accuracy and predictive power of the model, so that it can better describe and predict the actual scattering field. By solving the nonlinear regression objective function, the parameter estimates of the model can be obtained. These parameter estimates are the parameter values ​​that make the model fit the measured data best under the constraints of the optimization objective function, and can more accurately describe the characteristics and behavior of the scattering field. After obtaining the estimated value data of the free parameters, these values ​​can be used to represent the parameters in the model. These estimates provide specific numerical values ​​for the parameters of the model, which is helpful for quantitative analysis and study of the characteristics of the scattering field. By substituting the free parameter estimation data into the nonlinear scattering model, intrinsic scattering cloud map data can be generated. These cloud map data are calculated based on the estimated model parameters and reflect the characteristics and behavior of the scattering field. They provide a visual representation of the scattering field and help understand and analyze the scattering phenomenon. The generated intrinsic scattering cloud map data can be used to further study the characteristics of the scattering field. By analyzing the cloud map data, the pattern, structure, distribution and other information in the scattering field can be revealed, and the essence of the scattering phenomenon can be deeply understood. This is of great significance for the design and optimization of scattering field applications.

[0138] Preferably, the acoustic energy flow around the center is identified on the intrinsic scattering cloud map data, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; the position information of the stationary source and the mobile source is distinguished and integrated according to the noise source motion trajectory data and the potential noise source position data to obtain the moving source path data, including:

[0139] The acoustic energy flow around the center is identified on the intrinsic scattering cloud image data, and the noise source position is matched to obtain the potential noise source position data;

[0140] The embodiment of the present invention performs preprocessing operations such as noise removal and filtering on the intrinsic scattering cloud image data to improve the signal-to-noise ratio and data quality. Image processing technology and object recognition algorithms are used to identify the center position of the acoustic energy flow in the cloud image. The center position of the acoustic energy flow is matched with the position of the actual scatterer to obtain the position data of the potential noise source.

[0141] According to the potential noise source position data and the filtered acoustic energy flow field data, the acoustic energy flow gradient vector is calculated based on the source position to obtain the instantaneous motion direction data of the source point;

[0142] The embodiment of the present invention applies a suitable filtering algorithm, such as a Gaussian filter, to the acoustic energy flow field data to smooth the data and remove noise. Based on the potential noise source position data and the filtered acoustic energy flow field data, the acoustic energy flow gradient vector at the source point position is calculated to determine the instantaneous motion direction of the source point.

[0143] The instantaneous motion direction data of the source point is reconstructed based on the time series analysis of the source point motion trajectory to obtain the noise source motion trajectory data;

[0144] The embodiment of the present invention performs smoothing, filtering and other processing on the instantaneous motion direction data of the source point to remove noise and unnecessary fluctuations. Time series analysis methods such as Kalman filtering and spline interpolation are applied to analyze and reconstruct the processed data to obtain the motion trajectory data of the noise source.

[0145] According to the preset trajectory classification standard and the potential noise source position data, the noise source motion trajectory data is separated into stationary source and mobile source to obtain mobile source trajectory data and stationary source position data;

[0146] The embodiment of the present invention extracts features from the noise source motion trajectory data, calculates the speed and acceleration of each trajectory point, and records the change of the motion direction. Using the noise source classification algorithm, the extracted trajectory features are input into the classification model to classify the stationary source and the moving source.

[0147] The moving source trajectory data and the stationary source position data are merged to obtain the moving source path data.

[0148] In the embodiment of the present invention, for noise sources classified as stationary sources, traditional positioning methods such as Doppler positioning, cross-correlation positioning, etc. can be used for accurate positioning. For noise sources classified as mobile sources, methods based on the Doppler effect or sound source positioning technology, such as extended Kalman filtering (EKF) or particle filters, can be used for real-time or offline positioning estimation.

[0149] The present invention can determine the potential noise source position by performing acoustic energy flow around the center identification on the intrinsic scattering cloud map data. This helps to accurately locate the approximate position of the noise source and provides a basis for subsequent analysis. The step corresponding to the noise source position can pre-process the intrinsic scattering cloud map data and convert it into potential noise source position data that is more interpretable and analyzable. This can reduce the complexity of the data and reduce the amount of calculation for subsequent analysis. By calculating the acoustic energy flow gradient vector based on the source position, the instantaneous motion direction of the source point can be determined. This provides motion information of the source point, which can help understand the motion behavior and characteristics of the noise source. The instantaneous motion direction data of the source point can be used to analyze the changes in the motion mode, speed and motion trajectory of the noise source. This helps to reveal the motion laws and characteristics of the noise source and provides a basis for subsequent trajectory reconstruction and analysis. By performing time series analysis on the instantaneous motion direction data of the source point, the motion trajectory of the noise source can be reconstructed. This provides the motion path information of the noise source in space, which helps to conduct a more in-depth study and analysis of the motion behavior of the noise source. The noise source motion trajectory data can be used to understand the motion behavior and pattern of the noise source in space. By analyzing the trajectory data, the movement law, frequency, amplitude and other characteristics of the noise source can be revealed, providing a basis for subsequent classification and analysis. According to the preset trajectory classification standards and potential noise source location data, the noise source movement trajectory data can be separated into mobile source trajectory data and stationary source location data. This helps to classify and categorize the noise sources, making the subsequent analysis more accurate and targeted. The mobile source trajectory data provides the movement trajectory information of the mobile noise source, which can be used to analyze the movement law, path and speed and other characteristics of the mobile source. The stationary source location data provides the location information of the stationary noise source, which can be used to analyze the distribution and spatial characteristics of the stationary source. By merging the mobile source trajectory data and the stationary source location data, the complete moving source path data can be obtained. This can consider the location information of the mobile source and the stationary source at the same time, and comprehensively analyze the overall distribution and movement characteristics of the noise source. The moving source path data provides the movement path information of the noise source from a global perspective. This helps to understand the overall movement pattern, trend and evolution law of the noise source. At the same time, the moving source path data can be correlated with other environmental factors to reveal the formation mechanism and influencing factors of the noise source. In summary, these steps include noise source location identification, motion direction calculation, motion trajectory reconstruction, source point classification and separation, moving source and stationary source analysis, and global perspective analysis. These effects can more comprehensively and deeply understand and analyze the location, motion behavior, and dynamic characteristics of noise sources, providing valuable data and information support for research and application in related fields.

[0150] Preferably, the intrinsic scattering cloud image data is subjected to acoustic energy flow surrounding the center identification, and the noise source position is matched to obtain potential noise source position data, including:

[0151] According to a preset time step, the corresponding acoustic energy flow vector field is calculated for the intrinsic scattering cloud image data to obtain the acoustic energy flow vector field data;

[0152] The embodiment of the present invention loads the acquired intrinsic scattering cloud map data into a computing environment; determines a suitable time step according to the data acquisition frequency and actual needs; and uses an acoustic model or an acoustic finite element method to calculate and simulate the acoustic energy flow vector field according to the intrinsic scattering cloud map data.

[0153] Filtering the acoustic energy flow vector field data and extracting the effective energy transmission path to obtain filtered acoustic energy flow field data;

[0154] The embodiment of the present invention applies a filtering method to filter the acoustic energy flow vector field data to remove noise and smooth the data. A threshold or a specific energy transmission path extraction method is used to extract transmission paths with higher energy, which may correspond to the main transmission direction of the acoustic energy flow.

[0155] The curl tensor of the filtered acoustic energy flow field data is calculated, and the acoustic energy flow surrounding the central area is identified to obtain the potential noise source position estimation data;

[0156] The embodiment of the present invention uses a curl calculation method to calculate a curl tensor based on filtered acoustic energy flow field data to describe the rotation characteristics in the acoustic energy flow field. Image processing technology and object recognition algorithms are applied to identify the surrounding center area of ​​the acoustic energy flow in the filtered acoustic energy flow field, which may correspond to the location of a potential noise source.

[0157] A polar sound line tracking model is constructed based on the potential noise source position estimation data, and the sound path from each spatial point to the potential noise source position is calculated to obtain the sound path field data;

[0158] The embodiment of the present invention prepares the potential noise source position estimation data obtained by step S513 to ensure that the format and structure of the data are suitable for subsequent processing. Based on the acoustic propagation theory and physical parameters, a polar sound ray tracing model for simulating the sound energy propagation path is constructed. The model can take into account the sound velocity, density, attenuation and other characteristics of the medium. Using the polar sound ray tracing model, the acoustic path from each spatial point to the potential noise source position is calculated. The acoustic path represents the path length that a sound wave travels through in a medium.

[0159] The matching degree of the potential source position is evaluated based on the intrinsic scattering cloud map data and the sound path field sound path data to obtain the potential noise source position data.

[0160] The embodiment of the present invention loads the acquired intrinsic scattering cloud map data into the computing environment to ensure that the format and structure of the data are suitable for subsequent processing. Prepare the sound path field data calculated by step S514 to ensure that the format and structure of the data are suitable for subsequent processing. Use appropriate methods to compare the intrinsic scattering cloud map data and the sound path field sound path data to evaluate the degree of match between the spatial point and the potential noise source position. Common methods include correlation analysis, difference measurement, statistical methods, etc.

[0161] The present invention can obtain the propagation direction and energy flow of acoustic energy in space by calculating the acoustic energy flow vector field of the intrinsic scattering cloud map data. This helps to understand the propagation path and distribution characteristics of acoustic energy. The acoustic energy flow vector field data provides the flow information of acoustic energy in space, which can be used to analyze the propagation path, speed and aggregation area of ​​acoustic energy. This helps to identify the propagation range and impact area of ​​the noise source. By filtering the acoustic energy flow vector field data, noise and interference can be removed, and real and effective acoustic energy flow field information can be extracted. This helps to improve the quality and reliability of the data. The filtered acoustic energy flow field data can be used to extract the effective energy transmission path, that is, the main channel and path of acoustic energy propagation. This helps to determine the main direction and path of acoustic energy propagation, and provides a basis for subsequent analysis. By calculating the curl tensor of the filtered acoustic energy flow field data, the curl information of the acoustic energy flow field, that is, the rotation and vortex distribution of acoustic energy in space, can be obtained. This helps to reveal the center-surrounding characteristics and vortex structure of the acoustic energy flow field. By analyzing the curl information of the acoustic energy flow field, the center-surrounding area of ​​the acoustic energy flow can be identified, and the potential noise source position can be inferred. This helps to preliminarily determine the location of the noise source and provide a reference for subsequent location matching and evaluation. By constructing a polar sound line tracing model based on the potential noise source location estimation data, the propagation path of sound energy in space and the direction of the sound line can be simulated. This helps to understand the path and propagation characteristics of sound energy propagation. Using the polar sound line tracing model, the sound path from each spatial point to the potential noise source location, that is, the distance of sound propagation, can be calculated. This helps to evaluate the intensity and attenuation of sound propagation and provide a basis for subsequent location matching and evaluation. By comparing and analyzing the intrinsic scattering cloud map data and the sound path field sound path data, the matching degree of the potential noise source location can be evaluated. This helps to determine the exact location of the potential noise source and verify the validity of the inference results in the previous steps. Through the matching degree evaluation, the potential noise source location data can be obtained, that is, the specific location of the noise source in space can be determined. This helps to locate the noise source and provides a basis for subsequent noise control and management. In summary, the above steps include calculation of acoustic energy flow vector field, acoustic energy flow analysis, noise filtering, energy transmission path extraction, curl tensor calculation, noise source position estimation, polar sound line tracking model construction, sound path calculation, position matching evaluation and noise source position data acquisition, etc. These effects are helpful to understand the characteristics of acoustic energy propagation, determine the location of noise sources and evaluate the degree of position matching, providing an important basis for noise control and management.

[0162] Preferably, the modal contribution analysis of the dynamic source path data is performed according to the vibration mode of the substation equipment, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution diagram, including:

[0163] Perform vibration modal analysis on substation equipment information and design parameters to obtain equipment vibration modal data;

[0164] The embodiment of the present invention collects detailed information and design parameters of substation equipment, including equipment type, size, material properties, etc. Ensure the accuracy and integrity of the data. Select an appropriate vibration modal analysis method, such as finite element analysis (FEA); establish a finite element model of the equipment, including geometric shape, material properties, boundary conditions, etc.; perform vibration modal analysis to calculate the resonant frequency, vibration shape and vibration modal data of the equipment in different modes.

[0165] Match the equipment vibration mode with the source path data in the space and frequency domain to obtain a matching score matrix;

[0166] The embodiment of the present invention collects dynamic source path data related to the substation, such as sensor data, noise source location data, etc. The equipment vibration modal data is matched with the dynamic source path data. Methods such as correlation analysis and spectrum analysis can be used; in space, the positional relationship between the equipment vibration mode and the dynamic source path is compared to determine the degree of matching; in the frequency domain, the spectrum characteristics of the equipment vibration mode and the dynamic source path are compared to determine the degree of matching. Based on the spatial and frequency domain matching results, a matching score matrix is ​​constructed to record the degree of matching between the equipment vibration mode and the dynamic source path.

[0167] According to the matching score matrix, the main contributing vibration modes and their contribution coefficients of each noise source are determined to obtain the modal contribution vector data;

[0168] The embodiment of the present invention determines the main contributing vibration mode of each noise source according to the matching score matrix. The mode with the highest matching score can be selected as the main contributing vibration mode. The contribution coefficient of each noise source is calculated to reflect its contribution to the vibration mode of the device. The contribution coefficient can be calculated according to the value in the matching score matrix, such as normalization processing.

[0169] The radiation field of each noise source is separated into different modal components according to the modal contribution vector data, and the modal component data of the stationary source and the modal component data of the moving source are obtained;

[0170] The embodiment of the present invention separates the radiation field of each noise source into different modal components; the separated modal component data can be classified into stationary sources and mobile sources using appropriate separation methods, such as modal filtering, modal superposition, etc. The stationary source modal component data represents the radiation field distribution generated by the vibration of the device itself. The mobile source modal component data represents the radiation field distribution of external noise sources related to the vibration of the device.

[0171] Perform modal superposition on the modal component data of the stationary source and the modal component data of the mobile source, and synthesize the total radiation field distribution of each radiation source to obtain the total radiation field data of the stationary source and the total radiation field data of the mobile source;

[0172] The embodiment of the present invention sums up the modal component data of the stationary source, that is, adds the radiation field data of each modal component; this can be achieved by superimposing the amplitude and phase of each modal component. Usually, the amplitude can be added directly, while the phase needs to be reasonably adjusted; for each modal component, the adjustment of the amplitude and phase can be determined based on the contribution coefficient obtained from the modal contribution vector data. The contribution coefficient reflects the contribution of each modal component to the total radiation field. The modal component data of the mobile source are summed, that is, the radiation field data of each modal component is also added; similar to the modal superposition of the stationary source, it can be achieved by superimposing the amplitude and phase of each modal component; the adjustment of the amplitude and phase can also be determined based on the contribution coefficient obtained from the modal contribution vector data.

[0173] According to the geometric topological relationship matrix, the total radiation field data of the stationary source and the total radiation field data of the mobile source are projected into a three-dimensional substation environment to obtain the separated noise source distribution map.

[0174] According to the geometric body topological relationship matrix, the embodiment of the present invention performs a three-dimensional substation environment projection on the total radiation field data of the stationary source and the total radiation field data of the mobile source to obtain a noise source distribution map after separation. Prepare a matrix that describes the topological relationship between the various geometric bodies of the substation; the rows and columns of the matrix represent different geometric bodies, respectively, and the matrix elements represent the connection relationship between the corresponding geometric bodies (such as adjacent, connected, etc.). Three-dimensional environmental projection of the total radiation field data of the stationary source: determine the position and distribution of the total radiation field data of the stationary source in the substation environment according to the geometric body topological relationship matrix; project the total radiation field data of the stationary source into the three-dimensional space of the substation, considering the positional relationship and boundary conditions between the geometric bodies; use three-dimensional visualization software or programming tools to correspond and project the total radiation field data of the stationary source with the geometric model of the substation to generate a noise source distribution map. Three-dimensional environmental projection of the total radiation field data of mobile sources: determine the position and distribution of the total radiation field data of mobile sources in the substation environment based on the topological relationship matrix of geometric bodies; project the total radiation field data of mobile sources into the three-dimensional space of the substation, considering the positional relationship and boundary conditions between geometric bodies; similarly, three-dimensional visualization software or programming tools can be used to correspond and project the total radiation field data of mobile sources with the geometric model of the substation to generate a noise source distribution map.

[0175] By analyzing the vibration modes of substation equipment, the present invention can understand the vibration characteristics and response frequencies of the equipment under different vibration modes. This helps to determine the natural vibration frequency and modal morphology of the equipment, and provides a basis for the subsequent matching and separation process. By matching the vibration mode of the equipment with the dynamic source path data, the correlation and matching degree between each vibration mode and the corresponding dynamic source path can be determined. This helps to determine the correlation between the noise source and the vibration mode of the equipment, and provides a basis for the subsequent modal contribution calculation. By analyzing the matching score matrix, the main contribution mode of each noise source to the vibration mode of the equipment and the corresponding contribution coefficient can be determined. This helps to understand the degree of influence of each noise source on the vibration of the equipment, and provides a basis for the subsequent modal separation and radiation field analysis. By separating the radiation field of the noise source into different modal components according to the modal contribution vector data, the radiation characteristics of the noise can be associated with the vibration mode of the equipment. This helps to understand the contribution of different modes to the radiation field, and provides a basis for the subsequent modal superposition and total radiation field analysis. By superimposing the modal component data of each stationary source and the modal component data of each mobile source, the total radiation field distribution of each radiation source can be synthesized. This helps to comprehensively consider the contribution of different modes to the radiation field and obtain more accurate total radiation field data. By projecting the total radiation field data of the stationary source and the total radiation field data of the mobile source into a three-dimensional substation environment according to the geometric topological relationship matrix, the position and distribution of the separated noise sources in the substation environment can be visualized. This helps to intuitively understand the distribution of noise sources and provide guidance for further noise control and optimization. In summary, through the execution of the above steps, the substation noise sources can be separated and analyzed, and the equipment vibration modal data, modal contribution vector data, modal component data of stationary sources and mobile sources, total radiation field data, and noise source distribution maps can be obtained. These results are of great significance for the identification, positioning and control of noise sources, and help to improve the noise problems in the substation environment and improve the reliability and operation efficiency of equipment.

[0176] The present invention also provides a low-frequency noise source positioning device, comprising:

[0177] The three-dimensional environment modeling module is used to obtain the three-dimensional environment model of each structure in the substation, obtain the geometric body topological relationship matrix in the three-dimensional environment model, perform on-site measurement of the substation, and obtain noise measurement data; based on the measurement position, the noise measurement data is divided into blocks corresponding to the three-dimensional environment model, and registered to obtain local noise image data; this module includes the following submodules:

[0178] A three-dimensional modeling acquisition module is used to obtain three-dimensional environmental model data of each structure in the substation;

[0179] A geometry extraction module is used to extract and represent the geometry of the three-dimensional environment model, and the geometry represents the model data;

[0180] A topological relationship extraction module is used to extract the topological relationship and boundary conditions of the geometric bodies in the model according to the three-dimensional environment model data and the geometric body representation model data, and obtain the geometric body topological relationship matrix;

[0181] The field measurement module is used to conduct field measurement of the substation to obtain noise measurement data, including noise decibel data and measurement point location data;

[0182] A wavelet filtering module is used to perform wavelet transform filtering on the noise decibel data to obtain noise image data;

[0183] A data registration module is used to correspond the noise image data to the three-dimensional environment model in blocks according to the measurement point position data to obtain block noise data;

[0184] An image registration module is used to perform image registration according to the block noise data and the model units in the three-dimensional environment model to obtain local noise image data;

[0185] The noise measurement module is used to perform outer convolution fusion on the noise images between adjacent units according to the geometric topological relationship matrix and the local noise image data, and iteratively propagate them to the entire substation area to generate the initial sound source distribution estimation data of the entire substation; this module includes the following submodules:

[0186] The spatial discretization module is used to discretize the substation space into small spatial unit grids according to the geometric topological relationship matrix to obtain spatial grid unit data;

[0187] A noise assignment module is used to assign local noise image data to the spatial unit of the spatial grid unit data according to the measurement point position data to obtain initial grid noise distribution data;

[0188] Convolution fusion module, which is used to perform outer convolution of the initial grid noise distribution data and the geometric body topological relationship matrix on the noise of adjacent grid cells based on boundary conditions to obtain fusion result data;

[0189] The noise propagation module is used to use the fusion result data as the initial value and repeat the outer convolution operation based on the boundary conditions until it propagates to the entire substation area to obtain the grid noise distribution data;

[0190] An interpolation and denoising module is used to interpolate and denoise the grid noise distribution data to obtain the initial sound source distribution estimation data;

[0191] The noise distribution estimation module is used to separate the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data; this module includes the following submodules:

[0192] A parameterized modeling module is used to perform parameterized modeling on the initial sound source distribution estimation data using a generalized coordinated filtering algorithm, and divide the model into two parts: a direct sound component and a multipath sound component, to obtain a direct sound model and a multipath sound model;

[0193] An optimization solution module is used to construct a joint optimization objective function based on the direct sound model and the multipath sound model, weigh the two models through a coordination factor, and solve the objective function using a generalized least squares method to obtain separated direct sound component data and multipath sound component data;

[0194] The path tracking module is used to construct a random state space model of the multipath echo path based on the multipath acoustic component data, and filter and track each path based on the Kalman particle filter framework, and perform statistics on the delay, attenuation and direction of the path to obtain preliminary multipath path parameter data;

[0195] A noise sparse representation module is used to sparsely represent multipath acoustic component data on a wavelet basis to obtain a noise sparse representation model;

[0196] An adaptive modeling module is used to adaptively model the multipath noise according to the noise sparse representation model and the preliminary multipath path parameter data, and to subtract the multipath noise component from the direct sound component data to obtain the multipath noise-removed field data;

[0197] A residual noise manifold module is used to construct a residual noise manifold for the residual noise of the multipath denoised field data using a manifold learning technique, and to map the multipath denoised field data onto the noise manifold to obtain reconstructed residual multipath noise component data;

[0198] A noise decoupling module is used to subtract the reconstructed residual multipath noise component data from the multipath de-noised field data to obtain echo decoupling data;

[0199] The echo decoupling module is used to perform dimensional compression on the echo decoupling data and extract the eigenmodes of the low-order scattering field to obtain the eigenmode set of the scattering field; perform parameter estimation on the eigenmode set of the scattering field to obtain the intrinsic scattering cloud map data; this module includes the following submodules:

[0200] A core tensor representation module is used to discretely sample the echo decoupling data in the time and space domains and arrange them into a core tensor form to obtain core tensor representation data;

[0201] A low-rank decomposition module is used to perform low-rank decomposition approximation on the core tensor representation data using supersymmetric matrix decomposition technology to obtain low-dimensional embedded vector data;

[0202] A mode reconstruction module is used to map the low-dimensional embedded vector data onto the differential manifold and reconstruct the eigenmode of the scattering field to obtain the eigenmode set of the scattering field;

[0203] A parameter estimation module is used to estimate the parameters of the field mode of the scattering field eigenmode set through a nonlinear regression model to obtain eigenscattering cloud map data;

[0204] The dynamic source path analysis module is used to identify the center of the acoustic energy flow around the intrinsic scattering cloud map data, and correspond the noise source position to obtain the potential noise source position data; according to the potential noise source position data, the instantaneous motion direction of the source point is locked, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; according to the noise source motion trajectory data and the potential noise source position data, the position information of the stationary source and the mobile source is distinguished and integrated to obtain the dynamic source path data; this module includes the following submodules:

[0205] The acoustic energy flow analysis module is used to identify the center of the acoustic energy flow around the intrinsic scattering cloud image data, and to correspond the position of the noise source to obtain the potential noise source position data;

[0206] The motion direction locking module is used to calculate the acoustic energy flow gradient vector based on the source position according to the potential noise source position data and the filtered acoustic energy flow field data, and obtain the instantaneous motion direction data of the source point;

[0207] The motion trajectory reconstruction module is used to reconstruct the motion trajectory of the source point based on the time series analysis of the instantaneous motion direction data of the source point to obtain the motion trajectory data of the noise source;

[0208] The static and dynamic source separation module is used to separate the static source and the mobile source of the noise source motion trajectory data according to the preset trajectory classification standard and the potential noise source position data, and obtain the mobile source trajectory data and the static source position data;

[0209] A path integration module is used to merge the moving source trajectory data and the stationary source position data to obtain the moving source path data;

[0210] The modal contribution analysis module is used to perform modal contribution analysis on the dynamic source path data according to the vibration mode of the substation equipment, separate the radiation modes of the stationary and moving noise sources, project them into the scalar sound pressure field, and obtain the separated noise source distribution map; this module includes the following submodules:

[0211] The vibration modal analysis module is used to perform vibration modal analysis on the equipment information and design parameters of the substation to obtain equipment vibration modal data;

[0212] The modal matching module is used to match the equipment vibration mode with the source path data in the space and frequency domain to obtain a matching score matrix;

[0213] A contribution calculation module is used to determine the main contribution vibration mode and its contribution coefficient of each noise source according to the matching score matrix, and obtain the modal contribution vector data;

[0214] A modal separation module is used to separate the radiation field of each noise source into different modal components according to the modal contribution vector data, and obtain the modal component data of the stationary source and the modal component data of the moving source;

[0215] The radiation field synthesis module is used to perform modal superposition on the modal component data of the stationary source and the modal component data of the mobile source, and synthesize the total radiation field distribution of each radiation source to obtain the total radiation field data of the stationary source and the total radiation field data of the mobile source;

[0216] The three-dimensional projection module is used to perform three-dimensional substation environment projection on the total radiation field data of the stationary source and the total radiation field data of the mobile source according to the geometric topological relationship matrix to obtain the separated noise source distribution map.

[0217] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0218] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for locating a low-frequency noise source in a substation, characterized in that: include: Acquire a three-dimensional environment model of each structure in the substation, and obtain a topological relationship matrix of geometric bodies in the three-dimensional environment model; Conduct on-site measurements of the substation to obtain noise measurement data; match the noise measurement data to the three-dimensional environment model based on the measurement location, and perform registration to obtain local noise image data; According to the geometric body topological relationship matrix and the local noise image data, the noise images between adjacent units are fused by outer convolution, and iteratively propagated to the entire substation area to generate initial sound source distribution estimation data of the entire substation; Separating the direct sound signal and the multipath echo components according to the initial sound source distribution estimation data to generate echo decoupling data; The echo decoupling data is dimensional compressed and the eigenmode of the low-order scattered field is extracted to obtain the eigenmode set of the scattered field; Estimating the parameters of the field mode for the eigenmode set of the scattering field to obtain the intrinsic scattering cloud map data; The acoustic energy flow around the center is identified on the intrinsic scattering cloud image data, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; According to the noise source motion trajectory data and potential noise source location data, the location information of the stationary source and the moving source are distinguished and integrated to obtain the moving source path data; The modal contribution analysis of the dynamic source path data is performed according to the vibration modes of the substation equipment, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution map.

2. The method for locating a low-frequency noise source in a substation according to claim 1, characterized in that: Acquire a three-dimensional environmental model of each structure in the substation, and obtain a topological relationship matrix of geometric bodies in the three-dimensional environmental model; perform on-site measurement of the substation to obtain noise measurement data; Based on the measurement position, the noise measurement data is divided into blocks and mapped to the 3D environment model, and then registered to obtain local noise image data, including: Obtain the 3D environmental model data of each structure in the substation; Extract and represent the geometrical body of the three-dimensional environment model, and the geometrical body represents the model data; Extract the topological relationship and boundary conditions of the geometric bodies in the model according to the three-dimensional environment model data and the geometric body representation model data, and obtain the geometric body topological relationship matrix; Conduct on-site measurements of substations to obtain noise measurement data, including noise decibel data and measurement point location data; Perform wavelet transform filtering on the noise decibel data to obtain noise image data; According to the measurement point position data, the noise image data is divided into blocks and matched to the three-dimensional environment model to obtain the block noise data; Image registration is performed based on the block noise data and the model units in the three-dimensional environment model to obtain local noise image data.

3. The method for locating a low-frequency noise source in a substation according to claim 2, characterized in that: The topological relationship and boundary conditions of the geometric bodies in the model are extracted according to the 3D environment model data and the geometric body representation model data, and the geometric body topological relationship matrix is ​​obtained, including: Performing basic shape type recognition on the geometric body representation model data to obtain geometric body type data; Extract the acoustic impedance and absorption coefficient based on the material information according to the three-dimensional environment model data and the geometric representation model data to obtain the geometric boundary condition data; Performing topological relationship analysis of adjacency, inclusion, separation and intersection between geometric bodies according to geometric body representation model data and geometric body type data to obtain geometric body topological relationship data; Based on the geometric body topological relationship data and the geometric body boundary condition data, the topological relationship and boundary conditions are integrated into a matrix form to obtain a geometric body topological relationship matrix.

4. The method for locating a low-frequency noise source in a substation according to claim 3, characterized in that: The noise images between adjacent units are fused by outer convolution according to the geometric topological relationship matrix and the local noise image data, and iteratively propagated to the entire substation area to generate the initial sound source distribution estimation data of the entire substation, including: According to the geometric topological relationship matrix, the substation space is discretized into small space unit grids to obtain space grid unit data; Assigning local noise image data to the spatial unit of spatial grid unit data according to the measurement point position data to obtain initial grid noise distribution data; The initial grid noise distribution data and the geometric body topological relationship matrix perform outer convolution on the noise of adjacent grid cells based on boundary conditions to obtain the fusion result data; The fusion result data is used as the initial value, and the outer convolution operation based on the boundary conditions is repeated until it propagates to the entire substation area to obtain the grid noise distribution data; The grid noise distribution data is interpolated and denoised to obtain the initial sound source distribution estimation data.

5. The method for locating a low-frequency noise source in a substation according to claim 4, characterized in that: The direct sound signal and multipath echo components are separated according to the initial sound source distribution estimation data to generate echo decoupling data, including: The generalized coordinated filtering algorithm is used to perform parameterized modeling on the initial sound source distribution estimation data, and the model is divided into two parts: the direct sound component and the multipath sound component, thus obtaining the direct sound model and the multipath sound model. A joint optimization objective function is constructed based on the direct sound model and the multipath sound model, the two models are weighed by the coordination factor, and the objective function is solved using the generalized least squares method to obtain the separated direct sound component data and multipath sound component data; A random state space model of the multipath echo path is constructed based on the multipath acoustic component data, and each path is filtered and tracked based on the Kalman particle filter framework. The path delay, attenuation and direction are statistically analyzed to obtain preliminary multipath path parameter data. The multipath acoustic component data is sparsely represented on the wavelet basis to obtain a noise sparse representation model; Adaptively model the multipath noise according to the noise sparse representation model and preliminary multipath path parameter data, and subtract the multipath noise component from the direct sound component data to obtain the multipath noise-removed field data; The residual noise manifold is constructed for the residual noise of the multipath denoised field data using manifold learning technology, and the multipath denoised field data is mapped onto the noise manifold to obtain the reconstructed residual multipath noise component data; The reconstructed residual multipath noise component data is subtracted from the multipath de-noised field data to obtain the echo decoupling data.

6. The method for locating a low-frequency noise source in a substation according to claim 5, characterized in that: The echo decoupling data is dimensional compressed and the eigenmode of the low-order scattered field is extracted to obtain the eigenmode set of the scattered field; The parameters of the field pattern are estimated for the eigenmode set of the scattered field to obtain the intrinsic scattering cloud map data, including: The echo decoupling data is discretely sampled in the time and space domains and arranged in a nuclear tensor form to obtain nuclear tensor representation data; The supersymmetric matrix decomposition technology is used to perform low-rank decomposition approximation on the core tensor representation data to obtain low-dimensional embedded vector data; Mapping the low-dimensional embedded vector data onto the differential manifold and reconstructing the eigenmode of the scattering field to obtain the eigenmode set of the scattering field; The parameters of the field pattern of the eigenmode set of the scattering field are estimated through a nonlinear regression model to obtain the intrinsic scattering cloud map data.

7. The method for locating a low-frequency noise source in a substation according to claim 6, characterized in that: The parameters of the field pattern are estimated for the eigenmode set of the scattering field through the nonlinear regression model to obtain the intrinsic scattering cloud map data, including: The nonlinear scattering model of eigenmode theory is established according to the eigenmode set of the scattered field and the topological relationship matrix of the geometric body; The free parameters of wave number, attenuation rate and phase in the nonlinear scattering model are identified to obtain a free parameter set; According to the preset measured pattern data, the scattering field intrinsic mode set and the nonlinear scattering model difference are constructed as an optimization objective function to obtain a nonlinear regression objective function; Solve the model parameters of the nonlinear regression objective function according to the free parameter set to obtain the free parameter estimation value data; Substituting the free parameter estimation data into the nonlinear scattering model, the intrinsic scattering cloud map data is generated.

8. The method for locating a low-frequency noise source in a substation according to claim 7, characterized in that: The acoustic energy flow around the center is identified on the intrinsic scattering cloud image data, and the noise source position is matched to obtain the potential noise source position data; the instantaneous motion direction of the source point is locked according to the potential noise source position data, and the motion trajectory of the source point is reconstructed to obtain the noise source motion trajectory data; According to the noise source motion trajectory data and potential noise source location data, the location information of the stationary source and the moving source is distinguished and integrated to obtain the moving source path data, including: The acoustic energy flow around the center is identified on the intrinsic scattering cloud image data, and the noise source position is matched to obtain the potential noise source position data; According to the potential noise source position data and the filtered acoustic energy flow field data, the acoustic energy flow gradient vector is calculated based on the source position to obtain the instantaneous motion direction data of the source point; The instantaneous motion direction data of the source point is reconstructed based on the time series analysis of the source point motion trajectory to obtain the noise source motion trajectory data; According to the preset trajectory classification standard and the potential noise source position data, the noise source motion trajectory data is separated into stationary source and mobile source to obtain mobile source trajectory data and stationary source position data; The moving source trajectory data and the stationary source position data are merged to obtain the moving source path data.

9. The method for locating a low-frequency noise source in a substation according to claim 8, characterized in that: The acoustic energy flow around the center is identified on the intrinsic scattering cloud image data, and the noise source position is matched to obtain the potential noise source position data, including: According to a preset time step, the corresponding acoustic energy flow vector field is calculated for the intrinsic scattering cloud image data to obtain the acoustic energy flow vector field data; Filtering the acoustic energy flow vector field data and extracting the effective energy transmission path to obtain filtered acoustic energy flow field data; The curl tensor of the filtered acoustic energy flow field data is calculated, and the acoustic energy flow surrounding the central area is identified to obtain the potential noise source position estimation data; A polar sound line tracking model is constructed based on the potential noise source position estimation data, and the sound path from each spatial point to the potential noise source position is calculated to obtain the sound path field data; The matching degree of the potential source position is evaluated based on the intrinsic scattering cloud map data and the sound path field sound path data to obtain the potential noise source position data.

10. The method for locating a low-frequency noise source in a substation according to claim 9, characterized in that: According to the vibration modes of the substation equipment, the modal contribution analysis of the dynamic source path data is performed, and the radiation modes of the stationary and moving noise sources are separated and projected into the scalar sound pressure field to obtain the separated noise source distribution map, including: Perform vibration modal analysis on substation equipment information and design parameters to obtain equipment vibration modal data; Match the equipment vibration mode with the source path data in the space and frequency domain to obtain a matching score matrix; According to the matching score matrix, the main contributing vibration modes and their contribution coefficients of each noise source are determined to obtain the modal contribution vector data; The radiation field of each noise source is separated into different modal components according to the modal contribution vector data, and the modal component data of the stationary source and the modal component data of the moving source are obtained; Perform modal superposition on the modal component data of the stationary source and the modal component data of the mobile source, and synthesize the total radiation field distribution of each radiation source to obtain the total radiation field data of the stationary source and the total radiation field data of the mobile source; According to the geometric topological relationship matrix, the total radiation field data of the stationary source and the total radiation field data of the mobile source are projected into a three-dimensional substation environment to obtain the separated noise source distribution map.

Images