CN114487742B - High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis - Google Patents
High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis Download PDFInfo
- Publication number
- CN114487742B CN114487742B CN202210386936.7A CN202210386936A CN114487742B CN 114487742 B CN114487742 B CN 114487742B CN 202210386936 A CN202210386936 A CN 202210386936A CN 114487742 B CN114487742 B CN 114487742B
- Authority
- CN
- China
- Prior art keywords
- discharge
- connected domain
- real
- coefficient
- voltage shell
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000009413 insulation Methods 0.000 title claims abstract description 45
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 238000001514 detection method Methods 0.000 title claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 52
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 230000002265 prevention Effects 0.000 claims abstract description 6
- 238000011156 evaluation Methods 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims description 28
- 230000007123 defense Effects 0.000 claims description 24
- 238000000034 method Methods 0.000 claims description 23
- 238000004891 communication Methods 0.000 claims description 8
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000007599 discharging Methods 0.000 description 24
- 230000007547 defect Effects 0.000 description 10
- 238000005286 illumination Methods 0.000 description 3
- HCHKCACWOHOZIP-UHFFFAOYSA-N Zinc Chemical compound [Zn] HCHKCACWOHOZIP-UHFFFAOYSA-N 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 238000005507 spraying Methods 0.000 description 2
- 229910052725 zinc Inorganic materials 0.000 description 2
- 239000011701 zinc Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1218—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing using optical methods; using charged particle, e.g. electron, beams or X-rays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J2005/0077—Imaging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
Abstract
The invention relates to the field of artificial intelligence, in particular to a high-voltage shell discharge insulation performance detection system based on multi-mode texture analysis. The image acquisition module is used for processing the acquired thermal infrared image to obtain a plurality of real connected domains; the real connected domain dividing module is used for calculating the temperature abnormal rate of the real connected domain; calculating the depth abnormal rate of a real connected domain, and dividing the real connected domain into a discharge connected domain and a non-discharge connected domain; the device comprises a distribution coefficient calculation module, a discharge coefficient determination module, a prevention coefficient determination module and a control module, wherein the distribution coefficient calculation module calculates the distribution coefficients of a discharge connected domain and a non-discharge connected domain; and the evaluation module is used for establishing a binary group to evaluate the discharge insulation performance. The invention collects different image data of the high-voltage shell to perform multi-mode analysis, and can obtain accurate discharge insulation performance.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a high-voltage shell discharge insulation performance detection system based on multi-mode texture analysis.
Background
The high-voltage shell is one of shells used on gas insulated metal enclosed switchgear (GIS), is an extremely important part, and has the main functions of: when the GIS runs, insulating media are loaded in a voltage-stabilizing and density-stabilizing mode, parts mounted inside the GIS normally work under high voltage without mutual influence, effective and safe connection among functional components of the GIS is achieved, and therefore the insulating performance of the GIS needs to be guaranteed.
At present, the insulation performance of the high-voltage shell is mainly tested manually or detected by using an instrument, but the manual testing efficiency is low, and the detection instrument is generally expensive and has high cost; in the prior art, surface defects are detected through technologies such as threshold segmentation and the like so as to determine the insulation performance of the high-voltage shell, but the insulation performance of the high-voltage shell is judged to be too single through defect detection, so that the obtained insulation performance is not accurate enough.
Therefore, the invention provides a high-voltage shell discharge insulation performance detection system based on multi-mode analysis, which collects RGB data, depth data and thermal infrared data and integrates the advantages of various data to obtain a more accurate discharge insulation performance detection result.
Disclosure of Invention
The invention provides a high-voltage shell discharge insulation performance detection system based on multi-mode texture analysis, which aims to solve the existing problems and comprises the following steps: the image acquisition module is used for processing the acquired thermal infrared image to obtain a plurality of real connected domains; the real connected domain dividing module is used for calculating the temperature abnormal rate of the real connected domain; calculating the depth abnormal rate of a real connected domain, and dividing the real connected domain into a discharge connected domain and a non-discharge connected domain; the device comprises a distribution coefficient calculation module, a discharge coefficient determination module, a prevention coefficient determination module and a control module, wherein the distribution coefficient calculation module calculates the distribution coefficients of a discharge connected domain and a non-discharge connected domain; and the evaluation module is used for establishing a binary group to evaluate the discharge insulation performance.
According to the technical means provided by the invention, RGB data, depth data and thermal infrared data of the high-voltage shell are collected, multi-mode analysis is carried out by utilizing the characteristics among different data, the thermal infrared image is utilized to calculate the temperature abnormal rate, the discharging condition of the high-voltage shell can be accurately judged, the depth abnormal rate is calculated by utilizing the depth image, the discharging position of the high-voltage shell can be accurately obtained, the insulating property of the high-voltage shell is further calculated, and the more accurate discharging insulating property of the high-voltage shell can be obtained.
The invention adopts the following technical scheme that a high-voltage shell discharge insulation performance detection system based on multi-mode texture analysis comprises:
and the image processing module is used for processing the collected thermal infrared image to obtain a plurality of real connected domains.
The real connected domain dividing module is used for calculating the temperature abnormal rate of the real connected domain according to the average value of the temperature values of the real connected domain pixel points output by the image processing module; and using the abnormal rate of the temperature value to enable the real connected domain to be a discharge connected domain or an undischarge connected domain, and obtaining all divided discharge connected domains and undischarge connected domains.
A distribution coefficient calculation module: and clustering all the discharging connected domains/non-discharging connected domains respectively, and obtaining the distribution coefficients of the discharging connected domains/non-discharging connected domains by using the total area of all the clustering results of the discharging connected domains/non-discharging connected domains.
And the discharge coefficient determining module is used for calculating the discharge coefficient of the high-voltage shell according to the maximum temperature abnormal rate in all the discharge communication domains and the distribution coefficient of the discharge communication domains.
And the defense coefficient determining module is used for acquiring the minimum value of the distances between all the non-discharge connected domains and the discharge connected domains and calculating the defense coefficient of the high-voltage shell by combining the distribution coefficients of the non-discharge connected domains.
And the evaluation module is used for evaluating the discharge insulation performance of the high-voltage shell according to the discharge coefficient and the defense coefficient determined by the discharge coefficient determination module and the defense coefficient determination module.
Further, a high voltage housing discharge insulation performance detecting system based on multi-modal texture analysis, the image processing module still includes: respectively carrying out threshold segmentation on the acquired gray level image and the acquired depth image, acquiring a texture connected domain in the gray level image and a depth connected domain in the depth image, superposing the gray level image and the depth image, and taking the superposed connected domain of the texture connected domain and the depth connected domain as a real connected domain.
Further, a method for acquiring a plurality of real connected domains in a thermal infrared image in the image processing module is as follows: and projecting the superposed connected domains of the texture connected domain and the depth connected domain into the thermal infrared image to obtain a plurality of real connected domains in the thermal infrared image.
Further, a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis, the method for calculating the temperature abnormal rate of the real connected domain comprises the following steps:
performing OTSU threshold segmentation on the thermal infrared image to obtain a segmentation threshold k, calculating a difference value between a mean value of temperature values of pixel points in each real connected domain of the thermal infrared image and the segmentation threshold, and taking a ratio of the difference value to the mean value of the temperature values of the pixel points in each real connected domain of the thermal infrared image as a temperature anomaly rate of each real connected domain.
Further, in a system for detecting the discharge insulation performance of a high-voltage shell based on multi-modal texture analysis, a method for using the abnormal rate of the temperature value to make a real connected domain be a discharge connected domain or a non-discharge connected domain in a real connected domain division module is as follows:
when the abnormal rate of the temperature value of the real connected domain is greater than a first threshold value, the real connected domain is a discharge connected domain;
when the temperature value abnormal rate of the real connected domain is smaller than a first threshold and larger than a second threshold, acquiring the depth value abnormal rate of the real connected domain in the depth image, when the depth value abnormal rate of the real connected domain in the depth image is larger than the threshold, the real connected domain is a discharge connected domain, otherwise, the real connected domain is a non-discharge connected domain;
and when the abnormal rate of the temperature value of the real connected domain is smaller than a second threshold value, the real connected domain is the undischarged connected domain.
Further, a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis, and the method for acquiring the depth value abnormal rate of the real connected domain in the depth image comprises the following steps:
acquiring a communication domain corresponding to the real communication domain in the high-pressure shell depth image and a segmentation threshold for performing OTSU threshold segmentation on the high-pressure shell depth image; and taking the difference value between the maximum value of the depth values of the pixels in the corresponding connected domain and the segmentation threshold value and the ratio of the maximum value of the depth values of the pixels in the real connected domain in the depth image as the depth abnormal rate of the real connected domain.
Further, a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis, wherein the method for determining the discharge coefficient in the discharge coefficient determination module comprises the following steps: and acquiring the maximum temperature abnormal rate of all the discharge connected domains, and taking the ratio of the maximum temperature abnormal rate to the distribution coefficient of the discharge connected domains as the discharge coefficient of the high-voltage shell.
Further, a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis, the method for calculating the high-voltage shell defense coefficient comprises the following steps:
and acquiring Euclidean distances between all the discharge connected domains and each undischarged connected domain, calculating a ratio of the minimum distance between the undischarged connected domains and the discharge connected domains to the number of rows and columns of pixel points in the gray level image, and taking the product of the distribution coefficient of the undischarged connected domains and the ratio as the defense coefficient of the high-voltage shell.
Further, a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis, and the method for evaluating the discharge insulation performance of the high-voltage shell comprises the following steps:
establishing a binary group (p 1, p 2) according to the discharge coefficient and the defense coefficient of the high-voltage shell, wherein p1 represents the discharge coefficient of the high-voltage shell, and p2 represents the defense coefficient of the high-voltage shell;
when any one of the discharge coefficient p1 of the high-voltage shell and the defense coefficient p2 of the high-voltage shell exceeds the threshold range, the high-voltage shell has poor insulating performance and needs to be repaired.
The invention has the beneficial effects that: according to the technical means provided by the invention, RGB data, depth data and thermal infrared data of the high-voltage shell are collected, multi-mode analysis is carried out by utilizing the characteristics among different data, the thermal infrared image is utilized to calculate the temperature abnormal rate, the discharging condition of the high-voltage shell can be accurately judged, the depth abnormal rate is calculated by utilizing the depth image, the discharging position of the high-voltage shell can be accurately obtained, the insulating property of the high-voltage shell is further calculated, and the more accurate discharging insulating property of the high-voltage shell can be obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis according to an embodiment of the invention;
fig. 2 is a schematic flow chart of a specific partitioning method of the real connected domain partitioning module in fig. 1.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a schematic structural diagram of a high-voltage casing discharge insulation performance detection system based on multi-modal texture analysis according to an embodiment of the present invention is provided, including:
and the image processing module is used for processing the collected thermal infrared image to obtain a plurality of real connected domains.
The image processing module further comprises: respectively carrying out threshold segmentation on the acquired gray level image and the acquired depth image, acquiring a texture connected domain in the gray level image and a depth connected domain in the depth image, superposing the gray level image and the depth image, and superposing the superposed texture connected domain and depth connected domain.
The method for acquiring a plurality of real connected domains in the thermal infrared image comprises the following steps: projecting the superposed connected domains with the overlapped texture connected domains and the overlapped depth connected domains into the thermal infrared image to obtain a plurality of real connected domains in the thermal infrared image
Three types of data are obtained by an RGB-D camera and a thermal infrared camera: RGB data, depth data and thermal infrared data. RGB data and depth data are obtained by an RGB-D camera, and thermal infrared image data are obtained by a thermal infrared camera.
After the image data is acquired, the present invention identifies the objects in the segmented image by means of DNN semantic segmentation.
The relevant content of the DNN network is as follows:
the data set used is a high-pressure shell image data set acquired side-looking, and the high-pressure shell is diversified in style.
The pixels needing to be segmented are divided into two types, namely the labeling process of the corresponding labels of the training set is as follows: and in the single-channel semantic label, the pixel at the corresponding position belongs to the background class and is marked as 0, and the pixel belonging to the high-voltage shell is marked as 1.
The task of the network is classification, and all used loss functions are cross entropy loss functions.
The 0-1 mask image obtained by semantic segmentation is multiplied by the original image, and the obtained image only contains the image of the high-voltage shell, so that the interference of the background is removed.
The outer surface of the high-voltage shell is subjected to zinc spraying treatment or other conductive paint spraying, so that the outer surface of the high-voltage shell is reliably grounded to reach the degree of contact with a human body, therefore, the surface of the high-voltage shell is greatly influenced by illumination, false defects similar to cracks caused by uneven illumination exist on RGB images, the false defects are easily confused with real defects such as scratches and the like on the galvanized surface, and on the basis, the defects are distinguished through depth images.
Defects caused by illumination do not exist on the surface of the zinc coating actually, so that the distance value and the texture value (the texture value refers to the pixel value on the texture level image) do not have a corresponding relation; the actual defects of the galvanized surface often affect the depth of the position, so that the distance value and the texture value of the position have corresponding relations.
Firstly, calculating a gray level co-occurrence matrix in each 9 x 9 region in a gray level image, obtaining a combination of a central point pixel value and a central point eight neighborhood pixel value in each 9 x 9 range through statistics, obtaining the frequency of each combination, calculating the product of the difference value and the frequency of each combination as a texture characteristic value of the central point, and referring an image with the gray level value of each pixel value in the image representing the texture characteristic value as a texture image, wherein the size of the texture image is consistent with the size of the gray level image.
And performing multi-threshold segmentation on the gray value of the texture image (performing multi-threshold segmentation on the texture image by using the principle of maximum inter-class variance and minimum intra-class variance according to the Fisher criterion) to obtain different gray levels, so as to obtain a gray level image, wherein the gray value of each pixel point in the gray level image is the gray level mean value of the gray level of the original pixel point.
The purpose of multi-threshold segmentation is to make the pixel values with similar texture gray levels become the same gray level, thereby obtaining a gray level image. The gray levels in the belonging gray level image correspond to pixel values on the texture image. The images are divided into different categories by belonging gray scale images. The gray-scale image corresponding to the texture image is a texture-level image for distinguishing from the gray-scale image corresponding to the RGB image.
On the depth image, different categories are obtained through a multi-threshold segmentation method, connected domain analysis is respectively carried out on the different categories to obtain different connected domains, the connected domains are called depth connected domains, and the boundary pixel value of the depth connected domains is represented by 1.
And respectively carrying out connected domain analysis on different texture levels on the texture level image to obtain connected domains of different texture levels, namely texture connected domains, and expressing the boundary pixel value of the depth connected domain by 1.
The superposition of the depth connected domain and the texture connected domain is realized through the superposition of images, and if the number of pixel points with the pixel value of 2 in the boundary points of the connected domain exceeds 80 percent, the corresponding connected domain is considered as the real texture by calculating the attribute condition of each point on the boundary of each connected domain.
The real connected domain dividing module is used for calculating the temperature abnormal rate of the real connected domain according to the average value of the temperature values of the real connected domain pixel points output by the image processing module; and dividing the real connected domain into a discharge connected domain or a non-discharge connected domain by using the temperature value abnormal rate, and obtaining all the divided discharge connected domains and non-discharge connected domains.
As shown in fig. 2, a schematic flow chart of a specific dividing method for the discharging connected domain and the non-discharging connected domain in the module is given, which includes:
the method for calculating the temperature abnormal rate of each real connected domain comprises the following steps:
performing OTSU threshold segmentation on the thermal infrared image to obtain a segmentation threshold k, calculating a difference value between a mean value of temperature values of pixel points in each real connected domain of the thermal infrared image and the segmentation threshold, and taking a ratio of the difference value to the mean value of the temperature values of the pixel points in each real connected domain of the thermal infrared image as a temperature anomaly rate of each real connected domain.
The method for dividing the real connected domain comprises the following steps:
when the abnormal rate of the temperature value of the real connected domain is more than 0.8, the corresponding real connected domain is a discharge connected domain;
when the abnormal rate of the temperature value of the real connected domain is less than 0.8 and more than 0.5, calculating the abnormal rate of the depth value of the corresponding real connected domain, and when the abnormal rate of the depth value of the real connected domain is more than 0.7, the real connected domain is the discharging connected domain;
and when the abnormal rate of the temperature value of the real connected domain is less than 0.5, the real connected domain is the undischarged connected domain.
The method for calculating the depth abnormal rate of each real connected domain comprises the following steps:
performing OTSU threshold segmentation on the depth image to obtain a segmentation threshold k, calculating a difference value between the maximum value of the depth value of the pixel point in each real connected domain of the depth image and the segmentation threshold, and taking the ratio of the difference value to the maximum value of the depth value of the pixel point in each real connected domain of the depth image as the depth abnormal rate of each real connected domain.
The discharge defect refers to a defect which causes electric field disorder and affects the discharge insulation capability, so the real connected domain is divided into two types according to whether the discharge insulation capability is affected, one type is called a discharge connected domain, and the other type is called an undischarged connected domain.
A distribution coefficient calculation module: and clustering all the discharging connected domains/non-discharging connected domains respectively, and obtaining the distribution coefficients of the discharging connected domains/non-discharging connected domains by using the total area of all the clustering results of the discharging connected domains/non-discharging connected domains.
For the discharge connected domain, firstly, calculating to obtain central points of different connected domains, then clustering the central points of the connected domains through k-means to obtain different clustering categories, then calculating to obtain the area of each clustering category through a convex hull algorithm, calculating the sum of the areas of all the clustering categories, and taking the ratio of the sum of the areas of all the clustering categories to the total area of an image as the distribution coefficient of the discharge connected domain.
And similarly, calculating the distribution coefficient of the undischarged connected domain.
And the discharge coefficient determining module is used for calculating the discharge coefficient of the high-voltage shell according to the maximum temperature abnormal rate in all the discharge communication domains and the distribution coefficient of the discharge communication domains.
And acquiring the maximum temperature abnormal rate of all the discharge connected domains, and taking the ratio of the maximum temperature abnormal rate to the distribution coefficient of the discharge connected domains as the discharge coefficient of the high-voltage shell.
The discharge coefficient of the high-voltage shell is calculated by screening the maximum temperature abnormal rate of the discharge connected domain, namely when the discharge coefficient of the high-voltage shell is overlarge due to the existence of any temperature abnormal rate in a thermal infrared image of the high-voltage shell, the insulation performance of the current high-voltage shell is poor, and repair or replacement is needed.
And the defense coefficient determining module is used for acquiring the minimum value of the distances between all the non-discharge connected domains and the discharge connected domains and calculating the defense coefficient of the high-voltage shell by combining the distribution coefficients of the non-discharge connected domains.
The undischarged connected domain is a candidate of the discharged connected domain, and can be changed into the discharged connected domain along with the increase of the service time of the high-voltage shell, and the undischarged connected domain is more densely distributed and the hidden danger coefficient is larger; the closer the non-discharge connected domain is to the discharge connected domain, the larger the hidden danger coefficient is.
The method for calculating the high-pressure shell defense coefficient comprises the following steps:
and acquiring Euclidean distances between all the discharge connected domains and each undischarged connected domain, calculating a ratio of the minimum distance between the undischarged connected domains and the discharge connected domains to the number of rows and columns of pixel points in the gray level image, and taking the product of the distribution coefficient of the undischarged connected domains and the ratio as the defense coefficient of the high-voltage shell.
And the evaluation module is used for judging the discharge insulation performance of the high-voltage shell according to the discharge coefficient and the prevention coefficient determined by the discharge coefficient determination module and the prevention coefficient determination module.
After obtaining the discharge insulation performance of the high-voltage shell, evaluating the discharge insulation performance of the high-voltage shell, including:
establishing a binary group (p 1, p 2) according to the discharge coefficient and the defense coefficient of the high-voltage shell, wherein p1 represents the discharge coefficient of the high-voltage shell, and p2 represents the defense coefficient of the high-voltage shell;
when p1 is greater than 0.6, the discharge coefficient of the high-voltage shell exceeds a threshold value, the discharge insulation performance is very poor, and a discharge area needs to be repaired;
when p1<0.6, p2>0.8, the defense coefficient of the high-voltage shell exceeds a threshold value, the discharge insulation performance of the high-voltage shell is poor, and a repair basis is needed.
According to the technical means provided by the invention, RGB data, depth data and thermal infrared data of the high-voltage shell are collected, multi-mode analysis is carried out by utilizing the characteristics among different data, the thermal infrared image is utilized to calculate the temperature abnormal rate, the discharging condition of the high-voltage shell can be accurately judged, the depth abnormal rate is calculated by utilizing the depth image, the discharging position of the high-voltage shell can be accurately obtained, the insulating property of the high-voltage shell is further calculated, and the more accurate discharging insulating property of the high-voltage shell can be obtained.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (9)
1. A high-voltage shell discharge insulation performance detection system based on multi-modal texture analysis is characterized by comprising:
the image processing module is used for processing the collected thermal infrared image to obtain a plurality of real connected domains;
the real connected domain dividing module is used for calculating the temperature abnormal rate of the real connected domain according to the average value of the temperature values of the real connected domain pixel points output by the image processing module; dividing the real connected domain into a discharge connected domain or a non-discharge connected domain by using the temperature value abnormal rate, and obtaining all divided discharge connected domains and non-discharge connected domains;
a distribution coefficient calculation module: clustering all the discharge connected domains/non-discharge connected domains respectively, and obtaining the distribution coefficients of the discharge connected domains/non-discharge connected domains by using the total area of all the clustering results of the discharge connected domains/non-discharge connected domains;
the discharge coefficient determining module is used for calculating the discharge coefficient of the high-voltage shell according to the maximum temperature abnormal rate in all the discharge connected domains and the distribution coefficient of the discharge connected domains;
the defense coefficient determining module is used for acquiring the minimum value of the distances between all the undischarged connected domains and the discharged connected domains and calculating the defense coefficient of the high-voltage shell by combining the distribution coefficients of the undischarged connected domains;
and the evaluation module is used for evaluating the discharge insulation performance of the high-voltage shell according to the discharge coefficient and the prevention coefficient determined by the discharge coefficient determination module and the prevention coefficient determination module.
2. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis as claimed in claim 1, wherein the image processing module further comprises: respectively carrying out threshold segmentation on the acquired gray level image and the acquired depth image, acquiring a texture connected domain in the gray level image and a depth connected domain in the depth image, superposing the gray level image and the depth image, and taking the superposed connected domain of the texture connected domain and the depth connected domain as a real connected domain.
3. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis as claimed in claim 2, wherein the method for acquiring the plurality of real connected domains in the thermal infrared image in the image processing module comprises the following steps: and projecting the superposed connected domains of the texture connected domain and the depth connected domain into the thermal infrared image to obtain a plurality of real connected domains in the thermal infrared image.
4. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis as claimed in claim 1, wherein the method for calculating the temperature anomaly rate of the real connected domain comprises the following steps:
performing OTSU threshold segmentation on the thermal infrared image to obtain a segmentation threshold k, calculating a difference value between a mean value of temperature values of pixel points in each real connected domain of the thermal infrared image and the segmentation threshold, and taking a ratio of the difference value to the mean value of the temperature values of the pixel points in each real connected domain of the thermal infrared image as a temperature anomaly rate of each real connected domain.
5. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis according to claim 1, wherein the method for dividing the real connected domain into the discharge connected domain or the undischarged connected domain by using the abnormal rate of the temperature value in the real connected domain dividing module is as follows:
when the abnormal rate of the temperature value of the real connected domain is greater than a first threshold value, the real connected domain is a discharge connected domain;
when the temperature value abnormal rate of the real connected domain is smaller than a first threshold and larger than a second threshold, acquiring the depth value abnormal rate of the real connected domain in the depth image, when the depth value abnormal rate of the real connected domain in the depth image is larger than the threshold, the real connected domain is a discharge connected domain, otherwise, the real connected domain is a non-discharge connected domain;
and when the abnormal rate of the temperature value of the real connected domain is smaller than the second threshold value, the real connected domain is the undischarged connected domain.
6. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis as claimed in claim 5, wherein the method for obtaining the abnormal rate of the depth value of the real connected domain in the depth image comprises:
acquiring a communication domain corresponding to the real communication domain in the high-pressure shell depth image and a segmentation threshold for performing OTSU threshold segmentation on the high-pressure shell depth image; and according to the difference value between the maximum value of the depth value of the pixel point in the corresponding connected domain and the segmentation threshold value, taking the ratio of the maximum value of the depth value of the pixel point in the real connected domain in the depth image as the depth abnormal rate of the real connected domain.
7. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis according to claim 1, wherein the method for determining the discharge coefficient in the discharge coefficient determination module is as follows: and acquiring the maximum temperature abnormal rate of all the discharge connected domains, and taking the ratio of the maximum temperature abnormal rate to the distribution coefficient of the discharge connected domains as the discharge coefficient of the high-voltage shell.
8. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis as claimed in claim 2, wherein in the defense factor determining module, the method for calculating the defense factor of the high-voltage shell comprises:
and acquiring Euclidean distances between all the discharge connected domains and each undischarged connected domain, calculating a ratio of the minimum distance between the undischarged connected domains and the discharge connected domains to the number of rows and columns of pixel points in the gray level image, and taking the product of the distribution coefficient of the undischarged connected domains and the ratio as the defense coefficient of the high-voltage shell.
9. The system for detecting the discharge insulation performance of the high-voltage shell based on the multi-modal texture analysis according to claim 1, wherein the method for evaluating the discharge insulation performance of the high-voltage shell comprises the following steps:
establishing a duplet (p 1, p 2) according to the discharge coefficient and the defense coefficient of the high-voltage shell, wherein p1 represents the discharge coefficient of the high-voltage shell, and p2 represents the defense coefficient of the high-voltage shell;
when any one of the discharge coefficient p1 of the high-voltage shell and the defense coefficient p2 of the high-voltage shell exceeds the threshold range, the high-voltage shell has poor insulating performance and needs to be repaired.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210386936.7A CN114487742B (en) | 2022-04-14 | 2022-04-14 | High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210386936.7A CN114487742B (en) | 2022-04-14 | 2022-04-14 | High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114487742A CN114487742A (en) | 2022-05-13 |
| CN114487742B true CN114487742B (en) | 2022-07-05 |
Family
ID=81487298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210386936.7A Active CN114487742B (en) | 2022-04-14 | 2022-04-14 | High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114487742B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114839495B (en) * | 2022-06-30 | 2022-09-20 | 江苏苏能森源电气有限公司 | Transformer partial discharge abnormity detection method based on artificial intelligence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011033538A (en) * | 2009-08-04 | 2011-02-17 | Mitsubishi Electric Corp | Method and apparatus for diagnosis of partial discharge in gas insulated switchgear |
| CN105004972A (en) * | 2015-06-25 | 2015-10-28 | 华北电力大学（保定） | Porcelain insulator insulation state evaluation method based on solar-blind ultraviolet imaging image feature |
| CN110346699A (en) * | 2019-07-26 | 2019-10-18 | 国网山东省电力公司电力科学研究院 | Insulator arc-over information extracting method and device based on ultraviolet image processing technique |
| CN110992306A (en) * | 2019-11-04 | 2020-04-10 | 国网河北省电力有限公司检修分公司 | Method and device for segmenting deteriorated insulator in infrared image based on deep learning |
| CN112379231A (en) * | 2020-11-12 | 2021-02-19 | 国网浙江省电力有限公司信息通信分公司 | Equipment detection method and device based on multispectral image |
-
2022
- 2022-04-14 CN CN202210386936.7A patent/CN114487742B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011033538A (en) * | 2009-08-04 | 2011-02-17 | Mitsubishi Electric Corp | Method and apparatus for diagnosis of partial discharge in gas insulated switchgear |
| CN105004972A (en) * | 2015-06-25 | 2015-10-28 | 华北电力大学（保定） | Porcelain insulator insulation state evaluation method based on solar-blind ultraviolet imaging image feature |
| CN110346699A (en) * | 2019-07-26 | 2019-10-18 | 国网山东省电力公司电力科学研究院 | Insulator arc-over information extracting method and device based on ultraviolet image processing technique |
| CN110992306A (en) * | 2019-11-04 | 2020-04-10 | 国网河北省电力有限公司检修分公司 | Method and device for segmenting deteriorated insulator in infrared image based on deep learning |
| CN112379231A (en) * | 2020-11-12 | 2021-02-19 | 国网浙江省电力有限公司信息通信分公司 | Equipment detection method and device based on multispectral image |
Non-Patent Citations (1)
| Title |
|---|
| 输变电设备电晕放电紫外图谱量化参数提取;李炼炼 等;《高压电器》;20171216;第229-235页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114487742A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105004972B (en) | Porcelain insulator Condition assessment of insulation method based on day blind ultraviolet imagery characteristics of image | |
| Davari et al. | Intelligent diagnosis of incipient fault in power distribution lines based on corona detection in UV-visible videos | |
| CN103442209B (en) | Video monitoring method of electric transmission line | |
| CN108765373A (en) | A kind of insulator exception automatic testing method based on integrated classifier on-line study | |
| Davari et al. | Corona detection and power equipment classification based on GoogleNet-AlexNet: An accurate and intelligent defect detection model based on deep learning for power distribution lines | |
| CN106950472B (en) | insulator detection method based on infrared and ultraviolet imaging | |
| CN111652857B (en) | Infrared detection method for insulator defects | |
| CN111079955A (en) | GIS (geographic information System) equipment defect detection method based on X-ray imaging | |
| CN115797641B (en) | Electronic equipment gas leakage detection method | |
| CN112017173B (en) | Power equipment defect detection method based on target detection network and structured positioning | |
| CN113592828B (en) | Nondestructive testing method and system based on industrial endoscope | |
| CN114487742B (en) | High-voltage shell discharge insulation performance detection system based on multi-mode texture analysis | |
| CN111753794B (en) | Fruit quality classification method, device, electronic equipment and readable storage medium | |
| CN113065484A (en) | Insulator contamination state assessment method based on ultraviolet spectrum | |
| CN113344475B (en) | Transformer bushing defect identification method and system based on sequence modal decomposition | |
| CN115018838A (en) | Method for identifying pitting defects on surface of oxidized steel pipe material | |
| CN112801949A (en) | Method and device for determining discharge area in ultraviolet imaging detection technology | |
| CN108345898A (en) | A kind of novel line insulator Condition assessment of insulation method | |
| CN113850330A (en) | Power distribution network fault cause detection method based on short-time Fourier transform and convolutional neural network | |
| CN113888462A (en) | Crack identification method, system, readable medium and storage medium | |
| Huang et al. | Study on hydrophobicity detection of composite insulators of transmission lines by image analysis | |
| CN108596196B (en) | Pollution state evaluation method based on insulator image feature dictionary | |
| CN110672988A (en) | Partial discharge mode identification method based on hierarchical diagnosis | |
| CN115239646A (en) | Defect detection method and device for power transmission line, electronic equipment and storage medium | |
| CN115855961B (en) | Distribution box fault detection method used in operation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |