Power supply fault detection method and system based on data analysis

Abstract

The invention relates to the technical field of power transmission fault detection, in particular to a power supply fault detection method and system based on data analysis, and the method comprises the steps: carrying out the inspection of a to-be-detected overhead line through an unmanned plane, obtaining a to-be-processed image of the overhead line, carrying out the noise reduction, color enhancement and other processing of the to-be-processed image, and obtaining a to-be-processed image; comparing the to-be-processed image with the first contrast subset, screening out images containing the obvious characteristics of insulators, stockbridge dampers, spacers and the like in the to-be-processed image, comparing the screened to-be-processed image with the second contrast subset, and further obtaining whether the obvious characteristics of the insulators, the stockbridge dampers, the spacers and the like have defects or not; and converting the significant features in the to-be-processed image to form a picture in which the significant features of the to-be-processed image are the same as the significant features in the second contrast subset in size, so that the significant features in the to-be-processed image can be quickly compared, and a key component with a fault in the significant features of the to-be-processed image is screened out. And the fault detection efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power transmission fault detection, and in particular to a power supply fault detection method and system based on data analysis. Background Art

[0002] The power grid is an important infrastructure related to the national economy, people's livelihood and national energy security. As the infrastructure of the power grid, the overhead transmission line is an important part of the power grid. With the rapid development of society, the demand for electricity from industrial facilities in various industries is increasing. In order to ensure the normal demand of electrical equipment, relevant departments in the power industry are constantly building transmission lines.

[0003] Currently, power transmission lines are mainly inspected by manual inspection, robot inspection, helicopter inspection, remote sensing satellite inspection and drone inspection. However, in the daily operation of power lines, key components such as insulators, spacers and shock-absorbing hammers may fall off due to harsh natural environments or mechanical factors such as loose bolts. This situation will have a serious impact on the normal operation of the power grid and may even cause failures, threatening the safety of the power system.

[0004] Therefore, there is an urgent need for a method that can improve the accuracy of key component identification to detect key components of transmission lines. Summary of the invention

[0005] In order to improve the accuracy of identifying key components in a tie-point line, the present invention provides a power supply fault detection method and system based on data analysis.

[0006] In a first aspect, the present invention provides a power supply fault detection method based on data analysis, which adopts the following technical solution: A power supply fault detection method based on data analysis comprises the following steps: Acquire an original image data set through drone inspection; process images in the original image data set to obtain a first control subset formed by n first feature images; perform significant feature recognition on the first feature images in the first control subset to obtain m second control subsets; An image to be processed is obtained, the image to be processed is identified according to the first comparison subset and the second comparison subset, and significant features in the image to be processed are compared.

[0007] The original image data set is a large number of photos obtained by taking pictures of overhead lines through drone inspections in different environments. These photos must include key components in the overhead lines. For example, the photos include normal insulators, broken insulators, normal anti-vibration hammers, broken anti-vibration hammers, normal spacers, broken spacers, etc. These photos are processed by noise reduction, color enhancement, etc. to form a first feature image, and these first feature images form a first comparison set. After that, the first feature image is processed by noise reduction, color enhancement, etc. to obtain significant features in the first features for identification. The significant features here refer to key components such as normal insulators, broken insulators, normal anti-vibration hammers, broken anti-vibration hammers, normal spacers, broken spacers, etc., and m second comparison subsets are obtained (for example: the first second comparison subset is the feature image of the insulator, and the second second comparison subset is the feature image of the spacer); Afterwards, the overhead lines to be inspected are inspected by drones to obtain images to be processed of the overhead lines. The images to be processed are first subjected to noise reduction, color enhancement and other processing. Then, the images to be processed are compared with the first reference subset to screen out images to be processed that contain significant features such as insulators, shock absorbers, and spacers. Then, the screened images to be processed are compared with the second reference subset to determine whether the significant features such as insulators, shock absorbers, and spacers have defects. Optionally, when acquiring the original image data set through drone inspection, a first inspection path of the drone is acquired, and when acquiring the image to be processed, the drone acquires the image to be processed along the first inspection path.

[0008] When the drone acquires the original image data set, the inspection path of the drone is planned, that is, a three-dimensional spatial coordinate system is established to plan the flight path of the drone, wherein the information of the first inspection path includes the starting position, the end position, the flight altitude, and the distance between the drone and the overhead line. When the drone acquires the image to be processed, the drone still performs inspection according to the information of the first inspection path. The image to be processed obtained in this way makes the image to be processed more standardized and reduces the processing workload of the image to be processed.

[0009] Optionally, when identifying and comparing the significant features in the image to be processed, the significant features in the image to be processed are transformed and the size of the significant features is adjusted, and the significant features in the second comparison subset are transformed and the size of the significant features is adjusted.

[0010] Here, the significant features in the second comparison subset are first transformed. The significant features in the second comparison subset are key components such as normal insulators, damaged insulators, normal shockproof hammers, damaged shockproof hammers, normal spacers, and damaged spacers. These key components are converted into jpg format images with a fixed size of 540*540, and the significant features in the image to be processed are converted by this conversion method. At this time, a photo with the same size of the significant features of the image to be processed and the significant features in the second comparison subset is formed. In this way, the significant features in the image to be processed can be quickly compared, and the key components with faults in the significant features of the image to be processed can be screened out, thereby improving the fault detection efficiency.

[0011] Optionally, after transforming the significant features in the image to be processed, the position of the significant features in the image to be processed is further adjusted. When adjusting the position of the significant features in the image to be processed, the image of the significant features is rotated according to the position of the significant features in the second comparison subset.

[0012] When acquiring the image to be processed, the drone may encounter bad weather, such as strong winds, during the inspection and shooting, which makes it impossible for the drone to shoot according to the first inspection route. At this time, the acquired image to be processed may be offset. At this time, a coordinate system is established with the length of the image to be processed as the x-axis and the length of the image to be processed as the y-axis, and the image to be processed is rotated. For example, a point P (x1, y1) in the image to be processed is rotated a certain angle to point Q (x2, y2). ; ; Among them, the distance between P and the far point is A, and the angle between the line between P and the origin and the x-axis is , the rotation angle of point Q relative to point P is , and the angle between the line from point Q to the origin and the x-axis is By rotating the image to be processed, the significant features in the image to be processed can be quickly identified, thereby improving the detection efficiency of the significant features.

[0013] In a second aspect, the present invention provides a power supply fault detection system based on data analysis, which adopts the following technical solution: A power supply fault detection system based on data analysis, comprising a drone and a processing module, wherein the drone is electrically connected to the processing module; Drone: Obtain the original image data set and the images to be processed through drone inspection; Processing module: Process the images in the original image data set to obtain a first control subset formed by n first feature images; identify significant features of the first feature images in the first control subset to obtain m second control subsets, identify the image to be processed based on the first control subset and the second control subset, and compare the significant features in the image to be processed.

[0014] Optionally, when the drone acquires the original image data set, a first inspection path of the drone is acquired, and when acquiring the image to be processed, the drone acquires the image to be processed along the first inspection path.

[0015] Optionally, when identifying and comparing the significant features in the image to be processed in the processing module, the significant features in the image to be processed are transformed and the size of the significant features is adjusted, and the significant features in the second comparison subset are transformed and the size of the significant features is adjusted.

[0016] Optionally, after transforming the significant features in the image to be processed in the processing module, the position of the significant features in the image to be processed is also adjusted. When adjusting the position of the significant features in the image to be processed, the image of the significant features is rotated according to the position of the significant features in the second comparison subset.

[0017] In a third aspect, the present invention provides a power supply fault detection device based on data analysis, A device comprises a processor and a storage, wherein the storage is used to store a computer program, and the processor is used to execute the computer program stored in the storage, so that the device executes the method as described in the first aspect.

[0018] In a fourth aspect, the present invention provides a computer storage medium based on data analysis, which adopts the following technical solution: A medium having a computer program stored thereon; when the computer program is executed by a processor, the method described in the first aspect is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of Example 1 of the present application; Figure 2 This is a system structure diagram of Example 2 of the present application; Figure 3 This is a schematic diagram of the bus communication structure of Example 3 of the present application. DETAILED DESCRIPTION

[0020] The following combination Figures 1 to 3 The present invention is described in further detail.

[0021] Embodiment 1: This embodiment of the present application provides a power supply fault detection method based on data analysis. The following technical solutions are adopted: Reference Figure 1 , a power supply fault detection method based on data analysis, comprising the following steps: S1: Acquire an original image data set through drone inspection; when acquiring the original image data set through drone inspection, acquire a first inspection path of the drone, and when acquiring the image to be processed, the drone acquires the image to be processed along the first inspection path.

[0022] The original image data set is a large number of photos obtained by taking pictures of overhead lines through drone inspections in different environments. These photos must include key components in the overhead lines. For example, the photos include normal insulators, broken insulators, normal anti-vibration hammers, broken anti-vibration hammers, normal spacers, broken spacers, etc. These photos are processed by noise reduction, color enhancement, etc. to form a first feature image, and these first feature images form a first comparison set. After that, the first feature image is processed by noise reduction, color enhancement, etc. to obtain significant features in the first features for identification. The significant features here refer to key components such as normal insulators, broken insulators, normal anti-vibration hammers, broken anti-vibration hammers, normal spacers, broken spacers, etc., and m second comparison subsets are obtained (for example: the first second comparison subset is the feature image of the insulator, and the second second comparison subset is the feature image of the spacer); Afterwards, the overhead lines to be inspected are inspected by drones to obtain images to be processed of the overhead lines. The images to be processed are first subjected to noise reduction, color enhancement and other processing. Then, the images to be processed are compared with the first reference subset to screen out images to be processed that contain significant features such as insulators, shock absorbers, and spacers. Then, the screened images to be processed are compared with the second reference subset to determine whether the significant features such as insulators, shock absorbers, and spacers have defects.

[0023] When the drone acquires the original image data set, the inspection path of the drone is planned, that is, a three-dimensional spatial coordinate system is established to plan the flight path of the drone, wherein the information of the first inspection path includes the starting position, the end position, the flight altitude, and the distance between the drone and the overhead line. When the drone acquires the image to be processed, the drone still performs inspection according to the information of the first inspection path. The image to be processed obtained in this way makes the image to be processed more standardized and reduces the processing workload of the image to be processed.

[0024] S2: Processing the images in the original image data set to obtain a first control subset formed by n first feature images; performing significant feature recognition on the first feature images in the first control subset to obtain m second control subsets; S3: Acquire the image to be processed, identify the image to be processed according to the first control subset and the second control subset, compare the significant features in the image to be processed, transform the significant features in the image to be processed and adjust the size of the significant features, transform the significant features in the second control subset and adjust the size of the significant features; after the comparison, adjust the position of the significant features in the image to be processed, and when adjusting the position of the significant features in the image to be processed, rotate the image of the significant features according to the position of the significant features in the second control subset.

[0025] Here, the significant features in the second comparison subset are first transformed. The significant features in the second comparison subset are key components such as normal insulators, damaged insulators, normal shockproof hammers, damaged shockproof hammers, normal spacers, and damaged spacers. These key components are converted into jpg format images with a fixed size of 540*540, and the significant features in the image to be processed are converted by this conversion method. At this time, a photo with the same size of the significant features of the image to be processed and the significant features in the second comparison subset is formed. In this way, the significant features in the image to be processed can be quickly compared, and the key components with faults in the significant features of the image to be processed can be screened out, thereby improving the fault detection efficiency.

[0026] When acquiring the image to be processed, the drone may encounter bad weather, such as strong winds, during the inspection and shooting, which makes it impossible for the drone to shoot according to the first inspection route. At this time, the acquired image to be processed may be offset. At this time, a coordinate system is established with the length of the image to be processed as the x-axis and the length of the image to be processed as the y-axis, and the image to be processed is rotated. For example, a point P (x1, y1) in the image to be processed is rotated a certain angle to point Q (x2, y2). ; ; Among them, the distance between P and the far point is A, and the angle between the line between P and the origin and the x-axis is , the rotation angle of point Q relative to point P is , and the angle between the line from point Q to the origin and the x-axis is By rotating the image to be processed, the significant features in the image to be processed can be quickly identified, thereby improving the detection efficiency of the significant features.

[0027] The implementation principle of a power supply fault detection method based on data analysis in the embodiment of the present application is as follows: The original image dataset is a large number of photos obtained by taking drone inspection photos of overhead lines in different environments. When the drone obtains the original image dataset, the inspection path of the drone is planned, that is, a three-dimensional space coordinate system is established to plan the flight path of the drone. The information of the first inspection path includes the starting position, the end position, the flight altitude, and the distance between the drone and the overhead line. The original image dataset is processed by noise reduction, color enhancement, etc. to form a first feature image, and these first feature images are formed into a first comparison set. After that, the first feature image is processed by noise reduction, color enhancement, etc. to obtain the significant features in the first feature for identification. The significant features here refer to normal Insulators, damaged insulators, normal shockproof hammers, damaged shockproof hammers, normal spacers, damaged spacers and other key components are obtained to obtain m second comparison subsets (for example, the first second comparison subset is the characteristic image of the insulator, and the second second comparison subset is the characteristic image of the spacer); when the UAV obtains the image to be processed, the UAV still performs inspections according to the information of the first inspection path to obtain the image to be processed of the overhead line, first performs noise reduction, color enhancement and other processing on the image to be processed, and then compares the image to be processed with the first comparison subset to screen out images containing significant features such as insulators, shockproof hammers, spacers, etc., and then compares the screened image to be processed with the second comparison subset.

[0028] When acquiring the image to be processed, the drone may encounter bad weather, such as strong winds, during inspection and shooting, which may make it impossible for the drone to shoot according to the first inspection route. At this time, the acquired image to be processed may be offset. At this time, a coordinate system is established with the length of the image to be processed as the x-axis and the length of the image to be processed as the y-axis, and the image to be processed is rotated. For example, a point P (x1, y1) in the image to be processed is rotated a certain angle to point Q (x2, y2). By rotating the image to be processed, the significant features in the image to be processed can be quickly identified, thereby improving the detection efficiency of the significant features.

[0029] Embodiment 2, the present invention provides a power supply fault detection system based on data analysis, which adopts the following technical solution: Reference Figure 2 , a power supply fault detection system based on data analysis, comprising a drone and a processing module, wherein the drone is electrically connected to the processing module; Drone: The original image data set and the image to be processed are obtained through drone inspection. When the drone obtains the original image data set, the first inspection path of the drone is obtained. When obtaining the image to be processed, the drone obtains the image to be processed along the first inspection path. Processing module: Process the images in the original image data set to obtain a first control subset formed by n first feature images; identify significant features of the first feature images in the first control subset to obtain m second control subsets, identify the image to be processed based on the first control subset and the second control subset, and compare the significant features in the image to be processed.

[0030] When identifying and comparing the significant features in the image to be processed in the processing module, the significant features in the image to be processed are transformed and the size of the significant features is adjusted, and the significant features in the second comparison subset are transformed and the size of the significant features is adjusted.

[0031] After the significant features in the image to be processed are transformed in the processing module, the position of the significant features in the image to be processed is also adjusted. When the position of the significant features in the image to be processed is adjusted, the image of the significant features is rotated according to the position of the significant features in the second comparison subset.

[0032] Embodiment 3: This embodiment discloses a power supply fault detection device based on data analysis, referring to Figure 3 , a power supply fault detection device based on data analysis comprises: Storage for storing computer programs; The processor is used to execute the computer program stored in the memory, thereby implementing the method described in Example 1.

[0033] The storage may include a large capacity storage for storing data or instructions. By way of example and not limitation, the storage may include a hard disk, a floppy disk, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a combination of two or more of these. Where appropriate, the storage may include a removable or non-removable (or fixed) medium. Where appropriate, the storage may be inside or outside the data processing device. In a particular embodiment, the storage is a non-volatile solid-state storage. In a particular embodiment, the storage includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EBROM), or a combination of two or more of these.

[0034] Embodiment 4: This embodiment discloses a computer storage medium based on data analysis, wherein the computer storage medium stores a program, and when the program is executed, it can implement part or all of the steps of the method described in Embodiment 1.

[0035] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A power supply fault detection method based on data analysis, characterized in that: The following steps are involved: Obtaining raw image datasets through drone inspections; Processing the images in the original image data set to obtain a first control subset formed by n first feature images; performing significant feature recognition on the first feature images in the first control subset to obtain m second control subsets; An image to be processed is obtained, the image to be processed is identified according to the first comparison subset and the second comparison subset, and significant features in the image to be processed are compared.

2. The power supply fault detection method based on data analysis according to claim 1, characterized in that: When the original image data set is obtained through the drone inspection, a first inspection path of the drone is obtained, and when the image to be processed is obtained, the drone obtains the image to be processed along the first inspection path.

3. The power supply fault detection method based on data analysis according to claim 2, characterized in that: When identifying and comparing the significant features in the image to be processed, the significant features in the image to be processed are transformed and the size of the significant features is adjusted, and the significant features in the second comparison subset are transformed and the size of the significant features is adjusted.

4. The power supply fault detection method based on data analysis according to claim 3 is characterized in that: After the significant features in the image to be processed are transformed, the position of the significant features in the image to be processed is further adjusted. When the position of the significant features in the image to be processed is adjusted, the image of the significant features is rotated according to the position of the significant features in the second comparison subset.

5. A power supply fault detection system based on data analysis, applying the power supply fault detection method based on data analysis according to any one of claims 1 to 4, characterized in that: It includes a drone and a processing module, wherein the drone is connected to the processing module by electrical signals; Drone: Obtain the original image data set and the images to be processed through drone inspection; Processing module: Process the images in the original image data set to obtain a first control subset formed by n first feature images; identify significant features of the first feature images in the first control subset to obtain m second control subsets, identify the image to be processed based on the first control subset and the second control subset, and compare the significant features in the image to be processed.

6. A real-time video surveillance anomaly detection system according to claim 5, characterized in that: When the drone acquires the original image data set, a first inspection path of the drone is acquired, and when acquiring the image to be processed, the drone acquires the image to be processed along the first inspection path.

7. A real-time video surveillance anomaly detection system according to claim 6, characterized in that: When identifying and comparing the significant features in the image to be processed in the processing module, the significant features in the image to be processed are transformed and the size of the significant features is adjusted, and the significant features in the second comparison subset are transformed and the size of the significant features is adjusted.

8. A real-time video surveillance anomaly detection system according to claim 7, characterized in that: After the significant features in the image to be processed are transformed in the processing module, the position of the significant features in the image to be processed is also adjusted. When the position of the significant features in the image to be processed is adjusted, the image of the significant features is rotated according to the position of the significant features in the second comparison subset.

9. A power supply fault detection device based on data analysis, characterized in that: The device comprises a processor and a storage device, wherein the storage device is used to store a computer program, and the processor is used to execute the computer program stored in the storage device, so that the device performs the method as described in any one of potential requirements 1-4.

10. A computer storage medium based on data analysis, having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images