Intelligent monitoring method and system for power distribution cabinet

Abstract

The invention relates to an intelligent monitoring method for a power distribution cabinet. A model is used for simulating the moving direction angle and the moving distance of a heat center in a sampling area between two adjacent detection areas; and performing vectorization processing on the reference grid in the sampling area between the two adjacent detection areas, and calculating the theoretical moving direction angle and distance of the heat center. And judging whether a suspected abnormal component exists in the two groups of components corresponding to the adjacent detection area or not by comparing a simulation result with a calculation result, and sending a prompt to operation and maintenance personnel. The method is suitable for performing automatic batch evaluation on multiple groups of power distribution cabinets in a power distribution room, does not need to manually perform fault inspection on components in each power distribution cabinet one by one, can quickly screen out components with potential faults, reminds maintenance personnel to detect and confirm the components with potential faults and the power distribution cabinets to which the components with potential faults belong, and improves the efficiency of the power distribution room. And all components do not need to be detected manually, so that the fault detection efficiency is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of power distribution cabinet monitoring technology, specifically to a method and system for intelligent monitoring of power distribution cabinets. Background Technology

[0002] Distribution cabinets are divided into power distribution cabinets, lighting distribution cabinets, and metering cabinets. They are the final stage equipment in the power distribution system. Distribution cabinets are a general term for motor control centers. Distribution cabinets are used in situations where the load is relatively dispersed and there are fewer circuits. Motor control centers are used in situations where the load is concentrated and there are more circuits. They distribute the electrical energy of a certain circuit of the previous level power distribution equipment to the nearest load. This level of equipment should provide protection, monitoring and control for the load.

[0003] Distribution cabinets typically house multiple components, such as air switches and circuit breakers. These components are prone to malfunctions during operation. When components fail or malfunction, they often exhibit significant temperature fluctuations. Therefore, detecting temperature is a commonly used auxiliary detection method to determine if a component is malfunctioning.

[0004] In existing technologies, the inspection of components inside distribution cabinets typically involves manual handheld temperature measuring devices to check the temperature of each component one by one. The measured temperature is then compared with a set high-temperature value; if the measured temperature is higher than the set value, the component is considered to be abnormal. While this method can detect faults in components within the distribution cabinet, it is inefficient due to the need for manual inspection of all components individually. It cannot meet the requirements for rapid screening of anomalies in large numbers of distribution cabinets and needs improvement. Summary of the Invention

[0005] Based on the above description, the present invention provides a method and system for intelligent monitoring of power distribution cabinets to improve the detection efficiency of abnormal states of power distribution cabinets.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A method for intelligent monitoring of a power distribution cabinet is characterized by: placing the power distribution cabinet in a constant temperature environment; installing an infrared thermal imaging module on the inner side of the cabinet door to capture and generate infrared thermal images of the cabinet; uniformly dividing the image into a grid array; defining rectangular detection areas for each component based on the coverage of the components, wherein each detection area contains multiple grid cells; defining grid cells outside each detection area as reference grids; and defining quadrilateral areas formed by connecting the opposite sides of adjacent detection areas as sampling areas; continuously capturing and generating infrared thermal images of the cabinet at certain intervals during the operation of the power distribution cabinet; selecting two adjacent infrared thermal images; obtaining the front and rear temperature values ​​of each detection area and each reference grid through image brightness; and then calculating the temperature difference values ​​of each detection area and each reference grid. The temperature difference values ​​of each detection area are compared with those of its adjacent detection areas in turn. Then, based on the coverage area of ​​the two adjacent detection areas, the distance between the center points of the areas, the type of components, the typical operating temperature, and the range of temperature difference values, the angle α and the distance L0 of the heat center movement direction in the sampling area between the two detection areas are simulated. Simultaneously, the average temperature difference value of each reference grid in the sampling area is calculated, and the reference grid with a temperature difference value greater than the average value is defined as a type A reference grid, and the reference grid with a temperature difference value less than the average value is defined as a type B reference grid. Each Class A reference grid is converted into a vector by assigning a value to it based on the difference between the temperature difference value and the average value of the Class A reference grid, and defining this value as the magnitude of the vector; the angle between the line connecting the center point of the grid and the center point of the detection area with a large temperature difference value and the horizontal line is defined as the angle of the vector, thus obtaining the vectors of all Class A reference grids. Each type B reference grid is converted into a vector by assigning a value to it based on the difference between the temperature difference value and the average value of the type B reference grid, and defining this value as the magnitude of the vector; the angle between the line connecting the center point of the grid and the center point of the detection area with the smaller temperature difference value and the horizontal line is defined as the angle of the vector, thus obtaining the vectors of all type B reference grids. Then, the sum of all vectors is calculated, followed by the direction angle β and magnitude V of the sum vector. Finally, the corresponding theoretical distance value L1 (i.e., the theoretical distance the heat center has moved) is calculated: L1 = V k, where k is a preset conversion factor; Pre-set angle ratio range and distance ratio range. When the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area. The two sets of components are marked as "potential faults" and a prompt is sent to the operation and maintenance personnel to remind them to go to the site for troubleshooting.

[0007] As a preferred approach: Modeling is pre-established based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the two detection areas, the types of the two sets of corresponding components, and the typical operating temperature. Multiple experiments are conducted to ensure that the temperatures of the two sets of corresponding components have different temperature differences from the typical operating temperature. Multiple temperature difference ranges are set, and the brightness change characteristics of the sampling area in the infrared thermal images during the experiment are extracted and analyzed within these different temperature difference ranges to obtain the heat migration characteristics of the sampling area. These heat migration characteristics include the angle α of the heat center's movement direction and the movement distance L0. These migration characteristics are the output of the model. The model is trained and its parameters are fine-tuned using multiple sets of experimental data to complete the model training. During the operation of the distribution cabinet, two adjacent detection areas are selected. The coverage area of ​​these two detection areas, the distance between their center points, the types of the two sets of corresponding components, the typical operating temperature, and the temperature difference range are modeled and input into the model. The model automatically outputs the corresponding heat migration characteristics, namely the angle α of the heat center's movement direction and the movement distance L0.

[0008] As a preferred solution: when assigning values ​​to type A grids, a calculation coefficient m1 is matched to it, and the assigned value is corrected using the calculation coefficient m1; when assigning values ​​to type B grids, a calculation coefficient m2 is matched to it, and the assigned value is corrected using the calculation coefficient m2; by default, the values ​​of m1 and m2 are both 1. When maintenance personnel go to the site to investigate, they input the investigation results and adjust the values ​​of m1 and m2 according to the investigation results.

[0009] A smart monitoring system for power distribution cabinets, comprising: Infrared thermal imaging module, used to capture and generate infrared thermal images of the cabinet; The image acquisition module is used to acquire infrared thermal images output by the infrared thermal imaging module and transmit the image data to the microprocessor module; The microprocessor module is used to process infrared thermal images and evaluate the two groups of components based on the temperature difference between two adjacent groups of components in the infrared thermal image and the temperature difference between each sampling grid in the sampling area between the two groups of components, in order to determine whether there are any "potential faults" in the two groups of components. Storage module, which is used to store data; An alarm module is used to issue an alarm signal when the microprocessor module determines that there is a "potentially faulty" component in the two groups of components; The power supply module is used to supply power to the various modules.

[0010] As a preferred embodiment: the microprocessor module includes an image processing unit, a model unit, and a computing unit; The image processing unit is used to perform gridding processing on the infrared thermal image to determine the temperature value of each grid cell, as well as the temperature value of the corresponding detection area of ​​each component; The model unit is used to model the heat migration characteristics based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the areas, the types of the two sets of corresponding components and the typical operating temperature, as well as the temperature difference range, and output the corresponding heat migration characteristics, namely the heat center movement direction angle α and the movement distance L0. The calculation unit is used to vectorize the reference grid in the sampling area between two adjacent detection areas according to the temperature difference value, and calculate the theoretical moving direction angle β and moving distance L1 of the heat center. When the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area, and the two sets of components are marked as "potential faults".

[0011] As a preferred embodiment, the microprocessor module further includes a parameter adjustment unit, which is used to adjust preset parameters within the computing unit.

[0012] As a preferred embodiment, it also includes a communication module, which is used to communicate with a host computer or cloud platform and transmit data.

[0013] Compared with the prior art, the technical solution of this application has the following beneficial technical effects: This method is suitable for automated batch evaluation of multiple sets of distribution cabinets in a power distribution room. It eliminates the need for manual fault inspection of each component in each distribution cabinet, and can quickly screen out components that may have "potential faults". It reminds maintenance personnel to test and confirm the "potentially faulty" components and their respective distribution cabinets. It eliminates the need for manual testing of all components, greatly improving fault detection efficiency. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the internal structure of the power distribution cabinet in Example 1; Figure 2 This is a schematic diagram of the meshing process in Example 1; Figure 3 This is a schematic diagram of the detection area in Example 1; Figure 4 This is a schematic diagram of the sampling area in Example 1; Figure 5 This is a schematic diagram of the reference grid in Example 1; Figure 6 This is a schematic diagram of the vectorization process in Example 1; Figure 7 This is a schematic diagram of the system in Example 2.

[0015] The attached diagram lists the components represented by each number as follows: 1. Cabinet; 2. Components. Detailed Implementation

[0016] Example 1: A method for intelligent monitoring of power distribution cabinets, specifically: This method is applicable to constant temperature power distribution rooms, which are now equipped with air conditioning systems that can control the indoor temperature.

[0017] The distribution cabinet is placed in a temperature-controlled environment within the power distribution room, and an infrared thermal imaging module (not shown in the figure) is installed on the inside side of the cabinet door. (Refer to...) Figure 1 After the power distribution cabinet is installed and debugged, the position distribution of each component 2 inside the cabinet 1 will not change, and the type of each component 2 can also be determined.

[0018] When the cabinet door is closed, the components inside the cabinet (i.e., the components to be monitored) are exactly within the sensing range of the infrared thermal imaging module, which can then capture and generate an infrared thermal image of the cabinet. When the distribution cabinet is in operation, under normal circumstances, the distribution of the highlighted areas in the captured infrared thermal image should correspond to the installation location of each component.

[0019] In this embodiment, the position and coverage area of ​​each component need to be pre-calibrated. For example... Figure 2 As shown, the infrared thermal image is first uniformly divided into a grid array. Based on the location and coverage of each component, the grid cells covered by each component are merged. The rectangular area formed by merging these grid cells is defined as the detection area corresponding to that component. For example... Figure 3 As shown, regions S1, S2, S3, S4, and S5 represent the detection areas corresponding to each group of components. The component temperature will be determined based on the image brightness within the detection area.

[0020] Grid cells outside each detection area are defined as reference grids.

[0021] The quadrilateral region formed by connecting the opposite sides of adjacent detection regions is defined as the sampling region. (Refer to...) Figure 4 For example, the quadrilateral region P formed by connecting the opposite sides of detection regions S1 and S2 is the sampling region between regions S1 and S2.

[0022] The sampling area between each detection area can be determined using the above method.

[0023] The infrared thermal image is converted into a Cartesian coordinate graph. From this graph, the coordinates of the four corner points of each detection area can be determined, allowing for the calculation of the center point coordinates and area of ​​each detection area. Furthermore, the type of each component needs to be identified, and its temperature under various typical operating conditions needs to be determined experimentally.

[0024] Then, modeling is performed based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the two detection areas, the types of the two sets of corresponding components, and the typical operating temperature.

[0025] Through multiple sets of experiments, the temperatures of the two corresponding components were made to have different temperature differences from the typical operating temperature. Multiple temperature difference ranges were set, and the brightness change characteristics of the sampling area in the infrared thermal image of the experiment were extracted and analyzed within different temperature difference ranges to obtain the heat migration characteristics of the sampling area. The heat migration characteristics include the angle α of the heat center movement direction and the movement distance L0. The migration characteristics are the output of the model. The model was trained and the parameters were tuned through multiple sets of experimental data to complete the model training.

[0026] The working principle of this method is as follows: During the operation of the power distribution cabinet, infrared thermal images of the cabinet are continuously captured and generated at certain intervals. The temperature values ​​before and after each detection area and the temperature values ​​before and after each reference grid are obtained by the brightness changes of the two images. Then, the temperature difference values ​​of each detection area and the temperature difference values ​​of each reference grid are calculated.

[0027] Simultaneously, the current operating current of each component is collected to determine its current operating condition. Based on the current operating condition of each component, the typical operating temperature values ​​for each group of components under their respective conditions are retrieved.

[0028] Two adjacent detection areas are selected in sequence. The size of the coverage area of ​​the two detection areas, the distance between the center points of the areas, the types of the two sets of corresponding components and the typical operating temperature, as well as the temperature difference range are used as inputs and input into the model. The model automatically outputs the predicted heat migration characteristics, namely the angle α of the heat center movement direction and the movement distance L0.

[0029] Calculate the average temperature difference of each reference grid within the sampling area between two adjacent detection areas. For example... Figure 5 As shown, reference grids with temperature differences greater than the average value are defined as Class A reference grids, and reference grids with temperature differences less than the average value are defined as Class B reference grids.

[0030] Reference Figure 6Each type A reference grid is converted into a vector by assigning a value to it based on the difference between the temperature difference value and the average value of the type A reference grid. This assigned value is defined as the magnitude of the vector. The larger the difference between the temperature difference value and the average value, the larger the assigned value; conversely, the smaller the difference, the smaller the value.

[0031] The angle γ between the line connecting the center point of the Class A grid and the center point of the detection area with a large temperature difference (for example, the temperature difference between the front and back of area S1 is greater than that between the front and back of area S2, i.e., area S1 has a larger temperature difference and area S2 has a smaller temperature difference) and the horizontal line is defined as the angle of the vector, thus obtaining the vector of all Class A reference grids.

[0032] Similarly, each type B reference grid is converted into a vector, that is, a value is assigned to it based on the difference between the temperature difference value and the average value of the type B reference grid, and this assignment is defined as the magnitude of the vector. The larger the difference between the temperature difference value and the average value, the larger the assignment value; conversely, the smaller the difference, the smaller the assignment value.

[0033] The angle between the line connecting the center point of the grid and the center point of the detection area with a smaller temperature difference and the horizontal line is defined as the angle of the vector, thus obtaining the vectors of all B-type reference grids.

[0034] Then, the sum of all vectors is calculated, followed by the direction angle β and magnitude V of the sum vector. Finally, the corresponding theoretical distance value L1 is calculated, where L1 = V. k, where k is a preset conversion factor.

[0035] Under normal circumstances, the temperature changes of each grid within the sampling area (i.e., the temperature difference between two samplings) can reflect the heat migration characteristics of the sampling area to a certain extent, that is, can reflect the direction angle α and distance L0 of the heat center movement to a certain extent.

[0036] In this embodiment, the heat center's theoretical moving direction angle β and moving distance L1 are obtained by vectorizing each grid cell in the sampling area according to the change of its temperature value, and by setting an appropriate conversion coefficient k through a pre-simulation experiment, and converting the sum vector of all reference grids in the sampling area.

[0037] When the distribution cabinet is functioning normally, the calculated theoretical movement angle β and movement distance L1 of the heat center should be close to the model-predicted movement angle α and movement distance L0. Otherwise, the distribution cabinet can be considered to be in an abnormal state, indicating a potential fault.

[0038] In this embodiment, the angle ratio range and the distance ratio range are preset. When the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area. The two sets of components are marked as "potential faults" and a prompt is sent to the operation and maintenance personnel to remind them to go to the site for troubleshooting.

[0039] This method allows for the assessment of two sets of components within a distribution cabinet before any obvious faults appear. It involves evaluating the temperature difference between two consecutive sampling measurements of different blocks within the sampling area between two adjacent components. This enables the timely detection of potentially faulty components before a fault actually occurs, prompting maintenance personnel to inspect and confirm these components on-site, thus preventing potential problems.

[0040] This method is suitable for automated batch evaluation of multiple sets of distribution cabinets in a power distribution room. It eliminates the need for manual fault inspection of each component in each distribution cabinet, and can quickly screen out components that may have "potential faults". It reminds maintenance personnel to test and confirm the "potentially faulty" components and their respective distribution cabinets, greatly improving fault detection efficiency.

[0041] This embodiment also includes a step of correcting the assigned values. Specifically: when assigning values ​​to type A grids, a calculation coefficient m1 is matched and used to correct the assigned value; when assigning values ​​to type B grids, a calculation coefficient m2 is matched and used to correct the assigned value. By default, both m1 and m2 are 1. When maintenance personnel conduct on-site investigations, they input the investigation results and adjust the values ​​of m1 and m2 accordingly.

[0042] If the investigation results indicate that the component is indeed faulty, then the "potential fault" assessment result is correct, and there is no need to adjust parameters m1 and m2.

[0043] If the troubleshooting results show that the component is not faulty, then the "potential fault" assessment result is incorrect. In this case, maintenance personnel can adjust parameters m1 and m2 based on experience to improve the accuracy of subsequent assessments.

[0044] Example 2: Reference Figure 7 A smart monitoring system for power distribution cabinets, comprising: Infrared thermal imaging module, used to capture and generate infrared thermal images of the cabinet; The image acquisition module is used to acquire infrared thermal images output by the infrared thermal imaging module and transmit the image data to the microprocessor module; The microprocessor module is used to process infrared thermal images and evaluate the two groups of components based on the temperature difference between two adjacent groups of components in the infrared thermal image and the temperature difference between each sampling grid in the sampling area between the two groups of components, in order to determine whether there are any "potential faults" in the two groups of components. Storage module, which is used to store data; An alarm module is used to issue an alarm signal when the microprocessor module determines that there is a "potentially faulty" component in the two groups of components; The power supply module is used to supply power to the various modules.

[0045] The microprocessor module in this embodiment includes an image processing unit, a model unit, and a computing unit.

[0046] The image processing unit is used to perform gridding processing on the infrared thermal image to determine the temperature value of each grid cell and the temperature value of the corresponding detection area of ​​each component. The model unit is used to model the heat migration characteristics based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the areas, the types of two sets of corresponding components and the typical operating temperature, as well as the temperature difference range, namely the heat center movement direction angle α and the movement distance L0. The calculation unit is used to vectorize the reference grid in the sampling area between two adjacent detection areas according to the temperature difference value, and calculate the theoretical moving direction angle β and moving distance L1 of the heat center. When the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area, and the two sets of components are marked as "potential faults".

[0047] The microprocessor module in this embodiment also includes a parameter adjustment unit, which is used to adjust the preset parameters in the calculation unit.

[0048] The intelligent monitoring system for the power distribution cabinet in this embodiment also includes a communication module, which is used to communicate with the host computer or cloud platform and transmit data.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring of power distribution cabinets, characterized by: The power distribution cabinet is placed in a constant temperature environment. An infrared thermal imaging module is installed on the inner side of the cabinet door to capture and generate infrared thermal images of the cabinet. The image is evenly divided into a grid array. A rectangular detection area is defined for each component based on its coverage area. Each detection area contains multiple grid cells. Grid cells outside each detection area are defined as reference grids. The quadrilateral area formed by connecting the opposite sides of adjacent detection areas is defined as the sampling area. During the operation of the power distribution cabinet, infrared thermal images of the cabinet are continuously captured and generated at certain intervals. Two adjacent infrared thermal images are selected, and the front and rear temperature values ​​of each detection area and each reference grid are obtained by measuring the image brightness. Then, the temperature difference values ​​of each detection area and each reference grid are calculated. The temperature difference values ​​of each detection area are compared with those of its adjacent detection areas in turn. Then, based on the coverage area of ​​the two adjacent detection areas, the distance between the center points of the areas, the type of components, the typical operating temperature, and the range of temperature difference values, the angle α and the distance L0 of the heat center movement direction in the sampling area between the two detection areas are simulated. Simultaneously, the average temperature difference value of each reference grid in the sampling area is calculated, and the reference grid with a temperature difference value greater than the average value is defined as a type A reference grid, and the reference grid with a temperature difference value less than the average value is defined as a type B reference grid. Each Class A reference grid is converted into a vector by assigning a value to it based on the difference between the temperature difference value and the average value of the Class A reference grid, and defining this value as the magnitude of the vector; the angle between the line connecting the center point of the grid and the center point of the detection area with a large temperature difference value and the horizontal line is defined as the angle of the vector, thus obtaining the vectors of all Class A reference grids. Each type B reference grid is converted into a vector by assigning a value to it based on the difference between the temperature difference value and the average value of the type B reference grid, and defining this value as the magnitude of the vector; the angle between the line connecting the center point of the grid and the center point of the detection area with the smaller temperature difference value and the horizontal line is defined as the angle of the vector, thus obtaining the vectors of all type B reference grids. Then, the sum of all vectors is calculated, followed by the direction angle β and magnitude V of the sum vector. Finally, the corresponding theoretical distance value L1 is calculated, where L1 = V. k, where k is a preset conversion factor; Pre-set angle ratio range and distance ratio range. When the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area. The two sets of components are marked as "potential faults" and a prompt is sent to the operation and maintenance personnel to remind them to go to the site for troubleshooting.

2. The intelligent monitoring method for power distribution cabinets according to claim 1, characterized in that: A model was pre-built based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the two detection areas, the types of two sets of corresponding components, and the typical operating temperature. Through multiple sets of experiments, different temperature differences were established between the temperatures of the two sets of corresponding components and the typical operating temperature. Multiple temperature difference ranges were set, and the brightness change characteristics of the sampling area in the infrared thermal images of the experiment were extracted and analyzed within different temperature difference ranges to obtain the heat migration characteristics of the sampling area. These heat migration characteristics include the angle α of the heat center's movement direction and the movement distance L0. Transfer features are the output of the model; The model was trained and its parameters were optimized using multiple sets of experimental data. During the operation of the distribution cabinet, two adjacent detection areas were selected. The size of the coverage area of ​​the two detection areas, the distance between the center points of the areas, the types of the corresponding components and the typical operating temperature, as well as the temperature difference range were modeled and input into the model. The model automatically outputs the corresponding heat migration characteristics, namely the angle α of the heat center movement direction and the movement distance L0.

3. The intelligent monitoring method for distribution cabinets according to claim 1, characterized in that: in When assigning values ​​to type A grids, a calculation coefficient m1 is matched and used to correct the assigned value; when assigning values ​​to type B grids, a calculation coefficient m2 is matched and used to correct the assigned value; by default, the values ​​of m1 and m2 are both 1. When maintenance personnel go to the site to investigate, they input the investigation results and adjust the values ​​of m1 and m2 according to the investigation results.

4. An intelligent monitoring system for power distribution cabinets, characterized in that, include: Infrared thermal imaging module, used to capture and generate infrared thermal images of the cabinet; The image acquisition module is used to acquire infrared thermal images output by the infrared thermal imaging module and transmit the image data to the microprocessor module; The microprocessor module is used to process infrared thermal images and evaluate the two groups of components based on the temperature difference between two adjacent groups of components in the infrared thermal image and the temperature difference between each sampling grid in the sampling area between the two groups of components, in order to determine whether there are any "potential faults" in the two groups of components. Storage module, which is used to store data; An alarm module is used to issue an alarm signal when the microprocessor module determines that there is a "potentially faulty" component in the two groups of components; The power supply module is used to supply power to the various modules.

5. The intelligent monitoring system for distribution cabinets according to claim 4, characterized in that: The microprocessor module includes an image processing unit, a model unit, and a computing unit; The image processing unit is used to perform gridding processing on the infrared thermal image to determine the temperature value of each grid cell, as well as the temperature value of the corresponding detection area of ​​each component; The model unit is used to model the heat migration characteristics based on the coverage area of ​​two adjacent detection areas, the distance between the center points of the areas, the types of two sets of corresponding components and the typical operating temperature, as well as the temperature difference range, namely the heat center movement direction angle α and the movement distance L0. The calculation unit is used to vectorize the reference grid in the sampling area between two adjacent detection areas according to the temperature difference value, and calculate the theoretical moving direction angle β and moving distance L1 of the heat center; when the ratio of β to α exceeds the preset angle ratio range and the ratio of L1 to L0 exceeds the preset distance ratio range, it is considered that there are suspected abnormal components in the two sets of components corresponding to this adjacent detection area, and the two sets of components are marked as "potential faults".

6. The intelligent monitoring system for distribution cabinets according to claim 5, characterized in that: The microprocessor module also includes a parameter adjustment unit, which is used to adjust the preset parameters within the computing unit.

7. The intelligent monitoring system for distribution cabinets according to claim 4, characterized in that: It also includes a communication module, which is used to communicate with a host computer or cloud platform and transmit data.

Images