Insulator pollution degree detection method and system based on multi-source spectrum sensing

Abstract

The invention relates to the technical field of multi-source spectrum sensing, and provides an insulator pollution degree detection method and system based on multi-source spectrum sensing, and the method comprises the steps: obtaining the spectral image data of an insulator at a plurality of focal lengths and the scattered spectrum data of the surface of the insulator; extracting dirty layer thickness information from the spectral image data; the filthy layer thickness information and the scattering spectrum data are fused, a fusion result is input into a graph neural network, the spatial incidence relation between position units is determined through the graph neural network, and the position units correspond to spatial partition areas of the insulator; and according to the spatial association relationship, generating three-dimensional pollution distribution data of the insulator, so as to solve the problems of low spatial resolution and low accuracy of insulator pollution detection in the prior art, and improve the spatial resolution and accuracy of insulator pollution detection.

Description

Technical Field

[0001] The present application relates to the technical field of multi-source spectral sensing, and in particular to a method and system for detecting the contamination level of an insulator based on multi-source spectral sensing. Background Art

[0002] High-voltage transmission line insulators operate in complex environments for extended periods, and contamination accumulates on their surfaces, leading to degradation of insulation performance and even flashover accidents. Traditional inspection methods struggle to accurately measure the uneven surface contamination accumulation on complex insulator configurations, such as flashover sheds. A high-precision inspection technology is urgently needed that combines contamination thickness distribution with compositional characteristics to enable three-dimensional visual assessment of contamination levels.

[0003] At present, a method for detecting insulator contamination based on single hyperspectral imaging is widely used. This method uses a fixed-focal-length hyperspectral camera to collect insulator surface reflectance spectral data, uses spectral characteristics to invert the contamination components, and combines image segmentation technology to estimate the contamination coverage area.

[0004] This method relies on spectral data from a single perspective, making it difficult to accurately reflect the three-dimensional thickness distribution of the contamination layer. At the same time, it does not integrate the scattering characteristic data of the contamination, resulting in insufficient ability to distinguish between surface and subsurface contamination. In addition, the lack of modeling of the complex physical structure of the insulator causes the analysis of the contamination diffusion path to deviate from the actual operating conditions. Summary of the Invention

[0005] The present application provides a method and system for detecting the degree of insulator contamination based on multi-source spectral sensing, which is used to solve the problems of low spatial resolution and low accuracy in insulator contamination detection in the prior art.

[0006] In a first aspect, the present application provides a method for detecting the degree of insulator contamination based on multi-source spectral sensing, comprising: Acquire spectral image data of the insulator at multiple focal lengths and scattered spectrum data of the insulator surface; Extracting contamination layer thickness information from the spectral image data; fusing the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial association relationship between position units through the graph neural network, wherein the position units correspond to spatial partition areas of the insulator; Based on the spatial correlation, three-dimensional pollution distribution data of the insulator is generated.

[0007] Optionally, fusing the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial correlation relationship between position units through the graph neural network includes: Aligning the contamination layer thickness information with the scattering spectrum data according to position units to form a joint feature vector for each position unit; Inputting the joint feature vector into a graph neural network, and calculating the target spatial connection weights between adjacent position units through the graph neural network; The spatial association relationship between the location units is determined according to the target spatial connection weight.

[0008] Optionally, calculating the target spatial connection weights between adjacent location units by the graph neural network includes: Identify the physical connection topology between location units from the physical spatial structure of the insulator; Extracting the pollution thickness similarity and spectral feature similarity between the adjacent position units from the joint feature vector to form a two-dimensional similarity index; fusing the physical connection topology structure and the two-dimensional similarity index into a spatial constraint factor; The spatial constraint factor is coupled with the joint feature vector to generate target spatial connection weights between adjacent location units.

[0009] Optionally, coupling the spatial constraint factor with the joint feature vector to generate a target spatial connection weight between adjacent location units includes: Converting the spatial constraint factor into an edge constraint condition of a graph neural network; injecting the physical configuration characteristics of the spatial constraint factor into the joint feature vector to generate an enhanced feature vector; Performing constraint-guided processing on the enhanced feature vector and the edge constraint condition through the graph neural network, and outputting an initial spatial connection weight value; According to the pollution diffusion characteristics of the insulator, the initial spatial connection weight value is optimized and adjusted to generate a target spatial connection weight.

[0010] Optionally, extracting the contamination layer thickness information from the spectral image data includes: Calculating an optical depth parameter of the contamination layer according to a spectral focus change difference of the spectral image data; Based on the optical depth parameter and in combination with a preset law of image definition change, the contamination layer thickness information is generated.

[0011] Optionally, generating the contamination layer thickness information based on the optical depth parameter and in combination with a preset image clarity variation rule includes: Matching and mapping the optical depth parameter with the image clarity variation rule; According to the matching mapping results, the quantitative value of the contamination layer thickness is calculated by linear interpolation method; Aggregate the quantized values of the pollution layer thickness of all position units to form the pollution layer thickness information.

[0012] Optionally, generating three-dimensional pollution distribution data of insulators according to the spatial correlation relationship includes: Constructing a pollution transfer path between location units based on the spatial association relationship; Performing thickness diffusion calculation on the contamination layer thickness information according to the contamination transmission path; The calculation results are mapped to the preset three-dimensional insulator model to generate three-dimensional pollution distribution data.

[0013] In a second aspect, the present application provides an insulator contamination degree detection system based on multi-source spectral sensing, comprising: An acquisition module is used to acquire spectral image data of the insulator at multiple focal lengths and scattered spectrum data of the insulator surface; An extraction module, configured to extract contamination layer thickness information from the spectral image data; a fusion module, configured to fuse the contamination layer thickness information and the scattering spectrum data, input the fusion result into a graph neural network, and determine the spatial association relationship between position units through the graph neural network, wherein the position units correspond to spatial partitioned areas of the insulator; A generating module is used to generate three-dimensional pollution distribution data of the insulator according to the spatial correlation relationship.

[0014] In a third aspect, the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any of the methods for detecting the degree of insulator contamination based on multi-source spectral perception described in the first aspect.

[0015] In a fourth aspect, the present application provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement an insulator contamination degree detection method based on multi-source spectral perception as described in any one of the first aspects.

[0016] The present application provides a method for detecting the contamination level of an insulator based on multi-source spectral perception, the method comprising: obtaining spectral image data of the insulator at multiple focal lengths and scattered spectrum data of the insulator surface; extracting contamination layer thickness information from the spectral image data; fusing the contamination layer thickness information and the scattered spectrum data, inputting the fusion result into a graph neural network, determining the spatial correlation relationship between position units through the graph neural network, wherein the position units correspond to spatial partitioned areas of the insulator; and generating three-dimensional contamination distribution data of the insulator based on the spatial correlation relationship.

[0017] The technical solution provided by this application has the following beneficial effects: This application uses multi-focal length spectral imaging and simultaneous acquisition of scattering spectra to achieve complementary acquisition of contamination layer depth information and surface / subsurface composition information, providing a multi-source data basis for three-dimensional contamination analysis. Based on the focus change differences of multi-focal length spectra, the optical thickness of the contamination layer is quantified to solve the thickness measurement problem of uneven contamination on the surface of insulators with complex configurations. Through the collaborative analysis of thickness and spectral characteristics, the comprehensive characterization capability of the physical properties and chemical composition of contamination is enhanced, providing multi-dimensional data support for spatial correlation modeling. Graph neural networks are used to capture the diffusion law of contamination between spatial partition units of insulators, and a contamination distribution correlation model that conforms to actual physical processes is established. Combining the contamination thickness and spatial diffusion path, the three-dimensional contamination distribution morphology is reconstructed to achieve a visual assessment of the degree of contamination.

[0018] Furthermore, the present application also aligns the pollution layer thickness information and the scattering spectrum data according to the position unit to form a joint feature vector, inputs the graph neural network to calculate the target spatial connection weights between adjacent units, and determines the spatial correlation relationship of pollution diffusion based on the weights.

[0019] In addition, through multi-source data fusion and spatial modeling of graph neural networks, the diffusion path of contamination on the complex surface of insulators can be accurately depicted, improving the physical rationality and accuracy of the three-dimensional distribution analysis of contamination.

[0020] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1A flowchart of a method for detecting the degree of insulator contamination based on multi-source spectral sensing provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of an insulator contamination degree detection system based on multi-source spectral sensing provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0024] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0025] Existing methods for insulator contamination detection based on single hyperspectral imaging have limitations: they rely on spectral data with a fixed focal length, resulting in insufficient analysis of the three-dimensional thickness distribution of the contamination layer. Furthermore, they lack the ability to collaboratively analyze surface and subsurface scattering spectra, making it difficult to distinguish the depth characteristics of contamination components. Furthermore, they fail to consider the impact of the complex insulator configuration on contamination diffusion paths, resulting in systematic deviations between detection results and actual operating conditions. These shortcomings stem from the dual constraints of a single data dimension and a lack of physical modeling. A new detection method is urgently needed that integrates multi-source optical features with spatial topological correlations.

[0026] In response to the above problems, this application proposes a method for detecting the degree of insulator contamination based on multi-source spectral perception. Its innovation lies in the simultaneous acquisition of multi-focal length spectral imaging and scattered spectrum, combined with spatial correlation modeling of graph neural networks, to achieve three-dimensional reconstruction of the contamination distribution. Specifically, the thickness of the contamination layer is obtained by extracting the depth information of spectral images with different focal lengths, the surface / subsurface scattered spectrum data are fused to form a joint feature vector, and the graph neural network is used to analyze the contamination diffusion correlation between the spatial partition units of the insulator, and finally generate three-dimensional contamination distribution data reflecting the actual physical process. This method breaks through the dimensional limitations of single spectral detection. Through the collaborative calculation of multi-source optical data and topological structure, it not only solves the problem of accurate measurement of contamination thickness and composition, but also corrects the diffusion path deviation caused by complex configuration, providing a high-precision three-dimensional visualization solution for the assessment of insulator contamination status.

[0027] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0028] Figure 1 A flowchart of a method for detecting the degree of insulator contamination based on multi-source spectral sensing is provided in an embodiment of the present application, such as Figure 1 As shown, the method includes: Step 101: Acquire spectral image data of an insulator at multiple focal lengths and scattering spectrum data of the insulator surface.

[0029] In step 101, spectral image data represents the insulator surface reflectance spectrum information collected by a hyperspectral camera at different focal lengths. This data includes the correspondence between wavelength and reflection intensity and is used to analyze the optical properties of the contamination layer. Scattering spectrum data represents the scattered light signals from the insulator surface and subsurface, acquired by a laser scatterometer. This data reflects the size, composition, and distribution characteristics of contamination particles and includes both surface and subsurface scattering spectrum data.

[0030] In this embodiment, a hyperspectral camera is moved along the optical axis at preset focal length intervals to capture multiple sets of spectral images of the insulator surface at different focus states. Simultaneously, a laser scatterometer illuminates the insulator surface at a fixed angle, receives scattered light signals, and converts them into spectral data. The multi-focal spectral images are used for subsequent depth information extraction, while the scattered light spectral data is used for contamination composition analysis. Together, these two forms the foundation of multi-source optical inspection.

[0031] For example, during the inspection of FXBW4-110 insulators at substation A, a hyperspectral camera with a focal length step of 0.2 mm was used to capture 20 spectral images, while a 532 nm laser scatterometer was used to obtain scattered spectra. The focal length interval was calculated using the depth of field formula Δz = 2λ / (NA)^2, where λ = 500 nm is the center wavelength of visible light and NA = 0.25 is the numerical aperture of the lens. This ensures that adjacent images cover a continuous depth range.

[0032] Step 102: extracting the contamination layer thickness information from the spectral image data.

[0033] In step 102, the contamination layer thickness information represents the physical thickness of contamination deposited on the insulator surface and is obtained by inverting the focus difference of the multi-focal length spectral image.

[0034] In an embodiment of the present application, a clarity analysis is performed on a multi-focal length spectral image sequence, the image gradient value of each position unit at different focal lengths is calculated, and a clarity-focal length variation curve is constructed; the optical depth parameter is calculated based on the focal length difference corresponding to the peak of the curve, and combined with a preset clarity-thickness mapping relationship, it is converted into the actual thickness data of the contamination layer.

[0035] For example, for the B line insulator, the spectral image gradient value Σ|G(x,y)| / n of the shed edge position unit is extracted, where G(x,y) is the pixel gradient and n is the total number of pixels. According to the clarity threshold calibrated in the laboratory, when the gradient value drops to 50% of the peak value, the corresponding focal length difference is determined to be 0.3mm. The thickness T=0.2mm is calculated using the interpolation formula T=k·Δf (k=0.67 is the calibration coefficient).

[0036] Step 103: Fusing the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial correlation relationship between position units through the graph neural network, wherein the position units correspond to the spatial partition areas of the insulator.

[0037] In step 103, a graph neural network (GNN) is a deep learning model used to process topologically structured data. Its nodes represent the spatially partitioned units of the insulator, and its edges represent the pollution diffusion correlation between units. By aggregating the feature information of adjacent nodes, spatial connection weights are calculated to model the diffusion patterns of pollution on the insulator surface. Spatial correlation refers to the degree of mutual influence of pollution diffusion between units at different locations on the insulator surface. The connection weights calculated by the GNN quantify this relationship, with larger weights indicating a higher probability of pollution diffusion along that path. Spatially partitioned regions refer to the number of detection units divided into based on the insulator's physical structural features (such as shed spacing and curvature). These units are obtained through 3D scanning or discretization of CAD models. Each unit corresponds to a node in the GNN, which is used for local pollution feature extraction and correlation analysis.

[0038] In an embodiment of the present application, the thickness information and the scattering spectrum are aligned by coordinates and combined into a feature vector containing thickness and dual-wavelength scattering rate; after inputting into the graph neural network, the network calculates the connection weight based on the node feature similarity and physical proximity, where the node represents the insulator partition unit and the edge weight represents the pollution diffusion correlation strength; finally, the spatial correlation matrix of the entire graph is output.

[0039] For example, for the shed partitioning of the insulators in power station C, the thickness of 0.18 mm and the scattering rate of 0.35 / 0.28 (532 nm / 1064 nm) are combined into a feature vector; the graph neural network adopts a three-layer structure and calculates the connection weight through the weight formula W=exp(-d^2 / 2σ^2)*S, where d=8 mm is the unit spacing, σ=5 mm is the attenuation coefficient, and S=0.82 is the feature similarity.

[0040] Step 104: Generate three-dimensional pollution distribution data of the insulator based on the spatial correlation relationship.

[0041] In step 104, the three-dimensional pollution distribution data represents a quantitative result reflecting the accumulation form and thickness variation of pollution in the three-dimensional space on the insulator surface.

[0042] In an embodiment of the present application, a pollution diffusion path network is constructed based on spatial connection weights, and thickness values are superimposed along strongly correlated paths (weight > 0.7) to generate a three-dimensional contour distribution map; spatial mapping is performed in combination with the insulator CAD model to achieve visualization of the pollution distribution.

[0043] For example, for the D line insulator, 12 main diffusion paths are extracted based on the weight threshold of 0.7, and the thickness of the downstream unit is recursively calculated along the path according to the thickness attenuation coefficient of 0.9. Finally, a three-dimensional contour distribution map with an interval of 0.1 mm is generated, where the thickness of the red area is ≥0.3 mm.

[0044] This method accurately extracts the contamination thickness and composition characteristics through the coordinated acquisition of multi-focal spectra and scattered spectra; uses graph neural networks to model the spatial correlation of contamination diffusion, and combines physical configuration constraints to generate three-dimensional distribution data. This achieves high-precision visual detection of uneven contamination on insulators with complex configurations, providing a reliable basis for contamination level assessment.

[0045] To solve the problem of multi-source data fusion and spatial correlation modeling in insulator contamination detection, in some embodiments, step 103: fusing the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial correlation relationship between position units through the graph neural network includes: Step 201: Align the contamination layer thickness information and the scattering spectrum data according to position units to form a joint feature vector for each position unit.

[0046] In step 201, the joint feature vector refers to multidimensional data formed by combining the contamination layer thickness value and the scattering spectrum characteristics according to the same position unit, where the thickness value reflects the height of the contamination accumulation and the scattering spectrum characteristics include the optical properties of the surface and subsurface contamination. The combination of the two can comprehensively characterize the physical and chemical properties of the contamination.

[0047] In an embodiment of the present application, the thickness data extracted from the spectral image of each position unit is first matched and aligned with the scattering spectrum data of the corresponding coordinates to ensure data spatial consistency; then the thickness value and the dual-wavelength scattering rate are combined in a fixed order to form a vector containing three feature dimensions; finally, the same operation is performed on all position units to construct a complete set of joint feature vectors to provide standardized input for subsequent graph neural network processing.

[0048] Step 202: Input the joint feature vector into a graph neural network, and calculate the target spatial connection weights between adjacent position units through the graph neural network.

[0049] In step 202, the spatial connection weights between adjacent cells specifically refer to the local connection strengths between directly adjacent cells; whereas the "spatial association relationship between cells" encompasses the global association relationship between adjacent and non-adjacent cells, representing the overall topological relationship derived through the transitivity of adjacent connection weights. These two represent a progressive relationship between the local and global.

[0050] In an embodiment of the present application, the joint feature vector is input into a pre-trained graph neural network. The network first calculates the physical proximity based on the spatial coordinates of the location unit; then compares the similarity of the feature vectors of adjacent units; finally, the proximity and similarity are fused and calculated through the neural network layer, and a connection weight value between 0 and 1 is output. The larger the weight, the stronger the pollution diffusion correlation.

[0051] Step 203: Determine the spatial association relationship between the location units according to the target spatial connection weight.

[0052] In an embodiment of the present application, a weight threshold is first set to filter associated unit pairs; then a directed connection relationship is constructed according to the weight value; finally, all effective connections are integrated to form a complete spatial association network, which can intuitively display the possible diffusion paths and core accumulation areas of contamination.

[0053] Here's a specific example: In the inspection of FXBW4-110 insulators at substation A, the obtained thickness data and scattering spectrum data were first aligned according to the coordinates of the location unit. The thickness measurement of a unit at the edge of the shed was 0.2 mm, and the corresponding scattering rates at 532 nm and 1064 nm were 0.35 and 0.28, respectively. These data were combined into a three-dimensional joint feature vector in the order of thickness + dual-wavelength scattering rate. The joint feature vector of all location units was then input into a graph neural network. This network adopts a three-layer computing structure. The first layer expands the feature vector dimension to 64 dimensions, the second layer performs 128-dimensional feature transformation, and the third layer outputs scalar weight values. For two adjacent units with a spacing of 8 mm, the network calculates a connection weight of 0.75 based on their feature similarity S=0.82 and physical distance d=8 mm using the weight calculation formula W=exp(-d² / 2σ²)*S, where σ=5 mm is the distance attenuation coefficient determined by experimental calibration. Finally, based on the preset weight threshold of 0.7, the adjacent unit pairs are screened to form a valid spatial association relationship.

[0054] In the embodiment of the present application, through the fusion of multi-source data and spatial correlation analysis of graph neural networks, the coordinated use of contamination thickness and composition characteristics is achieved, the diffusion law of contamination on the surface of complex configuration insulators is accurately portrayed, and a reliable correlation relationship basis is provided for the reconstruction of three-dimensional contamination distribution, thereby improving the physical rationality and practicality of the detection results.

[0055] To further improve the accuracy of insulator contamination diffusion path analysis, in some embodiments, step 202: calculating target spatial connection weights between adjacent location units using the graph neural network includes: Step 301: Identify the physical connection topology between location units from the physical spatial structure of the insulator.

[0056] In step 301, the insulator's physical spatial structure is derived from the actual 3D geometry of the insulator object. This includes physical parameters such as shed spacing, inclination, and curvature, directly obtained from insulator design drawings or 3D scanning data. The physical connection topology refers to the connection relationship between various positional units on the insulator surface based on the actual physical configuration. This includes spatial features such as the overlap between sheds and the spacing between adjacent sheds, reflecting the potential diffusion path of contamination on the actual physical structure.

[0057] In an embodiment of the present application, the spatial coordinates and geometric parameters of each shed unit are first extracted based on the three-dimensional structural model of the insulator. Then, the actual physical connection between the units is analyzed, including the contact area of adjacent sheds, the air gap distance, etc. Finally, a topological relationship diagram representing the physical connection strength between the units is constructed to provide a physical basis for subsequent spatial constraints.

[0058] Step 302: extracting the pollution thickness similarity and spectral feature similarity between the adjacent position units from the joint feature vector to form a two-dimensional similarity index.

[0059] In step 302, thickness similarity reflects the consistency of the pollution accumulation degree. Spectral feature similarity reflects the similarity of the pollution components. The dual-dimensional similarity index is a comprehensive evaluation index that considers both pollution thickness similarity and spectral feature similarity.

[0060] In an embodiment of the present application, the thickness difference and spectral feature difference of adjacent units are respectively extracted from the joint feature vector; the thickness similarity is calculated using the relative difference method, and the spectral feature similarity is calculated using the vector angle method; the two similarities are combined into a comprehensive evaluation index according to preset weights to comprehensively reflect the similarity of the pollution characteristics of adjacent units.

[0061] Step 303: Fusing the physical connection topology structure and the two-dimensional similarity index into a spatial constraint factor.

[0062] In step 303, the spatial constraint factor refers to a comprehensive parameter that integrates the physical connection topology and the two-dimensional similarity index, which takes into account both the limitations of the actual physical structure and the similarity of the pollution characteristics.

[0063] In an embodiment of the present application, the physical connection strength and the two-dimensional similarity index are weightedly fused, and regions with strong physical connections are given a higher fusion weight, ensuring that the final spatial constraint factor not only conforms to the actual physical structure characteristics, but also reflects the possibility of pollution diffusion.

[0064] Step 304: Couple the spatial constraint factor and the joint feature vector to generate target spatial connection weights between adjacent location units.

[0065] In an embodiment of the present application, the spatial constraint factor and the joint feature vector are input together into a special processing layer of the graph neural network. Through multi-level feature transformation and nonlinear calculation, a standardized connection weight value between 0 and 1 is output as the final basis for pollution diffusion analysis.

[0066] Here's a specific example: In the inspection of FXBW4-110 insulators at substation A, the physical connection relationship between the position units was first identified based on the insulator 3D model. The actual spacing between shed unit A and unit B was 5 mm, and there was a partial overlap. The physical connection strength was determined to be 0.3 based on the 30% overlapping area. The pollution characteristic data of the two units were then extracted from the joint feature vector. The thickness of unit A was 0.2 mm, and the dual-wavelength scattering rate was 0.35 and 0.28. The thickness of unit B was 0.18 mm, and the dual-wavelength scattering rate was 0.33 and 0.26. The thickness similarity was calculated using the formula 1-|T_A-T_B| / T_ma The thickness similarity of x is 0.9, where T_max=0.2mm is the maximum thickness difference. The similarity of 0.95 is obtained by the spectral feature similarity calculation formula (F_A·F_B) / (||F_A||×||F_B||). These two similarities are combined into a two-dimensional similarity index of 0.93 with a weight of 4:6; then the physical connection strength of 0.3 and the similarity index of 0.93 are fused with a weight of 3:7 to obtain a spatial constraint factor of 0.75; finally, the constraint factor and the joint feature vector are input into the graph neural network together. After processing through a three-layer computing structure, the network outputs the target spatial connection weight of 0.82 between units A and B.

[0067] In the embodiment of the present application, through the dual constraints of physical connection topology and similarity of pollution characteristics, the calculated spatial connection weight is made more consistent with the actual working conditions. It not only takes into account the limitations of the insulator physical structure on pollution diffusion, but also integrates the influence of the pollution's own characteristics, thereby improving the accuracy and reliability of the pollution diffusion path analysis and providing a more accurate correlation relationship basis for the subsequent three-dimensional pollution distribution reconstruction.

[0068] To further improve the accuracy of the pollution diffusion weight calculation, in some embodiments, step 304: coupling the spatial constraint factor with the joint feature vector to generate target spatial connection weights between adjacent location units includes: Step 401: Convert the spatial constraint factor into an edge constraint condition of the graph neural network.

[0069] In step 401, the edge constraint condition refers to converting the spatial constraint factor into a restriction parameter that controls the edge connection strength in the graph neural network, which is used to force the maintenance or weakening of the connection relationship between specific position units.

[0070] In an embodiment of the present application, the spatial constraint factors are first divided into different levels according to preset rules, then the corresponding constraint strength is set according to the level, and finally converted into edge connection control parameters that can be recognized by the graph neural network to ensure that the network follows the actual physical connection restrictions during the processing process.

[0071] Step 402: Inject the physical configuration features of the spatial constraint factors into the joint feature vector to generate an enhanced feature vector.

[0072] In step 402, the physical configuration characteristics of the spatial constraint factor quantify key parameters of the insulator's physical spatial structure (such as shed overlap area and overhang gap) into computable constraints, which are formed by encoding the spatial relationship between the shed's specific geometric features and the location unit. The enhanced feature vector adds a new dimension reflecting the physical configuration characteristics to the original joint feature vector, allowing the feature expression to simultaneously incorporate both contamination characteristics and spatial structure information.

[0073] In an embodiment of the present application, physical configuration-related parameters are extracted from the spatial constraint factors, standardized and used as new feature dimensions, and then concatenated with the original joint feature vector to form an enhanced feature vector after dimensional expansion, providing more comprehensive input information for subsequent network processing.

[0074] Step 403: Perform constraint-guided processing on the enhanced feature vector and the edge constraint condition through the graph neural network, and output an initial spatial connection weight value.

[0075] In step 403, constraint-guided processing refers to a special computational mode in which the graph neural network considers both feature data and edge constraints during its calculations. The initial spatial connection weights refer to the preliminary strength of associations between adjacent cells, calculated through forward propagation by the graph neural network, considering only the enhanced feature vectors and edge constraints. This value reflects the network's initial prediction of the likelihood of contamination diffusion based on the current input features and physical constraints, and has not yet been optimized for the actual diffusion characteristics of contamination.

[0076] In an embodiment of the present application, the enhanced feature vector is input into the first hidden layer of the graph neural network for feature transformation, and edge constraints are introduced during processing of the second hidden layer to screen the connection relationship. Finally, the initial spatial connection weight value is generated through the output layer. The entire process ensures that the network calculation is guided by physical constraints.

[0077] Step 404: Optimize and adjust the initial spatial connection weight value according to the pollution diffusion characteristics of the insulator to generate a target spatial connection weight.

[0078] In step 404, the pollution diffusion characteristics of the insulator refer to the physical laws of the diffusion of pollutants along the surface of the insulator in the actual environment, including the deposition of salt spray in the shed grooves and the accumulation of hydrophilic contamination along the edges. They are obtained through long-term field observation data or laboratory accelerated pollution accumulation tests.

[0079] In an embodiment of the present application, a pollution diffusion characteristic library is established based on historical detection data, the initial spatial connection weight values are compared with the typical patterns in the characteristic library, and the weight values that do not conform to the actual diffusion laws are adjusted in a targeted manner, and finally the target spatial connection weights that conform to physical reality are output.

[0080] Here's a specific example: In the detection of FXBW4-110 insulators at substation A, the spatial constraint factor of 0.75 was first converted into an edge constraint condition for the graph neural network. Specifically, this value was mapped into a constraint coefficient for the network edge calculation layer. When the constraint coefficient was greater than 0.7, the connection was retained; when it was less than 0.5, the connection was disconnected; and if it was in between, it was scaled linearly. Next, based on the original three-dimensional joint feature vector, the shed skirt inclination angle feature and the overlap area ratio feature were added to form a five-dimensional enhanced feature vector. In this case, the inclination angle of unit A was 15 degrees and the overlap ratio was 30%, while the inclination angle of unit B was 12 degrees and the overlap ratio was 25%. The enhanced feature vector is then input into the graph neural network. The first layer of the network maps the five-dimensional features to a 64-dimensional space through the weight matrix W_1, and the second layer transforms it to 128 dimensions through W_2. W_1 and W_2 are parameter matrices obtained through network training. Edge constraints are introduced during the calculation of the second layer to shield connections that do not meet the constraints. The initial spatial connection weight value of the output units A and B after the third layer is 0.78. Finally, based on the characteristic that contamination of this type of insulator is easily deposited in the overlapping area of the shed, the initial weight value is multiplied by a correction coefficient of 1.1 to obtain the final target spatial connection weight of 0.86.

[0081] In the embodiment of the present application, by introducing edge constraints and physical configuration features, the calculation process of the graph neural network is made more consistent with actual physical laws. Combined with the optimization adjustment of the pollution diffusion characteristics, the target space connection weights finally obtained not only maintain the accuracy of data-driven, but also ensure consistency with actual working conditions, providing a reliable quantitative basis for the three-dimensional distribution analysis of pollution.

[0082] To further improve the accuracy of contamination thickness measurement, in some embodiments, step 102: extracting contamination layer thickness information from the spectral image data includes: Step 501: Calculate the optical depth parameter of the contamination layer according to the spectral focus change difference of the spectral image data.

[0083] In step 501, the spectral focus variation of the spectral image data refers to the difference in focal position caused by the contamination layer's influence on the spectral reflection / transmission characteristics in spectral images acquired at different focal lengths. This difference originates from the optical path difference caused by the refractive index change of the contaminant medium for light waves. The raw data is directly obtained through multi-focal length scanning. The optical depth parameter, obtained by analyzing the focus variation characteristics of multi-focal length spectral images, reflects the refraction and absorption characteristics of the contamination layer for light waves and is used to characterize the optical penetration depth of contamination.

[0084] In an embodiment of the present application, the clarity of the spectral image sequence at different focal lengths is first analyzed, and the clarity index of each position unit in the image at each focal length is calculated; then the optimal focal length corresponding to the peak clarity of each unit is determined; finally, the optical depth parameter is calculated based on the change gradient of the clarity between adjacent focal lengths, and this parameter corresponds to the actual thickness of the contamination layer.

[0085] Step 502: Generate contamination layer thickness information based on the optical depth parameter and a preset image clarity variation rule.

[0086] In step 502, the image clarity variation rule refers to the mapping relationship between the clarity index and the contamination thickness established in advance through experiments, which is used to convert the optical parameters into actual thickness values.

[0087] In an embodiment of the present application, a pre-calibrated clarity-thickness comparison table is called to find the corresponding thickness interval according to the calculated optical depth parameter; the precise thickness value is calculated using linear interpolation; and finally, the same operation is performed on all position units to generate complete contamination layer thickness distribution information.

[0088] Here's a specific example: In the inspection of FXBW4-110 insulators at substation A, a hyperspectral camera with a focal length interval of 0.2 mm was used to collect 20 sets of spectral image sequences. Analysis of a unit at a certain position on the edge of the shed found that the clarity of the unit reached its peak in the sixth set of images, with a clarity index of 0.85. The clarity of the adjacent fifth and seventh sets of images was 0.82 and 0.81, respectively. The optical depth parameter was obtained by calculating the clarity change rate ΔC / Δf=(0.85-0.81) / 0.4mm=0.1mm^-1. The number is 0.25, where Δf is the focal length change. According to the calibration curve pre-established in the laboratory, when the optical depth parameter of this type of insulator is 0.2, the corresponding thickness is 0.1mm, and when the parameter is 0.3, it corresponds to 0.2mm. The linear interpolation formula T=0.1mm+(0.25-0.2)×(0.2mm-0.1mm) / (0.3-0.2) is used to calculate that the pollution thickness of this unit is 0.15mm. After processing the data of all position units using the same method, a complete pollution thickness distribution map is generated.

[0089] In the embodiment of the present application, non-contact and accurate measurement of the thickness of the contamination layer is achieved through multi-focal length spectral analysis and clarity-thickness mapping conversion, which solves the problem of thickness detection of uneven contamination accumulation on the surface of insulators with complex configurations and provides key parameter basis for contamination degree assessment.

[0090] To further improve the accuracy and reliability of contamination thickness measurement, in some embodiments, step 502: generating contamination layer thickness information based on the optical depth parameter and a preset image clarity variation pattern includes: Step 601: Match and map the optical depth parameter with the image clarity variation rule.

[0091] In step 601, matching mapping refers to the process of comparing the calculated optical depth parameters with the pre-calibrated clarity-thickness correspondence to determine the thickness range of the optical parameters.

[0092] In an embodiment of the present application, a clarity-thickness comparison table established in advance through standard sample experiments is first loaded, and then the upper and lower limit parameter values closest to the current optical depth parameters are searched in the table to determine the corresponding thickness reference interval, providing basic data for subsequent precise calculations.

[0093] Step 602: According to the matching mapping result, a quantized value of the dirt layer thickness is calculated by a linear interpolation method.

[0094] In step 602, linear interpolation is a mathematical method that calculates the thickness of an intermediate point using a linear relationship, given the parameter values and thickness values of two reference points. The quantized contamination layer thickness value, calculated using linear interpolation, reflects the actual thickness of the contamination layer at a specific location on the insulator surface. This value is determined based on the mapping relationship between the optical depth parameter and a preset definition-thickness rule and serves as the fundamental data unit for subsequently constructing a three-dimensional contamination distribution.

[0095] In an embodiment of the present application, a linear relationship between optical parameters and thickness is established based on the upper and lower limits of the thickness reference interval determined by the matching mapping. The actually measured optical depth parameters are substituted into the relationship to calculate the precise contamination thickness value, thereby ensuring the continuity of the measurement results.

[0096] Step 603: Aggregate the quantized values of the pollution layer thickness of all location units to form pollution layer thickness information.

[0097] In step 603, all positions refer to the position units corresponding to the spatial partitioned areas of the insulator, that is, all the local areas of the insulator surface that are divided and detected and covered by the multiple focal length spectrum images.

[0098] In an embodiment of the present application, the thickness values calculated for each unit are systematically organized and stored according to the arrangement order of the insulator surface position units, and a complete data set including spatial position coordinates and thickness values is constructed to provide a basis for subsequent analysis.

[0099] Here's a specific example: During the inspection of FXBW4-110 insulators at Power Station B, the optical depth parameter measured for a unit in the middle of the shed was 0.28. The calibration data table for this type of insulator shows that a parameter of 0.25 corresponds to a thickness of 0.15 mm, and a parameter of 0.3 corresponds to 0.2 mm. Using the interpolation formula T = 0.15 mm + (0.28 - 0.25) × (0.2 mm - 0.15 mm) / (0.3 - 0.25), the contamination thickness of this unit was calculated to be 0.18 mm. The numerator in the interpolation formula represents the difference between the current parameter and the lower limit parameter, and the denominator represents the parameter range. The same method was used to process all 128 units on the insulator surface. The units at the edge of the shed had a thickness of 0.2-0.25 mm, and the units at the top of the shed had a thickness of 0.1-0.15 mm. Finally, the thickness values of all units were integrated according to the inspection position coordinates to form a complete contamination thickness information table containing the position number and thickness value.

[0100] In the embodiment of the present application, a reliable conversion from optical parameters to thickness values is achieved through precise matching mapping and interpolation calculations, and then complete thickness distribution information is formed through system integration, providing an accurate and comprehensive thickness data basis for the assessment of the degree of insulator contamination, and solving the difficult problem of measuring the contamination thickness on complex surfaces.

[0101] To further improve the accuracy of the reconstruction of the three-dimensional pollution distribution, in some embodiments, step 104: generating the three-dimensional pollution distribution data of the insulator according to the spatial correlation relationship includes: Step 701: Constructing a pollution transfer path between location units based on the spatial association relationship.

[0102] In step 701, the pollution transmission path refers to the possible diffusion route of pollution between various position units on the insulator surface determined based on the spatial correlation relationship, reflecting the actual propagation direction and intensity of the pollution.

[0103] In an embodiment of the present application, first, the unit pairs whose connection weights in the spatial association relationship exceed the set threshold are screened, then the path direction is determined according to the weight value, and finally, the unit pairs that meet the conditions are connected to form a complete contamination transmission network, providing a path basis for subsequent thickness diffusion calculation.

[0104] Step 702: performing thickness diffusion calculation on the contamination layer thickness information according to the contamination transmission path.

[0105] In step 702, thickness diffusion calculation refers to the calculation process of transferring and attenuating the contamination thickness value according to the correlation strength along the contamination transfer path.

[0106] In an embodiment of the present application, starting from the core area of contamination, the thickness value of the upstream unit is multiplied by the corresponding attenuation coefficient and then transmitted to the downstream unit in the direction of the transmission path. At the same time, the influence of the correlation strength on the path on the attenuation degree is considered, and finally the thickness distribution of all position units under the influence of contamination diffusion is calculated.

[0107] Step 703: Map the calculation results to a preset three-dimensional insulator model to generate three-dimensional pollution distribution data.

[0108] In step 703, the insulator three-dimensional model refers to a digital three-dimensional spatial model established based on the actual physical structure of the insulator (including geometric parameters such as shed size, spacing, and inclination angle), which is used to map and visualize pollution distribution data.

[0109] In an embodiment of the present application, the calculated thickness value is assigned to the corresponding area of the three-dimensional model according to the spatial coordinates of the position unit, and different colors are used to represent different thickness ranges to generate a three-dimensional map that intuitively shows the distribution of dirt.

[0110] Here's a specific example: In the inspection of FXBW4-110 insulators in substation B, 15 pairs of adjacent position units with weights greater than 0.7 were first selected based on spatial correlation to construct the main pollution transfer path. The connection weight from shed unit A to unit B was 0.82, and the weight from unit B to unit C was 0.75. Then, starting from shed groove unit A with a maximum thickness of 0.28 mm, the thickness diffusion was calculated along the transfer path according to the formula T_downstream=T_upstream×α, where the attenuation coefficient α=0.9-(0.9-0.7)×(1-W), where W is the connection weight value. The thickness of unit B was calculated to be 0.28×[0.9-0.2×(1-0.82 )]=0.25mm, and the thickness of unit C is 0.25×[0.9-0.2×(1-0.75)]=0.22mm. Finally, the thickness calculation results of all position units are mapped to the insulator 3D model to generate a color distribution map with an interval of 0.1mm. The red area is the shed groove and adjacent area with a thickness of ≥0.25mm, the yellow area is the transition area of 0.15-0.25mm, and the green area is the clean area with a thickness of <0.15mm. This fully presents the three-dimensional distribution characteristics of the pollution diffusion from the shed groove. The 0.9 in the attenuation coefficient formula is the basic attenuation value, and 0.7 is the minimum attenuation limit. The larger the weight value, the smaller the attenuation, which is in line with the physical law of pollution diffusion.

[0111] In the embodiment of the present application, by constructing the pollution transfer path and thickness diffusion calculation, the spatial reconstruction of the pollution distribution is achieved. Then, through three-dimensional visualization mapping, the accumulation of pollution on the insulator surface is intuitively presented, providing a reliable basis for pollution level assessment and maintenance decision-making.

[0112] Figure 2 A schematic diagram of the structure of an insulator contamination degree detection system based on multi-source spectral sensing provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the system includes: The acquisition module 21 is used to acquire spectral image data of the insulator at multiple focal lengths and scattering spectrum data of the insulator surface.

[0113] The extraction module 22 is used to extract the contamination layer thickness information from the spectral image data.

[0114] A fusion module 23 is used to fuse the contamination layer thickness information and the scattering spectrum data, input the fusion result into a graph neural network, and determine the spatial correlation relationship between position units through the graph neural network, where the position units correspond to the spatial partition areas of the insulator.

[0115] The generating module 24 is configured to generate three-dimensional pollution distribution data of the insulator according to the spatial correlation relationship.

[0116] Figure 2 The insulator contamination degree detection system based on multi-source spectral sensing can be performed Figure 1 The implementation principles and technical effects of the multi-source spectral sensing-based insulator contamination detection method described in the illustrated embodiment are not further elaborated. The specific manner in which the various modules and units in the multi-source spectral sensing-based insulator contamination detection system described in the aforementioned embodiment perform their operations has been described in detail in the related embodiments and will not be further elaborated here.

[0117] In one possible design, Figure 2 The insulator contamination degree detection system based on multi-source spectral sensing of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0118] The processing component 32 is used to perform the above Figure 1 The embodiment provides a method for detecting the degree of insulator contamination based on multi-source spectral sensing.

[0119] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0120] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0121] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0122] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0123] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0124] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0125] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment provides a method for detecting the degree of insulator contamination based on multi-source spectral sensing.

[0126] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0127] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0128] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting the degree of insulator contamination based on multi-source spectral sensing, characterized in that: include: Acquire spectral image data of the insulator at multiple focal lengths and scattered spectrum data of the insulator surface; Extracting contamination layer thickness information from the spectral image data; fusing the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial association relationship between position units through the graph neural network, wherein the position units correspond to spatial partition areas of the insulator; Based on the spatial correlation, three-dimensional pollution distribution data of the insulator is generated.

2. The method according to claim 1, characterized in that The fusing of the contamination layer thickness information and the scattering spectrum data, inputting the fusion result into a graph neural network, and determining the spatial correlation relationship between the position units through the graph neural network includes: Aligning the contamination layer thickness information with the scattering spectrum data according to position units to form a joint feature vector for each position unit; Inputting the joint feature vector into a graph neural network, and calculating the target spatial connection weights between adjacent position units through the graph neural network; The spatial association relationship between the location units is determined according to the target spatial connection weight.

3. The method according to claim 2, characterized in that Calculating the target spatial connection weights between adjacent location units by the graph neural network includes: Identify the physical connection topology between location units from the physical spatial structure of the insulator; Extracting the pollution thickness similarity and spectral feature similarity between the adjacent position units from the joint feature vector to form a two-dimensional similarity index; fusing the physical connection topology structure and the two-dimensional similarity index into a spatial constraint factor; The spatial constraint factor is coupled with the joint feature vector to generate target spatial connection weights between adjacent location units.

4. The method according to claim 3, characterized in that The coupling of the spatial constraint factor and the joint feature vector to generate target spatial connection weights between adjacent location units includes: Converting the spatial constraint factor into an edge constraint condition of a graph neural network; injecting the physical configuration characteristics of the spatial constraint factor into the joint feature vector to generate an enhanced feature vector; Performing constraint-guided processing on the enhanced feature vector and the edge constraint condition through the graph neural network, and outputting an initial spatial connection weight value; According to the pollution diffusion characteristics of the insulator, the initial spatial connection weight value is optimized and adjusted to generate a target spatial connection weight.

5. The method according to claim 1, wherein The step of extracting the contamination layer thickness information from the spectral image data includes: Calculating an optical depth parameter of the contamination layer according to a spectral focus change difference of the spectral image data; Based on the optical depth parameter and in combination with a preset law of image definition change, the contamination layer thickness information is generated.

6. The method according to claim 5, characterized in that The generating of the contamination layer thickness information based on the optical depth parameter and the preset image clarity variation rule includes: Matching and mapping the optical depth parameter with the image clarity variation rule; According to the matching mapping results, the quantitative value of the contamination layer thickness is calculated by linear interpolation method; Aggregate the quantized values of the pollution layer thickness of all position units to form the pollution layer thickness information.

7. The method according to claim 1, characterized in that Generating three-dimensional pollution distribution data of the insulator according to the spatial correlation relationship includes: Constructing a pollution transfer path between location units based on the spatial association relationship; Performing thickness diffusion calculation on the contamination layer thickness information according to the contamination transmission path; The calculation results are mapped to the preset three-dimensional insulator model to generate three-dimensional pollution distribution data.

8. An insulator contamination degree detection system based on multi-source spectral sensing, characterized in that: include: An acquisition module is used to acquire spectral image data of the insulator at multiple focal lengths and scattered spectrum data of the insulator surface; An extraction module, configured to extract contamination layer thickness information from the spectral image data; a fusion module, configured to fuse the contamination layer thickness information and the scattering spectrum data, input the fusion result into a graph neural network, and determine the spatial association relationship between position units through the graph neural network, wherein the position units correspond to spatial partitioned areas of the insulator; A generating module is used to generate three-dimensional pollution distribution data of the insulator according to the spatial correlation relationship.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the insulator contamination degree detection method based on multi-source spectral perception as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for detecting the contamination degree of an insulator based on multi-source spectral perception as claimed in any one of claims 1 to 7 is implemented.

Images