Power supply equipment insulator cleanliness detection method and system

Abstract

The invention relates to a power supply equipment insulator cleanliness detection method and system. The power supply equipment insulator cleanliness detection method comprises the steps of obtaining a hyperspectral image and a speckle image of the surface of an insulator through cooperative work of a hyperspectral imaging module and a laser speckle detection module, calculating a salt crystallization characteristic coefficient according to hyperspectral image data, converting a salt substance characteristic absorption peak into a quantifiable index, and calculating the cleanliness of the insulator according to the quantifiable index. According to the method, the salt crystallization identification accuracy is improved, the root-mean-square roughness and the skewness coefficient are calculated according to speckle image data, laser speckle images are converted into quantifiable surface topography parameters, the identification accuracy of insulator surface contamination types is improved, and dual detection of contamination components and surface topography is achieved; by establishing the coupling relation among the salt crystallization characteristic coefficient, the root-mean-square roughness and the skewness coefficient, precise classification of the dirt is effectively realized, the adaptability of a cleaning strategy is improved, and the cleaning cost is reduced.

Description

Technical Field

[0001] This application relates to the field of cleanliness testing technology, and in particular to a method and system for testing the cleanliness of insulators of power supply equipment. Background Technology

[0002] The safe and stable operation of power systems largely depends on the cleanliness of insulators. As key components in power systems that perform both electrical insulation and mechanical fixation, insulators are inevitably affected by environmental pollution during operation. With the acceleration of industrial development and urbanization, the problem of insulator surface contamination is becoming increasingly serious, directly affecting the safe operation of the power grid. To address power outages caused by insulator contamination on high-voltage transmission lines, a remote insulator contamination monitoring system has been developed. By studying the mechanisms of contamination formation, natural cleaning, wetting, and flashover, a theoretical basis for insulator contamination monitoring has been provided.

[0003] Currently, insulator contamination detection mainly employs traditional methods such as the equivalent salt density method, surface contamination conductivity method, and leakage current method. While these methods are simple and easy to implement, they suffer from time-consuming and inefficient processes. More importantly, these methods cannot effectively distinguish between easily cleanable dust and stubborn salt crystals on the insulator surface. Dust, typically composed of airborne dirt and grime, has weak adhesion and is easily removed by simple cleaning methods. Salt crystals, on the other hand, are formed by industrial emissions, sea salt, or other environmental factors that condense on the insulator surface. They have strong adhesion, a stable structure, and require more complex treatment methods for effective removal. Existing technology cannot differentiate between these two types of contamination, leading maintenance personnel to often adopt a "one-size-fits-all" cleaning strategy. This can result in wasted resources and may fail to completely remove stubborn salt crystals, leaving safety hazards.

[0004] Therefore, it is of great significance to develop a detection method and system that can accurately distinguish between easily cleanable dust and stubborn salt crystals, improve the cleaning efficiency of insulators, reduce maintenance costs, and ensure the safe operation of the power system. Summary of the Invention

[0005] To address or partially address the problems existing in related technologies, this application provides a method and system for detecting the cleanliness of insulators in power supply equipment, aiming to solve the problem of how to accurately distinguish different types of contamination and assess their cleaning difficulty.

[0006] The first aspect of this application provides a method for detecting the cleanliness of insulators in power supply equipment, including: The hyperspectral imaging module and the laser speckle detection module are controlled to perform a coordinated scan of the insulator to acquire hyperspectral and speckle images of the insulator surface. Perform radiometric and atmospheric corrections on hyperspectral images, and calculate salt crystallization characteristic coefficients; Multi-scale denoising was performed on the speckle image, surface roughness parameters were calculated, and root mean square roughness and skewness coefficient were extracted. Based on the threshold for determining the type of contamination on the insulator surface, the type of contamination on the insulator surface is determined according to the coupling relationship between the salt crystallization characteristic coefficient, the root mean square roughness, and the skewness coefficient. Based on the type of dirt, the system matches cleaning strategies within a pre-defined database of cleaning methods and outputs cleaning recommendations.

[0007] Optionally, in some embodiments of the first aspect, calculating the salt crystallization characteristic coefficient includes: The characteristic absorption peaks of salt substances in hyperspectral data are converted into quantifiable indicators, and the formula for calculating the salt crystallization characteristic coefficient is as follows: (2) In the formula, This represents the corrected spectral reflectance function. Indicates spectral wavelength, This indicates the threshold value set for the spectral wavelength. This indicates the threshold value set for the spectral wavelength. Represents the spectral compensation coefficient. Indicates the wavelength sampling interval. This represents the weighting function for salt characteristics calibrated in the experiment. satisfy .

[0008] Optionally, in some embodiments of the first aspect, the extraction of root mean square roughness and skewness coefficient includes: The formula for calculating root mean square roughness is: (3) The formula for calculating the skewness coefficient is as follows: (4) In the formula, This represents the root mean square roughness, which is the square mean of the surface height deviation. Indicates the number of sampling points. Indicates the first Surface height deviation at each sampling point Indicates the skewness coefficient; By accurately calculating the root mean square roughness and skewness coefficient, the laser speckle image is transformed into quantifiable surface morphology parameters, enabling in-depth analysis of the fouling structure and facilitating subsequent differentiation between salt crystals and loose salt particles.

[0009] Optionally, in some embodiments of the first aspect, the threshold for determining the type of contamination on the insulator surface includes: Dynamic compensation is applied to the threshold for determining the type of contamination on the insulator surface based on ambient temperature and relative humidity. (5) In the formula, This indicates the judgment threshold after compensation. This represents the calibrated value of the judgment threshold. This represents the temperature compensation coefficient, where T represents the ambient temperature and RH represents the relative humidity.

[0010] Optionally, in some embodiments of the first aspect, determining the calibration value of the threshold includes: The threshold for determining the first feature coefficient is 0.75, the threshold for determining the second feature coefficient is 0.6, and the threshold for determining the third feature coefficient is 0.4. The first roughness threshold is 1.2. The second roughness threshold is 0.8. ; The threshold for determining the first skewness coefficient is 0.3, and the threshold for determining the second skewness coefficient is 0.5.

[0011] Optionally, in some embodiments of the first aspect, determining the type of contamination on the insulator surface includes: If the salt crystallization characteristic coefficient is greater than or equal to the first characteristic coefficient judgment threshold, the root mean square roughness is greater than or equal to the first roughness judgment threshold, and the absolute value of the skewness coefficient is less than the first skewness coefficient judgment threshold, it is judged as salt crystallization fouling. Salt crystallization characteristic coefficient ≥ second characteristic coefficient judgment threshold, first roughness judgment threshold > root mean square roughness ≥ second roughness judgment threshold, absolute value of skewness coefficient > second skewness coefficient judgment threshold, judged as loose salt particle contamination. If the salt crystallization characteristic coefficient is less than the third characteristic coefficient judgment threshold and the root mean square roughness is less than the second roughness judgment threshold, it is judged as floating dust and dirt. Among them, the first characteristic coefficient determination threshold is greater than the second characteristic coefficient determination threshold and the third characteristic coefficient determination threshold.

[0012] Optionally, in some embodiments of the first aspect, controlling the hyperspectral imaging module and the laser speckle detection module to perform coordinated scanning of the insulator includes: The hyperspectral imaging module and the laser speckle detection module perform coordinated scanning through a spatiotemporal synchronization device, and the synchronization error satisfies: (1) In the formula, For scanning spacing, For scanning speed, For optical path difference, The speed of light; By controlling the time synchronization error of spatial alignment to within 1ms, the positional deviation between hyperspectral and laser scanning is reduced, thereby improving the accuracy of salt crystal identification.

[0013] A second aspect of this application provides a system for detecting the cleanliness of insulators in power supply equipment, comprising: The system comprises a hyperspectral imaging module, a laser speckle detection module, a data fusion processing unit, and a cleaning strategy decision-making module. The hyperspectral imaging module is used to acquire hyperspectral images of the insulator surface and extract the compositional characteristics of the pollutants. The laser speckle detection module is used to emit pulsed lasers onto the surface of insulators, acquire speckle images, and calculate the surface roughness parameters of the insulators. The data fusion processing unit is connected to the hyperspectral imaging module and the laser speckle detection module respectively, and determines the type of contamination on the insulator surface based on the coupling relationship between the salt crystallization characteristic coefficient, root mean square roughness and skewness coefficient. The cleaning strategy decision module is equipped with a cleaning method database, which matches cleaning strategy decisions within the preset cleaning method database based on the type of dirt.

[0014] Optionally, in some embodiments of the second aspect, the power supply equipment insulator cleanliness detection system is integrated into the unmanned aerial vehicle (UAV) platform.

[0015] The technical solution provided in this application may include the following beneficial effects: By working in tandem with the hyperspectral imaging module and the laser speckle detection module, hyperspectral and speckle images of the insulator surface are acquired. Salt crystallization characteristic coefficients are calculated based on the hyperspectral image data, converting the characteristic absorption peaks of salt substances into quantifiable indicators, thereby improving the accuracy of salt crystallization identification. Root mean square roughness and skewness coefficients are calculated based on the speckle image data, converting the laser speckle images into quantifiable surface morphology parameters, thereby improving the accuracy of identifying contamination types on the insulator surface and achieving dual detection of contamination components and surface morphology. By establishing the coupling relationship between salt crystallization characteristic coefficients, root mean square roughness, and skewness coefficients, accurate classification of contamination is effectively achieved, improving the adaptability of cleaning strategies and reducing cleaning costs.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0018] Figure 1 This is a schematic flowchart illustrating the method for detecting the cleanliness of insulators in power supply equipment, as shown in an embodiment of this application. Figure 2This is a schematic diagram of the structure of a power supply equipment insulator cleanliness detection system shown in an embodiment of this application. Detailed Implementation

[0019] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0020] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0021] In the description of this application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0022] Unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0023] Currently, insulator contamination detection mainly employs traditional methods such as the equivalent salt density method, surface contamination conductivity method, and leakage current method. While these methods are simple and easy to implement, they suffer from time-consuming and inefficient processes. More importantly, these methods cannot effectively distinguish between easily cleanable dust and stubborn salt crystals on the insulator surface. Dust, typically composed of airborne dirt and grime, has weak adhesion and is easily removed by simple cleaning methods. Salt crystals, on the other hand, are formed by industrial emissions, sea salt, or other environmental factors that condense on the insulator surface. They have strong adhesion, a stable structure, and require more complex treatment methods for effective removal. Existing technology cannot differentiate between these two types of contamination, leading maintenance personnel to often adopt a "one-size-fits-all" cleaning strategy. This can result in wasted resources and may fail to completely remove stubborn salt crystals, leaving safety hazards.

[0024] To address the aforementioned issues, this application provides a method and system for detecting the cleanliness of power supply equipment insulators. This system, through the collaborative operation of a hyperspectral imaging module and a laser speckle detection module, acquires hyperspectral and speckle images of the insulator surface. Based on the hyperspectral image data, it calculates salt crystallization characteristic coefficients, converting the characteristic absorption peaks of salt substances into quantifiable indicators, thus improving the accuracy of salt crystallization identification. Based on the speckle image data, it calculates root mean square roughness and skewness coefficients, converting the laser speckle image into quantifiable surface morphology parameters, thereby improving the accuracy of identifying the type of contamination on the insulator surface and achieving dual detection of contamination components and surface morphology. By establishing a coupling relationship between the salt crystallization characteristic coefficients, root mean square roughness, and skewness coefficients, it effectively achieves accurate classification of contamination, improves the adaptability of cleaning strategies, and reduces cleaning costs.

[0025] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0026] Figure 1 This is a schematic flowchart illustrating a method for detecting the cleanliness of insulators in power supply equipment, as shown in an embodiment of this application.

[0027] See Figure 1 A method for detecting the cleanliness of insulators in power supply equipment, comprising: S101. Control the hyperspectral imaging module and the laser speckle detection module to perform a coordinated scan of the insulator to obtain a hyperspectral image and speckle image of the insulator surface; Specifically, the hyperspectral imaging module and the laser speckle detection module perform coordinated scanning through a spatiotemporal synchronization device, and the synchronization error satisfies: (1) In the formula, For scanning spacing, For scanning speed, For optical path difference, The speed of light; Controlling the time synchronization error of spatial alignment to within 1ms can reduce the positional deviation between hyperspectral and laser scanning, thereby improving the accuracy of salt crystal identification. Specifically, for the mechanical synchronization scanning item ( In terms of physical meaning, it refers to the distance the scanning probe moves. The required time, and also for the optical path difference compensation item ( In terms of physical meaning, it refers to the time difference between laser and spectral signal transmission. This allows for speed matching not only through closed-loop motor control but also provides a safety margin of three orders of magnitude. ).

[0028] S102. Perform radiometric and atmospheric corrections on the hyperspectral image and calculate the salt crystallization characteristic coefficients. Specifically, the calculation of salt crystallization characteristic coefficients includes: Salt crystals exhibit strong absorption valleys in specific wavelength ranges (e.g., 1350 nm), while dust exhibits gentle reflection. Formula (2) amplifies this difference through second-order derivative calculations. For example, salt crystals produce steep absorption valleys near 1350 nm due to ion vibrations, resulting in extreme pulses in the second derivative at these points. In contrast, the spectrum of dust is smoother, leading to smaller second-order derivatives. The characteristic coefficient of salt crystals is more than twice that of dust, thus reliably distinguishing between dust and salt crystals. By defining the characteristic coefficient of salt crystals... The calculation formula (including spectral reflectance function) Compensation coefficient and weight function ), to extract the characteristic absorption peaks of salts from hyperspectral data (e.g., It is 1200nm. The characteristic coefficient of salt crystallization (1350nm) is converted into a quantifiable index, thereby improving the accuracy of salt crystallization identification. The formula for calculating the characteristic coefficient of salt crystallization is: (2) In the formula, This represents the corrected spectral reflectance function; Indicates spectral wavelength, This indicates the threshold value set for the spectral wavelength. This indicates the threshold value set for the spectral wavelength, limiting the wavelength. to Adjustments can be made based on local conditions. For example, in coastal areas, the main salt crystals are generally sodium chloride, while in inland areas near factories, the main salt crystals are generally sodium sulfate. and The specific values ​​can be adjusted adaptively according to the type of crystallization of the main salt; This represents the spectral compensation coefficient, which is mainly used to balance dimensions. At the same time, it also allows for adjustments to the formula by changing the specific value of the spectral compensation coefficient according to the actual situation. Indicates the wavelength sampling interval; This represents the weighting function for salt characteristics calibrated in the experiment. satisfy , The specific function form and values ​​are determined through experiments, giving higher weight to the characteristic fluctuations of salts. For example, it can be implemented using a Gaussian weighting function based on the characteristic absorption spectrum of salt crystallization in a certain spectral band.

[0029] S103. Perform multi-scale denoising on the speckle image, calculate surface roughness parameters, and extract root mean square roughness and skewness coefficient; Specifically, the root mean square roughness and skewness coefficient are extracted, including: The formula for calculating root mean square roughness is: (3) The formula for calculating the skewness coefficient is as follows: (4) In the formula, This represents the root mean square roughness, which is the square mean of the surface height deviation. Indicates the number of sampling points. Indicates the first Surface height deviation at each sampling point Indicates the skewness coefficient; The laser speckle image is converted into quantifiable surface topography parameters, which are then analyzed using root mean square roughness. The overall roughness of the insulator surface due to contamination is reflected, i.e., the degree of dispersion of surface height deviation is quantified. Simultaneously, the skewness coefficient is used to... The asymmetry in the height distribution of contaminants on the surface of the insulator is detected by measuring the skewed height distribution and identifying localized aggregation characteristics of salt particles. This is then combined with the root mean square roughness. and skewness coefficient This allows for in-depth analysis of the contamination structure, facilitating the subsequent differentiation between salt crystals and loose salt particles, and improving the accuracy of identifying contamination types on the insulator surface.

[0030] It should be noted that salt crystallization results in the formation of a continuous hard layer, leading to... higher and Approaching zero, i.e., approaching a symmetrical distribution; for loose salt particles, this results in non-uniform protrusions, making... The absolute value is relatively large.

[0031] S104. Based on the threshold for determining the type of contamination on the insulator surface, the type of contamination on the insulator surface is determined according to the coupling relationship between the salt crystallization characteristic coefficient, the root mean square roughness, and the skewness coefficient. Optionally, in some embodiments of the first aspect, the threshold for determining the type of contamination on the insulator surface includes: Dynamic compensation is applied to the threshold for determining the type of contamination on the insulator surface based on ambient temperature and relative humidity. (5) In the formula, This indicates the judgment threshold after compensation. This represents the calibrated value of the judgment threshold. This represents the temperature compensation coefficient, where T represents the ambient temperature and RH represents the relative humidity.

[0032] Temperature compensation is used to regulate the softening effect of salt crystallization and loose salt particles during phase transition, while humidity compensation is used to regulate the dynamic equilibrium of deliquescence and crystallization in salt crystallization and loose salt particles. The logarithmic function ln is used to match the nonlinear trend of moisture absorption. Through the coupled compensation of humidity and temperature, the threshold values ​​for the first, second, and third characteristic coefficients, the first and second roughness threshold values, and the first and second skewness coefficient threshold values ​​can be adjusted in real time according to the environment, improving environmental adaptability.

[0033] It should be noted that for the temperature compensation item ( In terms of temperature, when the temperature is above 25℃, both salt crystals and loose salt particles will soften to a certain extent, and the spectral characteristics will weaken, thus requiring a reduction in the corresponding judgment threshold. When the temperature is below 25℃, both salt crystals and loose salt particles will harden to a certain extent, and the spectral characteristics will be more pronounced, thus requiring an increase in the corresponding judgment threshold.

[0034] For humidity compensation items ( In terms of relative humidity, when the relative humidity is greater than 50%, both salt crystals and loose salt particles will deliquesce to a certain extent, and their surface roughness will decrease to a certain extent, so the corresponding judgment threshold needs to be increased; when the relative humidity is less than 50%, both salt crystals and loose salt particles will dry to a certain extent, and their surface roughness will increase to a certain extent, so the corresponding judgment threshold needs to be increased.

[0035] It is understandable that the calibrated value of the judgment threshold is... The calibrated values ​​of the judgment threshold in this application can be obtained experimentally under standard conditions (25℃, RH50%), including: The threshold for determining the first feature coefficient is 0.75, the threshold for determining the second feature coefficient is 0.6, and the threshold for determining the third feature coefficient is 0.4. The first roughness threshold is 1.2. The second roughness threshold is 0.8. ; The threshold for determining the first skewness coefficient is 0.3, and the threshold for determining the second skewness coefficient is 0.5.

[0036] Specifically, determining the type of contamination on the insulator surface includes: When the salt crystallization characteristic coefficient is greater than or equal to the first characteristic coefficient judgment threshold, the root mean square roughness is greater than or equal to the first roughness judgment threshold, and the absolute value of the skewness coefficient is less than the first skewness coefficient judgment threshold, it is judged as salt crystallization fouling. When the salt crystallization characteristic coefficient is greater than or equal to the second characteristic coefficient judgment threshold, the first roughness judgment threshold is greater than the root mean square roughness and the second roughness judgment threshold, and the absolute value of the skewness coefficient is greater than the second skewness coefficient judgment threshold, it is judged as loose salt particle contamination. When the salt crystallization characteristic coefficient is less than the third characteristic coefficient judgment threshold and the root mean square roughness is less than the second roughness judgment threshold, it is judged as floating dust and dirt. Among them, the first characteristic coefficient determination threshold is greater than the second characteristic coefficient determination threshold and the third characteristic coefficient determination threshold.

[0037] The judgment rule is based on the salt crystallization characteristic coefficient. Root mean square roughness and skewness coefficient The three-parameter coupling enables accurate separation of salt crystals, loose salt particles, and dust. Meanwhile, the gradient design of the first to third feature coefficient judgment thresholds can form a decision tree logic, avoiding the misjudgments that may occur with the traditional binary search method.

[0038] S105. Based on the type of dirt, match the cleaning strategy decision within the preset cleaning method database and output cleaning suggestions.

[0039] Specifically, cleaning recommendations include: For salt crystallization contamination, it is recommended to use insulator cleaner for rinsing with water, which can accurately and thoroughly clean the contaminants and save water. For loose salt particles, it is recommended to use plasma water for rinsing. Because the loose salt particles are relatively loose, they can be effectively removed by the rinsing water, while also avoiding the pollution of the environment caused by chemical agents. For floating dust and dirt, it is recommended to use airflow with a purging pressure greater than 0.7 MPa for flushing. This can not only effectively ensure the flushing effect, but also effectively save water resources and energy consumption.

[0040] Corresponding to the aforementioned application function implementation device embodiments, this application also provides a power supply equipment insulator cleanliness detection system and corresponding embodiments.

[0041] A second aspect of this application provides a system for detecting the cleanliness of insulators in power supply equipment, comprising: The system comprises a hyperspectral imaging module, a laser speckle detection module, a data fusion processing unit, and a cleaning strategy decision-making module. The hyperspectral imaging module is used to acquire hyperspectral images of the insulator surface and extract the compositional characteristics of the pollutants. The laser speckle detection module is used to emit pulsed lasers onto the surface of insulators, acquire speckle images, and calculate the surface roughness parameters of the insulators. The data fusion processing unit is connected to the hyperspectral imaging module and the laser speckle detection module respectively, and determines the type of contamination on the insulator surface based on the coupling relationship between the salt crystallization characteristic coefficient, root mean square roughness and skewness coefficient. The cleaning strategy decision module is equipped with a cleaning method database, which matches cleaning strategy decisions within the preset cleaning method database based on the type of dirt.

[0042] The power supply equipment insulator cleanliness detection system is integrated into the drone platform.

[0043] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for detecting the cleanliness of insulators in power supply equipment, characterized in that, include: The hyperspectral imaging module and the laser speckle detection module are controlled to perform a coordinated scan of the insulator to acquire hyperspectral and speckle images of the insulator surface. Perform radiometric and atmospheric corrections on hyperspectral images, and calculate salt crystallization characteristic coefficients; Multi-scale denoising was performed on the speckle image, surface roughness parameters were calculated, and root mean square roughness and skewness coefficient were extracted. Based on the threshold for determining the type of contamination on the insulator surface, the type of contamination on the insulator surface is determined according to the coupling relationship between the salt crystallization characteristic coefficient, the root mean square roughness, and the skewness coefficient. Based on the type of dirt, the system matches cleaning strategies within a pre-defined database of cleaning methods and outputs cleaning recommendations.

2. The method for detecting the cleanliness of insulators in power supply equipment according to claim 1, characterized in that, The calculation of the salt crystallization characteristic coefficient includes: The characteristic absorption peaks of salt substances in hyperspectral data are converted into quantifiable indicators, and the formula for calculating the salt crystallization characteristic coefficient is as follows: （2） In the formula, This represents the corrected spectral reflectance function. Indicates spectral wavelength, This indicates the threshold value set for the spectral wavelength. This indicates the threshold value set for the spectral wavelength. Represents the spectral compensation coefficient. Indicates the wavelength sampling interval. This represents the weighting function for salt characteristics calibrated in the experiment. satisfy .

3. The method for detecting the cleanliness of insulators in power supply equipment according to claim 1, characterized in that, The extraction of root mean square roughness and skewness coefficient includes: The formula for calculating root mean square roughness is: （3） The formula for calculating the skewness coefficient is as follows: （4） In the formula, This represents the root mean square roughness, which is the square mean of the surface height deviation. Indicates the number of sampling points. Indicates the first Surface height deviation at each sampling point Indicates the skewness coefficient; By accurately calculating the root mean square roughness and skewness coefficient, the laser speckle image is transformed into quantifiable surface morphology parameters, enabling in-depth analysis of the fouling structure and facilitating subsequent differentiation between salt crystals and loose salt particles.

4. The method for detecting the cleanliness of insulators in power supply equipment according to claim 1, characterized in that, The threshold for determining the type of contamination on the insulator surface includes: Dynamic compensation is applied to the threshold for determining the type of contamination on the insulator surface based on ambient temperature and relative humidity. （5） In the formula, This indicates the judgment threshold after compensation. This represents the calibrated value of the judgment threshold. This represents the temperature compensation coefficient, where T represents the ambient temperature and RH represents the relative humidity.

5. The method for detecting the cleanliness of insulators in power supply equipment according to claim 4, characterized in that, The calibration values ​​of the determination threshold include: The threshold for determining the first feature coefficient is 0.75, the threshold for determining the second feature coefficient is 0.6, and the threshold for determining the third feature coefficient is 0.

4. The first roughness threshold is 1.

2. The second roughness threshold is 0.

8. ; The threshold for determining the first skewness coefficient is 0.3, and the threshold for determining the second skewness coefficient is 0.

5.

6. The method for detecting the cleanliness of insulators in power supply equipment according to claim 1, characterized in that, The determination of the type of contamination on the insulator surface includes: If the salt crystallization characteristic coefficient is greater than or equal to the first characteristic coefficient judgment threshold, the root mean square roughness is greater than or equal to the first roughness judgment threshold, and the absolute value of the skewness coefficient is less than the first skewness coefficient judgment threshold, it is judged as salt crystallization fouling. Salt crystallization characteristic coefficient ≥ second characteristic coefficient judgment threshold, first roughness judgment threshold > root mean square roughness ≥ second roughness judgment threshold, absolute value of skewness coefficient > second skewness coefficient judgment threshold, judged as loose salt particle contamination. If the salt crystallization characteristic coefficient is less than the third characteristic coefficient judgment threshold and the root mean square roughness is less than the second roughness judgment threshold, it is judged as floating dust and dirt. Among them, the first characteristic coefficient determination threshold is greater than the second characteristic coefficient determination threshold and the third characteristic coefficient determination threshold.

7. The method for detecting the cleanliness of insulators in power supply equipment according to claim 1, characterized in that, The control module for hyperspectral imaging and the laser speckle detection module perform coordinated scanning of the insulator, including: The hyperspectral imaging module and the laser speckle detection module perform coordinated scanning through a spatiotemporal synchronization device, and the synchronization error satisfies: （1） In the formula, For scanning spacing, For scanning speed, For optical path difference, The speed of light; By controlling the time synchronization error of spatial alignment to within 1ms, the positional deviation between hyperspectral and laser scanning is reduced, thereby improving the accuracy of salt crystal identification.

8. A power supply equipment insulator cleanliness detection system, applied to the power supply equipment insulator cleanliness detection method according to any one of claims 1-7, characterized in that, include: The system comprises a hyperspectral imaging module, a laser speckle detection module, a data fusion processing unit, and a cleaning strategy decision-making module. The hyperspectral imaging module is used to acquire hyperspectral images of the insulator surface and extract the compositional characteristics of the pollutants. The laser speckle detection module is used to emit pulsed laser light onto the surface of the insulator, acquire speckle images, and calculate the surface roughness parameters of the insulator. The data fusion processing unit is connected to the hyperspectral imaging module and the laser speckle detection module respectively, and determines the type of contamination on the insulator surface based on the coupling relationship between the salt crystallization characteristic coefficient, the root mean square roughness and the skewness coefficient. The cleaning strategy decision module is equipped with a cleaning method database, and makes cleaning strategy decisions based on the type of dirt by matching the cleaning method database within the preset cleaning method database.

9. The power supply equipment insulator cleanliness detection system according to claim 8, characterized in that: The power supply equipment insulator cleanliness detection system is integrated into the UAV platform.

Images