Method for identifying impurity types in 10kV cable intermediate joint

Abstract

The 10kV cable intermediate joint impurity type identification method provided in the application comprises the following steps: a microwave signal with a frequency of 22GHz-30GHz is emitted radially to a 10kV intermediate joint of a cable to be identified, and the obtained microwave reflection amplitude curve and phase curve are converted into a Smith circle image; a normalized gray level gradient co-occurrence matrix corresponding to the Smith circle image is constructed, and a large gradient advantage feature, a gradient average feature and a hybrid entropy feature are calculated according to the normalized gray level gradient co-occurrence matrix; a type factor used for representing the impurity type of the wrapping tape of the intermediate joint of the cable is calculated according to the large gradient advantage feature, the gradient average feature and the hybrid entropy feature; and the impurity type of the intermediate joint of the cable to be identified is determined according to the type factor. In this way, the internal impurity type of the intermediate joint can be quickly and accurately identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable joint detection and evaluation, and particularly relates to a 10kV cable joint impurity type identification method. BACKGROUND

[0002] The cable joint plays an important role in connecting cable sections and ensuring the continuous transmission of power in the power system, and needs to have good electrical insulation performance and environmental adaptability to ensure long-term stable operation under various laying conditions. The manufacturing and installation quality of the cable joint is directly related to the safety and reliability of system operation.

[0003] In the actual construction process, impurities are often mixed into the cable joint due to unclean external environment or improper operation. For example, there are dust, semi-conductive particles and other pollutants on the construction site, which are easy to adhere to the inside of the joint if not completely removed; improper operation when stripping the cable may cause damage to the insulation layer; and loose wrapping of the insulation tape may also form gaps. These impurities or defects can destroy the uniformity of the electric field inside the joint, cause electric field distortion, trigger partial discharge, and further accelerate the aging of the insulation material, thereby reducing the sealing performance of the joint. Over time, these hidden dangers may cause insulation breakdown, resulting in cable failure, shortening the service life, and even affecting the safe operation of the entire line in severe cases. Therefore, it is of great significance to quickly and accurately identify the specific type of impurities inside the cable joint for improving the safety and stability of the cable operation. SUMMARY

[0004] The purpose of the present application is to at least solve one of the above technical defects, in particular the technical defect of how to quickly and accurately identify the specific type of impurities inside the cable joint in the prior art.

[0005] In a first aspect, the present application provides a 10kV cable joint impurity type identification method, which comprises:

[0006] radially emitting a microwave signal of a frequency band of 22GHz-30GHz to a 10kV cable joint to be identified, and converting the obtained microwave reflection amplitude curve and phase curve into a Smith circle image;

[0007] constructing a normalized gray level gradient co-occurrence matrix corresponding to the Smith circle image, and calculating a large gradient advantage feature, a gradient average feature and a hybrid entropy feature according to the normalized gray level gradient co-occurrence matrix;

[0008] calculating a type factor for representing the type of the wrapping tape impurities of the cable joint according to the large gradient advantage feature, the gradient average feature and the hybrid entropy feature;

[0009] determining the type of the impurities of the cable joint to be identified according to the type factor.

[0010] In one embodiment, the step of constructing the normalized gray level gradient co-occurrence matrix corresponding to the Smith chart image comprises:

[0011] Reading the gray level value of each pixel in the Smith chart image, and for each pixel, calculating the square root of the sum of squares of the gray level difference between the pixel and the adjacent right pixel and the adjacent lower pixel to obtain the gradient value corresponding to the pixel, and generating a gradient matrix;

[0012] After the gradient matrix is discretized, an initial gray level gradient co-occurrence matrix is constructed;

[0013] The initial gray level gradient co-occurrence matrix is normalized to obtain a normalized gray level gradient co-occurrence matrix.

[0014] In one embodiment, the calculation expression of the large gradient dominant feature is:

[0015]

[0016] wherein, represents the large gradient dominant feature, represents the sum of all elements in the normalized gray level gradient co-occurrence matrix, represents each element in the normalized gray level gradient co-occurrence matrix, represents the level gray level, represents the level gradient.

[0017] In one embodiment, the calculation expression of the gradient average feature is:

[0018]

[0019] wherein, represents the gradient average feature, represents the sum of all elements in the normalized gray level gradient co-occurrence matrix, represents each element in the normalized gray level gradient co-occurrence matrix, represents the level gray level, represents the level gradient.

[0020] In one embodiment, the calculation expression of the hybrid entropy feature is:

[0021]

[0022] wherein, represents the hybrid entropy feature, represents each element in the normalized gray level gradient co-occurrence matrix, represents the first level gray scale, represents the first level gradient, represents a positive number for avoiding taking logarithm of zero.

[0023] In one of the embodiments, the step of calculating a type factor for representing the impurity type of the cable joint around the tape according to the large gradient dominant feature, the gradient average feature and the mixed entropy feature comprises:

[0024] The type factor is calculated according to the following expression:

[0025]

[0026] wherein, represents the type factor, represents the large gradient dominant feature, represents the gradient average feature, represents the mixed entropy feature.

[0027] In one of the embodiments, the step of determining the impurity type of the cable joint to be identified according to the type factor comprises:

[0028] If the type factor is greater than zero and not greater than 0.24, the impurity type of the cable joint to be identified is air;

[0029] If the type factor is greater than 0.24 and not greater than 0.63, the impurity type of the cable joint to be identified is sand;

[0030] If the type factor is greater than 0.63 and not greater than 1, the impurity type of the cable joint to be identified is metal residual particles.

[0031] In a second aspect, the application provides a 10kV cable joint impurity type identification device, the device comprising:

[0032] A Smith circle image determination module is configured to radially emit a microwave signal of a 10kV cable joint to be identified in a frequency band of 22GHz~30GHz, and convert the obtained microwave reflection amplitude curve and phase curve into a Smith circle image;

[0033] A feature calculation module is configured to construct a normalized gray scale gradient co-occurrence matrix corresponding to the Smith circle image, and calculate a large gradient dominant feature, a gradient average feature and a mixed entropy feature according to the normalized gray scale gradient co-occurrence matrix;

[0034] A type factor calculation module is configured to calculate a type factor for representing the impurity type of the cable joint around the tape according to the large gradient dominant feature, the gradient average feature and the mixed entropy feature;

[0035] The impurity type determination module is configured to determine the impurity type of the intermediate joint of the cable to be identified according to the type factor.

[0036] In a third aspect, the present application provides a storage medium, the storage medium storing computer readable instructions, the computer readable instructions being executed by one or more processors to cause the one or more processors to perform the steps of the method for identifying the impurity type of the intermediate joint of the 10kV cable according to any one of the above embodiments.

[0037] In a fourth aspect, the present application provides a computer device, comprising: one or more processors, and a memory;

[0038] The memory stores computer readable instructions, the computer readable instructions being executed by one or more processors to perform the steps of the method for identifying the impurity type of the intermediate joint of the 10kV cable according to any one of the above embodiments.

[0039] From the above technical solutions, the embodiments of the present application have the following advantages:

[0040] The method for identifying the impurity type of the intermediate joint of the 10kV cable provided by the present application can realize rapid and accurate identification of the impurity type inside the intermediate joint. Specifically, by radially emitting a microwave signal in the 22GHz-30GHz frequency band and constructing a Smith circle image, the change of the electromagnetic characteristics inside the joint is effectively reflected. Then, based on the image, a normalized gray level gradient co-occurrence matrix is constructed, and large gradient dominant features, gradient average features and hybrid entropy features are extracted, which helps to comprehensively depict the microscopic interference features caused by the impurities to the internal structure of the joint. By calculating the type factor and determining the impurity type, the characteristic differences caused by different impurities can be distinguished, thereby providing support for subsequent fault warning and quality tracing, improving the intelligent and refined level of intermediate joint defect detection, and enhancing the safety and reliability of the cable line operation. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0042] Figure 1 A flowchart of the method for identifying the impurity type of the intermediate joint of the 10kV cable provided by the embodiments of the present application is shown in the figure.

[0043] Figure 2 A structural diagram of the device for identifying the impurity type of the intermediate joint of the 10kV cable provided by the embodiments of the present application is shown in the figure.

[0044] Figure 3 An internal structure schematic diagram of a computer device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0046] The present application provides a method for identifying the type of impurities in a 10kV cable intermediate joint. The following embodiments take the application of the method to a computer device as an example for illustration. It can be understood that the computer device can be various devices with data processing functions, which can be but are not limited to a single server, a server cluster, a personal notebook computer, a desktop computer, etc. As shown in Figure 1 The method can include the following steps:

[0047] S101: radially emit a microwave signal with a frequency range of 22GHz~30GHz to the 10kV cable intermediate joint to be identified, and convert the obtained microwave reflection amplitude curve and phase curve into a Smith chart image.

[0048] The radial emission of the microwave signal means that the microwave signal is vertically emitted to the cross section of the cable intermediate joint, so as to ensure that the electromagnetic wave propagates radially in the internal medium of the cable. The 10kV cable intermediate joint refers to a middle connecting device for connecting two sections of 10kV cross-linked polyethylene (XLPE) cables, which has the functions of electrical insulation and sealing protection. The microwave signal with a frequency range of 22GHz~30GHz refers to a high-frequency electromagnetic wave signal with a frequency range of 22GHz~30GHz, which can effectively penetrate the internal structure of the cable and produce a sensitive response to the dielectric properties thereof. The microwave reflection amplitude curve and the phase curve respectively represent the amplitude change and the phase change of the microwave signal in the reflection process, which reflect the electromagnetic response characteristics of the internal structure of the cable joint. The Smith chart image is a graphical representation of the mapping of the complex reflection coefficient to a two-dimensional circular chart, which is used to analyze the impedance change and matching of the high-frequency signal in the medium.

[0049] Specifically, the microwave port is attached to the surface of the cable, and a continuous wave microwave signal with a frequency range of 22-30 GHz is transmitted into the 10 kV cable intermediate joint to be identified by controlling the vector network analyzer. In order to realize radial transmission, the microwave signal is coupled to the rectangular waveguide port through a coaxial cable, and a fixed clamp is used to ensure that the transmission port is vertically attached to the center position of the surface of the cable joint to realize a stable and repeatable radial incidence angle. At the same time of signal transmission, the reflected echo is received synchronously, and the complex reflection coefficient data corresponding to each frequency point is collected by the vector network analyzer.

[0050] Subsequently, the computer device normalizes the collected data in sequence of frequency, and decomposes them into amplitude response curves and phase response curves respectively. The amplitude curve describes the reflection intensity characteristics of different frequency points, and the phase curve depicts the phase shift of electromagnetic waves in the signal propagation path. This process is automatically completed by the data preprocessing module to ensure the stability and consistency of the input data.

[0051] Further, the computer device maps the above-mentioned complex reflection coefficient data (including amplitude and phase) into a Smith circle image. For this purpose, each complex reflection coefficient is projected onto the Smith circle in the complex plane, thereby obtaining a complete and continuous two-dimensional image. The image indicates the impedance change trend with different colors or gray values, and intuitively reflects the influence of the internal medium change of the cable joint on the microwave reflection characteristics.

[0052] It can be understood that by performing radial microwave transmission on the cable intermediate joint in the frequency range of 22-30 GHz, the electromagnetic response characteristics of the internal materials of the cable can be fully excited, and the obtained microwave reflection amplitude and phase curves truly reflect the distribution state of the internal medium structure. Further converting these curves into a Smith circle image can map the original one-dimensional signal curve into a two-dimensional image expression form with spatial characteristics, so that the subsequent image processing algorithm can use the spatial information in the image structure to mine the small difference characteristics of impurities or defects. Therefore, this step not only improves the characterization ability of impurities in the cable joint, but also enhances the diversity and discriminability of data input, provides a data basis and image support for realizing high-precision defect identification, and thus improves the detection efficiency and identification accuracy.

[0053] S102: Construct a normalized gray level gradient co-occurrence matrix corresponding to the Smith circle image, and calculate the large gradient dominant feature, the gradient average feature, and the hybrid entropy feature according to the normalized gray level gradient co-occurrence matrix.

[0054] The gray scale gradient co-occurrence matrix refers to a two-dimensional matrix constructed based on the spatial relationship between the gray scale gradients of pixels in an image, and is used to quantify the structural distribution characteristics of the image texture. The normalized gray scale gradient co-occurrence matrix refers to a matrix obtained by normalizing the original gray scale gradient co-occurrence matrix according to the total frequency, and reflects the proportion of each gradient pair in the entire image. The large gradient dominant feature is a numerical index for measuring the intensity of the strong gradient distribution in the image. The gradient average feature represents the average level of the overall gray scale change of the image. The hybrid entropy feature is used to measure the uncertainty and complexity of the information in the image texture.

[0055] Specifically, after completing the Smith chart image generation, the computer device converts the image into a gray scale image and calculates the gray scale gradient value of each pixel point in the image in the horizontal and vertical directions, respectively. In order to improve the stability of the image gradient feature, the system uses the Sobel operator for edge gradient enhancement processing to obtain the horizontal gradient matrix and the vertical gradient matrix, and calculates the gradient amplitude of each pixel based on the same.

[0056] Then, the computer device constructs a gray scale gradient co-occurrence matrix based on the gradient amplitude matrix. This process includes setting the discrete gradient level interval, mapping the gradient values of all pixel points to the preset gradient level interval, and counting the frequency of different gradient value combinations within a given direction and distance range, thereby forming a two-dimensional gradient co-occurrence relationship matrix. Here, the adjacent relationship with angles of 0° and 90° can be used to fully reflect the horizontal and vertical texture changes in the image.

[0057] Subsequently, the computer device normalizes the gray scale gradient co-occurrence matrix, which specifically includes dividing each element in the matrix by the total sum of the gradient combination frequencies, so that each item in the matrix represents the relative probability distribution of its occurrence in the image. This normalized matrix reduces the interference caused by the difference in gradient amplitude distribution between different images.

[0058] Further, the computer device extracts the large gradient dominant feature, the gradient average feature, and the hybrid entropy feature based on the normalized gray scale gradient co-occurrence matrix. The large gradient dominant feature reflects whether there is a local strong reflection area in the image by accumulating the probability value of the high gradient area and weighted calculation; the gradient average feature reflects the stationarity of the overall reflection field texture distribution by weighted mean of the matrix; the hybrid entropy feature is used to measure the texture complexity and randomness in the image by traversing all non-zero elements in the normalized matrix and calculating the logarithm of the probability value and then weighted sum.

[0059] It can be understood that by constructing a normalized gray level gradient co-occurrence matrix on the basis of the Smith chart image, and further extracting the large gradient dominant feature, the gradient average feature and the hybrid entropy feature, the texture change pattern caused by impurities in the image can be effectively quantified, and the originally subjective image features are converted into quantitative indicators that can be used for identification modeling. The large gradient dominant feature can identify the local strong impurity caused by the electric field disturbance area, the gradient average feature helps to determine the uniformity of the overall insulation structure, and the hybrid entropy feature improves the perception ability of the texture complexity difference caused by different types of impurities. Therefore, the fine-grained analysis ability and robustness of image feature extraction are improved, a reliable numerical basis is provided for accurate identification of the type of impurities in the cable joint, and the identification accuracy is significantly enhanced.

[0060] S103: According to the large gradient dominant feature, the gradient average feature and the hybrid entropy feature, a type factor for representing the type of the wrapping tape impurities in the cable intermediate joint is calculated.

[0061] Among them, the type factor is a composite numerical index calculated by combining multiple image features, which is used to depict the specific type or morphological characteristics of the impurities in the wrapping tape of the cable intermediate joint in the feature space.

[0062] Specifically, the computer device receives and loads the large gradient dominant feature, the gradient average feature and the hybrid entropy feature. In this stage, it is preferred to store the three types of features in a group of feature vectors, and to normalize them to eliminate the differences in numerical dimensions of different features, and to ensure the numerical consistency and physical interpretation consistency of subsequent combination calculation.

[0063] Then, the three types of image features can be weighted and fused. The fusion method can be based on principal component analysis, weighted linear combination, support vector embedding, etc. In a specific embodiment, the large gradient dominant feature weight can be set to be relatively high to highlight the local disturbance caused by the impurity edge; at the same time, the gradient average feature and the hybrid entropy feature are given different degrees of secondary weight, and then a weighted formula is constructed.

[0064] Further, the computer device can perform normalization mapping processing on the initially generated type factor to make its distribution more concentrated and improve the clarity of the classification boundary. At the same time, multiple type factors can be mapped to a multi-dimensional feature space to construct a feature cluster model, and compared with a sample library corresponding to known impurity types to form a clear feature-type label mapping relationship. In order to improve the calculation accuracy, fuzzy clustering or distance measurement algorithms such as K-Means or Mahalanobis distance can also be introduced to improve the discrimination ability of the type factor, so that it not only has numerical representation, but also has classification and judgment potential.

[0065] It can be understood that by weighting and fusing the large gradient advantage feature, the gradient average feature and the mixed entropy feature to calculate the type factor, the local strong disturbance caused by impurities in the image, the overall uniformity change and the texture complexity information can be effectively integrated to form a numerical expression with high discrimination for the type of impurities. The type factor not only enhances the clustering separability of each type of impurities in the feature space, but also provides a stable and quantifiable input basis for subsequent classification recognition algorithms, thereby significantly improving the accuracy of impurity type recognition.

[0066] S104: Determine the impurity type of the cable intermediate joint to be identified according to the type factor.

[0067] The impurity type refers to different properties of foreign matter or defects that may be mixed in the wrapping tape structure of the cable intermediate joint, including dust, semi-conductive particles, oil stains, water vapor residues or loose wrapping, etc., which have different dielectric properties and image texture performances.

[0068] Specifically, the computer device receives and analyzes the calculated type factor, which can be input into a pre-constructed impurity recognition model. The model can be a supervised classifier such as support vector machine (SVM), decision tree, random forest, or a deep neural network model. The recognition model has learned the mapping relationship between different type factors and impurity types based on a large number of labeled samples in the training stage, and has the ability to quickly classify new input factors.

[0069] Then, the recognition model projects the input type factor in the multi-dimensional feature space and calculates its Euclidean distance or Mahalanobis distance from the center of the known sample, or calculates its classification probability after activation in the neural network. A confidence threshold can be set. When the recognition confidence is greater than the preset threshold, the specific impurity type label is output; otherwise, it is marked as "uncertain" or "manual review" to ensure the accuracy of recognition and system stability.

[0070] Further, the computer device can compare and verify the recognition result with the typical impurity samples in the database, and simultaneously display the recognized impurity type and its corresponding image feature information on the graphical user interface, so that the engineering personnel can quickly judge whether it needs to be rectified or retested on site. In a preferred embodiment, the recognition model is also supported based on the incremental learning mechanism for real-time optimization. That is, after the recognition result is manually confirmed, the confirmed type factor and its corresponding label can be fed back to the model for fine-tuning of the model parameters, thereby realizing self-learning and continuous evolution of the impurity type recognition system.

[0071] It can be understood that by determining the impurity type of the cable intermediate joint according to the type factor, complex image feature information can be converted into structured and classifiable specific impurity labels, thereby realizing explicit identification of impurity properties.

[0072] In the above embodiment, rapid and accurate identification of the type of impurities inside the intermediate joint can be achieved. Specifically, by radially emitting a microwave signal in the frequency band of 22-30 GHz and constructing a Smith circle image, the change in electromagnetic characteristics inside the joint is effectively reflected. Then, based on the image, a normalized gray level gradient co-occurrence matrix is constructed, and large gradient dominant features, gradient average features, and hybrid entropy features are extracted, which helps to fully characterize the microscopic interference characteristics caused by impurities to the internal structure of the joint. By calculating the type factor and determining the type of impurities, the differences in characteristics caused by different impurities can be distinguished, thereby providing support for subsequent fault warning and quality tracing, improving the intelligent and refined level of intermediate joint defect detection, and enhancing the safety and reliability of cable line operation.

[0073] In one embodiment, the step of constructing the normalized gray level gradient co-occurrence matrix corresponding to the Smith circle image comprises:

[0074] Reading the gray level value of each pixel in the Smith circle image, and for each pixel, calculating the square root of the sum of squares of the gray level value difference between the pixel and the adjacent right and lower pixels to obtain the gradient value corresponding to the pixel, generating a gradient matrix;

[0075] After discretization processing of the gradient matrix, an initial gray level gradient co-occurrence matrix is constructed.

[0076] The initial gray level gradient co-occurrence matrix is normalized to obtain a normalized gray level gradient co-occurrence matrix.

[0077] The gray level value refers to the brightness intensity represented by each pixel in the image, usually an integer between 0 and 255, used to represent the light and dark degree of the pixel. The gradient value refers to the change in gray level value between a pixel and its adjacent pixels in the image, used to reflect the local texture change of the image. The gradient matrix is a two-dimensional numerical matrix, where each element represents the gray level gradient intensity of the corresponding pixel in the image. The initial gray level gradient co-occurrence matrix is a two-dimensional matrix constructed by counting the frequency relationship of different gradient values appearing adjacent to each other in the image, used to capture the directionality and repetitive patterns of the image texture.

[0078] Specifically, the computer device loads the Smith circle image data to be processed from the memory or cache, and calls the image reading module to convert the image to a two-dimensional gray level matrix. Each element in the matrix corresponds to the gray level value of a pixel in the image. In this embodiment, 8-bit gray level coding can be preferably used for image representation to ensure that the image details are rich enough and facilitate subsequent calculation and processing.

[0079] Next, a gradient calculation operation is performed on each pixel in the grayscale image in turn. For each non-edge pixel in the image, the current grayscale value and the grayscale values of the right and lower pixels are read, and the square root of the sum of the squares of the grayscale differences between the current pixel and the two adjacent pixels is calculated, i.e.

[0080]

[0081] wherein, represents the grayscale value at the position , is the corresponding gradient value. The formula can effectively reflect the intensity of the grayscale change direction of the current pixel and form a gradient matrix consistent with the size of the original image.

[0082] Subsequently, the computer device performs discretization processing on the gradient matrix. The gradient values can be divided into several levels (such as 0-15, 16-31, etc.) according to the preset interval boundary, and each gradient value is replaced by the level number of the interval in which it is located, to generate a discrete gradient matrix. This step facilitates the subsequent construction of a statistical co-occurrence matrix and avoids statistical sparseness caused by too fine numerical differences.

[0083] On this basis, all adjacent pixel pairs in the discrete gradient matrix are scanned, such as horizontal and vertical directions, the number of occurrences of each adjacent pixel gradient level combination is counted, and an initial grayscale gradient co-occurrence matrix is constructed. Each row and each column of the matrix corresponds to a gradient level, and the element value in the matrix is the co-occurrence number of the level combination.

[0084] Finally, the computer device performs normalization processing on the initial grayscale gradient co-occurrence matrix, i.e., each element in the matrix is divided by the total co-occurrence number sum, so that the sum of all elements in the matrix is 1, thereby obtaining a normalized grayscale gradient co-occurrence matrix. Each element of the normalized matrix can be regarded as the relative probability distribution of a specific gradient combination in the entire image, facilitating subsequent feature calculation with comparability.

[0085] In one example, each pixel coordinate on the Smith chart image can be calculated by the formula , wherein is the modulus of the microwave reflection coefficient, is the phase, in radians, and are the horizontal and vertical coordinates of the image, respectively. After reading the Smith chart image, for each pixel, the square root of the sum of the squares of the grayscale value differences between the right and lower adjacent pixels is calculated to obtain the corresponding gradient value, forming a gradient matrix. After discretization of the gradient matrix, a grayscale gradient co-occurrence matrix is constructed, and the co-occurrence matrix is normalized to obtain the relative distribution information of different gradient combinations in the image.

[0086] In this embodiment, by reading the gray value of each pixel in the Smith chart image, and calculating the gray difference with the adjacent pixels to form the gradient matrix, the local gray change information of the image can be accurately extracted, reflecting the dielectric disturbance trace caused by the small impurities in the cable intermediate joint; the discretization of the gradient matrix and the construction of the initial gray gradient co-occurrence matrix are helpful to capture the texture directionality and structural repeatability; the normalized gray gradient co-occurrence matrix obtained after further normalization not only unifies the scale difference between different images, but also provides a stable probability basis for subsequent feature extraction and type identification. Therefore, the image texture structure information can be effectively converted into a quantifiable statistical model, improving the accuracy of subsequent classification and judgment and the robustness of the model, and enhancing the reliability of identifying impurities in complex working conditions.

[0087] In one embodiment, the calculation expression of the large gradient advantage feature is:

[0088]

[0089] wherein, represents the large gradient advantage feature, represents the sum of all elements in the normalized gray gradient co-occurrence matrix, represents each element in the normalized gray gradient co-occurrence matrix, represents the th gray level, represents the th gradient level.

[0090] Specifically, represents the matrix element value corresponding to the th gray level and the th gradient level in the normalized gray gradient co-occurrence matrix, i.e. the co-occurrence probability of gray and gradient ; represents the value of the current gradient level; represents the cumulative occurrence probability of all gray levels at the th gradient level.

[0091] This formula multiplies the co-occurrence probability of each gradient level by the corresponding gradient level value, and then adds all the gradient levels to finally obtain a numerical index that reflects whether the high gradient region in the image is dominant. The larger the value, the more obvious the high gradient texture feature in the image, and there may be more prominent impurity edges or structural discontinuity.

[0092] Therefore, the calculation method can effectively extract the gradient enhanced region features caused by impurities in the Smith circle image of the cable intermediate joint, thereby enhancing the difference of different impurity types in the image texture space, improving the perception ability of the impurity classification and recognition model to abnormal disturbance regions, and finally helping to improve the accuracy of impurity recognition and the stability of the system.

[0093] In one embodiment, the calculation expression of the gradient average feature is:

[0094]

[0095] wherein, represents the gradient average feature, represents the sum of all elements in the normalized gray gradient co-occurrence matrix, represents each element in the normalized gray gradient co-occurrence matrix, represents the gray level, represents the gradient level.

[0096] It can be understood that the formula reflects whether the gray gradient distribution of the whole image is balanced by calculating the weighted average of all gray levels. When the gradient average feature value is high, it means that the overall gray change in the image is intense, and there may be a large area of impurity disturbance or complex texture region; otherwise, it means that the overall image is relatively smooth, and the gray texture change is weak. Therefore, the calculation method can comprehensively reflect the influence of impurities on the overall texture distribution of the Smith circle image of the cable intermediate joint, assist in identifying impurity types that do not form obvious edges but have disturbance characteristics in a statistical sense, thereby improving the sensitivity of impurity detection and the integrity of image feature expression, and enhancing the recognition ability of the system in a low contrast scene.

[0097] In one embodiment, the calculation expression of the mixed entropy feature is:

[0098]

[0099] wherein, represents the mixed entropy feature, represents each element in the normalized gray gradient co-occurrence matrix, represents the gray level, represents the gradient level, represents a positive number used to avoid taking logarithm of zero.

[0100] It can be understood that the formula finally obtains the complexity of the texture distribution of the image by calculating the entropy of the cumulative value of the co-occurrence probability of all gradient levels under each gray level. The greater the value, the more dispersed the gradient distribution corresponding to each gray level in the image, the higher the information entropy, and the more complex the texture structure. The smaller the value, the more concentrated the texture distribution, and the simpler the structure. Therefore, using the hybrid entropy feature as an auxiliary feature helps to depict the change in texture complexity in the cable intermediate joint image caused by different impurities. By quantitatively reflecting the order degree of information distribution in the image, the perception ability of the impurity recognition model to abnormal features such as edge blur and non-concentrated distribution can be further improved, thereby improving the overall recognition robustness and adaptability in complex impurity interference scenarios.

[0101] In one embodiment, according to the large gradient advantage feature, the gradient average feature, and the hybrid entropy feature, the step of calculating a type factor for representing the type of the wrapping tape impurity of the cable intermediate joint includes:

[0102] The type factor is calculated according to the following expression:

[0103]

[0104] wherein, represents the type factor, represents the large gradient advantage feature, represents the gradient average feature, represents the hybrid entropy feature.

[0105] It can be understood that this calculation method fuses multiple texture statistical features through linear weighting, extracts a single scalar index that can be used for classification and discrimination while preserving the multi-dimensional information of the image. The positive weight coefficient in the formula is used to amplify the response to the target feature, and the negative weight coefficient is used to suppress the interference of the feature with low discrimination degree, thereby achieving the cooperative optimization of feature dimension reduction and discrimination ability. Therefore, through the calculation of the type factor, not only can the quantification of the influence degree of impurities in the cable intermediate joint image be realized, but also the discrimination degree between different impurity types can be effectively improved, which helps to realize subsequent accurate classification. Compared with the traditional method relying on a single image index, this method takes into account the overall gradient distribution, texture complexity, and strong edge features of the image, has stronger robustness and adaptability, can significantly improve the recognition accuracy of different types of impurities, and enhances the practicality and reliability of the cable quality intelligent detection system.

[0106] In one embodiment, according to the type factor, the step of determining the type of the impurity of the cable intermediate joint to be identified includes:

[0107] If the type factor is greater than zero and not greater than 0.24, the type of the impurity of the cable intermediate joint to be identified is air;

[0108] if the type factor is greater than 0.24 and not greater than 0.63, the impurity type of the cable intermediate joint to be identified is sand;

[0109] if the type factor is greater than 0.63 and not greater than 1, the impurity type of the cable intermediate joint to be identified is metal residual particles.

[0110] Specifically, the computer device obtains the type factor value calculated in the foregoing, and before performing the identification judgment, calls a preset type identification rule to perform interval matching judgment on the factor value. When the type factor is greater than 0 and not greater than 0.24, the identification result is marked as "air type impurity", which corresponds to defects such as wrapping not tight and gas mixing; when the type factor is greater than 0.24 and not greater than 0.63, the identification result is marked as "sand type impurity", which is usually caused by external dust pollution; when the type factor is greater than 0.63 and not greater than 1, the identification result is marked as "metal particle type impurity", which may be caused by residual wire stripping tool, construction debris and the like. Among them, each classification threshold value can be obtained by statistical modeling or supervised learning on a large number of labeled sample data. The computer device compares the type factor with the interval rule one by one, and after matching to the interval that meets the condition, automatically outputs the corresponding impurity type result.

[0111] It can be understood that by establishing a one-to-one correspondence between the value range of the type factor and the impurity type, the classification and identification of the internal impurities of the cable intermediate joint can be quickly and definitely completed after the feature calculation is completed, thereby improving the identification efficiency and the intelligent level of the system. The method of numerical interval determination avoids subjective interpretation or repeated processing of the original image, enhances the consistency and reproducibility of the identification result. In addition, the factor determination method based on image features has good expansibility and generalization ability, which can adapt to the change of impurity type under different batches of cables and different construction scenes, and finally improves the accuracy, stability and engineering practical value of the quality evaluation of the cable intermediate joint.

[0112] The 10kV cable intermediate joint impurity type identification device provided by the embodiment of the application is described below. The 10kV cable intermediate joint impurity type identification device described below can be correspondingly referred to the 10kV cable intermediate joint impurity type identification method described above. As shown in Figure 2 The present application provides a 10kV cable intermediate joint impurity type identification device, which comprises:

[0113] The Smith circle image determination module 201 is used for radially emitting a microwave signal of 22GHz~30GHz of the 10kV cable intermediate joint to be identified, and converts the obtained microwave reflection amplitude curve and phase curve into a Smith circle image;

[0114] The feature calculation module 202 is configured to construct a normalized gray gradient co-occurrence matrix corresponding to the Smith circle image, and calculate a large gradient dominant feature, a gradient average feature and a mixed entropy feature according to the normalized gray gradient co-occurrence matrix.

[0115] The type factor calculation module 203 is configured to calculate a type factor for representing the type of the wrapping tape impurity of the cable intermediate joint according to the large gradient dominant feature, the gradient average feature and the mixed entropy feature.

[0116] The impurity type determination module 204 is configured to determine the type of the impurity of the cable intermediate joint to be identified according to the type factor.

[0117] In an embodiment, the feature calculation module comprises:

[0118] The gradient matrix generation unit is configured to read the gray value of each pixel in the Smith circle image, and calculate, for each pixel, the square root of the sum of squares of the gray value difference between the pixel and the adjacent right pixel and the adjacent lower pixel to obtain the gradient value corresponding to the pixel, and generate a gradient matrix.

[0119] The initial gray gradient co-occurrence matrix construction unit is configured to construct an initial gray gradient co-occurrence matrix after discretization processing of the gradient matrix.

[0120] The normalized gray gradient co-occurrence matrix determination unit is configured to perform normalization processing on the initial gray gradient co-occurrence matrix to obtain a normalized gray gradient co-occurrence matrix.

[0121] In an embodiment, the calculation expression of the large gradient dominant feature is:

[0122]

[0123] wherein, represents the large gradient dominant feature, represents the sum of all elements in the normalized gray gradient co-occurrence matrix, represents each element in the normalized gray gradient co-occurrence matrix, represents the i-th gray level, represents the j-th gradient.

[0124] In an embodiment, the calculation expression of the gradient average feature is:

[0125]

[0126] wherein, represents the gradient average feature, represents the sum of all elements in the normalized gray gradient co-occurrence matrix, ​​denotes each element in the normalized gray level gradient co-occurrence matrix, denotes the gray level of the i-th order, denotes the gradient of the i-th order. denotes the gradient of the i-th order.

[0127] In one embodiment, the calculation expression of the hybrid entropy feature is:

[0128]

[0129] wherein, denotes the hybrid entropy feature, denotes each element in the normalized gray level gradient co-occurrence matrix, denotes the gray level of the i-th order, denotes the gradient of the i-th order, denotes the gradient of the i-th order, denotes a positive number for avoiding taking logarithm of zero. In one embodiment, the type factor calculation module comprises:

[0130] a type factor calculation unit, configured to calculate the type factor according to the following expression:

[0131]

[0132] wherein,

[0133] denotes the type factor, denotes the large gradient dominant feature, denotes the gradient average feature, denotes the hybrid entropy feature. In one embodiment, the impurity type determination module comprises:

[0134] a first impurity type determination unit, configured to determine that the impurity type of the cable intermediate joint to be identified is air if the type factor is greater than zero and not greater than 0.24;

[0135] a second impurity type determination unit, configured to determine that the impurity type of the cable intermediate joint to be identified is sand if the type factor is greater than 0.24 and not greater than 0.63;

[0136] a third impurity type determination unit, configured to determine that the impurity type of the cable intermediate joint to be identified is metal residual particle if the type factor is greater than 0.63 and not greater than 1.

[0137]

[0138] ​​In one embodiment, the present application also provides a storage medium having stored computer readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the method for identifying impurity types of 10kV cable intermediate joint as described in any of the above embodiments.

[0139] In one embodiment, the present application also provides a computer device having stored computer readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the method for identifying impurity types of 10kV cable intermediate joint as described in any of the above embodiments.

[0140] As shown in Figure 3 , Figure 3 An internal structure schematic diagram of a computer device provided by an embodiment of the present application is shown in FIG. 3. The computer device 300 can be provided as a server. As shown in Figure 3 , the computer device 300 includes a processing assembly 302, which further includes one or more processors, and a memory resource represented by a memory 301, for storing instructions, such as an application program, executable by the processing assembly 302. The application program stored in the memory 301 can include one or more than one module each corresponding to a set of instructions. In addition, the processing assembly 302 is configured to execute the instructions to perform the method for identifying impurity types of 10kV cable intermediate joint of any of the above embodiments.

[0141] The computer device 300 can also include a power supply assembly 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input output (I / O) interface 305. The computer device 300 can operate based on an operating system stored in the memory 301, such as Windows Server TM, Mac OS X TM, Unix TM, Linux TM, Free BSD TM, or the like.

[0142] Those skilled in the art can understand Figure 3 that the structure shown in FIG. 3 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0143] Finally, it should be noted that the terms "first" and "second", and the like, herein do not denote any order, quantity, combination or importance, but are used to identify one element from another, and do not imply that the specific identities thereof are essential or that the identities are chronological or related in their occurrence. Also, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element. Also, "a", "an", "the", and "said" are used to refer to one or more than one (i.e., to "at least one") of the referenced elements, unless otherwise specified. A plurality also means two or more, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of one or more of the associated listed items.

[0144] The various embodiments described in this specification are presented by way of example, and are not intended to limit the scope of the application. Each embodiment is presented in a way that emphasizes its particular features, and the embodiments can be combined according to the needs of the user. The same or similar parts are cross-referenced.

[0145] The above description of disclosed embodiments provides enabling disclosure sufficient for one of ordinary skill in the art to practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying the type of impurities in a 10kV cable joint, characterized in that, The method includes: A microwave signal with a radio frequency band of 22GHz~30GHz is inserted radially into the middle of a 10kV cable to be identified, and the obtained microwave reflection amplitude curve and phase curve are converted into a Smith circle image. Construct the normalized gray-level gradient co-occurrence matrix corresponding to the Smith circle image, and calculate the large gradient dominance feature, gradient average feature, and mixed entropy feature based on the normalized gray-level gradient co-occurrence matrix. Based on the large gradient dominance feature, the average gradient feature, and the mixed entropy feature, a type factor is calculated to represent the type of impurities in the cable joint wrapping tape, wherein the type factor is calculated according to the following expression: in, Indicates the type factor, This indicates the large gradient advantage feature. This represents the gradient average feature. This represents the hybrid entropy feature; The type of impurity in the cable joint to be identified is determined based on the type factor.

2. The method for identifying impurity types in a 10kV cable joint according to claim 1, characterized in that, The step of constructing the normalized gray-level gradient co-occurrence matrix corresponding to the Smith circle image includes: Read the grayscale value of each pixel in the Smith circle image, and for each pixel, calculate the square root of the sum of the squares of the grayscale value differences between the pixel and its adjacent right and bottom pixels to obtain the gradient value corresponding to the pixel and generate a gradient matrix. After discretizing the gradient matrix, an initial gray-level gradient co-occurrence matrix is ​​constructed. The initial gray-level gradient co-occurrence matrix is ​​normalized to obtain the normalized gray-level gradient co-occurrence matrix.

3. The method for identifying impurity types in a 10kV cable joint according to claim 1, characterized in that, The calculation expression for the large gradient advantage feature is as follows: in, This indicates the large gradient advantage feature. This represents the sum of all elements in the normalized gray-level gradient co-occurrence matrix. This represents each element in the normalized gray-level gradient co-occurrence matrix. Indicates the first grayscale, Indicates the first Level gradient.

4. The method for identifying impurity types in a 10kV cable intermediate joint according to claim 1, characterized in that, The expression for calculating the gradient average feature is as follows: in, This represents the gradient average feature. This represents the sum of all elements in the normalized gray-level gradient co-occurrence matrix. This represents each element in the normalized gray-level gradient co-occurrence matrix. Indicates the first grayscale, Indicates the first Level gradient.

5. The method for identifying impurity types in a 10kV cable joint according to claim 1, characterized in that, The calculation expression for the mixed entropy feature is as follows: in, This represents the mixed entropy feature. This represents each element in the normalized gray-level gradient co-occurrence matrix. Indicates the first grayscale, Indicates the first Level gradient, This indicates a positive number used to avoid taking the logarithm of zero.

6. The method for identifying impurity types in a 10kV cable joint according to claim 1, characterized in that, The step of determining the impurity type of the cable joint to be identified based on the type factor includes: If the type factor is greater than zero and not greater than 0.24, then the impurity type of the cable joint to be identified is air. If the type factor is greater than 0.24 and not greater than 0.63, then the impurity type of the cable joint to be identified is gravel; If the type factor is greater than 0.63 and not greater than 1, then the impurity type of the cable joint to be identified is metal residual particles.

7. A device for identifying the type of impurities in a 10kV cable joint, characterized in that, The device includes: The Smith circle image determination module is used to radially analyze microwave signals with a radio frequency band of 22GHz~30GHz that are transmitted between the midpoint of a 10kV cable to be identified, and convert the obtained microwave reflection amplitude curve and phase curve into a Smith circle image. The feature calculation module is used to construct the normalized gray-level gradient co-occurrence matrix corresponding to the Smith circle image, and to calculate the large gradient dominance feature, gradient average feature and mixed entropy feature based on the normalized gray-level gradient co-occurrence matrix. The type factor calculation module is used to calculate a type factor representing the impurity type of the cable joint wrapping tape based on the large gradient dominance feature, the gradient average feature, and the mixed entropy feature, wherein the type factor is calculated according to the following expression: in, Indicates the type factor, This indicates the large gradient advantage feature. This represents the gradient average feature. This represents the hybrid entropy feature; The impurity type determination module is used to determine the impurity type of the cable intermediate joint to be identified based on the type factor.

8. A storage medium, characterized in that: The storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the 10kV cable joint impurity type identification method as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the 10kV cable joint impurity type identification method as described in any one of claims 1 to 6.

Images