Method and system for evaluating thermal deicing effect of post insulator of transformer substation

Abstract

The invention provides a method and system for evaluating the thermal deicing effect of a post insulator of a transformer substation, and the method comprises the following steps: obtaining the temperature distribution of a deicing region based on a fully-focused infrared image of the deicing region of the post insulator of the transformer substation; acquiring working parameters of the deicing robot and environmental parameters of the post insulator of the transformer substation; based on the temperature distribution, the working parameters and the environment parameters, a hot air deicing jet flow heat exchange model is established; the jet flow heat exchange model is solved, and deicing time is obtained; based on the deicing time and the temperature distribution of the deicing area, constructing a thermal deicing effect evaluation model; and based on the thermal deicing evaluation model, completing thermal deicing effect evaluation of the post insulator of the transformer substation. According to the method, the deicing efficiency and safety can be remarkably improved by evaluating the thermal deicing effect of the post insulator of the transformer substation, and a powerful guarantee is provided for stable operation of a power system.

Description

Technical Field

[0001] The invention belongs to the technical field of thermal deicing of transformer substations, and in particular relates to a method and system for evaluating the thermal deicing effect of a transformer substation support insulator. Background Art

[0002] In recent years, the icing of power equipment caused by rain, snow and freezing weather has brought disasters to the power grid. The pillar insulators in many substations have been completely frozen and bridged. There have also been many cases of coupling capacitors, CTs, busbar pillar porcelain bottles, knife switch pillar porcelain bottles and other equipment in substations flashing or even bursting due to icing. In addition, the icing of the substation structure is likely to cause the deformation and collapse of the tower, causing the substation to stop operating and causing large-scale power outages in the surrounding areas. Therefore, after the substation is covered with ice, safe and effective methods should be adopted to remove the ice to prevent serious icing failures in the substation.

[0003] At present, deicing technologies for substation support insulators mainly include mechanical deicing, coating and thermal deicing. In view of the advantages of thermal deicing in deicing efficiency and effect, it is particularly important to accurately evaluate its effect. However, there are some obvious defects in the existing technology, which may affect the accuracy and reliability of the evaluation. For example, the limitations of data acquisition and processing, the complexity of model construction and solution (existing technology jet heat transfer model is over-simplified), the one-sidedness of evaluation indicators (single indicator evaluation makes the evaluation results not comprehensive and accurate), the lack of intelligence and automation (many manual interventions), and poor adaptability and scalability.

[0004] The present invention proposes a method and system for evaluating the thermal deicing effect of a transformer substation support insulator, which mainly studies the detection of the surface temperature distribution of the insulator during the deicing process, adjusts the deicing control strategy and the robot safe deicing technology strategy according to the real-time temperature data, and analyzes in real time the influence of the hot air velocity, nozzle angle, hot air temperature, distance from the hot air outlet to the insulator surface, ice thickness, ambient temperature, etc. on the deicing effect. By improving the existing technology, a more efficient and accurate thermal deicing effect evaluation can be achieved. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for evaluating the thermal deicing effect of substation post insulators, which realizes accurate evaluation and optimized control of the thermal deicing effect, and provides a strong guarantee for the safe operation of the substation.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for evaluating the thermal deicing effect of a transformer substation support insulator comprises the following steps:

[0008] Based on the fully focused infrared image of the deicing area of ​​the substation post insulator, the temperature distribution of the deicing area is obtained;

[0009] Collect the working parameters of the de-icing robot and the environmental parameters of the substation support insulators;

[0010] Based on the temperature distribution, the working parameters and the environmental parameters, a jet heat transfer model for hot air deicing is established; the jet heat transfer model is solved to obtain the deicing time;

[0011] Based on the deicing time and the temperature distribution of the deicing area, a thermal deicing effect evaluation model is constructed;

[0012] Based on the thermal deicing evaluation model, the thermal deicing effect evaluation of the substation post insulator is completed.

[0013] Preferably, the method for obtaining the temperature distribution of the deicing area is:

[0014] Mobile acquisition of infrared images of the de-icing area to obtain a multi-focus infrared image sequence;

[0015] Performing denoising processing on the multi-focus infrared image sequence to obtain a denoised infrared image sequence;

[0016] The existing multi-focus infrared image dataset is synthesized to obtain paired foreground and background focused images;

[0017] Based on the paired foreground and background clustered images, a training set is constructed, and an image fusion network is trained based on the training set to obtain a focus planning map;

[0018] Based on the aggregation planning map, guiding the denoised infrared image sequence to perform image fusion to obtain a fully focused infrared image;

[0019] Based on the DFF 3D reconstruction algorithm, the 3D surface height map of the deicing area of ​​the substation post insulator is reconstructed from the multi-focus infrared image sequence with focus position information.

[0020] The temperature distribution of the deicing area is obtained based on the fully focused infrared image and the three-dimensional surface height map.

[0021] Preferably, the method for constructing a training set based on the paired foreground and background clustered images is:

[0022] Perform semantic segmentation of foreground objects on the original images in the existing multi-focus infrared image dataset, binarize the semantic segmentation mask, and obtain a mask binary map;

[0023] Performing image segmentation based on the mask binary image and the original image to obtain a first foreground image and a first background image;

[0024] Based on the diffusion function, obtaining a second foreground image and a second background image of the original image;

[0025] Adding the first foreground image and the second background image to obtain a foreground aggregated image;

[0026] Adding the second foreground image and the first background image to obtain a background aggregated image;

[0027] The training set is obtained based on the foreground clustered image and the background clustered image.

[0028] Preferably, the image fusion network includes a convolutional layer, a residual block, a global perception fusion module and a bilinear upsampling layer; the method of training the image fusion network based on the training set to obtain a focus planning map is:

[0029] Based on two convolutional neural network branches with shared weights, multi-scale features of the foreground clustered image and the background clustered image are respectively extracted;

[0030] Based on the global perception fusion module, the multi-scale features extracted by the two convolutional neural network branches are cascaded to obtain a cascade feature map;

[0031] Performing average pooling on the cascade feature map to obtain a multi-scale pooling feature map;

[0032] Upsampling the multi-scale pooling feature map to a preset same spatial resolution to obtain a fused feature map;

[0033] The fused feature map is input into the decoder of the residual block to obtain the clustering planning map.

[0034] Preferably, the working parameters of the deicing robot include hot air velocity, nozzle angle, hot air temperature, and the distance from the hot air outlet to the insulator surface;

[0035] The environmental parameters of substation post insulators include ice thickness and ambient temperature.

[0036] Preferably, based on the temperature distribution, the working parameters and the environmental parameters, a method for establishing a jet heat exchange model for hot air deicing is:

[0037] Based on the temperature distribution, obtaining the surface temperature of the insulator;

[0038] Based on the working parameters, obtaining a convective heat transfer coefficient;

[0039] Based on the convection heat transfer coefficient and the surface temperature of the insulator, obtaining the heat flux density per unit area of ​​the insulator;

[0040] An energy balance equation is constructed based on the surface area of ​​the insulator directly acted upon by the hot wind, the deicing time, and the ice thickness;

[0041] Based on the energy balance equation, the jet heat transfer model is obtained.

[0042] The present invention also provides a thermal deicing effect evaluation system for transformer substation post insulators, which is used to implement the method, and comprises:

[0043] A temperature distribution acquisition module, used to obtain the temperature distribution of the deicing area based on the fully focused infrared image of the deicing area of ​​the substation support insulator;

[0044] Parameter collection module, used to collect the working parameters of the de-icing robot and the environmental parameters of the substation support insulators;

[0045] A jet heat exchange model building module is used to establish a jet heat exchange model for hot air deicing based on the temperature distribution, the working parameters and the environmental parameters; solve the jet heat exchange model to obtain the deicing time;

[0046] An evaluation model building module, used to build a thermal deicing effect evaluation model based on the deicing time and the temperature distribution of the deicing area;

[0047] The evaluation module is used to complete the evaluation of the thermal deicing effect of the substation post insulator based on the thermal deicing evaluation model.

[0048] Preferably, it also includes a background display module for recording the images before and after the insulators are de-iced, establishing a three-dimensional map, and dynamically displaying the de-icing conditions of the substation support insulators.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows: by establishing an accurate jet heat exchange model and a thermal deicing effect evaluation model, a scientific evaluation of the thermal deicing technology is achieved. This helps to discover potential problems in the deicing process and propose improvement measures. The evaluation results can optimize the deicing parameter settings and improve the deicing efficiency and safety. This helps to reduce energy consumption and operation and maintenance costs. The evaluation results provide important decision-making support for substation operation and maintenance personnel. They can choose appropriate deicing solutions and equipment configurations based on the evaluation results to ensure the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0051] Figure 1This is a flow chart of a method for evaluating the thermal deicing effect of a post insulator of a substation according to an embodiment of the present invention;

[0052] Figure 2 A curve diagram showing the relationship between the hot air speed and the deicing time at different hot air temperatures according to an embodiment of the present invention;

[0053] Figure 3 A curve diagram showing the relationship between hot air temperature and deicing time at different hot air speeds according to an embodiment of the present invention;

[0054] Figure 4 A flow chart of obtaining a jet heat transfer model for an embodiment of the present invention;

[0055] Figure 5 This is a flowchart of image denoising according to an embodiment of the present invention;

[0056] Figure 6 A flowchart of constructing a training set based on paired foreground and background clustered images according to an embodiment of the present invention;

[0057] Figure 7 The present invention is a flowchart of obtaining a clustering planning diagram according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] 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.

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] Embodiment 1

[0061] like Figure 1 As shown, a method for evaluating the thermal deicing effect of a transformer substation support insulator comprises the following steps:

[0062] S1: Based on the fully focused infrared image of the deicing area of ​​the substation support insulator, the temperature distribution of the deicing area is obtained; a further implementation method is that the method for obtaining the temperature distribution of the deicing area is:

[0063] S11: A mechanical arm-infrared thermal imager is used to move and collect infrared images of the deicing area to obtain a multi-focus infrared image sequence; in this embodiment, according to the characteristic that the infrared image can accurately reflect the temperature information of the measured object under focusing conditions, a multi-focus image fusion technology is used to obtain a fully focused infrared image, and the infrared camera is moved in a direction perpendicular to the surface of the measured object, and a series of infrared images are collected regularly, with the distance interval not exceeding a depth of field. Among them, the working principle of the infrared thermal imager is that as long as the temperature of the object is higher than absolute zero (273.15℃), it will emit infrared radiation outward. The energy of this radiation is proportional to the temperature of the object. The higher the temperature, the stronger the emitted infrared radiation. The core component of the infrared thermal imager is the infrared detector, which can be a detector based on different technologies, such as an uncooled type (such as a microbolometer) or a cooled type (such as a quantum well detector). The uncooled detector has the characteristics of low cost, small size, low power consumption, etc. The advantages of the infrared thermal imager used in the present invention are: (1) all-weather operation, not limited by light conditions, and can work normally even in a completely dark environment; (2) non-contact measurement, without direct contact with the object to measure its surface temperature. (3) Visualization can intuitively display the temperature distribution of an object, making it easier to analyze and record.

[0064] S12: Figure 5 As shown in the figure, the multi-focus infrared image sequence is denoised to obtain a denoised infrared image sequence. Specifically, according to the characteristics of the infrared image and the actual noise situation, the spatial domain standard deviation (σ d ) and the intensity domain standard deviation (σ r ). These two parameters control the weights of spatial smoothness and intensity similarity respectively. During the filtering process, the weight of each pixel is calculated according to the spatial distance and intensity difference, and the weighted average is calculated as the output pixel value. Each frame of the denoised image is stored in a new storage location or overwrites the original image. The quality of the denoised infrared image sequence is evaluated. In this embodiment, the signal-to-noise ratio (SNR) and peak signal-to-noise ratio (PSNR) indicators are calculated to verify the denoising effect.

[0065] S13: synthesizing the existing multi-focus infrared image data set to obtain paired foreground and background focused images; constructing a training set based on the paired foreground and background aggregated images, and training an image fusion network based on the training set to obtain a focusing planning map;

[0066] S14: Based on the aggregation planning map, guide the denoised infrared image sequence to perform image fusion to obtain a fully focused infrared image;

[0067] S15: Based on the DFF (depth of focus) 3D reconstruction algorithm, a 3D surface height map of the deicing area of ​​the substation post insulator is reconstructed from the multi-focus infrared image sequence with focus position information;

[0068] S16: Based on the fully focused infrared image and the three-dimensional surface height map, the temperature distribution of the deicing area is obtained.

[0069] A further embodiment is that Figure 6 As shown, in step S13, the method for constructing a training set based on paired foreground and background clustered images is:

[0070] The original image in the existing multi-focus infrared image dataset is semantically segmented for the foreground object, and the mask of the semantic segmentation is binarized to obtain a mask binary map, in which the pixel value of the foreground is set to 1 and the pixel value of the background is set to 0.

[0071] Perform image segmentation based on the mask binary image and the original image to obtain a first foreground image and a first background image;

[0072] Based on the diffusion function, a second foreground image and a second background image of the original image are obtained;

[0073] Adding the first foreground image and the second background image to obtain a foreground aggregated image;

[0074] Adding the second foreground image to the first background image to obtain a background aggregated image;

[0075] Based on the foreground clustered image and the background clustered image, a training set is obtained.

[0076] A further embodiment is that Figure 7 As shown, in step S13, the image fusion network adopts a U-shaped network structure, and the image fusion network includes a convolutional layer, a residual block, a global perception fusion module, and a bilinear upsampling layer; the method of training the image fusion network based on the training set to obtain the focus planning map is:

[0077] Based on two convolutional neural network branches (encoders) with shared weights, multi-scale features of foreground clustered images and background clustered images are extracted respectively. Multi-scale features include rich information from low-level to high-level, from local to global.

[0078] Based on the global perception fusion module, the multi-scale features extracted by the two convolutional neural network branches are cascaded to obtain a cascade feature map; global semantic prior constraints and multi-scale information can be retained at the same time. The cascade feature map forms a hierarchical structure by connecting multiple models or algorithms in series. After each model or algorithm processes the feature map, its output is used as the input of the next model, thereby gradually improving the representation ability of the features and the accuracy of the task. This method allows more network parameters to be shared between tasks, and each task can assist each other to improve the overall performance.

[0079] The cascade feature maps are average pooled to obtain a multi-scale pooled feature map; the multi-scale pooled feature map includes useful feature information extracted from different image scales.

[0080] The multi-scale pooled feature map is upsampled to the preset same spatial resolution to obtain a fused feature map; the fused feature map contains both global semantic information and retains multi-scale details.

[0081] The fused feature map is input into the decoder of the residual block to obtain a clustering planning map, which represents the clustering areas and distribution of the foreground and background in the image. Specifically, based on the clustering planning map, the formula guiding the image fusion of the denoised infrared image sequence is as follows:

[0082] F=I A ×M+I B ×(1-M),

[0083] F stands for fully focused infrared image, I A and I B is the original image, and M represents the focus planning map.

[0084] In step S15, the specific implementation process of the DFF (depth of focus) three-dimensional reconstruction algorithm is as follows: for each pixel point, the Laplace operator is used to calculate its focus in the images at different focus positions; for each pixel point, the image with the highest focus is selected as the best focus image of the point. This means that the focus depth of the image at this point is closest to the actual depth. Based on the infrared camera calibration and the insulator geometric model, the best focus position of each pixel point is mapped to the corresponding depth value to determine the correspondence between the focus position and the depth. Using the depth map, the two-dimensional coordinates and depth values ​​of each pixel point are converted into points in three-dimensional space to construct point cloud data. Using the point cloud data, the three-dimensional surface is fitted by the triangulation method, and the reconstructed three-dimensional surface is optimized and smoothed to eliminate noise and unevenness to obtain a three-dimensional surface height map.

[0085] S2: Collecting the working parameters of the deicing robot and the environmental parameters of the substation support insulator; in a further implementation method, the working parameters of the deicing robot include hot air velocity, nozzle angle, hot air temperature and the distance from the hot air outlet to the insulator surface;

[0086] The environmental parameters of substation post insulators include ice thickness and ambient temperature.

[0087] S3: Figure 4As shown, based on the temperature distribution, working parameters and environmental parameters, a jet heat transfer model for hot air deicing is established; the jet heat transfer model is solved to obtain the deicing time; a further implementation method is that based on the temperature distribution, working parameters and environmental parameters, a method for establishing the jet heat transfer model for hot air deicing is:

[0088] Based on the temperature distribution, the surface temperature of the insulator is obtained;

[0089] Based on the operating parameters, the convective heat transfer coefficient is obtained;

[0090] Based on the convection heat transfer coefficient and the surface temperature of the insulator, the heat flux density per unit area of ​​the insulator is obtained; the specific calculation formula is as follows:

[0091] q=h(T h -T s ),

[0092] Among them, q represents the heat flux density per unit area of ​​the insulator, h represents the convection heat transfer coefficient (affected by wind speed, distance, angle, etc.), T s Represents the surface temperature of the insulator (close to the temperature of the ice layer).

[0093] Based on the insulator surface area directly affected by hot air, deicing time and ice thickness, the energy balance equation is constructed:

[0094]

[0095] Where A is the surface area of ​​the insulator directly affected by the hot air, t is the time, and ρ i is the density of ice, L f is the latent heat of melting of ice, L0 is the initial ice thickness, and L(t) is the remaining ice thickness that varies with time.

[0096] Based on the energy balance equation, the jet heat transfer model is obtained. Specifically, the heat flux density q is substituted into the energy balance equation to obtain the expression of the jet heat transfer model:

[0097]

[0098] S4: Based on the deicing time and the temperature distribution in the deicing area, a thermal deicing effect evaluation model is constructed; Figure 2 , Figure 3 shown.

[0099] In this embodiment, the formula of the thermal deicing effect evaluation model is as follows:

[0100]

[0101] In the formula, Q Nrepresents the heat required to de-ice each section of the post insulator, J, m0 represents the mass of melted ice in each section, kg; T ice Represents the initial temperature of the ice layer, ℃; m fall represents the mass of ice that has not melted into water and has fallen off in each segment; c ice represents the specific heat of ice, J (kg·℃); γ represents the latent heat of phase change between ice and water, J / kg.

[0102] Specifically, in this embodiment, based on the jet heat transfer model and the thermal deicing effect evaluation model, the influence of hot air velocity, nozzle angle, hot air temperature, distance from the hot air outlet to the insulator surface, ice thickness, ambient temperature, etc. on the deicing effect is analyzed. In this embodiment, a sample library of insulators with different degrees of ice coating is also constructed, and the deicing effect of different influencing factors on insulator samples with different degrees of ice coating is obtained through the thermal deicing effect evaluation model. Among them, the degree of ice coating is obtained by measuring the thickness and density of ice coating.

[0103] Hot air speed: Increasing the wind speed can improve the convective heat transfer coefficient, thereby speeding up deicing, but it may also cause the hot air to stay on the insulator surface for a shorter time, affecting the heat transfer efficiency.

[0104] Nozzle Angle: Optimizing the angle can ensure that the hot air jet better covers the insulator surface and improves de-icing efficiency.

[0105] Hot air temperature: Increasing the hot air temperature can increase the amount of heat transferred to the ice layer, but care must be taken to avoid overheating the insulators.

[0106] Distance from the hot air outlet to the insulator surface: Reducing the distance can enhance the concentration of the jet and the heat transfer efficiency, but too close may cause local overheating.

[0107] Ice thickness: Thicker ice requires more heat to melt, so de-icing will take longer.

[0108] Ambient temperature: Although the ambient temperature has little effect on the direct deicing effect, it will affect the temperature difference driving force of the heat exchange process, thereby affecting the heat exchange efficiency.

[0109] S5: Based on the thermal deicing evaluation model, complete the thermal deicing effect evaluation of the substation post insulators.

[0110] In this embodiment, the response surface analysis method of the Box-Behnken Designs central composite design model is used to fit the factors affecting the thermal deicing effect and the deicing time.

[0111] The method to obtain the significant relationship between the influencing factors and the deicing time is:

[0112] Based on the response surface analysis method, the response values ​​of the influencing factors and the regression analysis equation are obtained;

[0113] Based on the regression analysis equation, one-way ANOVA was performed on the influencing factors and the response values ​​of the influencing factors to obtain the F value and probability P value of the F statistic;

[0114] Based on the F value and the probability P value, the significance relationship between the deicing time and the influencing factors is determined.

[0115] The embodiment of the present invention constructs a response surface model through regression analysis to conduct experiments, effectively estimate the parameters of the model and evaluate the interaction between various factors, and record the input variables (such as hot air speed, temperature, etc.) and output responses (such as deicing time) of each experiment. Ensure the stability and repeatability of the experimental conditions. By drawing response surface diagrams and contour maps, the influence of various factors on the deicing effect and the interaction between factors can be intuitively observed. At the same time, a genetic algorithm is used to search for the optimal combination of influencing factors under the constraints of the model. Through the optimal combination of influencing factors, the working parameters of the deicing robot are optimized to achieve efficient deicing.

[0116] Embodiment 2

[0117] The present invention also provides a thermal deicing effect evaluation system for transformer substation post insulators, which is used to implement the method, including:

[0118] A temperature distribution acquisition module, used to obtain the temperature distribution of the deicing area based on the fully focused infrared image of the deicing area of ​​the substation support insulator;

[0119] Parameter collection module, used to collect the working parameters of the de-icing robot and the environmental parameters of the substation support insulators;

[0120] The jet heat transfer model building module is used to establish a jet heat transfer model for hot air deicing based on temperature distribution, working parameters and environmental parameters; solve the jet heat transfer model to obtain the deicing time;

[0121] An evaluation model building module is used to build a thermal deicing effect evaluation model based on the deicing time and the temperature distribution of the deicing area;

[0122] The evaluation module is used to complete the thermal deicing effect evaluation of the substation post insulator based on the thermal deicing evaluation model.

[0123] A further implementation method is to further include a background display module for recording images before and after the insulators are de-iced, creating a three-dimensional map, and dynamically displaying the de-icing conditions of the substation support insulators.

[0124] The evaluation system provided by the present invention can be seamlessly integrated with the deicing robot system. The deicing robot can evaluate the working effect while working, thereby realizing automatic deicing process control and optimization.

[0125] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for evaluating the thermal deicing effect of a transformer substation support insulator, characterized in that: The following steps are involved: Based on the fully focused infrared image of the deicing area of ​​the substation post insulator, the temperature distribution of the deicing area is obtained; Collect the working parameters of the de-icing robot and the environmental parameters of the substation support insulators; Based on the temperature distribution, the working parameters and the environmental parameters, a jet heat exchange model for hot air deicing is established; and the jet heat exchange model is solved to obtain the deicing time. Based on the deicing time and the temperature distribution of the deicing area, a thermal deicing effect evaluation model is constructed; Based on the thermal deicing evaluation model, the thermal deicing effect evaluation of the substation post insulator is completed.

2. The method for evaluating the thermal deicing effect of a substation post insulator according to claim 1, characterized in that: The method to obtain the temperature distribution in the de-icing area is: Mobile acquisition of infrared images of the de-icing area to obtain a multi-focus infrared image sequence; Performing denoising processing on the multi-focus infrared image sequence to obtain a denoised infrared image sequence; The existing multi-focus infrared image dataset is synthesized to obtain paired foreground and background focused images; Based on the paired foreground and background clustered images, a training set is constructed, and an image fusion network is trained based on the training set to obtain a focus planning map; Based on the aggregation planning map, guiding the denoised infrared image sequence to perform image fusion to obtain a fully focused infrared image; Based on the DFF 3D reconstruction algorithm, the 3D surface height map of the deicing area of ​​the substation post insulator is reconstructed from the multi-focus infrared image sequence with focus position information. The temperature distribution of the deicing area is obtained based on the fully focused infrared image and the three-dimensional surface height map.

3. The method for evaluating the thermal deicing effect of a substation post insulator according to claim 2, characterized in that: Based on the paired foreground and background clustered images, the method for constructing a training set is: Perform semantic segmentation of foreground objects on the original images in the existing multi-focus infrared image dataset, binarize the semantic segmentation mask, and obtain a mask binary map; Performing image segmentation based on the mask binary image and the original image to obtain a first foreground image and a first background image; Based on the diffusion function, obtaining a second foreground image and a second background image of the original image; Adding the first foreground image and the second background image to obtain a foreground aggregated image; Adding the second foreground image and the first background image to obtain a background aggregated image; The training set is obtained based on the foreground clustered image and the background clustered image.

4. The method for evaluating the thermal deicing effect of a transformer substation post insulator according to claim 3, characterized in that: The image fusion network includes a convolutional layer, a residual block, a global perception fusion module and a bilinear upsampling layer; the method of training the image fusion network based on the training set to obtain a focus planning map is: Based on two convolutional neural network branches with shared weights, multi-scale features of the foreground clustered image and the background clustered image are respectively extracted; Based on the global perception fusion module, the multi-scale features extracted by the two convolutional neural network branches are cascaded to obtain a cascade feature map; Performing average pooling on the cascade feature map to obtain a multi-scale pooling feature map; Upsampling the multi-scale pooling feature map to a preset same spatial resolution to obtain a fused feature map; The fused feature map is input into the decoder of the residual block to obtain the clustering planning map.

5. The method for evaluating the thermal deicing effect of a transformer substation post insulator according to claim 1, characterized in that: The working parameters of the deicing robot include hot air velocity, nozzle angle, hot air temperature, and the distance from the hot air outlet to the insulator surface; The environmental parameters of substation post insulators include ice thickness and ambient temperature.

6. The method for evaluating the thermal deicing effect of a substation post insulator according to claim 5, characterized in that: Based on the temperature distribution, the working parameters and the environmental parameters, the method for establishing the jet heat transfer model of hot air deicing is: Based on the temperature distribution, obtaining the surface temperature of the insulator; Based on the working parameters, obtaining a convective heat transfer coefficient; Based on the convection heat transfer coefficient and the surface temperature of the insulator, obtaining the heat flux density per unit area of ​​the insulator; An energy balance equation is constructed based on the surface area of ​​the insulator directly acted upon by the hot wind, the deicing time, and the ice thickness; Based on the energy balance equation, the jet heat transfer model is obtained.

7. A thermal deicing effect evaluation system for substation post insulators, used to implement the method according to any one of claims 1 to 6, characterized in that: include: A temperature distribution acquisition module, used to obtain the temperature distribution of the deicing area based on the fully focused infrared image of the deicing area of ​​the substation support insulator; Parameter collection module, used to collect the working parameters of the de-icing robot and the environmental parameters of the substation support insulators; A jet heat exchange model building module is used to establish a jet heat exchange model for hot air deicing based on the temperature distribution, the working parameters and the environmental parameters; solve the jet heat exchange model to obtain the deicing time; An evaluation model building module, used to build a thermal deicing effect evaluation model based on the deicing time and the temperature distribution of the deicing area; The evaluation module is used to complete the evaluation of the thermal deicing effect of the substation post insulator based on the thermal deicing evaluation model.

8. The thermal deicing effect evaluation system for substation post insulators according to claim 7, characterized in that: It also includes a background display module, which is used to record the images before and after the insulators are de-iced, create a three-dimensional map, and dynamically display the de-icing status of the substation support insulators.

Images