Power distribution network fault positioning system

Abstract

The invention provides a power distribution network fault positioning system which comprises a power distribution network segmentation video monitoring module, a power distribution network segmentation environment data acquisition module, a power grid data feature processing module and a fault positioning module. The power distribution network sectioning video monitoring module detects to obtain power distribution network video information in a sectioning manner, and the power distribution network sectioning environment data acquisition module acquires fixed node environment parameter information; the power grid data feature processing module processes the power distribution network video information and the fixed node environment parameter information and then transmits the information to the fault positioning module for combination processing, the power distribution network positioning models are obtained through training of the fault positioning module, and the multiple power distribution network positioning models are used for sectional type fault comprehensive classification positioning. The fault position is accurately positioned; according to the method, the power distribution network is subjected to fault positioning through external environmental factors, comprehensive judgment is carried out through the classifiers suitable for various scenes, the fault position is positioned, and the risk and cost of manual inspection are reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of fault diagnosis and optimization, and in particular relates to a distribution network fault location system. Background Art

[0002] The significance of power supply is very important, as it is related to the national and regional economy, society, energy security and people's lives. Through effective power supply guarantee measures, the normal operation of the power system can be guaranteed, and the reliability and stability of power supply can be ensured, thereby promoting economic development, maintaining social stability and ensuring people's living needs.

[0003] Power lines are the lifeblood of power grid operation. While transmitting electric energy, they are also quite fragile. Any damage from external forces and staff's erroneous operation may cause large-scale power outages, bringing great harm to the social economy and people's lives. The distribution network directly faces a large number of end users, so its grid structure is complex and the probability of failure is high. A large amount of statistical data shows that 90% of power outages and fault disturbances occur in the distribution network. When a fault occurs, it is necessary to identify and locate the fault point so that timely measures can be taken to deal with it. If the fault cannot be accurately located in the first time, it will lead to the expansion of the power outage scope and the instability of the power system, affecting the quality and safety of users' electricity use, and having an adverse impact on the power supply.

[0004] The rapid diagnosis and location of distribution network line faults is an important part of distribution network operation and maintenance, and plays an important role in rapid power restoration and power supply. At present, the fault diagnosis and location of distribution network lines are still mainly based on manual inspections, supplemented by fault indicator analysis, and generally rely on the operating experience of operation and maintenance personnel, with relatively large uncertainties. In recent years, artificial intelligence has developed rapidly, providing new ideas and new tools for fault diagnosis of distribution network lines.

[0005] The Chinese patent publication number is CN104931857A, and the publication date is September 23, 2015. The name of the invention is a distribution network fault location method based on DS evidence theory. The scheme discloses a distribution network fault location method based on DS evidence theory. Its shortcomings are that the distribution network has many branches and the grounding points are hidden, resulting in a long time for locating the grounding fault, and the accuracy of locating the grounding line fault point is not high.

[0006] It can be seen that the existing technology has the following problems: the existing model processing distribution network fault location method does not take into account the environment of the distribution network area, such as temperature and smoke concentration in the air, and the algorithm model used fails to use the changes in surrounding environmental factors after the fault and the sudden change of image pixels to quickly locate. In addition, the monitoring video data of the distribution network is not well utilized, and the algorithm model used fails to obtain accurate fault location identification in a timely manner.

[0007] Therefore, there is an urgent need for a distribution network fault location system to obtain pixel mutation image data of monitoring videos for rapid fault location and identification, and to quickly locate distribution network faults in combination with the temperature and smoke conditions around the distribution network. Summary of the invention

[0008] In order to solve the above technical problems, the present invention proposes a distribution network fault location system.

[0009] In a first aspect of the present invention, a distribution network fault location system is provided, characterized in that the system comprises a distribution network segment video monitoring module, a distribution network segment environment data acquisition module, a power grid data feature processing module and a fault location module:

[0010] Distribution network segment video monitoring module: obtains the distribution network video of the first segment area, and is also used to obtain the distribution network video of the second segment area;

[0011] Distribution network segment environment data acquisition module: obtains the environmental parameters of the first segment area, and is also used to obtain the environmental parameters of the second segment area;

[0012] A power grid data feature processing module: obtaining a historical scenario fault area, wherein the historical scenario fault area includes a first scenario fault area, processing the first segment area distribution network video and the first segment area environmental parameters to obtain the first segment area distribution network fluctuation image and the first segment area environmental optimization parameter data, and also processing the second segment area distribution network video and the second segment area environmental parameters to obtain the second segment area distribution network fluctuation image and the second segment area environmental optimization parameter data;

[0013] Fault location module: obtain the processed undulation image of the distribution network in the first section area and the environmental optimization parameter data of the first section area and process them to obtain the first distribution network location feature, generate a distribution network location model according to the first distribution network location feature and the first scenario fault area, and also obtain the processed undulation image of the distribution network in the second section area and the environmental optimization parameter data of the second section area to obtain the second distribution network location feature, use the distribution network location model according to the second distribution network location feature to obtain the second scenario fault area, and locate the fault of the distribution network according to the second scenario fault area.

[0014] Furthermore, the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W The following is calculated by using the first section area distribution network video or the second section area distribution network video:

[0015]

[0016] Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, Then it means that the pixel value of the p*tth picture when the pixel value of the picture changes at the value y is the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W .

[0017] Furthermore, the distribution network positioning model is constructed using a machine learning method; the first section area distribution network video and the second section area distribution network video are obtained by monitoring the section areas fixed in the distribution network in a distribution network section video monitoring module.

[0018] Furthermore, the first segment area environmental parameter data or the second segment area environmental parameter data are the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental optimization parameter data or the second segment area environmental optimization parameter data are the average temperature around the line in the segment area and the air smoke concentration, and the first scenario fault area or the second scenario fault area is divided into three types of fault areas: distribution output end, distribution end and distribution input end, which are respectively the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

[0019] Furthermore, the machine learning method used is an SVM classifier improved according to the temperature of the fault area and the number of image fluctuations. Its specific function expression is:

[0020]

[0021] Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, T n is the average temperature around the distribution network, T max is the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

[0022] A method for locating a distribution network fault is also provided, wherein the method comprises:

[0023] S1. Obtain the distribution network video of the first subsection area, as well as the environmental parameters of the first subsection area at this time, and obtain the first scene fault area from the historical maintenance data;

[0024] S2. Obtaining the first section area distribution network fluctuation image from the first section area distribution network video, processing the first section area environmental parameters to obtain the first section area environmental optimization parameter data, and processing the first section area distribution network fluctuation image and the first section area environmental optimization parameter data to generate the first distribution network positioning feature;

[0025] S3. Generate a distribution network positioning model according to the first distribution network positioning feature and the first scenario fault area;

[0026] S4. Obtain the second sub-section area distribution network video and the second sub-section area environmental parameters at this time, process the second sub-section area distribution network video to obtain the second sub-section area distribution network undulation image, process the second sub-section area environmental parameters to obtain the second sub-section area environmental optimization parameter data, and combine the second sub-section area distribution network undulation image and the second sub-section area environmental optimization parameter data to obtain the second distribution network positioning feature;

[0027] S5. Using the distribution network positioning model, the second distribution network positioning feature is processed to generate a second scenario fault area, and based on the second scenario fault area, the fault of the current distribution network is located.

[0028] Furthermore, the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W The video of the distribution network in the first sub-section area is calculated as follows:

[0029]

[0030] Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, Then it means that the pixel value of the p*tth picture when the pixel value of the picture changes at the value y is the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W .

[0031] Furthermore, the distribution network positioning model is constructed using a machine learning method; the first section area distribution network video and the second section area distribution network video are obtained through section area monitoring of a fixed distribution network.

[0032] Furthermore, the first segment area environmental parameter data or the second segment area environmental parameter data are the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental parameter or the second segment area environmental parameter are the average temperature around the line in the segment area and the air smoke concentration, and the first scenario fault area or the second scenario fault area is divided into three types of fault areas: distribution output end, distribution end and distribution input end, which are respectively the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

[0033] Furthermore, the machine learning method used is an SVM classifier improved according to the temperature of the fault area and the number of image fluctuations. Its specific function expression is:

[0034]

[0035] Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, T n is the average temperature around the distribution network, T max is the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

[0036] The model constructed by the present invention takes into account factors such as the temperature and smoke concentration in the distribution network area, and the algorithm model used utilizes changes in surrounding environmental factors after a fault and pixel mutations in the image for rapid positioning. The monitoring video data of the distribution network is well positioned, and the algorithm model used can obtain accurate fault location identification in a timely manner. The present invention obtains pixel mutation image data of the monitoring video for rapid fault location identification, and combines the temperature and smoke conditions around the distribution network to quickly locate the distribution network fault. The constructed model is accurate and reliable, and at the same time, the SVM model is improved to an algorithm suitable for the application scenario of distribution network fault location, which can greatly reduce the time and labor costs consumed by manual inspections.

[0037] More embodiments and improved effects of the present invention will be further introduced in conjunction with the drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of a distribution network fault location method of the present invention;

[0039] Figure 2It is a schematic diagram of the image of the sudden change of the distribution network image of the present invention;

[0040] Figure 3 It is a schematic diagram of fault location of the distribution network location model of the present invention;

[0041] Figure 4 It is a schematic diagram of the principle of the SVM model of the present invention;

[0042] Figure 5 It is a schematic diagram of the structure of an electronic device of the method of the present invention. DETAILED DESCRIPTION

[0043] The invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0044] In order to solve the above technical problems, the first aspect of the present invention provides a distribution network fault location system, characterized in that the system includes a distribution network segment video monitoring module, a distribution network segment environment data acquisition module, a power grid data feature processing module and a fault location module:

[0045] Distribution network segment video monitoring module: obtains the distribution network video of the first segment area, and is also used to obtain the distribution network video of the second segment area;

[0046] Distribution network segment environment data acquisition module: obtains the environmental parameters of the first segment area, and is also used to obtain the environmental parameters of the second segment area;

[0047] A power grid data feature processing module: obtaining a historical scenario fault area, wherein the historical scenario fault area includes a first scenario fault area, processing the first segment area distribution network video and the first segment area environmental parameters to obtain the first segment area distribution network fluctuation image and the first segment area environmental optimization parameter data, and further processing the second segment area distribution network video and the second segment area environmental parameters to obtain the second segment area distribution network fluctuation image and the second segment area environmental optimization parameter data;

[0048] Fault location module: obtain the processed undulation image of the distribution network in the first section area and the environmental optimization parameter data of the first section area and process them to obtain the first distribution network location feature, generate a distribution network location model according to the first distribution network location feature and the first scenario fault area, and also obtain the processed undulation image of the distribution network in the second section area and the environmental optimization parameter data of the second section area and combine them to obtain the second distribution network location feature, obtain the second scenario fault area according to the second distribution network location feature using the distribution network location model, and locate the fault of the distribution network according to the second scenario fault area.

[0049] Furthermore, the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W The following is calculated by using the first section area distribution network video or the second section area distribution network video:

[0050]

[0051] Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, Then it means that the pixel value of the p*tth picture when the pixel value of the picture changes at the value y is the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W .

[0052] Furthermore, the distribution network positioning model is constructed using a machine learning method; the first section area distribution network video and the second section area distribution network video are obtained by monitoring the section areas fixed in the distribution network in a distribution network section video monitoring module.

[0053] Furthermore, the first segment area environmental parameter data or the second segment area environmental parameter data is the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental parameter or the second segment area environmental parameter is the average temperature around the line in the segment area and the air smoke concentration, the first scenario fault area or the second scenario fault area is divided into three types of fault areas, namely the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

[0054] Furthermore, the machine learning method used is an SVM classifier improved according to the temperature of the fault area and the number of image fluctuations. Its specific function expression is:

[0055]

[0056] Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, T n is the average temperature around the distribution network, T max is the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

[0057] A method for locating a distribution network fault is also provided, wherein the method comprises:

[0058] S1. Obtain the distribution network video of the first subsection area, as well as the environmental parameters of the first subsection area at this time, and obtain the first scene fault area from the historical maintenance data;

[0059] S2. Obtaining the first section area distribution network fluctuation image from the first section area distribution network video, processing the first section area environmental parameters to obtain the first section area environmental optimization parameter data, and processing the first section area distribution network fluctuation image and the first section area environmental optimization parameter data to generate the first distribution network positioning feature;

[0060] S3. Generate a distribution network positioning model according to the first distribution network positioning feature and the first scenario fault area;

[0061] S4. Obtain the second section area distribution network video and the second section area environmental parameters at the same time, process the second section area distribution network video to obtain the second section area distribution network fluctuation image, process the second section area environmental parameters to obtain the second section area environmental optimization parameter data, and combine the second section area distribution network fluctuation image and the second section area environmental optimization parameter data to obtain the second distribution network positioning feature;

[0062] S5. Using the distribution network positioning model, the second distribution network positioning feature is processed to generate a second scenario fault area, and based on the second scenario fault area, the fault of the current distribution network is located.

[0063] Furthermore, the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W The video of the distribution network in the first sub-section area is calculated as follows:

[0064]

[0065] Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, Then it means that the pixel value of the p*tth picture when the pixel value of the picture changes at the value y is the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W .

[0066] In this embodiment, the number of video frames is 25, and the video length is a video record of a certain fault-prone time period, which can be 10 minutes. The number of pictures is 25x10x60, a total of 15,000 pictures. If the pixel value of the 3000th picture is greater than or equal to the y value, 30-50 can be taken. Here, 30 is taken, that is, the pixel feature changes by more than 30. Then the 3000th picture is an undulating image, and the characteristic value of the picture is (151, 185, 215).

[0067] Furthermore, the distribution network positioning model is constructed using a machine learning method; the first section area distribution network video and the second section area distribution network video are obtained through section area monitoring of a fixed distribution network.

[0068] Furthermore, the first segment area environmental parameter data or the second segment area environmental parameter data are the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental optimization parameter or the second segment area environmental optimization parameter are the average temperature around the line in the segment area and the air smoke concentration, and the first scenario fault area or the second scenario fault area is divided into three types of fault areas: distribution output end, distribution end and distribution input end, which are respectively the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

[0069] In this embodiment, the normal fault-free ambient average temperature is 25°C. When a fault occurs, ionization will occur, and the temperature will change. The average temperature after the change is set to 30°C. The air smoke concentration is generally a constant value. When a fault occurs in the distribution network, if the fault is serious, a fire will occur. At this time, the air smoke concentration changes to 8000ppm. The air smoke concentration is sensed by the gas sensor installed at the stepped position of the distribution network. Since the air humidity is affected and has little correlation with the grid fault, the two parameter characteristics of temperature and smoke concentration are optimized.

[0070] Furthermore, the machine learning method used is an SVM classifier improved according to the temperature of the fault area and the number of image fluctuations. Its specific function expression is:

[0071]

[0072] Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, and the k value range is 0.01-0.05, T n is the average temperature around the distribution network, T maxis the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

[0073] In this embodiment, the input positioning feature is (151, 185, 215, 30, 8000), which is the first distribution network positioning feature obtained by processing the first sub-region distribution network undulation image and the first sub-region environmental optimization parameter data. At this time, the output (ω T X i +b) value is greater than If the output value is less than the set value of 30, the output is the second distribution terminal fault area; if the output value is greater than 30, the output is the third distribution input terminal fault area. The output is the first type of distribution output fault area. If the output value is equal to This indicates that the distribution network is normal.

[0074] The model constructed by the present invention takes into account factors such as the temperature and smoke concentration in the distribution network area, and the algorithm model used utilizes changes in surrounding environmental factors after a fault and pixel mutations in the image for rapid positioning. The monitoring video data of the distribution network is well positioned, and the algorithm model used can obtain accurate fault location identification in a timely manner. The present invention obtains pixel mutation image data of the monitoring video for rapid fault location identification, and combines the temperature and smoke conditions around the distribution network to quickly locate the distribution network fault. The constructed model is accurate and reliable, and at the same time, the SVM model is improved to an algorithm suitable for the application scenario of distribution network fault location, which can greatly reduce the time and labor costs consumed by manual inspections.

[0075] Of course, it can be understood that each embodiment of the present invention can achieve one of the effects alone, and a combination of multiple embodiments of the present invention can achieve all of the above effects, but it is not required that every embodiment of the present invention achieve all of the above advantages and effects, because each embodiment of the present invention can constitute a separate technical solution and make one or more contributions to the prior art.

[0076] The present invention does not particularly specify the structure of some modules, which shall be subject to the contents recorded in the prior art. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters. The scope of protection of the present invention shall be subject to the contents actually recorded in the claims.

Claims

1. A distribution network fault location system, characterized in that: The system includes a distribution network segment video monitoring module, a distribution network segment environment data acquisition module, a power grid data feature processing module and a fault location module: Distribution network segment video monitoring module: obtain the distribution network video of the first segment area and the distribution network video of the second segment area; Distribution network segment environment data acquisition module: obtains the first segment area environment parameters and the second segment area environment parameters; A power grid data feature processing module: after processing the first section area distribution network video and the first section area environmental parameters, an undulating image of the first section area distribution network and first section area environmental optimization parameter data are obtained; after processing the second section area distribution network video and the second section area environmental parameters, an undulating image of the second section area distribution network and second section area environmental optimization parameter data are obtained; Fault location module: obtain historical scenario fault areas, the historical scenario fault areas include first scenario fault areas, obtain first distribution network location features according to the first sub-section area distribution network fluctuation image and the first sub-section area environmental optimization parameter data, generate a distribution network location model according to the first distribution network location features and the first scenario fault area; obtain second distribution network location features according to the second sub-section area distribution network fluctuation image and the second sub-section area environmental optimization parameter data, obtain the second scenario fault area through the distribution network location model based on the second distribution network location features, and locate the fault of the distribution network according to the second scenario fault area.

2. A distribution network fault location system according to claim 1, characterized in that: The first section area distribution network fluctuation image or the second section area distribution network fluctuation image is calculated by the first section area distribution network video or the second section area distribution network video: Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, It means the pixel value of the p*tth image when the pixel value of the image changes at the y value, F W It is the first section area power distribution network fluctuation image or the second section area power distribution network fluctuation image.

3. A distribution network fault location system as claimed in claim 2, characterized in that: The distribution network positioning model is constructed using a machine learning method; the first segmented area distribution network video and the second segmented area distribution network video are obtained by monitoring the segmented areas fixed to the distribution network in a distribution network segmented video monitoring module.

4. A distribution network fault location system according to claim 1, characterized in that: The first segment area environmental parameter data or the second segment area environmental parameter data are the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental optimization parameter data or the second segment area environmental optimization parameter data are the average temperature around the line in the segment area and the air smoke concentration, the first scenario fault area or the second scenario fault area is divided into three types of fault areas: distribution output end, distribution end and distribution input end, which are respectively the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

5. A distribution network fault location system as claimed in claim 3, characterized in that: The machine learning method is an SVM classifier improved according to the temperature of the fault area and the number of image fluctuations, and its specific function expression is: Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, T n is the average temperature around the distribution network, T max is the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

6. A method for locating a distribution network fault, the method being implemented based on the system according to claim 1, characterized in that: The method comprises: S1. Obtain the distribution network video of the first subsection area and the environmental parameters of the first subsection area at the same time, and obtain the first scene fault area from the historical maintenance data; S2. Obtaining a first section area distribution network fluctuation image from the first section area distribution network video, obtaining first section area environment optimization parameter data based on the first section area environment parameter processing, and processing the first section area distribution network fluctuation image and the first section area environment optimization parameter data to generate a first distribution network positioning feature; S3. Generate a distribution network positioning model according to the first distribution network positioning feature and the first scenario fault area; S4. Acquire the second sub-section area distribution network video and the second sub-section area environmental parameters at the same time, process the second sub-section area distribution network video to obtain the second sub-section area distribution network fluctuation image, process the second sub-section area environmental parameters to obtain the second sub-section area environmental optimization parameter data, and process the second sub-section area distribution network fluctuation image and the second sub-section area environmental optimization parameter data to obtain the second distribution network positioning feature; S5. Using the distribution network positioning model, the second distribution network positioning feature is processed to generate a second scenario fault area, and based on the second scenario fault area, the fault of the current distribution network is located.

7. A distribution network fault location method according to claim 6, characterized in that: The first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W The following is calculated by using the first section area distribution network video or the second section area distribution network video: Where p is the number of frames of the distribution network video, t is the video duration in seconds, Indicates the p*tth ups and downs picture corresponding to when the pixel value of the picture increases or decreases by more than or equal to the y value, Then it means that the pixel value of the p*tth picture when the pixel value of the picture changes at the value y is the first section area distribution network fluctuation image or the second section area distribution network fluctuation image F W .

8. A distribution network fault location method according to claim 7, characterized in that: The distribution network positioning model is constructed using a machine learning method; the first segmented area distribution network video and the second segmented area distribution network video are obtained through segmented area monitoring of a fixed distribution network.

9. A distribution network fault location method according to claim 6, characterized in that: The first segment area environmental parameter data or the second segment area environmental parameter data are the average temperature around the line in the segment area, the air smoke concentration and the air humidity, the first segment area environmental optimization parameter or the second segment area environmental optimization parameter are the average temperature around the line in the segment area and the air smoke concentration, and the first scenario fault area or the second scenario fault area is divided into three types of fault areas: distribution output end, distribution end and distribution input end, which are respectively the first distribution output end fault area, the second distribution end fault area and the third distribution input end fault area.

10. A distribution network fault location method according to claim 8, characterized in that: The machine learning method used is an SVM classifier improved according to the temperature in the fault area and the number of image fluctuations. Its specific function expression is: Where ω, b are the normal vector and intercept of the hyperplane, respectively, X i is the first distribution network positioning feature or the second distribution network positioning feature, k is the temperature difference coefficient according to temperature change, T n is the average temperature around the distribution network, T max is the average temperature of the surrounding area after the distribution network failure, p*t is the p*tth number of the first undulating image of the first section area distribution network video or the second section area distribution network video, and N is the total number of images of the first section area distribution network video or the second section area distribution network video.

Images