Cable joint abnormity early warning method and system

Abstract

The invention relates to the technical field of cable joint diagnosis, in particular to a cable joint abnormity early warning method and system, and the method comprises the steps: carrying out the feature extraction of infrared thermal image data and acoustic vibration data, and obtaining a multi-modal feature data set; quantifying the coupling correlation degree between the hot spot diffusion gradient and the vibration frequency component according to the multi-modal characteristic data set to obtain a regional thermal vibration coupling anomaly index; calculating a regional defect risk score according to the regional defect type distribution; obtaining a modal fusion weight distribution value based on the regional defect risk score and the regional thermal vibration coupling anomaly index; performing weighted fusion on the infrared thermal image feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector; and obtaining the defect probability of the cable joint according to the regional defect coupling feature vector, and carrying out graded early warning according to the defect probability of the cable joint. According to the method, the infrared thermal image and the acoustic vibration data are fused, so that accurate early warning of the abnormity of the cable joint is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of cable joint diagnosis, and in particular to a cable joint abnormality early warning method and system. Background Art

[0002] As a core component in the power transmission system, the operating status of cable joints is directly related to the safety and stability of the power grid. Especially in the complex operating environment of high voltage and high load, the internal connection interface of the cable joint is very likely to cause failures due to problems such as poor contact or material aging. If these faults are not discovered and handled in time, they are likely to cause serious power accidents and pose a huge threat to the normal operation of the power system. Therefore, how to effectively improve the identification accuracy of cable joint faults has become a key issue to ensure the reliable operation of the power system.

[0003] However, current methods for cable joint condition monitoring mostly rely on the analysis of a single data source. For example, fault diagnosis is performed solely based on infrared thermal imaging data or acoustic vibration data. However, this single-data-source analysis approach often fails to fully reflect the complex operating conditions within the joint. Due to the complex internal structure of cable joints and the diverse manifestations of faults, a single data source often fails to fully and accurately reflect the true internal condition of the joint. In practical applications, different regions of cable joints have different structural and functional characteristics, such as the crimping area, the insulation coating area, and the outer sheath overlap area. When a fault occurs within the joint, the characteristic responses of different regions also vary. For example, an abnormal temperature distribution may manifest as an irregular diffusion of the shape of a hot spot on the joint surface. This diffusion further affects the axial temperature gradient inside the crimped sleeve, presenting complex nonlinear characteristics. Acoustic vibration data may also exhibit different characteristics depending on the fault location and type. A single data source cannot simultaneously capture these complex changing characteristics. In the identification of defects in multiple regions and multiple structural types, misjudgments or omissions are prone to occur, thus limiting the accuracy and reliability of fault diagnosis.

[0004] In summary, the existing cable joint status monitoring technology has many shortcomings in defect identification. Therefore, there is an urgent need to provide an intelligent diagnostic cable joint early warning method that can accurately capture the correlation characteristics of abnormal temperature and mechanical vibration inside the joint to meet the high requirements of the power system for the reliable operation of cable joints. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a cable joint abnormality early warning method and system.

[0006] In a first aspect, the present invention provides a cable joint abnormality early warning method, the method comprising the following steps: Collecting infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time, and performing feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; quantifying the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set, and obtaining a regional thermal-vibration coupling anomaly index of the cable joint; Identifying regional defect type distributions based on regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculating regional defect risk scores of the cable joint in different regions based on the regional defect type distributions; Performing multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value; Performing weighted fusion on the infrared thermal image feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; The cable joint defect probability is predicted based on the regional defect coupling characteristic vector, and a graded warning is performed based on the cable joint defect probability.

[0007] In a further embodiment, the step of extracting features from the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set includes: Filtering the infrared thermal image data to obtain infrared thermal image filtered data, and converting the grayscale values ​​in the infrared thermal image filtered data into actual temperature values ​​to obtain a cable joint surface temperature field matrix; Calculating the first-order gradient components of the cable joint surface temperature field matrix in the horizontal and vertical directions using an edge detection operator to obtain temperature gradient components; Identifying a temperature anomaly region based on the temperature gradient component and extracting a thermal space feature vector of the temperature anomaly region; Performing a fast Fourier transform on the acoustic vibration data to obtain an acoustic energy spectrum, and identifying, from the acoustic energy spectrum, acoustic vibration frequency components corresponding to amplitudes exceeding a preset characteristic peak threshold; The thermal spatial feature vector and the acoustic vibration frequency component are aligned by time stamp, and the aligned thermal spatial feature vector and the acoustic vibration frequency component are arranged in time series order to form a multimodal feature data set.

[0008] In a further embodiment, the thermal space feature vector includes at least the geometric center coordinates, boundary contour perimeter, region area and temperature peak of the temperature anomaly region.

[0009] In a further embodiment, the step of quantifying the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain the regional thermal-vibration coupling anomaly index of the cable joint includes: Analyzing the temperature gradient mutation points according to the multimodal feature data set to identify the boundary contours of the hot spot area; Calculating the hot spot evolution amount according to the hot spot area boundary contours at adjacent time points, and constructing the hot spot diffusion gradient vector based on the hot spot evolution amount; the hot spot evolution amount includes at least the hot spot area change rate and the hot spot centroid displacement speed; Determining a hot spot centroid position based on the hot spot diffusion gradient vector, and extracting an acoustic vibration frequency component amplitude corresponding to the hot spot centroid position from the multimodal feature data set; Calculate the Pearson correlation coefficient between the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude to obtain the temperature vibration correlation coefficient matrix; The temperature vibration correlation coefficient matrix is ​​used as the correlation weight, and the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude are weightedly fused to obtain a cross-modal fusion feature; Based on the cross-modal fusion features, the least squares linear regression algorithm is used to obtain the spatial thermal-vibration coupling anomaly distribution, and the thermal-vibration coupling horizontal gradient component, thermal-vibration coupling vertical gradient component and thermal-vibration coupling gradient amplitude are calculated according to the thermal-vibration coupling anomaly distribution between adjacent time points; The thermal vibration coupling horizontal gradient component, the thermal vibration coupling vertical gradient component and the thermal vibration coupling gradient amplitude are weightedly summed to obtain a regional thermal vibration coupling anomaly index of the cable joint.

[0010] In a further embodiment, the hot spot diffusion gradient vector includes a diffusion rate scalar value and a diffusion direction unit vector; wherein, the diffusion rate scalar value is the vector modulus of the hot spot area change rate and the hot spot center of mass displacement velocity; the diffusion direction unit vector is the normalized vector of the hot spot center of mass displacement direction.

[0011] In a further embodiment, the step of identifying the regional defect type distribution based on the regional structural characteristic parameters and the material mechanical resonance characteristics of the cable joint includes: Divide the regional space boundary coordinates of the conductor connection area, the insulation layer area and the outer sheath area according to the regional structural characteristic parameters of the cable joint; Mapping the regional thermal-vibration coupling anomaly index to the regional spatial boundary coordinates according to the spatial position, and calculating the arithmetic mean and maximum value of the regional thermal-vibration coupling anomaly index in each area of ​​the cable joint to identify the defective area of ​​the cable joint; Determining a characteristic frequency band corresponding to the defective area of ​​the cable joint based on the mechanical resonance characteristics of the material, and obtaining the characteristic frequency band of the defective area of ​​the cable joint; Calculate the relative change rate of the frequency band energy of the characteristic frequency band of each cable joint defect area relative to the reference value to obtain the vibration energy change rate of the cable joint defect area; The regional defect type of each cable joint defect area is identified according to the vibration energy change rate, and the regional defect type distribution is formed.

[0012] In a further embodiment, the step of calculating regional defect risk scores of cable joints in different regions based on the regional defect type distribution includes: Extract the regional defect types corresponding to the conductor connection area, insulation layer area, and outer sheath area from the regional defect type distribution, and obtain the defect duration exceeding limit and defect exceeding limit ratio corresponding to each regional defect type; Calculate the ratio of the duration of the defect exceeding the limit to the total monitoring time, generate a time enhancement factor, and use the time enhancement factor to correct the defect exceeding limit ratio value to obtain an effective exceeding limit value; Calculating a regional defect score based on the effective limit value and a preset defect type basic weight coefficient; When defects exist simultaneously in adjacent areas of the cable joint, the regional defect scores are corrected using a preset cross-region coupling coefficient to obtain a corrected regional defect score; The corrected regional defect scores are normalized to obtain the regional defect risk scores of the cable joints in each area.

[0013] In a further embodiment, the modal fusion weight distribution value includes a temperature modal weight distribution value and an acoustic modal weight distribution value, and the step of performing multimodal weight distribution based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain the modal fusion weight distribution value includes: Determine the key monitoring area range of cable joints in each area based on the regional defect risk score; The regional thermal-vibration coupling anomaly index is used to calculate the information entropy of the temperature anomaly gradient and the amplitude of the acoustic vibration frequency component within the important monitoring area, and the corresponding temperature modal weight distribution value and acoustic modal weight distribution value are obtained.

[0014] In a further embodiment, the step of predicting the cable joint defect probability based on the regional defect coupling characteristic vector includes: Inputting the regional defect coupling feature vector into a random forest classifier integrating multiple decision trees, performing classification and judgment on the regional defect coupling feature vector through each decision tree, and obtaining the defect probability votes of each decision tree; The highest defect votes are selected from the defect probability votes of each decision tree, and the proportion of the highest defect votes to the total number of decision trees is calculated to obtain the cable joint defect probability.

[0015] In a second aspect, the present invention provides a cable joint abnormality warning system, the system comprising: a feature extraction module for collecting infrared thermal imaging data of the cable connector surface and acoustic vibration data generated by the vibration of the cable connector in real time, and performing feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; A temperature analysis module is used to quantify the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution based on the multimodal feature data set, and obtain a regional thermal-vibration coupling anomaly index of the cable joint; A defect analysis module is used to identify the regional defect type distribution based on the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculate the regional defect risk score of the cable joint in different areas based on the regional defect type distribution; A weight allocation module, configured to perform multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value; A modal fusion module, configured to perform weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; The defect warning module is used to predict the cable joint defect probability based on the regional defect coupling characteristic vector and perform graded warning based on the cable joint defect probability.

[0016] The present invention provides a cable joint abnormality warning method and system, which collects infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time, and extracts features from the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; quantifies the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain a regional thermal vibration coupling abnormality index of the cable joint; identifies the regional defect type distribution according to the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculates the regional defect risk scores of the cable joint in different regions according to the regional defect type distribution; performs multimodal weight distribution based on the regional defect risk score to obtain a modal fusion weight distribution value; performs weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; predicts the cable joint defect probability according to the regional defect coupling feature vector, and performs a graded warning according to the cable joint defect probability. Compared with the existing technology, this method realizes multimodal feature weighted fusion and defect probability prediction by fusing infrared thermal images and acoustic vibration data, achieving accurate early warning of cable joint abnormalities, and effectively improving the accuracy of cable joint abnormality detection and the timeliness of early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a cable joint abnormality warning method provided by an embodiment of the present invention; Figure 2 This is a block diagram of a cable joint abnormality warning system provided by an embodiment of the present invention.

[0018] Explanation of the accompanying drawings: 101, feature extraction module; 102, temperature analysis module; 103, defect analysis module; 104, weight allocation module; 105, modal fusion module; 106, defect warning module. DETAILED DESCRIPTION

[0019] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings. The embodiments are provided for illustrative purposes only and are not to be construed as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention. Many changes may be made to the present invention without departing from the spirit and scope of the present invention.

[0020] refer to Figure 1 , the embodiment of the present invention provides a cable joint abnormality warning method, such as Figure 1 As shown, the method includes the following steps: S1. Collect infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time, and perform feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set.

[0021] In some embodiments, the step of extracting features from the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature dataset includes: Filtering the infrared thermal image data to obtain infrared thermal image filtered data, and converting the grayscale values ​​in the infrared thermal image filtered data into actual temperature values ​​to obtain a cable joint surface temperature field matrix; Calculating the first-order gradient components of the cable joint surface temperature field matrix in the horizontal and vertical directions using an edge detection operator to obtain temperature gradient components; Identifying a temperature anomaly region based on the temperature gradient component and extracting a thermal space feature vector of the temperature anomaly region; Performing a fast Fourier transform on the acoustic vibration data to obtain an acoustic energy spectrum, and identifying, from the acoustic energy spectrum, acoustic vibration frequency components corresponding to amplitudes exceeding a preset characteristic peak threshold; The thermal spatial feature vector and the acoustic vibration frequency component are aligned by time stamp, and the aligned thermal spatial feature vector and the acoustic vibration frequency component are arranged in time series order to form a multimodal feature data set.

[0022] Specifically, this embodiment uses a high-resolution infrared thermal imager to scan the surface of the cable connector in real time. The resolution of the infrared thermal imager can be set to 640×480 pixels, and the scanning frequency is 30Hz, ensuring that tiny temperature changes on the cable connector surface can be accurately captured. In this embodiment, the infrared thermal imager is installed at a position close to the cable connector to ensure that its field of view can completely cover the cable connector and a certain range of areas around it, avoiding data loss due to visual obstruction. The infrared thermal imager collects infrared radiation data from the cable connector surface in a non-contact manner, converts it into a digital signal, and stores it as raw infrared thermal image data. At the same time, this embodiment uses a high-sensitivity acoustic sensor installed on the housing of the cable connector or on the equipment structure connected to the cable connector. The frequency response range of the high-sensitivity acoustic sensor should cover the main frequency range of the cable connector vibration, ensuring that the acoustic vibration signal generated by the cable connector due to vibration can be stably and reliably obtained. The sampling frequency of the acoustic sensor can be set to 20kHz to ensure that high-frequency vibration signals can be captured. The collected acoustic vibration signal is converted into a digital signal by a data acquisition card and stored as raw acoustic vibration data.

[0023] In this embodiment, a Gaussian filter algorithm is used to perform denoising on the original infrared thermal image data to eliminate the influence of environmental noise and sensor noise, thereby obtaining infrared thermal image filtered data. The kernel size of the Gaussian filter can be set to 5×5 pixels with a standard deviation of 1.5. Then, according to the calibration curve of the infrared thermal imager (calibrated in advance by a blackbody radiation source), the grayscale value of each pixel in the infrared thermal image filtered data is converted into an actual temperature value. The converted data forms a cable joint surface temperature field matrix. The dimension of the temperature field matrix is ​​consistent with the image size of the infrared thermal image data. Each element of the cable joint surface temperature field matrix corresponds to the temperature value of a pixel point on the cable joint surface. In this embodiment, a Sobel edge detection operator is used to perform a convolution operation on the cable joint surface temperature field matrix to calculate the first-order horizontal direction (x direction) and vertical direction (y direction) respectively. Gradient component, obtain horizontal gradient component and vertical gradient component, and calculate the arithmetic square root of the sum of the squares of the horizontal gradient component and the vertical gradient component, that is, sum the squares of the horizontal gradient component and the vertical gradient component and take the square root to obtain the temperature gradient amplitude. The temperature gradient amplitude indicates the severity of the temperature change on the surface of the cable joint. In this embodiment, the temperature gradient amplitude is used as the temperature gradient component, and the connected region analysis algorithm is used to segment the temperature gradient component. The area where the temperature gradient amplitude exceeds the preset gradient amplitude threshold is marked as a temperature abnormality area. For each temperature abnormality area, thermal space features such as geometric center coordinates, boundary contour perimeter, regional temperature peak, regional average temperature, mean of regional temperature gradient amplitude, variance of regional temperature gradient amplitude and regional area (number of pixels) are extracted to form a thermal space feature vector.

[0024] At the same time, this embodiment performs Hanning windowing processing on the acoustic vibration data, converts the acoustic vibration data into a frequency domain signal through fast Fourier transform, and calculates the amplitude square of the fast Fourier transform result to obtain an acoustic energy spectrum diagram, which represents the energy distribution of different frequency components. Then, the frequency points in the acoustic energy spectrum diagram whose amplitude exceeds the preset characteristic peak threshold are identified, and the vibration center frequency and normalized vibration energy ratio corresponding to the frequency point are used as the acoustic vibration characteristic frequency component. In this embodiment, the preset characteristic peak threshold can be set to 50% of the maximum amplitude of the acoustic energy spectrum diagram. Finally, this embodiment adds a timestamp to each thermal space feature vector and acoustic vibration frequency component, uses nearest neighbor interpolation to align the thermal space feature vector and the acoustic vibration frequency component to the same time base, and arranges the aligned thermal space feature vector and acoustic vibration frequency component in time series order to form a multimodal feature data set.

[0025] S2. quantify the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution based on the multimodal feature data set to obtain the regional thermal-vibration coupling anomaly index of the cable joint.

[0026] In some embodiments, the step of quantifying the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain the regional thermal-vibration coupling anomaly index of the cable joint includes: Analyzing the temperature gradient mutation points according to the multimodal feature data set to identify the boundary contours of the hot spot area; Calculating the hot spot evolution amount according to the hot spot area boundary contours at adjacent time points, and constructing the hot spot diffusion gradient vector based on the hot spot evolution amount; the hot spot evolution amount includes at least the hot spot area change rate and the hot spot centroid displacement speed; Determining a hot spot centroid position based on the hot spot diffusion gradient vector, and extracting an acoustic vibration frequency component amplitude corresponding to the hot spot centroid position from the multimodal feature data set; Calculate the Pearson correlation coefficient between the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude to obtain the temperature vibration correlation coefficient matrix; The temperature vibration correlation coefficient matrix is ​​used as the correlation weight, and the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude are weightedly fused to obtain a cross-modal fusion feature; Based on the cross-modal fusion features, the least squares linear regression algorithm is used to obtain the spatial thermal-vibration coupling anomaly distribution, and the thermal-vibration coupling horizontal gradient component, thermal-vibration coupling vertical gradient component and thermal-vibration coupling gradient amplitude are calculated according to the thermal-vibration coupling anomaly distribution between adjacent time points; The thermal vibration coupling horizontal gradient component, the thermal vibration coupling vertical gradient component and the thermal vibration coupling gradient amplitude are weightedly summed to obtain a regional thermal vibration coupling anomaly index of the cable joint.

[0027] Specifically, this embodiment extracts the temperature gradient component of each time point from the multimodal feature data set, and performs a second-order difference operation on the temperature gradient component, marking the points exceeding the preset second-order difference threshold as temperature gradient mutation points. These temperature gradient mutation points reflect the edge of the hot spot or the abnormal change position on the heat conduction path. Then, this embodiment adopts a region growing algorithm, takes the temperature gradient mutation point as the seed point, and performs region growth according to the similarity of the temperature gradient value. Starting from the seed point, according to certain similarity criteria (such as the temperature gradient change is within a certain range), adjacent pixels are gradually classified into the same hot spot area until no pixels that meet the conditions can be found. The Canny edge detection algorithm is used to perform edge detection on the grown area to obtain the hot spot. The boundary contour of the region. For each hot spot region, this embodiment obtains the area of ​​adjacent time points by counting the number of pixels within the boundary contour of the hot spot region, and obtains the hot spot area change rate by calculating the ratio of the difference in the hot spot area between the current time point and the next time point to the area of ​​the hot spot region at the current time point. At the same time, the displacement velocity of the hot spot centroid is calculated based on the ratio of the displacement distance of the hot spot centroid between adjacent time points to the time interval. The position of the hot spot centroid can be obtained by averaging the coordinates of all pixel points in the hot spot region. This embodiment obtains the diffusion rate scalar value by calculating the vector modulus of the hot spot area change rate and the hot spot centroid displacement velocity, and determines the direction of the hot spot centroid displacement, which is used as the diffusion direction. The displacement vector is normalized, that is, the displacement vector is divided by its modulus to obtain the diffusion direction unit vector. The diffusion rate scalar value and the diffusion direction unit vector are combined to form the hot spot diffusion gradient vector. The hot spot diffusion gradient vector describes the diffusion of the hot spot in space.

[0028] This embodiment updates the hot spot centroid position coordinates based on the diffusion direction unit vector and the diffusion rate scalar value in the hot spot diffusion gradient vector. In each time step (such as 1 second), the current centroid position coordinates are moved along the diffusion direction unit vector by the distance corresponding to the diffusion rate scalar value to obtain new centroid position coordinates. Then, the acoustic vibration frequency component corresponding to the current time point is found from the multimodal feature data set, and the acoustic vibration frequency component amplitude that is closest to or corresponding to the new hot spot centroid position is extracted from the acoustic vibration frequency component. For the convenience of statistics, this embodiment can average the acoustic vibration frequency component amplitudes in a small area to obtain the acoustic vibration frequency component amplitude corresponding to the hot spot centroid position. It should be noted that the Pearson correlation coefficient is used to measure the linear correlation between two variables. For the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude, this embodiment considers the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude respectively. The correlation between the diffusion rate scalar value of the gradient vector and the diffusion direction unit vector and the acoustic vibration frequency component amplitude. For example, for the diffusion rate scalar value sequence and the acoustic vibration frequency component amplitude sequence, this embodiment first calculates the mean of the diffusion rate scalar value sequence and the acoustic vibration frequency component amplitude sequence, then calculates the covariance and standard deviation based on the mean of the diffusion rate scalar value sequence and the acoustic vibration frequency component amplitude sequence, and finally divides the covariance of the diffusion rate scalar value and the acoustic vibration frequency component amplitude by the product of the standard deviations of the two to calculate the Pearson correlation coefficient between the diffusion rate scalar value sequence and the acoustic vibration frequency component amplitude sequence. Repeat the above steps. This embodiment calculates the Pearson correlation coefficients between the diffusion rate scalar value and the acoustic vibration frequency component amplitude, the diffusion direction horizontal component and the acoustic vibration frequency component amplitude, and the diffusion direction vertical component and the acoustic vibration frequency component amplitude, thereby obtaining a temperature vibration correlation coefficient matrix.

[0029] For each hot spot, this embodiment uses the elements in the temperature vibration correlation coefficient matrix as the association weight, and performs weighted summation on the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude. This embodiment combines the hot spot diffusion gradient information and the acoustic vibration frequency component information through weighted fusion to obtain a cross-modal fusion feature. The cross-modal fusion feature reflects the coupling relationship between hot spot diffusion and vibration. Then, this embodiment grids the cross-modal fusion feature in space, and each grid point corresponds to a cross-modal fusion feature vector. The least squares linear regression algorithm is used to fit the cross-modal fusion features on the grid points to obtain the spatial thermal-vibration coupling anomaly distribution. This embodiment fits the spatial thermal-vibration coupling anomaly distribution with respect to the horizontal direction. The partial derivative (x-direction) is taken to obtain the horizontal gradient component of the thermal-vibration coupling. At the same time, the partial derivative of the spatial thermal-vibration coupling anomaly distribution with respect to the vertical direction (y-direction) is taken to obtain the vertical gradient component of the thermal-vibration coupling. The arithmetic square root of the sum of the squares of the horizontal gradient component of the thermal-vibration coupling and the vertical gradient component of the thermal-vibration coupling is calculated, that is, the square root of the sum of the squares of the horizontal gradient component of the thermal-vibration coupling and the vertical gradient component of the thermal-vibration coupling is taken to obtain the thermal-vibration coupling gradient amplitude. Finally, this embodiment uses a weighted summation method to integrate the horizontal gradient component of the thermal-vibration coupling, the vertical gradient component of the thermal-vibration coupling, and the thermal-vibration coupling gradient amplitude to obtain the regional thermal-vibration coupling anomaly index of the cable joint. The regional thermal-vibration coupling anomaly index can be used to assess the degree of thermal-vibration coupling anomaly in the cable joint region.

[0030] S3. Identify the regional defect type distribution based on the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculate the regional defect risk scores of the cable joint in different areas based on the regional defect type distribution.

[0031] In some embodiments, the step of identifying the regional defect type distribution based on the regional structural characteristic parameters and the material mechanical resonance characteristics of the cable joint includes: Divide the regional space boundary coordinates of the conductor connection area, the insulation layer area and the outer sheath area according to the regional structural characteristic parameters of the cable joint; Mapping the regional thermal-vibration coupling anomaly index to the regional spatial boundary coordinates according to the spatial position, and calculating the arithmetic mean and maximum value of the regional thermal-vibration coupling anomaly index in each area of ​​the cable joint to identify the defective area of ​​the cable joint; Determining a characteristic frequency band corresponding to the defective area of ​​the cable joint based on the mechanical resonance characteristics of the material, and obtaining the characteristic frequency band of the defective area of ​​the cable joint; Calculate the relative change rate of the frequency band energy of the characteristic frequency band of each cable joint defect area relative to the reference value to obtain the vibration energy change rate of the cable joint defect area; The regional defect type of each cable joint defect area is identified according to the vibration energy change rate, and the regional defect type distribution is formed.

[0032] Specifically, this embodiment obtains regional structural characteristic parameters of the cable joint from the structural parameter data set of the cable joint. The structural characteristic parameters of the cable joint may include the conductor cross-sectional area, the insulation layer thickness and the outer sheath material density, etc., and divides the surface of the cable joint into a conductor connection area, an insulation layer area and an outer sheath area according to the regional structural characteristic parameters, wherein the conductor connection area corresponds to the crimping part of the cable core wire, and the spatial boundary is determined by the coordinates of the metal shielding layer; the insulation layer area covers the surface of the insulating medium, and the boundary is the junction line between the outer semi-conductive layer and the sheath; the outer sheath area is the outermost protective structure, and the boundary is defined by the geometric dimensions of the joint shell, thereby determining the regional spatial boundary coordinates of the conductor connection area, the insulation layer area and the outer sheath area. This embodiment grids the cable joint in space, and each grid point corresponds to a spatial coordinate, and the grid points are divided according to the regional boundaries. The conductor connection area, the insulation layer area, and the outer sheath area are then mapped in this embodiment to the regional space boundary coordinates of the regional thermal vibration coupling anomaly index according to the pixel space position coordinates. For example, for each grid point, its regional thermal vibration coupling anomaly index value is assigned to the corresponding area. The regional thermal vibration coupling anomaly index in each area (conductor connection area, insulation layer area, and outer sheath area) is statistically analyzed, and the arithmetic mean and maximum value of the regional thermal vibration coupling anomaly index in each area are calculated to obtain the regional average value and regional maximum value of the regional thermal vibration coupling anomaly index in each area. The cable joint defect area is identified based on the regional average value and regional maximum value. For example, in this embodiment, a coupling anomaly threshold can be set. When the regional average value or regional maximum value of the regional thermal vibration coupling anomaly index in a certain area exceeds the coupling anomaly threshold, the area is considered to be a defective area.

[0033] At the same time, this embodiment calls a preset material mechanical resonance characteristic database according to the material type of the candidate defect area (such as copper conductor, XLPE insulation or PVC sheath), and obtains the material mechanical resonance characteristic parameters of each area of ​​the cable joint (conductor connection area, insulation layer area, outer sheath area) from the material mechanical resonance characteristic database. The material mechanical resonance characteristic parameters include natural frequency, damping ratio, etc., and according to the material mechanical resonance characteristics, the characteristic frequency band corresponding to the defective area of ​​the cable joint is determined. For example, the material mechanical resonance characteristics of the defective area may change, resulting in its natural frequency shift. By analyzing the material mechanical resonance characteristics of the defective area of ​​the cable joint, its characteristic frequency band can be determined. Assuming that the natural frequency of the normal area is , the natural frequency shift of the defect area is , then the characteristic frequency band is In this embodiment, a spectrum analysis is performed on the acoustic vibration signal of the cable joint, and the portion of the frequency in the spectrum that is in the characteristic frequency band is integrated or summed to obtain the frequency band energy of the characteristic frequency band of each defect area. In this embodiment, the relative energy change rate of the frequency band energy of the characteristic frequency band of the defect area relative to the baseline energy value of the historical normal state of the same area is calculated to obtain the vibration energy change rate. The defect type of the defect area is identified based on the magnitude of the vibration energy change rate. For example, in this embodiment, a vibration energy change rate range corresponding to different defect types is set. When the vibration energy change rate of the conductor connection area is greater than 200%, it is identified as a poor contact defect; when the vibration energy change rate of the insulation layer area is greater than 150%, it is identified as an insulation degradation defect; and when the vibration energy change rate of the outer sheath area is greater than 120%, it is identified as a mechanical looseness defect. In this embodiment, the identified defect type is mapped to the spatial area of ​​the cable joint and output in the form of spatial coordinate mapping to form a regional defect type distribution containing defect type labels. For areas that simultaneously meet the requirements of multiple defect types, they are classified according to the type corresponding to the maximum energy change rate.

[0034] In some embodiments, the step of calculating regional defect risk scores of cable joints in different regions based on the regional defect type distribution includes: Extract the regional defect types corresponding to the conductor connection area, insulation layer area, and outer sheath area from the regional defect type distribution, and obtain the defect duration exceeding limit and defect exceeding limit ratio corresponding to each regional defect type; Calculate the ratio of the duration of the defect exceeding the limit to the total monitoring time, generate a time enhancement factor, and use the time enhancement factor to correct the defect exceeding limit ratio value to obtain an effective exceeding limit value; Calculating a regional defect score based on the effective limit value and a preset defect type basic weight coefficient; When defects exist simultaneously in adjacent areas of the cable joint, the regional defect scores are corrected using a preset cross-region coupling coefficient to obtain a corrected regional defect score; The corrected regional defect scores are normalized to obtain the regional defect risk scores of the cable joints in each area.

[0035] Specifically, this embodiment extracts the regional defect types corresponding to the conductor connection area, the insulation layer area, and the outer sheath area from the regional defect type distribution data set. For each area, the cumulative time length of the continuous existence of the defect state exceeds the preset time threshold is counted to obtain the defect duration exceeding limit time corresponding to the defect type of each area. At the same time, this embodiment calculates the defect exceeding limit ratio value of each area. Assuming that the number of monitoring times for each area is N times within the monitoring period, and the number of times the defect exceeding limit is detected is n times, then the defect exceeding limit ratio value is For example, if the conductor connection area is monitored 100 times and the number of times the defect exceeds the limit is 30 times, the defect exceeding the limit ratio value is 0.3.

[0036] For each area, this embodiment calculates the square value of the ratio of the duration of the defect exceeding the limit to the total monitoring time, and generates a time enhancement factor. The time enhancement factor reflects the relative importance of the defect duration. The longer the duration, the larger the time enhancement factor. The time enhancement factor is used to correct the defect exceeding limit ratio value. The time enhancement factor and the defect exceeding limit ratio value are multiplied to obtain the effective exceeding limit value. When the effective exceeding limit value exceeds 100%, it is forcibly set to 100%. This embodiment sets basic weight coefficients for different defect types based on historical experience and an assessment of the severity of defects in various areas of the cable joint. The effective exceeding limit value of each area is multiplied by the corresponding basic weight coefficient of the defect type to calculate the regional defect score of each area. At the same time, this embodiment checks whether there is an adjacent relationship between the various areas of the cable joint, and determines whether the adjacent areas have defects. For example, the conductor connection area and When the insulation layer areas are adjacent and there are defects in the insulation layer areas, and there are defects in the adjacent areas of the cable joint at the same time, the cross-region coupling coefficient between the adjacent areas is set according to the coupling tightness between the areas. For example, the cross-region coupling coefficient between the conductor connection area and the insulation layer area is 0.5; the cross-region coupling coefficient between the insulation layer area and the outer sheath area is 0.4. For areas with adjacent defects, the cross-region coupling coefficient is used to correct the regional defect score. The regional defect score of the area is calculated by adding the product of the cross-region coupling coefficient and the regional defect score to obtain the corrected regional defect score. The cross-region coupling coefficient reflects the degree of mutual influence between the defects in adjacent areas. The larger the coupling coefficient, the greater the correction amplitude. The corrected regional defect score is normalized so that its range is between [0, 1], where 0 indicates no risk and 1 indicates extremely high risk.

[0037] S4. Perform multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value.

[0038] In some embodiments, the modal fusion weight distribution value includes a temperature modal weight distribution value and an acoustic modal weight distribution value, and the step of performing multi-modal weight distribution based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain the modal fusion weight distribution value includes: Determine the key monitoring area range of cable joints in each area based on the regional defect risk score; The regional thermal-vibration coupling anomaly index is used to calculate the information entropy of the temperature anomaly gradient and the amplitude of the acoustic vibration frequency component within the important monitoring area, and the corresponding temperature modal weight distribution value and acoustic modal weight distribution value are obtained.

[0039] Specifically, this embodiment traverses all areas of the cable joint and determines the important monitoring area range of each area according to the level of the regional defect risk score. For example, this embodiment sets a risk score threshold. When the regional defect risk score of a certain area exceeds the risk score threshold, the area is considered to be an important monitoring area. When the regional thermal vibration coupling anomaly index of a certain area exceeds the upper limit of the historical normal fluctuation range, the coordinates of the area are added to the important monitoring area. This embodiment compresses the regional thermal vibration coupling anomaly index of all pixels in the key monitoring area to between 0 and 1, maps the minimum regional thermal vibration coupling anomaly index to 0, and maps the maximum regional thermal vibration coupling anomaly index to 1. The values ​​of the intermediate regional thermal vibration coupling anomaly indices are linearly scaled to obtain the normalized regional thermal vibration coupling anomaly index. This embodiment calculates the arithmetic mean of the normalized regional thermal vibration coupling anomaly indices as the global coupling strength coefficient. This embodiment takes the value of 1 minus the global coupling strength coefficient as the value of the normalized regional thermal vibration coupling anomaly index. Entropy correction coefficient. It should be noted that the stronger the coupling, the smaller the correction coefficient, which suppresses the influence of the entropy value. In the important monitoring area, this embodiment counts the directional distribution of the temperature anomaly gradient of all pixels, calculates the proportion of the number of pixels in each directional partition to the total number of pixels, and obtains the probability distribution of the occurrence of temperature anomaly gradients in different directions. The original temperature gradient information entropy of the temperature anomaly gradient in the conductor connection area is calculated based on the probability distribution of the occurrence of temperature anomaly gradients in different directions. The original temperature gradient information entropy is equal to the probability distribution multiplied by the sum of the opposite numbers of the logarithms of the corresponding probability distribution with base 2. The larger the temperature gradient information entropy value, the more disordered the gradient direction. The original temperature gradient information entropy is multiplied by the entropy correction coefficient to correct the original temperature gradient information entropy by the entropy correction coefficient to obtain the corrected temperature entropy value. This embodiment calculates the initial temperature modal weight distribution value based on the regional defect risk score and the corrected temperature entropy value. The calculation formula of the initial temperature modal weight distribution value is: Where, Assign values ​​to the initial temperature modal weights; Score regional defect risk; is the corrected temperature entropy value.

[0040] At the same time, this embodiment calculates the ratio distribution between the amplitude of the acoustic vibration frequency component and the total spectral energy of all vibration measuring points in the important monitoring area, and calculates the original acoustic frequency information entropy of the amplitude of the acoustic vibration frequency component in the conductor connection area based on the ratio distribution. For each important monitoring area, this embodiment multiplies the original acoustic frequency information entropy and the entropy correction coefficient, and corrects the original acoustic frequency information entropy by the entropy correction coefficient to obtain a corrected acoustic entropy value. This embodiment calculates the initial acoustic modal weight distribution value based on the regional thermal vibration coupling anomaly index and the corrected acoustic entropy value. The calculation formula of the initial acoustic modal weight distribution value is: Where, Assign values ​​to the initial acoustic modal weights; is the regional thermal-vibration coupling anomaly index; is the corrected acoustic entropy value.

[0041] This embodiment calculates the modal weight sum of the initial temperature modal weight distribution value and the initial acoustic modal weight distribution value, and calculates the ratio of the initial temperature modal weight distribution value to the modal weight sum to obtain the final temperature modal weight distribution value, and calculates the ratio of the initial acoustic modal weight distribution value to the modal weight sum to obtain the final acoustic modal weight distribution value.

[0042] S5. Perform weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint.

[0043] S6. Predict the cable joint defect probability based on the regional defect coupling characteristic vector, and perform graded warning based on the cable joint defect probability.

[0044] In some embodiments, the step of predicting the cable joint defect probability based on the regional defect coupling characteristic vector includes: Inputting the regional defect coupling feature vector into a random forest classifier integrating multiple decision trees, performing classification and judgment on the regional defect coupling feature vector through each decision tree, and obtaining the defect probability votes of each decision tree; The highest defect votes are selected from the defect probability votes of each decision tree, and the proportion of the highest defect votes to the total number of decision trees is calculated to obtain the cable joint defect probability.

[0045] Specifically, the modal fusion weight distribution value is used to perform weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data. Specifically, the infrared thermal imaging feature data is multiplied by the temperature modal weight distribution value to obtain an infrared feature vector, and the acoustic vibration feature data is multiplied by the acoustic modal weight distribution value to obtain an acoustic feature vector. The infrared feature vector and the acoustic feature vector are connected end to end to generate a regional defect coupling feature vector. The dimension of the regional defect coupling feature vector is the sum of the dimensions of the infrared thermal imaging feature data and the acoustic vibration feature data. The weighted fused regional defect coupling feature vector integrates the information of infrared thermal imaging and acoustic vibration, and can more comprehensively reflect the defect status of the cable joint. At the same time, this embodiment constructs a random forest classifier that integrates multiple decision trees. Each decision tree is trained by random sampling and random feature selection. This embodiment inputs the regional defect coupling feature vector into the pre-trained random forest classifier, traverses from the root node to the leaf node of the decision tree, and each decision tree classifies and judges the regional defect coupling feature vector. The leaf node outputs the decision tree. The defect type prediction result is obtained. For example, each decision tree determines whether the cable joint has a defect based on the eigenvalue of the regional defect coupling eigenvector and outputs the defect probability votes. The defect probability votes obtained by each decision tree for the regional defect coupling eigenvector are counted. The highest defect votes are screened out from the defect probability votes of each decision tree. The proportion of the highest defect votes to the total number of decision trees is calculated to obtain the cable joint defect probability. The cable joint defect probability reflects the possibility of a defect in the cable joint. The higher the probability, the greater the defect risk. This embodiment performs graded warnings based on the level of the cable joint defect probability. For example, this embodiment can set the grading standard as follows: when the defect probability is less than 0.3, a level 1 warning (normal) is determined; when the defect probability is between 0.3 and 0.7, a level 2 warning (minor defect) is determined; and when the defect probability is greater than or equal to 0.7, a level 3 warning (serious defect) is determined. This embodiment outputs corresponding warning information based on the grading results of the cable joint defect probability, indicating that a serious defect exists and that immediate inspection and repair are required.

[0046] An embodiment of the present invention provides a cable joint abnormality warning method, which collects infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time, and extracts features from the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; quantifies the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain a regional thermal vibration coupling abnormality index of the cable joint; identifies the regional defect type distribution according to the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculates the regional defect risk scores of the cable joint in different regions according to the regional defect type distribution; performs multimodal weight allocation based on the regional defect risk score to obtain a modal fusion weight allocation value; performs weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight allocation value to obtain a regional defect coupling feature vector of the cable joint; predicts the cable joint defect probability according to the regional defect coupling feature vector, and performs a graded warning according to the cable joint defect probability. Compared with the existing technology, this method realizes multimodal feature weighted fusion and defect probability prediction by fusing infrared thermal images and acoustic vibration data, achieving accurate early warning of cable joint abnormalities, and effectively improving the accuracy of cable joint abnormality detection and the timeliness of early warning.

[0047] It should be noted that the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of this application.

[0048] In one embodiment, Figure 2 As shown, an embodiment of the present invention provides a cable joint abnormality warning system, the system comprising: The feature extraction module 101 is used to collect infrared thermal imaging data of the cable connector surface and acoustic vibration data generated by the cable connector vibration in real time, and perform feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; The temperature analysis module 102 is configured to quantify the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution based on the multimodal feature data set, and obtain a regional thermal-vibration coupling anomaly index of the cable joint; The defect analysis module 103 is configured to identify the regional defect type distribution based on the regional structural characteristic parameters and the material mechanical resonance characteristics of the cable joint, and calculate the regional defect risk scores of the cable joint in different regions based on the regional defect type distribution; A weight allocation module 104 is configured to perform multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value; The modal fusion module 105 is configured to perform weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; The defect warning module 106 is configured to predict the cable joint defect probability based on the regional defect coupling feature vector and perform graded warning based on the cable joint defect probability.

[0049] For the specific definition of a cable joint abnormality warning system, please refer to the above-mentioned definition of a cable joint abnormality warning method, which will not be repeated here. A person of ordinary skill in the art will appreciate that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0050] An embodiment of the present invention provides a cable joint abnormality warning system, wherein the system collects infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time through a feature extraction module, and performs feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; the temperature analysis module quantifies the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain a regional thermal vibration coupling abnormality index of the cable joint; the defect analysis module identifies the regional defect type distribution according to the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculates the regional defect risk score of the cable joint in different regions according to the regional defect type distribution; the weight distribution module performs multimodal weight distribution based on the regional defect risk score to obtain a modal fusion weight distribution value; the modal fusion module performs weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; the defect warning module predicts the cable joint defect probability according to the regional defect coupling feature vector, and performs a graded warning according to the cable joint defect probability. Compared with existing technologies, this system realizes multi-modal feature weighted fusion and defect probability prediction by fusing infrared thermal images and acoustic vibration data, achieving accurate early warning of cable joint abnormalities, and effectively improving the accuracy of cable joint abnormality detection and the timeliness of early warning.

[0051] The above-described embodiments merely represent several preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the scope of protection of the claims.

Claims

1. A cable joint abnormality warning method, characterized in that: The following steps are involved: Collecting infrared thermal imaging data of the cable joint surface and acoustic vibration data generated by the cable joint vibration in real time, and performing feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; quantifying the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set, and obtaining a regional thermal-vibration coupling anomaly index of the cable joint; Identifying regional defect type distributions based on regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculating regional defect risk scores of the cable joint in different regions based on the regional defect type distributions; Performing multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value; Performing weighted fusion on the infrared thermal image feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; The cable joint defect probability is predicted based on the regional defect coupling characteristic vector, and a graded warning is performed based on the cable joint defect probability.

2. A cable joint abnormality early warning method according to claim 1, characterized in that: The step of extracting features from the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set includes: Filtering the infrared thermal image data to obtain infrared thermal image filtered data, and converting the grayscale values ​​in the infrared thermal image filtered data into actual temperature values ​​to obtain a cable joint surface temperature field matrix; Calculating the first-order gradient components of the cable joint surface temperature field matrix in the horizontal and vertical directions using an edge detection operator to obtain temperature gradient components; Identifying a temperature anomaly region based on the temperature gradient component and extracting a thermal space feature vector of the temperature anomaly region; Performing a fast Fourier transform on the acoustic vibration data to obtain an acoustic energy spectrum, and identifying, from the acoustic energy spectrum, acoustic vibration frequency components corresponding to amplitudes exceeding a preset characteristic peak threshold; The thermal spatial feature vector and the acoustic vibration frequency component are aligned by time stamp, and the aligned thermal spatial feature vector and the acoustic vibration frequency component are arranged in time series order to form a multimodal feature data set.

3. A cable joint abnormality early warning method according to claim 2, characterized in that: The thermal space feature vector includes at least the geometric center coordinates, boundary contour perimeter, region area and temperature peak value of the temperature anomaly region.

4. A cable joint abnormality early warning method according to claim 1, characterized in that: The step of quantifying the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution according to the multimodal feature data set to obtain the regional thermal-vibration coupling anomaly index of the cable joint includes: Analyzing the temperature gradient mutation points according to the multimodal feature data set to identify the boundary contours of the hot spot area; Calculating the hot spot evolution amount according to the hot spot area boundary contours at adjacent time points, and constructing the hot spot diffusion gradient vector based on the hot spot evolution amount; the hot spot evolution amount includes at least the hot spot area change rate and the hot spot centroid displacement speed; Determining a hot spot centroid position based on the hot spot diffusion gradient vector, and extracting an acoustic vibration frequency component amplitude corresponding to the hot spot centroid position from the multimodal feature data set; Calculate the Pearson correlation coefficient between the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude to obtain the temperature vibration correlation coefficient matrix; The temperature vibration correlation coefficient matrix is ​​used as the correlation weight, and the hot spot diffusion gradient vector and the acoustic vibration frequency component amplitude are weightedly fused to obtain a cross-modal fusion feature; Based on the cross-modal fusion features, the least squares linear regression algorithm is used to obtain the spatial thermal-vibration coupling anomaly distribution, and the thermal-vibration coupling horizontal gradient component, thermal-vibration coupling vertical gradient component and thermal-vibration coupling gradient amplitude are calculated according to the thermal-vibration coupling anomaly distribution between adjacent time points; The thermal vibration coupling horizontal gradient component, the thermal vibration coupling vertical gradient component and the thermal vibration coupling gradient amplitude are weightedly summed to obtain a regional thermal vibration coupling anomaly index of the cable joint.

5. A cable joint abnormality early warning method according to claim 4, characterized in that: The hot spot diffusion gradient vector includes a diffusion rate scalar value and a diffusion direction unit vector; wherein, the diffusion rate scalar value is the vector modulus of the hot spot area change rate and the hot spot center of mass displacement velocity; the diffusion direction unit vector is the normalized vector of the hot spot center of mass displacement direction.

6. A cable joint abnormality warning method according to claim 1, characterized in that: The step of identifying the regional defect type distribution according to the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint includes: Divide the regional space boundary coordinates of the conductor connection area, the insulation layer area and the outer sheath area according to the regional structural characteristic parameters of the cable joint; Mapping the regional thermal-vibration coupling anomaly index to the regional spatial boundary coordinates according to the spatial position, and calculating the arithmetic mean and maximum value of the regional thermal-vibration coupling anomaly index in each area of ​​the cable joint to identify the defective area of ​​the cable joint; Determining a characteristic frequency band corresponding to the defective area of ​​the cable joint based on the mechanical resonance characteristics of the material, and obtaining the characteristic frequency band of the defective area of ​​the cable joint; Calculate the relative change rate of the frequency band energy of the characteristic frequency band of each cable joint defect area relative to the reference value to obtain the vibration energy change rate of the cable joint defect area; The regional defect type of each cable joint defect area is identified according to the vibration energy change rate, and the regional defect type distribution is formed.

7. The cable joint abnormality warning method according to claim 1, characterized in that: The step of calculating the regional defect risk scores of cable joints in different areas according to the regional defect type distribution includes: Extract the regional defect types corresponding to the conductor connection area, insulation layer area, and outer sheath area from the regional defect type distribution, and obtain the defect duration exceeding limit and defect exceeding limit ratio corresponding to each regional defect type; Calculate the ratio of the duration of the defect exceeding the limit to the total monitoring time, generate a time enhancement factor, and use the time enhancement factor to correct the defect exceeding limit ratio value to obtain an effective exceeding limit value; Calculating a regional defect score based on the effective limit value and a preset defect type basic weight coefficient; When defects exist simultaneously in adjacent areas of the cable joint, the regional defect scores are corrected using a preset cross-region coupling coefficient to obtain a corrected regional defect score; The corrected regional defect scores are normalized to obtain the regional defect risk scores of the cable joints in each area.

8. The cable joint abnormality warning method according to claim 4, characterized in that: The modal fusion weight distribution value includes a temperature modal weight distribution value and an acoustic modal weight distribution value. The step of performing multi-modal weight distribution based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain the modal fusion weight distribution value includes: Determine the key monitoring area range of cable joints in each area based on the regional defect risk score; The regional thermal-vibration coupling anomaly index is used to calculate the information entropy of the temperature anomaly gradient and the amplitude of the acoustic vibration frequency component within the important monitoring area, and the corresponding temperature modal weight distribution value and acoustic modal weight distribution value are obtained.

9. The cable joint abnormality warning method according to claim 1, characterized in that: The step of predicting the cable joint defect probability based on the regional defect coupling characteristic vector includes: Inputting the regional defect coupling feature vector into a random forest classifier integrating multiple decision trees, performing classification and judgment on the regional defect coupling feature vector through each decision tree, and obtaining the defect probability votes of each decision tree; The highest defect votes are selected from the defect probability votes of each decision tree, and the proportion of the highest defect votes to the total number of decision trees is calculated to obtain the cable joint defect probability.

10. A cable joint abnormality warning system, characterized in that: The system comprises: a feature extraction module for collecting infrared thermal imaging data of the cable connector surface and acoustic vibration data generated by the cable connector vibration in real time, and performing feature extraction on the infrared thermal imaging data and the acoustic vibration data to obtain a multimodal feature data set; A temperature analysis module is used to quantify the coupling correlation between the hot spot diffusion gradient and the vibration frequency component in different directions of the current infrared temperature distribution based on the multimodal feature data set, and obtain a regional thermal-vibration coupling anomaly index of the cable joint; A defect analysis module is used to identify the regional defect type distribution based on the regional structural characteristic parameters and material mechanical resonance characteristics of the cable joint, and calculate the regional defect risk score of the cable joint in different areas based on the regional defect type distribution; A weight allocation module, configured to perform multimodal weight allocation based on the regional defect risk score and the regional thermal-vibration coupling anomaly index to obtain a modal fusion weight allocation value; A modal fusion module, configured to perform weighted fusion on the infrared thermal imaging feature data and the acoustic vibration feature data according to the modal fusion weight distribution value to obtain a regional defect coupling feature vector of the cable joint; The defect warning module is used to predict the cable joint defect probability based on the regional defect coupling characteristic vector and perform graded warning based on the cable joint defect probability.

Images