Intelligent inspection system for power distribution

Abstract

The invention belongs to the technical field of power distribution detection, and particularly relates to an intelligent inspection system for power distribution. The intelligent inspection system for power distribution comprises a data acquisition module, a communication transmission module and a cloud service platform, the cloud service platform comprises an edge gateway module, a cloud computing module and a work management module; the inspection system builds and calculates health assessment indexes of equipment to which each area belongs through automatic acquisition and measurement data and a preprocessing mode of a data acquisition module, realizes graded inspection and screening, automatically generates maintenance tasks and allocates the maintenance tasks to appropriate maintenance personnel, so that the inspection efficiency is improved, real-time monitoring and early warning are realized, and the maintenance efficiency is improved. Potential faults are timely found and processed, the maintenance efficiency and accuracy of the power distribution inspection system are improved, the potential safety hazard of power system operation is accurately identified, and the power operation risk is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power distribution detection, and in particular, relates to an intelligent inspection system for power distribution. Background Art

[0002] In the current field of electrical installation engineering, especially in the operation and maintenance management of power distribution systems, traditional inspection methods have gradually revealed their limitations. With the rapid development of large-scale distribution network automation, low-voltage meter reading, distributed energy access and other businesses, the communication needs of various distribution network equipment and electricity users have increased dramatically. In modern power distribution systems, the rise of industrial Internet and wireless sensor technology has provided new opportunities for intelligent monitoring and management of power systems, and automation and intelligent inspections have emerged. However, existing technologies have the following deficiencies in equipment inspection and maintenance:

[0003] Traditional inspection systems mainly rely on manual records and judgments, experience and simple statistical methods, and lack intelligent analysis methods. They are difficult to accurately detect, analyze and evaluate the operating status of equipment, resulting in the inability to timely and accurately analyze and utilize inspection data, making it difficult to timely discover potential equipment failures or safety hazards.

[0004] Since the maintenance and management of the existing power system mainly relies on manual scheduling and experience judgment, and lacks a scientific task allocation and resource scheduling mechanism, manual inspections require a lot of time and manpower costs, and are limited by personnel experience and subjective judgment, it is difficult to improve inspection efficiency. In addition, the existing inspection system lacks real-time and early warning functions and cannot achieve real-time monitoring and early warning. This results in some potential equipment failures or safety hazards not being discovered and handled in a timely manner, increasing the risk of power system operation. Summary of the invention

[0005] In order to solve the problems of lack of data analysis means and lack of resource scheduling mechanism in existing inspection and maintenance systems, resulting in low maintenance efficiency, the present invention provides an intelligent inspection system for power distribution, which builds and calculates the health assessment index of equipment in each area through automated collection of measurement data and preprocessing methods; timely discovers and assigns tasks, handles potential faults, and improves the maintenance efficiency and accuracy of the distribution network.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An intelligent inspection system for power distribution, comprising a data acquisition module, a communication transmission module and a cloud service platform;

[0008] The data acquisition module is used to collect measurement data in the power distribution base station through mobile detection equipment, and obtain relevant parameters of aging of each factory equipment; the measurement data includes image, thermal imaging, temperature and noise data;

[0009] The communication transmission module is used to transmit the captured sensor measurement data to the cloud service platform;

[0010] The mobile detection device includes a robot or a drone.

[0011] Preferably, a high-definition camera is used to capture the image of the device; a thermal imager is used to capture the thermal image of the device; and both images are ensured to be captured at the same time point and at the same angle to ensure image consistency.

[0012] The cloud service platform includes an edge gateway module, a cloud computing module and a work management module; the cloud service platform is used to connect the data acquisition module, the edge gateway module, the cloud computing module and the work management module together;

[0013] The edge gateway module is used to receive the measurement data sent by the data acquisition module through the wireless personal area network and perform a series of pre-processing processes;

[0014] The cloud computing module is used to process the measurement data and calculate the comprehensive health assessment index;

[0015] The work management module is used to automatically generate maintenance tasks according to the comprehensive health assessment index of the power distribution base station and assign them to appropriate maintenance personnel; and track the progress of maintenance tasks in real time.

[0016] Preferably, the cloud computing module further comprises: outputting a conventional failure prediction probability of a corresponding device according to a health assessment index of multiple targeted measurement data;

[0017] The work management module also includes: matching tools and accessories required for maintenance tasks according to fault prediction results to ensure that maintenance personnel can complete tasks efficiently;

[0018] After the maintenance is completed, the maintenance personnel will update the task progress and re-collect the measurement data to ensure the real-time and accuracy of the system data;

[0019] Generate maintenance reports to record maintenance processes and results, providing a basis for subsequent analysis and decision-making.

[0020] Preferably, the steps for calculating the comprehensive health assessment index are as follows:

[0021] The Canny edge detection algorithm is used to extract key feature edges from the device image and thermal image, and the key feature edges of the device image and thermal image are matched.

[0022] According to the structure of the device image and the key feature edges, the device image and the thermal image are synchronously divided into multiple areas to ensure that the area positions of the two are consistent.

[0023] According to the temperature data, noise data and equipment aging data corresponding to each area in the equipment image or thermal imaging map, the health index model is used to evaluate the health status of the comprehensive health assessment index of each area in the power distribution equipment. The specific calculation method is as follows:

[0024] EHI=∑[W i ·f i (X i )]

[0025] Where: EHI represents the equipment health index; X i Indicates different types of measurement data; W i Represents the weight coefficient of the corresponding measurement data; f i Indicates the corresponding measurement data.

[0026] Preferably, the implementation steps of the Canny edge detection algorithm are as follows:

[0027] Gaussian filtering: Use a Gaussian filter to smooth the image and reduce noise interference.

[0028] Non-maximum suppression:

[0029] For each pixel, check the pixel values ​​in its neighborhood according to the gradient direction, retain the local maximum and suppress the non-maximum.

[0030] Double threshold processing:

[0031] Classify the gradient magnitude using high and low thresholds;

[0032] Pixels above the high threshold are marked as strong edges;

[0033] Pixels between the high and low thresholds are marked as weak edges;

[0034] Pixels below the low threshold are suppressed.

[0035] Edge Connection:

[0036] Through the hysteresis threshold method, the weak edges are connected to the strong edges to form a complete edge.

[0037] Preferably, the preprocessing process includes image alignment, the steps are as follows:

[0038] Image registration technology is used to perform geometric correction and alignment on the device image and thermal image to ensure that their spatial positions are consistent.

[0039] Preferably, the preprocessing process includes a temperature data extraction process, the steps of which are as follows:

[0040] Reading images: Use OpenCV to read thermal images.

[0041] Convert color space: Convert RGB color thermal imaging images to grayscale images.

[0042] Linear conversion: The temperature range and grayscale value range are obtained through actual measurement, and the grayscale value is converted to the temperature value through a linear relationship.

[0043] Formula: T = kG + b

[0044] Where T is the temperature value, G is the grayscale value, and k and b are linear conversion coefficients.

[0045] Calculate the temperature data for each area in the thermal image.

[0046] Temperature data normalization: Temperature data were normalized to the range of 0-1 using Z-score or Min-Max normalization methods.

[0047] Preferably, the preprocessing process includes noise data normalization: using Z-score or Min-Max normalization method to normalize the noise data to a range of 0-1.

[0048] Preferably, the preprocessing process includes normalization of equipment aging data: an exponential function is used to simulate the decay of equipment performance over time, which is expressed as:

[0049]

[0050] Where: λ represents the attenuation coefficient; C i Represents time-related parameters; calculated from the relevant parameters of factory equipment aging.

[0051] Beneficial effects of the present invention:

[0052] 1. Through automated data collection and preprocessing, the health assessment index of the equipment in each area is established and calculated; hierarchical inspection and screening are carried out according to the health assessment index to timely discover and handle potential faults, achieve real-time monitoring and early warning, timely discover and handle potential faults, improve the maintenance efficiency and accuracy of the power distribution inspection system and distribution network, accurately identify safety hazards in the operation of the power system, and reduce the risk of power operation.

[0053] 2. Extend the service life of equipment and reduce equipment replacement costs through regular health index assessment and preventive maintenance.

[0054] 3. Predict faults based on health assessment index and historical routine faults, match required maintenance resources, and improve maintenance efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0056] Figure 1 This is a structural framework diagram of an intelligent inspection system for electric power distribution according to the present invention.

[0057] Figure 2 The present invention is a flowchart of the implementation steps of an intelligent inspection system for power distribution.

[0058] Figure 3 The present invention provides a flow chart of a Canny edge detection algorithm in an intelligent inspection system for electric power distribution. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] In power grid projects, a distribution base station (or distribution substation) is a key facility used to convert high-voltage electric energy into low-voltage electric energy suitable for users and distribute it to various users. The main function of a distribution base station is to receive electric energy from high-voltage transmission lines, reduce the voltage through transformers, and distribute it to various low-voltage distribution lines. The layout of a distribution base station usually includes the following parts:

[0061] High voltage area: includes high voltage switchgear and high voltage side of transformer.

[0062] Transformer area: including distribution transformers, heat dissipation equipment, etc.

[0063] Low voltage area: includes the low voltage side of the transformer and low voltage switchgear.

[0064] See also Figure 1-Figure 3 As shown, an intelligent inspection system for power distribution includes a data acquisition module, a communication transmission module and a cloud service platform; the cloud service platform includes an edge gateway module, a cloud computing module and a work management module;

[0065] Data acquisition module: Use mobile detection equipment (such as robots or drones) to collect sensor measurement data in distribution base stations, including images, thermal imaging, temperature and noise data.

[0066] Optionally, the measurement data of this embodiment is not limited to the above-mentioned images, thermal imaging, temperature and noise data, and may also include current data, load data, vibration data, etc.

[0067] The data acquisition module also includes obtaining relevant parameters of aging of each factory equipment.

[0068] Optionally, use a high-definition camera to capture the device image; use a thermal imager to capture the device thermal image; ensure that the two images are captured at the same time and angle to ensure image consistency. The device image shows the appearance and structure of the device. The thermal image shows the temperature distribution of the device. The high-definition camera and thermal imager can be integrated in the same mobile detection device for synchronous acquisition.

[0069] The communication transmission module is used to transmit the captured sensor measurement data to the cloud service platform.

[0070] Specifically, low-power wireless personal area network (LPWAN) technology is used to ensure the stability and real-time performance of data transmission. LPWAN technology has the characteristics of low power consumption and long-distance transmission, which is suitable for the data transmission needs of power distribution base stations.

[0071] Cloud service platform: provides a cloud-based IoT operating system that connects the data acquisition module, edge gateway module, cloud computing module, and work management module together.

[0072] Edge gateway module: runs on the cloud service platform, receives the measurement data sent by the data acquisition module by building a stable low-power wireless personal area network, and performs a series of pre-processing processes.

[0073] Specifically, the use of edge computing technology can reduce cloud computing pressure and improve data processing efficiency. The edge gateway module receives and pre-processes data by building a stable, low-power wireless personal area network, including image alignment, temperature data extraction and standardization, noise data standardization, and equipment aging data standardization.

[0074] Cloud computing module: runs on the cloud service platform, is used to process measurement data, calculate the comprehensive health assessment index, and output the conventional failure prediction probability of the corresponding equipment;

[0075] Work management module: runs on the cloud service platform, automatically generates maintenance tasks based on the comprehensive health assessment index and fault prediction results of the distribution base station, and assigns them to appropriate maintenance personnel;

[0076] Match the tools and accessories required for maintenance tasks to ensure that maintenance personnel can complete tasks efficiently;

[0077] Track the progress of maintenance tasks in real time to ensure they are completed on time;

[0078] After the maintenance is completed, the maintenance personnel will update the task progress and re-collect the measurement data to ensure the real-time and accuracy of the system data;

[0079] Generate maintenance reports to record maintenance processes and results, providing a basis for subsequent analysis and decision-making.

[0080] Furthermore, the operation process of the power distribution intelligent inspection system is as follows:

[0081] S1) Data collection:

[0082] Use mobile monitoring equipment to collect equipment images, thermal images, and noise data in the power distribution base station. The equipment images and thermal images must be collected at the same time and angle to ensure consistency.

[0083] Obtain relevant parameters of aging of each device;

[0084] S2) Data transmission:

[0085] The collected data is transmitted to the edge gateway module of the cloud service platform through the communication transmission module.

[0086] S3) Data preprocessing:

[0087] The edge gateway module receives the data and performs a series of preprocessing, including the following:

[0088] Image alignment: Use image registration technology to perform geometric correction on the device image and thermal image to ensure that their spatial positions are consistent;

[0089] Temperature data extraction: Convert the RGB color thermal imaging image into a grayscale image, and convert the grayscale value into a temperature value through a linear relationship;

[0090] Standardize the temperature data, noise data and equipment aging data.

[0091] S4) Data analysis and calculation:

[0092] The cloud computing module calculates the preprocessed images and data and uses the Canny edge detection algorithm to extract key feature edges from the device images and thermal images;

[0093] Match the key feature edges of the device image and the thermal image, and divide them into multiple areas simultaneously to ensure that the area positions of the two are consistent;

[0094] A comprehensive health assessment index is calculated based on the temperature data, noise data and equipment aging data of each area in the equipment image or thermal imaging map.

[0095] S5) Task Scheduling:

[0096] The work management module automatically generates maintenance tasks based on the health assessment index and fault prediction results and assigns them to appropriate staff.

[0097] Match the tools and accessories required for maintenance tasks to ensure that maintenance personnel can complete their tasks efficiently.

[0098] S6) Task execution and feedback:

[0099] Workers perform maintenance tasks and provide updates on task progress.

[0100] After maintenance is completed, re-collect measurement data to ensure the real-time and accuracy of system data.

[0101] Further, the calculation steps of the comprehensive health assessment index are as follows:

[0102] The Canny edge detection algorithm is used to extract key feature edges from the device image and thermal image, and the key feature edges of the device image and thermal image are matched.

[0103] According to the structure of the device image and the key feature edges, the device image and the thermal image are synchronously divided into multiple areas to ensure that the area positions of the two are consistent.

[0104] According to the temperature data, noise data and equipment aging data corresponding to each area in the equipment image or thermal imaging map, the health index model is used to evaluate the health status of the comprehensive health assessment index of each area in the power distribution equipment. The specific calculation method is as follows:

[0105] EHI=∑[W i ·f i (X i )];

[0106] Where: EHI represents the equipment health index; X i Indicates different types of measurement data; W i Represents the weight coefficient of the corresponding measurement data; f i Indicates the corresponding measurement data.

[0107] Specifically, equipment health index assessment is a method to quantify the comprehensive operating status of power equipment. The health index ranges from 0 to 1, and a higher value means a better health status of the equipment. i Depending on the specific measurement data and correlation evaluation requirements, a function representation based on experience, statistics or machine learning methods can be used to convert the raw measurement data into indicators that can reflect the health status of the equipment. i Including temperature data, noise data and equipment aging data.

[0108] Further, the implementation steps of the Canny edge detection algorithm are as follows:

[0109] Gaussian filtering: Use a Gaussian filter to smooth the image and reduce noise interference. The Gaussian filter smoothes the noise in the image through convolution operations while retaining edge information; the calculation example is as follows:

[0110]

[0111] In the formula, H represents the matrix of the Gaussian filter, which is usually a 3x3 matrix used for convolution operation; A represents the pixel value matrix of the original image; e represents the pixel value of the image after Gaussian filtering; the two matrices are multiplied element-wise through the convolution operation, and then all the values ​​are added to obtain the new value of the target pixel.

[0112] Calculate the gradient: Use the Sobel operator to calculate the gradient magnitude and direction of the image. The Sobel operator calculates the gradient of the image in the x and y directions through convolution operations.

[0113] The calculation example is as follows: The gradient amplitude is:

[0114]

[0115] The gradient direction is:

[0116] θ=arctan(G y / G x );

[0117] In the above formula, G: gradient amplitude, which indicates the rate of change of brightness at a certain point in the image;

[0118] θ: Gradient direction, which indicates the direction of brightness change at a certain point in the image;

[0119] Gx: gradient in the x direction, reflecting the brightness change of the image in the horizontal direction;

[0120] Gy: gradient in the y direction, reflecting the brightness change of the image in the vertical direction;

[0121] Non-maximum suppression: For each pixel point, check the pixel values ​​in its neighborhood according to the gradient direction, retain the local maximum value, and suppress the non-maximum value. By comparing the gradient amplitude, suppress non-edge points, retain the local maximum value, and eliminate the stray response caused by edge detection.

[0122] Double thresholding: Use a high threshold and a low threshold to classify the gradient magnitude:

[0123] Pixels above the high threshold are marked as strong edges;

[0124] Pixels between the high and low thresholds are marked as weak edges;

[0125] Pixels below the low threshold are suppressed.

[0126] Edge connection: Through the hysteresis threshold method, weak edges are connected to strong edges to form a complete edge.

[0127] Furthermore, the preprocessing process includes image alignment, and the specific steps are as follows:

[0128] Image registration technology is used to perform geometric correction and alignment on the device image and thermal image to ensure that their spatial positions are consistent.

[0129] Preferably, the preprocessing process includes a temperature data extraction process, the steps of which are as follows:

[0130] Reading images: Use OpenCV to read thermal images.

[0131] Convert color space: Convert RGB color thermal imaging images to grayscale images.

[0132] Linear conversion: The temperature range and grayscale value range are obtained through actual measurement, and the grayscale value is converted to the temperature value through a linear relationship.

[0133] Formula: T = kG + b

[0134] Where T is the temperature value, G is the grayscale value, and k and b are linear conversion coefficients.

[0135] Calculate the temperature data for each area in the thermal image.

[0136] Temperature data standardization: Use Z-score or Min-Max standardization method to standardize temperature data to the range of 0-1. Using the characteristics of standard normal distribution, the data is converted to standard normal distribution through Z-score standardization method.

[0137] Furthermore, the preprocessing process includes noise data normalization: using Z-score or Min-Max normalization method to normalize the noise data to a range of 0-1.

[0138] Specifically,

[0139] Furthermore, the preprocessing process includes normalization of equipment aging data: an exponential function is used to simulate the decay of equipment performance over time, which is expressed as:

[0140]

[0141] Where: λ represents the attenuation coefficient; C iIndicates time-related parameters, which can be calculated from the relevant parameters of the equipment when it leaves the factory.

[0142] Specifically, the measurement data f i Including temperature data, noise data and equipment aging data, the standardized data format is as follows:

[0143] Temperature data: The temperature in a certain area is 75°C, which is 0.46 after standardization.

[0144] Noise data: The noise level in a certain area is 50dB, which is 0.28 after standardization.

[0145] Equipment aging data: The equipment in a certain area is 10 years old, which is 0.81 after standardization.

[0146] Then calculate the comprehensive health assessment index according to the health index model. The calculation example is as follows:

[0147] EHI=W1·0.46+W2·0.28+W3·0.81;

[0148] Where: EHI represents the equipment health index; W 1、 W 2、 W3 represents the weight coefficient of the corresponding measurement data.

[0149] Through the above calculations, the health index assessment of the power distribution equipment is completed.

[0150] Further, the cloud computing module also includes: outputting a conventional failure prediction probability of a corresponding device according to a health assessment index of a plurality of measurement data;

[0151] The work management module also includes: matching tools and accessories required for maintenance tasks according to fault prediction results to ensure that maintenance personnel can complete tasks efficiently;

[0152] After the maintenance is completed, the maintenance personnel will update the task progress and re-collect the measurement data to ensure the real-time and accuracy of the system data;

[0153] Generate maintenance reports, record maintenance processes and results, and provide a basis for subsequent analysis and decision-making. Output the predicted probability of failure of equipment in the area to ensure efficient operation and maintenance of the system.

[0154] Specifically, assuming that the equipment in this area is a transformer, the calculated health index is 0.6. According to historical data query, when the health assessment index exceeds 0.56, maintenance needs to be arranged. Therefore, the probability of conventional failure of the transformer is predicted by further using targeted multiple measurement data, for example:

[0155] Insulation aging: Predict failure probability using temperature and aging data.

[0156] Overload failure: Predict failure probability based on load data and temperature data.

[0157] Oil Leakage: Predicting Failure Probability Using Image and Temperature Data.

[0158] Other types of equipment include: The probability of conventional failure of circuit breakers:

[0159] Contact erosion: Predicting failure probability using current and temperature data.

[0160] Mechanical failure: Predict failure probability using vibration and noise data.

[0161] In this way, maintenance personnel can be matched with the tools and accessories required for maintenance tasks, ensuring that they can complete their tasks efficiently.

[0162] The invention provides a power distribution intelligent inspection system, provides a cloud-based Internet of Things operating system, connects a data acquisition module, an edge gateway module, a cloud computing module and a work management module together, and realizes centralized data processing and management. The Canny edge detection algorithm and the health index model are used to process the measurement data, calculate the comprehensive health assessment index, and output the conventional fault prediction probability of the corresponding equipment.

[0163] This power distribution intelligent system achieves efficient, accurate and intelligent inspection and maintenance through comprehensive monitoring and prediction, ensuring the stable operation of the distribution system and improving the overall reliability of the system.

[0164] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0165] In the several embodiments provided in the present application, it should be understood that the disclosed platforms, modules and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, platforms or modules, which can be electrical, mechanical or other forms.

[0166] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software functional modules.

[0167] If the module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

[0168] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0169] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent inspection system for power distribution, characterized in that: Including data acquisition module, communication transmission module and cloud service platform; The data acquisition module is used to collect measurement data in the power distribution base station through mobile detection equipment, and obtain relevant parameters of aging of each factory equipment; the measurement data includes image, thermal imaging, temperature and noise data; The communication transmission module is used to transmit the captured sensor measurement data to the cloud service platform; The cloud service platform includes an edge gateway module, a cloud computing module and a work management module; The cloud service platform is used to connect the data acquisition module, edge gateway module, cloud computing module, and work management module together; The edge gateway module is used to receive the measurement data sent by the data acquisition module through the wireless personal area network and perform a series of pre-processing processes; The cloud computing module is used to process the measurement data and calculate the comprehensive health assessment index; The work management module is used to automatically generate maintenance tasks according to the comprehensive health assessment index of the power distribution base station and assign them to appropriate maintenance personnel; and track the progress of maintenance tasks in real time.

2. The intelligent inspection system for power distribution according to claim 1, characterized in that: The mobile detection device includes a robot or a drone.

3. The intelligent inspection system for power distribution according to claim 1, characterized in that: The calculation steps of the comprehensive health assessment index are as follows: Use the Canny edge detection algorithm to extract key feature edges from the device image and thermal image, and match the key feature edges of the device image and thermal image; According to the structure and key feature edges of the device image, the device image and thermal image are synchronously divided into multiple areas to ensure that the area positions of the two are consistent; According to the temperature data, noise data and equipment aging data corresponding to each area in the equipment image or thermal imaging map, the health index model is used to evaluate the health status of the comprehensive health assessment index of each area in the power distribution equipment. The specific calculation formula is as follows: EHI=∑[W i f i (X i )]; Where: EHI represents the equipment health index; X i Indicates different types of measurement data; W i Represents the weight coefficient of the corresponding measurement data; f i Indicates the corresponding measurement data.

4. The intelligent inspection system for power distribution according to claim 3 is characterized in that: The implementation steps of the Canny edge detection algorithm are as follows: Gaussian filtering: Use a Gaussian filter to smooth the image; Non-maximum suppression: For each pixel point, check the pixel values ​​in its neighborhood according to the gradient direction, retain the local maximum value, and suppress the non-maximum value; Double threshold processing: Use high and low thresholds to classify the gradient magnitude, Pixels above the high threshold are marked as strong edges. Pixels between the high threshold and the low threshold are marked as weak edges. Pixels below the low threshold are suppressed; Edge connection: Through the hysteresis threshold method, weak edges are connected to strong edges to form a complete edge.

5. The intelligent inspection system for power distribution according to claim 1, characterized in that: The preprocessing process includes image alignment, and the steps are as follows: Image registration technology is used to perform geometric correction and alignment on the device image and thermal image to ensure that their spatial positions are consistent.

6. The intelligent inspection system for power distribution according to claim 1, characterized in that: The preprocessing process includes a temperature data extraction process, and the steps are as follows: Read image: Use OpenCV to read thermal images; Convert color space: Convert RGB color thermal imaging images to grayscale images; Linear conversion: The temperature range and gray value range are obtained through actual measurement, and the gray value is converted into the temperature value through a linear relationship; the conversion formula is: T = kG + b; Where T is the temperature value, G is the gray value, k and b are linear conversion coefficients; Calculate the temperature data of each area in the thermal imaging image according to the conversion formula; Temperature data normalization: Temperature data were normalized to the range of 0-1 using Z-score or Min-Max normalization methods.

7. The intelligent inspection system for power distribution according to claim 1, characterized in that: The preprocessing process includes normalization of noisy data: the noisy data is normalized to the range of 0-1 using Z-score or Min-Max normalization method.

8. The intelligent inspection system for power distribution according to claim 1, characterized in that: The preprocessing process includes equipment aging data normalization: an exponential function is used to simulate the decay of equipment performance over time, which is expressed as: Where: λ represents the attenuation coefficient; C i Indicates time-related parameters; calculated from the relevant parameters of factory equipment aging.

9. The intelligent inspection system for power distribution according to claim 3, characterized in that: The cloud computing module further includes: outputting a conventional failure prediction probability of a corresponding device according to a health assessment index of multiple targeted measurement data; The work management module also includes: matching tools and accessories required for maintenance tasks according to the fault prediction results; After the maintenance is completed, the maintenance personnel will update the task progress and re-collect the measurement data; Generate maintenance reports to record maintenance processes and results.

10. An intelligent inspection system for power distribution according to any one of claims 1 to 7, characterized in that: The operation process of the system is as follows: S1) Data collection: Use mobile monitoring equipment to collect equipment images, thermal images, and noise data in the power distribution base station. The equipment images and thermal images must be collected at the same time and angle to ensure consistency. Obtain relevant parameters of aging of each factory equipment; S2) Data transmission: The collected data is transmitted to the edge gateway module of the cloud service platform through the communication transmission module; S3) Data preprocessing: The edge gateway module receives the data and performs a series of preprocessing, including the following: Image alignment: Use image registration technology to perform geometric correction on the device image and thermal image to ensure that their spatial positions are consistent; Temperature data extraction: Convert the RGB color thermal imaging image into a grayscale image, and convert the grayscale value into a temperature value through a linear relationship; Standardize temperature data, noise data and equipment aging data; S4) Data analysis and calculation: The cloud computing module calculates the preprocessed images and data and uses the Canny edge detection algorithm to extract key feature edges from the device images and thermal images; Match the key feature edges of the device image and the thermal image, and divide them into multiple areas simultaneously to ensure that the area positions of the two are consistent; A comprehensive health assessment index is calculated based on the temperature data, noise data and equipment aging data of each area in the equipment image or thermal imaging map.

Images