Insulation state evaluation method and system based on space charge characteristics

Abstract

The invention relates to an insulation state evaluation method and system based on space charge characteristics, and the method comprises the steps: obtaining an original image of space charge density distribution, extracting a curve contour, and recognizing a curve type; based on the curve contour and the curve category, extracting a local average change rate and a peak value feature of the curve; and on the basis of the local average change rate and the peak value characteristics, in combination with the constructed sample database, performing insulation state evaluation in a weighted form to obtain an insulation state evaluation result. According to the invention, by introducing an image recognition technology, key characteristic quantities in a space charge distribution diagram can be extracted for insulating material samples in different aging states, and objective analysis and quantitative evaluation of space charge information are realized.

Description

Technical Field

[0001] This invention relates to the field of electrical insulation condition assessment technology, and in particular to an insulation condition assessment method and system based on space charge characteristics. Background Technology

[0002] Space charge detection technology, as an important means of assessing the internal condition of insulating materials, is widely used in insulation performance analysis and condition evaluation. By monitoring the accumulation and distribution of space charge within the insulator, it can effectively reflect the aging degree of the material, the electric field distortion, and potential insulation defects. However, the interpretation of existing space charge detection results relies on the experience and judgment of professionals, making the assessment process highly subjective and technically challenging. Non-professionals can only obtain the electric field distribution through numerical integration of space charge, calculate the electric field distortion rate (the percentage of the difference between the integrated electric field and the theoretical electric field to the theoretical electric field), and use this to evaluate the quality of the insulating material. This method suffers from a single evaluation system and ignores the insulation information implicit in a large amount of space charge distribution data.

[0003] In contrast, partial discharge detection technology has gradually incorporated image recognition and intelligent analysis methods into the field of insulation condition assessment. Through data-driven pattern recognition and feature extraction, it has effectively improved the automation and objectivity of condition assessment. However, research on image recognition of space charge detection results is relatively scarce, and systematic application methods and assessment tools are still lacking. The in-depth utilization and intelligent analysis of space charge image information are still in their infancy, and innovative technical means are urgently needed to scientifically and accurately assess space charge distribution data.

[0004] In the prior art, patent application CN114019329A discloses a multi-dimensional evaluation system and method for early insulation degradation of XLPE cables based on machine learning. This system mentions space charge as a feature parameter for insulation degradation assessment; however, the device used to measure space charge is an integrating capacitor, which only obtains a single value of accumulated space charge, not the one-dimensional space charge distribution within the material. Therefore, the evaluation index is singular, the amount of information extracted is limited, and it is difficult to deeply explore the correlation between space charge distribution and insulation state. Summary of the Invention

[0005] The purpose of this invention is to provide an insulation condition assessment method and system based on space charge characteristics that improves the accuracy of insulation condition assessment.

[0006] The objective of this invention can be achieved through the following technical solutions: An insulation state assessment method based on space charge characteristics includes the following steps: Obtain the original image of the space charge density distribution, extract the curve contour, and identify the curve category; Based on the curve profile and curve category, the local average rate of change and peak features of the curve are extracted; Based on the local average rate of change and peak characteristics, and combined with the constructed sample database, a weighted insulation status assessment is performed to obtain the insulation status assessment result.

[0007] Furthermore, the step of extracting the curve contour includes: The original image of the space charge density distribution is converted to grayscale to obtain a grayscale image; The grayscale image is denoised to obtain a denoised grayscale image; Based on the denoised grayscale image, binarization is performed to obtain a binarized image; The curve contour is extracted from the binarized image.

[0008] Furthermore, the step of extracting the local average rate of change and peak features includes: The extracted curve contours are divided into different groups according to curve category, and pixel coordinates are extracted from each group. Based on the pixel coordinates of each curve, its start and end coordinate positions are determined, and then curve fitting and numerical processing are performed to obtain the local average rate of change and peak features.

[0009] Furthermore, the curve fitting and numerical processing steps include: For each curve, the gradient descent method is used to detect extreme points, and the maximum and minimum points are obtained, which correspond to the peak values ​​of positive charge density and negative charge density, respectively, and are used as peak features. Divide the entire curve into n For each local interval, calculate the local average rate of change within that interval.

[0010] Further, the steps for obtaining the peak positive charge density and peak negative charge density are as follows: Initialization: Select a starting point x 0, and the learning rate η Substitute into the following iterative formula: , In the formula, To calculate the value of the next data point, For gradient; When gradient When the value is close to 0, the minimum point is obtained. P min As the peak value of negative charge density, the peak value of negative charge density is: , In the formula, x It corresponds to the position.D ( x ) is the space charge density function; For the maximum point P max As the peak value of positive charge density, its iterative formula is: , When gradient The maximum point is obtained. P max The peak value of the positive charge density is: , In the formula, is the space charge density function.

[0011] Furthermore, the gradient The central difference method is used for approximation: , In the formula, h It is a tiny step size.

[0012] Furthermore, the calculation steps for the local average rate of change include: Record the minimum point. P min and maximum point P max The difference between the charge density value and the corresponding geometric position is calculated and used as the sample thickness. Based on the sample thickness and resolution, for n For a local interval, calculate the local average rate of change.

[0013] Furthermore, the expression for calculating the local average rate of change is: , In the formula, The local average rate of change The number of pixels contained in each interval. This represents the vertical coordinate value of the pixel, i.e., the space charge density value.

[0014] Furthermore, the computational expression for the weighted insulation state assessment is as follows: , In the formula, ∈[0, 1], representing an insulating state. , As weight, w 1+ w 2=1, The degree of drastic change in charge density distribution, , These are the characteristic quantities of the original sample. To calculate the average slope, The peak value of positive charge density. The peak value of negative charge density. The number of data segments divided based on measured data. This represents the local average rate of change.

[0015] The present invention also provides an insulation state assessment system based on space charge characteristics, comprising: Curve extraction module: used to acquire the original image of space charge density distribution, extract curve contours and identify curve categories; Feature extraction module: used to extract the local average rate of change and peak features of the curve based on the curve profile and curve category; Evaluation module: Based on the local average rate of change and peak characteristics, combined with the constructed sample database, the module performs a weighted insulation status evaluation to obtain the insulation status evaluation result.

[0016] Compared with the prior art, the present invention has the following beneficial effects: (1) By introducing image recognition technology, this invention can extract key feature quantities from the space charge distribution map of insulating material samples under different aging conditions, thereby enabling objective analysis and quantitative evaluation of space charge information and improving the accuracy of evaluation.

[0017] (2) By extracting curves, classifying curves and extracting pixel coordinates from the acquired images, this invention achieves high-quality, low-noise, and computable data from noisy original images, providing a good foundation for subsequent feature extraction.

[0018] (3) The present invention can adapt to the change of curve curvature and improve the fitting accuracy through curve fitting and numerical processing steps, and uses the gradient descent method to iteratively find the peak value of positive charge density and the peak value of negative charge density from the starting position, which not only improves the feature extraction accuracy, but also improves the feature extraction efficiency.

[0019] (4) By combining the feature information in the sample database and weightedly fusing multiple key features, this invention can achieve flexible, objective and quantitative evaluation of the insulation status of insulating materials, and has strong engineering practical value.

[0020] (5) This invention not only reduces the difficulty of professional interpretation of space charge detection results, but also allows for in-depth exploration of the implicit information of space charge distribution characterizing insulation state. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a sample data diagram of space charge from the present invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0023] Example 1 This embodiment provides a method for evaluating insulation status based on space charge characteristics, specifically, as follows: Figure 1 As shown, the method includes the following steps: S1: Basic Image Data Acquisition: The raw images of space charge density distribution were obtained in experimental tests using the Pulsed Electro-Acoustic Method (PEA). The Pulsed Electro-Acoustic Method is a testing technique that detects the space charge distribution within a dielectric material using acoustic signals. Its principle involves applying a high-voltage pulsed electric field to the sample, exciting minute vibrations of the space charges and generating acoustic signals. These acoustic waves are converted into electrical signals by a piezoelectric sensor, and the one-dimensional spatial distribution information of the charges is then retrieved through time-domain analysis. Based on the linear relationship between charge vibration under force and sound wave propagation, this method offers high resolution (down to the μm level) and real-time dynamic monitoring capabilities, and is widely used for characterizing the space charge within insulating materials.

[0024] The image is converted to grayscale, denoised, and binarized to extract curve contours. Grayscale conversion transforms the original image into a grayscale image, reducing color interference and facilitating subsequent processing; denoising removes noise from the image, improving image quality and ensuring accuracy in subsequent processing; binarization converts the grayscale image into a binary image (black and white image), further simplifying image information and highlighting key features. Curve contours are extracted from the binarized image to prepare for subsequent feature extraction. In a parallel step, different categories of curves in the image are identified based on labels, allowing the extracted curve contours to be grouped according to curve category. After completion, curve contours are extracted from each group. x - y Coordinate pairs (pixel coordinates). The above processing can use common methods such as cvtColor, GaussianBlur, and threshold to complete image transformation and extract pixels by traversing contour points. x - y Coordinates, this is for the initial image preprocessing.

[0025] S2: Feature extraction: By extracting pixel coordinates from the preprocessed image, the start and end positions of the curves are determined. Then, mathematical methods are used to fit and numerically process the curves. Extreme point detection is performed on each curve, and the maximum value (peak value of positive charge density) and minimum value (peak value of negative charge density) are found using gradient descent. Initialization: A starting point is selected. x 0, and the learning rate η (For example, 0.1), substitute into the iterative formula: , Where, gradient The central difference method can be used to approximate this: , in h For a small step size, such as 0.5 μm. When the gradient... When the value is close to 0, a minimum point appears. P min For the maximum point P max When gradient Maximum point appears P max Its iterative formula is: , Record the charge density values ​​and corresponding geometric locations of these extreme points; the difference between these values ​​represents the sample thickness. Peak negative charge density: , in D ( x ) is the space charge density function. x It corresponds to the position. P max and P min These represent the peak values ​​of positive and negative charge densities, respectively.

[0026] Space charge detection has resolution limitations, typically around 20 μm. Based on the extracted sample thickness and resolution, the entire curve is divided into... n Within each local interval, the local average rate of change is calculated to extract the local rate of change feature. The local average rate of change for each local interval is as follows: , in k j The number of pixels contained in each interval. This represents the vertical coordinate value of the pixel, i.e., the space charge density value.

[0027] The aforementioned peak values ​​of positive charge density, peak values ​​of negative charge density, and local average rate of change are collectively referred to as space charge characteristics.

[0028] S3: Insulation condition assessment: Based on extensive experimental results, a sample database of insulation condition indices, peak values, and rates of change was established. Combining this database with characteristic data, a weighted insulation condition assessment method was developed. , In the formula, ∆ Peak This indicates the degree of drastic change in charge density distribution and is positively correlated with the degree of aging; mean local rate of change. It characterizes the frequency of charge abrupt changes, and the value is positively correlated with the degree of sample aging. This represents the number of data segments divided based on measured data. It is set according to existing sample data. w 1, w 2 represents the weight of the two indicators, and w 1+ w 2=1; This parameter can be set according to the experimental results of different materials. For example, the space charge amplitude of cross-linked polyethylene (XLPE), an insulating material, changes significantly, so it should be appropriately increased. w 1 proportional weight, w 1 and w 2 can be set to 0.7 and 0.3 respectively; there is no obvious priority between the change in the amplitude of silicone rubber and the change in the value of the internal charge of the sample. w 1 and w 2 can be set to 0.5. The Insulation Index ∈ [0, 1], the closer it is to 0, the better the insulation condition. ∆ Peak norm and These are the characteristic quantities of the original sample, which can be obtained from previous experimental parameters.

[0029] Combination Figure 2 The data shown is based on the raw data (A-), representing measurements of a 200 μm thick XLPE sample subjected to thermal aging for different durations. The test resolution is 20 μm. N The value is 10, and this set of data can be used as the raw data for state assessment to obtain ∆. Peak norm and Feature quantities. These can be obtained using the methods described above. P max and P min 2.77 C / m 3 and -5.79 C / m 3 , ∆ Peak norm and 8.56 C / m3 And 0.13. And the ∆ of the 90-day aging data for Ding-... Peak norm and They are 60.01 C / m 3 And 1.06. The calculated Insulation Index is 0.86. Similarly, the corresponding Insulation Index can also be obtained for samples B and C.

[0030] Furthermore, based on a large amount of sample data, the method described in this embodiment can classify the Insulation Index of samples at different aging stages. For example, 0.9~1 indicates severe aging, 0.5~0.9 indicates moderate aging, and 0~0.5 indicates slight aging, thereby establishing a lifespan assessment model. The remaining lifespan is estimated by relating the Insulation Index of different samples to the thermal aging time and the actual commissioning time.

[0031] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0032] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0033] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0034] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0035] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0036] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0037] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for assessing insulation state based on space charge characteristics, characterized in that, Includes the following steps: Obtain the original image of the space charge density distribution, extract the curve contour, and identify the curve category; Based on the curve profile and curve category, the local average rate of change and peak features of the curve are extracted; Based on the local average rate of change and peak characteristics, and combined with the constructed sample database, a weighted insulation status assessment is performed to obtain the insulation status assessment result.

2. The insulation state assessment method based on space charge characteristics according to claim 1, characterized in that, The step of extracting the curve contour includes: The original image of the space charge density distribution is converted to grayscale to obtain a grayscale image; The grayscale image is denoised to obtain a denoised grayscale image; Based on the denoised grayscale image, binarization is performed to obtain a binarized image; The curve contour is extracted from the binarized image.

3. The insulation state assessment method based on space charge characteristics according to claim 1, characterized in that, The steps for extracting the local average rate of change and peak features include: The extracted curve contours are divided into different groups according to curve category, and pixel coordinates are extracted from each group. Based on the pixel coordinates of each curve, its start and end coordinate positions are determined, and then curve fitting and numerical processing are performed to obtain the local average rate of change and peak features.

4. The insulation state assessment method based on space charge characteristics according to claim 3, characterized in that, The curve fitting and numerical processing steps include: For each curve, the gradient descent method is used to detect extreme points, and the maximum and minimum points are obtained, which correspond to the peak values ​​of positive charge density and negative charge density, respectively, and are used as peak features. Divide the entire curve into n For each local interval, calculate the local average rate of change within that interval.

5. The insulation state assessment method based on space charge characteristics according to claim 4, characterized in that, The steps for obtaining the peak positive charge density and peak negative charge density are as follows: Initialization: Select a starting point x 0, and the learning rate η Substitute into the following iterative formula: ， In the formula, To calculate the value of the next data point, For gradient; When gradient When the value is close to 0, the minimum point is obtained. P min As the peak value of negative charge density, the peak value of negative charge density is: ， In the formula, x It corresponds to the position. D ( x ) is the space charge density function; For the maximum point P max As the peak value of positive charge density, its iterative formula is: ， When gradient The maximum point is obtained. P max The peak value of the positive charge density is: ， In the formula, is the space charge density function.

6. The insulation state assessment method based on space charge characteristics according to claim 5, characterized in that, The gradient The central difference method is used for approximation: ， In the formula, h It is a tiny step size.

7. The insulation state assessment method based on space charge characteristics according to claim 4, characterized in that, The calculation steps for the local average rate of change include: Record the minimum point. P min and maximum point P max The difference between the charge density value and the corresponding geometric position is calculated and used as the sample thickness. Based on the sample thickness and resolution, for n For a local interval, calculate the local average rate of change.

8. The insulation state assessment method based on space charge characteristics according to claim 1, characterized in that, The expression for calculating the local average rate of change is: ， In the formula, The local average rate of change The number of pixels contained in each interval. This represents the vertical coordinate value of the pixel, i.e., the space charge density value.

9. The insulation state assessment method based on space charge characteristics according to claim 1, characterized in that, The computational expression for the weighted insulation state assessment is as follows: ， In the formula, ∈[0, 1], representing an insulating state. , As weight, w 1+ w 2=1, The degree of drastic change in charge density distribution, , These are the characteristic quantities of the original sample. To calculate the average slope, The peak value of positive charge density. The peak value of negative charge density. The number of data segments divided based on measured data. This represents the local average rate of change.

10. An insulation state assessment system based on space charge characteristics, characterized in that, include: Curve extraction module: used to acquire the original image of space charge density distribution, extract curve contours and identify curve categories; Feature extraction module: used to extract the local average rate of change and peak features of the curve based on the curve profile and curve category; Evaluation module: Based on the local average rate of change and peak characteristics, combined with the constructed sample database, the module performs a weighted insulation status evaluation to obtain the insulation status evaluation result.

Images