Intelligent substation patrol robot based on machine vision

Abstract

The invention belongs to the technical field of intelligent inspection, and particularly discloses and provides a transformer substation intelligent inspection robot based on machine vision, which comprises an area division module, a robot control module and a robot control module, and is characterized in that the area division module divides a transformer substation into areas according to preset rules; the priority degree analysis module analyzes the priority degree index of each region to set a routing inspection path by extracting a historical routing inspection path schematic diagram of the routing inspection robot in each region and collecting data such as soil humidity, total area and soil area. And the equipment state analysis module collects equipment images in each area and analyzes damage and corrosion degree indexes. The equipment temperature abnormity judgment module collects thermal imaging images, analyzes temperature abnormity degree indexes, judges whether the equipment temperature is abnormal or not, screens out abnormal equipment, and further judges whether a barrier exists in a ventilation opening or not so as to carry out early warning. And the safety early warning module analyzes the safety degree index of each region based on the damage, corrosion and temperature anomaly degree indexes, judges whether the region is safe and performs early warning.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent patrol, and relates to a substation intelligent patrol robot based on machine vision. Background Art

[0002] With the rapid development of the national economy, the demand for safe, stable and sufficient power supply is growing. The overall architecture of the power grid is becoming increasingly large and complex, and the scale of substation equipment is also expanding. However, the contradiction between the current operation and maintenance management model and the rapid growth of equipment is becoming increasingly prominent, and there are problems such as insufficient equipment monitoring intensity, insufficient operation and maintenance management details, and insufficient support and guarantee capabilities. Therefore, more intelligent and efficient operation and maintenance methods are needed to ensure the safe and stable operation of the power grid.

[0003] For example, the Chinese patent with publication number CN112405624A discloses an intelligent patrol robot system, including a sensing component, a control component, a mobile terminal, a communication component, a power component and a self-protection component; the sensing component, the communication component, the power component and the self-protection component are all connected to the control component, and the control component transmits data with the mobile terminal through the communication component. When the present invention is in use, the mobile terminal controls the patrol robot to work through the communication component. During the robot patrol process, the sensing component collects the surrounding environment, and the collected parameters are transferred to the control component for storage. At the same time, the parameters are also fed back to the mobile terminal through the communication component, which is convenient for the staff to view; at the same time, the self-protection component works, and when a malicious attack or damage is detected, the patrol robot is self-protected to avoid the destruction or theft of key equipment and machine data; the present invention meets the patrol needs of large shopping malls or warehouses and reduces manpower work.

[0004] The above existing technologies have the following problems: 1. The existing technologies lack an effective area division method, which leads to chaotic substation management. It is difficult for managers to fully grasp the overall layout and operation status. Due to the lack of clear area division, it is difficult to quickly locate the problem area when a problem occurs, which increases the difficulty of fault detection and handling and affects the stable operation of the substation.

[0005] 2. The existing technology does not combine the soil conditions and historical inspection conditions of the power station, and cannot fully evaluate the actual conditions of each area to determine a reasonable inspection priority. It may cause the inspection robot to inspect too early in areas with high soil moisture and large soil area, and get stuck in mud, greatly slowing down the inspection efficiency, and may even cause damage to the equipment. There are also situations where repeated inspections waste resources, or key areas are missed, and it is impossible to conduct efficient inspections on areas that require focus within a limited time, affecting the overall safe and stable operation and management efficiency of the substation.

[0006] 3. Existing technologies may not be able to effectively check whether there are obstacles at the ventilation openings of the equipment, and it is difficult to ensure that the equipment has good heat dissipation conditions, which may lead to overheating problems due to poor heat dissipation, thereby affecting the normal operation and life of the equipment. At the same time, it also increases the possibility of equipment failure and reduces the operational stability and safety of the substation.

[0007] 4. The existing technology lacks the means to quantitatively evaluate the degree of damage and corrosion of the equipment. It can only make subjective judgments, which are prone to errors and cannot accurately understand the actual status of the equipment. This means that when deciding whether to repair or replace the equipment, there is a lack of objective and accurate data support, which may lead to wrong decisions. Summary of the invention

[0008] In view of this, in order to solve the problems raised in the above background technology, a substation intelligent patrol robot based on machine vision is proposed.

[0009] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a substation intelligent patrol robot based on machine vision, including: an area division module: used to divide the substation into various areas based on preset area division rules to obtain various areas of the substation.

[0010] Priority analysis module: It is used to extract the historical inspection path diagram of the inspection robot in each area of ​​the substation from the database and the collected soil moisture, total area and soil area of ​​each area of ​​the substation, analyze the priority index of each area of ​​the substation, and then set the inspection path of the inspection robot according to the priority index of each area of ​​the substation.

[0011] Equipment status analysis module: used to collect images of each device in each area of ​​the substation, and analyze the damage degree index and corrosion degree index of each device in each area of ​​the substation.

[0012] Equipment temperature anomaly determination module: used to collect thermal imaging images of each device in each area of ​​the substation and analyze the temperature anomaly index of each device in each area of ​​the substation gj , where g represents the number corresponding to the g-th area of ​​the substation, g=1,2,3,...G, and j represents the number corresponding to the j-th device, j=1,2,3,...J. It is determined whether the temperature of each device in each area of ​​the substation is abnormal, and the temperature-abnormal devices in each area of ​​the substation are screened out. Then, it is determined whether there are obstacles in the ventilation openings of the temperature-abnormal devices in each area of ​​the substation, and an early warning is issued.

[0013] Safety warning module: It is used to analyze the safety index of each area of ​​the substation based on the damage index, corrosion index and temperature anomaly index of each equipment in each area of ​​the substation, judge whether each area of ​​the substation is safe, and then issue an early warning.

[0014] Database: used to store the historical inspection path diagrams of the inspection robot in each area of ​​the substation, store the standard images of each device in each area of ​​the substation, store the standard color component values ​​of each pixel in the standard image of each device, and store the predefined key features of each obstacle.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention makes the management of the substation more orderly through regional division, which helps the management personnel to better understand the overall layout and operation of the substation. The regional division makes the inspection work more targeted, and corresponding inspection plans and strategies can be formulated according to the characteristics of the region, thereby reducing the unnecessary movement and stay time of the inspection robot and improving the inspection efficiency.

[0016] (2) The present invention analyzes the priority index of each area of ​​the substation by extracting the historical inspection path diagram of the inspection robot in each area of ​​the substation from the database and collecting the soil moisture, total area and soil area of ​​each area of ​​the substation. Then, the inspection path of the inspection robot is set according to the priority index of each area of ​​the substation. The inspection path can give priority to visiting areas with higher risks or requiring more attention and reduce the stay time in unimportant or low-risk areas, thereby significantly improving the inspection efficiency. Reasonable inspection path planning can ensure that the inspection robot covers more key areas within a limited time, avoiding waste of resources and unnecessary duplication of work.

[0017] (3) The present invention collects images of each device in each area of ​​the substation and analyzes the damage degree index and corrosion degree index of each device in each area of ​​the substation. It can timely discover the damage signs and corrosion conditions on the surface of the equipment, and quantitatively evaluate the damage degree and corrosion degree of the equipment, providing objective and accurate data support, which helps to avoid errors caused by subjective judgment, timely discover and deal with equipment damage and corrosion problems, avoid safety accidents caused by equipment failure, and ensure the safe operation of the substation.

[0018] (4) The present invention can measure the temperature distribution on the surface of the equipment in a non-contact manner through thermal imaging technology, and timely detect temperature anomalies caused by internal faults, poor contact, overload, etc. This helps to take measures before the fault occurs to avoid equipment damage and power outages. Timely detection and treatment of temperature abnormalities can prevent the expansion of faults, extend the service life of the equipment, and improve the overall reliability of the equipment. By checking whether there are obstacles in the vents, it can ensure that the equipment has good heat dissipation conditions and avoid overheating caused by poor heat dissipation.

[0019] (5) The present invention can timely discover potential safety hazards and avoid equipment failures or accidents by real-time monitoring and analyzing the status parameters of equipment in various areas of the substation, thereby significantly improving the overall safety of the substation. Timely warning and intervention can significantly reduce the risk of accidents in the substation and ensure the stable operation of the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0021] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention. DETAILED DESCRIPTION

[0022] 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.

[0023] See also Figure 1 As shown, the present invention provides a substation intelligent patrol robot based on machine vision, including: an area division module, a priority analysis module, an equipment status analysis module, an equipment temperature anomaly determination module, a safety warning module and a database.

[0024] The priority analysis module is connected to the area division module, the equipment status analysis module and the database respectively; the equipment status analysis module is connected to the priority analysis module, the equipment temperature anomaly determination module and the database respectively; the equipment temperature anomaly determination module is connected to the equipment status analysis module, the safety warning module and the database respectively; the database is connected to the priority analysis module, the equipment status analysis module and the equipment temperature anomaly determination module respectively.

[0025] Area division module: used to divide the substation into various areas based on preset area division rules to obtain various areas of the substation.

[0026] As a preferred embodiment, the preset area division rule is specifically: all devices passing through the same bus are regarded as one area.

[0027] The present invention makes the management of substations more orderly through regional division, which helps managers to better understand the overall layout and operation of substations. Regional division makes inspection work more targeted, and corresponding inspection plans and strategies can be formulated according to regional characteristics, reducing unnecessary movement and stay time of inspection robots and improving inspection efficiency.

[0028] Priority analysis module: It is used to extract the historical inspection path diagram of the inspection robot in each area of ​​the substation from the database and the collected soil moisture, total area and soil area of ​​each area of ​​the substation, analyze the priority index of each area of ​​the substation, and then set the inspection path of the inspection robot according to the priority index of each area of ​​the substation.

[0029] Furthermore, the analysis of the priority index of each area of ​​the substation includes: according to the historical inspection path diagram of the inspection robot in each area of ​​the substation, counting the number of inspections of the inspection robot in each area of ​​the substation within a set monitoring time period, dividing the number of inspections by the monitoring time corresponding to the set monitoring time period to obtain the historical inspection frequency of the inspection robot in each area of ​​the substation within the set monitoring time period, recorded as frequency g , where g represents the number corresponding to the g-th area of ​​the substation, g = 1, 2, 3, ...G.

[0030] The soil moisture in each area of ​​the substation is collected through humidity sensors and recorded as humidity g , analyze the soil moisture index midity of each area of ​​the substation g , where humidity′ represents the permissible soil moisture in the area of ​​the substation.

[0031] By measuring and collecting the soil area of ​​each area of ​​the substation and the total area of ​​the substation, the soil area of ​​each area of ​​the substation is divided by the total area of ​​the substation to obtain the soil area ratio of each area of ​​the substation, which is recorded as Propor g .

[0032] Analyze the priority index of each area of ​​the substation g , Wherein ι1, ι2 and ι3 represent the weights corresponding to the soil moisture index, soil area ratio and historical inspection frequency of the substation area, respectively, and ι1+ι2+ι3=1.

[0033] As a preferred embodiment, the ι1, ι2 and ι3 can be set to 0.4, 0.3 and 0.3 respectively. Soil moisture is one of the key factors for the safe operation of the substation. Excessive soil moisture may cause the equipment to become damp, the insulation performance to deteriorate, and even cause faults such as short circuits. The soil area ratio reflects the actual size and importance of each area of ​​the substation. Although a large area does not necessarily mean a high risk, a larger area may contain more equipment and potential risk points. The historical inspection frequency provides the degree of attention paid by the inspection robot to each area in the past period of time. For areas where problems frequently occur, a higher historical inspection frequency may be reasonable; for relatively stable areas, the number of inspections can be appropriately reduced. Therefore, the weight corresponding to the soil moisture index of the substation area is greater than the weights corresponding to the soil area ratio and the historical inspection frequency.

[0034] Furthermore, the specific process of setting the inspection path of the patrol robot is: sorting the priority index of each area of ​​the substation in descending order to obtain the sorted areas of the substation, and setting the inspection path of the patrol robot based on the sorted areas of the substation.

[0035] The present invention analyzes the priority index of each area of ​​the substation by extracting the historical inspection path diagram of the inspection robot in each area of ​​the substation from the database and collecting the soil moisture, total area and soil area of ​​each area of ​​the substation, and then sets the inspection path of the inspection robot according to the priority index of each area of ​​the substation. It can give priority to visiting areas with higher risks or requiring more attention, and reduce the stay time in unimportant or low-risk areas, thereby significantly improving the inspection efficiency. Reasonable inspection path planning can ensure that the inspection robot covers more key areas within a limited time, avoiding waste of resources and unnecessary duplication of work.

[0036] Equipment status analysis module: used to collect images of each device in each area of ​​the substation, and analyze the damage degree index and corrosion degree index of each device in each area of ​​the substation.

[0037] Furthermore, the analysis of the damage degree index of each device in each area of ​​the substation includes: A1, extracting the standard image of each device in each area of ​​the substation from the database, and obtaining each feature point in the image of each device in each area of ​​the substation and each feature point in the standard image of each device through a feature extraction algorithm.

[0038] A2. Match each feature point in the image of each device in each area of ​​the substation with each feature point in the standard image of each device to obtain the feature point in the device standard image corresponding to each feature point in the image of each device in each area of ​​the substation, and then obtain the coordinates of each feature point in the image of each device in each area of ​​the substation and the coordinates of the feature point in the device standard image corresponding to it by image processing technology, substitute it into the Euclidean distance formula to obtain the displacement between each feature point in the image of each device in each area of ​​the substation and the feature point in the device standard image corresponding to it, select the maximum displacement as the displacement of the feature point in the image of each device in each area of ​​the substation and its standard image, recorded as displace gj .

[0039] As a preferred embodiment, the specific process of matching feature points in the standard image of the device corresponding to each feature point in the image of each device in each area of ​​the substation is: obtaining feature descriptors of each feature point in the image of each device in each area of ​​the substation and each feature point in the standard image of each device through a feature description algorithm, and then matching the feature descriptors of each feature point in the image of each device in each area of ​​the substation and each feature point in the standard image of each device through a feature matching algorithm, screening feature points with the same feature descriptors corresponding to the feature points in the image of each device in each area of ​​the substation and the feature points in the standard image of the device, and using them as the matching feature points in the standard image of the device corresponding to the feature points in the image of each device in each area of ​​the substation.

[0040] A3. Analyze the damage index of each equipment in each area of ​​the substation gj .

[0041] The specific formula for analyzing the damage degree index of each device in each area of ​​the substation is: Where displace′ represents the allowable displacement of feature points in the set substation equipment image and its standard image.

[0042] Furthermore, the analysis of the corrosion index of each device in each area of ​​the substation includes: obtaining the color component value of each pixel in the image of each device in each area of ​​the substation through the OpenCV library, and the color component value includes a hue component value, a saturation component value, and a brightness component value, which are respectively recorded as H gji , S gji and L gji , where i represents the number corresponding to the i-th pixel, i=1,2,3,...I.

[0043] Extract the hue component value H from the standard color component value of each pixel in the standard image of each device from the database j ' i , saturation component value Sj ' i , brightness component value L j ' i .

[0044] Analyze the corrosion index of each equipment in each area of ​​the substation gj , Wherein φ1, φ2 and φ3 represent the deviation coefficient of the hue component value, the deviation coefficient of the saturation component value and the deviation coefficient of the brightness component value of the pixel point in the image of the substation equipment respectively, and φ1+φ2+φ3=1.

[0045] As a preferred embodiment, the φ1, φ2 and φ3 can be set to 0.4, 0.3 and 0.3 respectively. Hue is one of the basic properties of color and is crucial for identifying color changes on the surface of the equipment. In the substation environment, the original color of the equipment may change due to corrosion. For example, rust on the metal surface may cause the color to darken or produce spots. Therefore, the deviation coefficient of the hue component can well reflect the subtle changes in this color and is an important indicator for evaluating the degree of corrosion; saturation reflects the purity or vividness of the color. During the corrosion process, the color of the surface of the equipment may become dim, that is, the saturation decreases. Brightness describes the brightness of the color. Although corrosion may cause the surface gloss to decrease, thereby affecting the brightness, the brightness is affected by many factors, including lighting conditions, equipment material and surface texture. Therefore, the weight corresponding to the deviation coefficient of the hue component value of the pixel point in the image of the equipment in the substation area is greater than the weight corresponding to the deviation coefficient of the saturation component value and the deviation coefficient of the brightness component value.

[0046] The present invention collects images of each device in each area of ​​the substation, analyzes the damage degree index and corrosion degree index of each device in each area of ​​the substation, can timely discover damage signs and corrosion conditions on the surface of the equipment, and quantitatively evaluate the damage degree and corrosion degree of the equipment, provide objective and accurate data support, help avoid errors caused by subjective judgment, timely discover and deal with equipment damage and corrosion problems, avoid safety accidents caused by equipment failure, and ensure the safe operation of the substation.

[0047] Equipment temperature anomaly determination module: used to collect thermal imaging images of each device in each area of ​​the substation and analyze the temperature anomaly index of each device in each area of ​​the substation gj, where g represents the number corresponding to the g-th area of ​​the substation, g=1,2,3,...G, and j represents the number corresponding to the j-th device, j=1,2,3,...J. It is determined whether the temperature of each device in each area of ​​the substation is abnormal, and the temperature-abnormal devices in each area of ​​the substation are screened out. Then, it is determined whether there are obstacles in the ventilation openings of the temperature-abnormal devices in each area of ​​the substation, and an early warning is issued.

[0048] Furthermore, the analysis of the temperature anomaly index of each device in each area of ​​the substation includes: collecting thermal imaging images of each device in each area of ​​the substation by a thermal imager, and then extracting the highest temperature of the surface of each device in each area of ​​the substation by image processing technology, and using it as the surface temperature of each device in each area of ​​the substation, recorded as htem gj .

[0049] As a preferred embodiment, the specific process of extracting the maximum temperature of the surface of each device in each area of ​​the substation through image processing technology is: segmenting the device area and the background area in the thermal imaging image of each device in each area of ​​the substation through the threshold segmentation method, traversing each pixel point of the device area in the thermal imaging image of each device in each area of ​​the substation, extracting the temperature value corresponding to each pixel point, and extracting the maximum temperature corresponding to the pixel point of the device area in the thermal imaging image of each device in each area of ​​the substation, and using it as the maximum temperature of the surface of each device in each area of ​​the substation.

[0050] Analyze the temperature anomaly index tem of each device in each area of ​​the substation gj , where tem′ j represents the reference temperature of the jth device in the area of ​​the substation.

[0051] Furthermore, the specific process of determining whether the temperature of each device in each area of ​​the substation is abnormal is as follows: comparing the temperature anomaly degree index of each device in each area of ​​the substation with a preset temperature anomaly degree index threshold; if the temperature anomaly degree index of a device in a certain area of ​​the substation is greater than the preset temperature anomaly degree index threshold, then the device in the area of ​​the substation is abnormal, and the temperature abnormal devices in each area of ​​the substation are screened out.

[0052] The determining whether there are obstacles at the ventilation openings of the devices with abnormal temperatures in the respective areas of the substation includes: selecting images of the devices with abnormal temperatures in the respective areas of the substation from images of the devices in the respective areas of the substation.

[0053] The key features of the vent area in the image of each temperature abnormal device in each area of ​​the substation are obtained through feature extraction technology.

[0054] The key features of each predefined obstacle are extracted from the database, and the key features of the vent area in the image of each temperature abnormal device in each area of ​​the substation are compared with the predefined key features of each obstacle to obtain the obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation.

[0055] The obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation is compared with the preset obstacle similarity threshold. If the obstacle similarity of the vent area in the image of a certain temperature abnormal device in a certain area of ​​the substation is greater than the preset obstacle similarity threshold, there is an obstacle at the vent of the temperature abnormal device in the area of ​​the substation. Similarly, it is judged whether there is an obstacle at the vent of each temperature abnormal device in each area of ​​the substation.

[0056] As a preferred embodiment, the key features include: shape, texture and color.

[0057] The specific process of obtaining the obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation is: extracting the feature vectors of the key features of the vent area in the image of each temperature abnormal device in each area of ​​the substation and the predefined key features of each obstacle, substituting them into the cosine calculation formula, and obtaining the obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation.

[0058] The present invention can measure the temperature distribution on the surface of the equipment in a non-contact manner through thermal imaging technology, and timely discover temperature anomalies caused by internal faults, poor contact, overload, etc. This helps to take measures before the fault occurs to avoid equipment damage and power outages. Timely discovery and treatment of temperature abnormalities can prevent the expansion of faults, extend the service life of the equipment, and improve the overall reliability of the equipment. By checking whether there are obstacles in the vents, it can ensure that the equipment has good heat dissipation conditions and avoid overheating problems caused by poor heat dissipation.

[0059] Safety warning module: It is used to analyze the safety index of each area of ​​the substation based on the damage index, corrosion index and temperature anomaly index of each equipment in each area of ​​the substation, judge whether each area of ​​the substation is safe, and then issue an early warning.

[0060] Furthermore, the specific formula for analyzing the safety index of each area of ​​the substation is: Where ε1, ε2 and ε3 represent the weights corresponding to the damage degree index, corrosion degree index and temperature anomaly degree index of the equipment in the substation area, respectively, and ε1+ε2+ε3=1, where e represents a natural constant, and J represents the total number of equipment.

[0061] As a preferred embodiment, the ε1, ε2 and ε3 can be set to 0.5, 0.3 and 0.2 respectively. Equipment damage is a key factor that directly affects the safe operation of the substation. Once serious damage occurs, the entire substation may be shut down. Although the degree of corrosion is not as direct as equipment damage, long-term corrosion may cause equipment performance to decline and increase the risk of failure. Temperature abnormality is usually a precursor to equipment failure, but its impact is smaller than direct damage and corrosion. Therefore, the weight corresponding to the damage degree index of the equipment in the substation area is greater than the weights corresponding to the corrosion degree index and the temperature abnormality degree index.

[0062] Furthermore, the specific process of judging whether each area of ​​the substation is safe is as follows: comparing the safety index of each area of ​​the substation with a preset safety index threshold; if the safety index of an area of ​​the substation is less than the preset safety index threshold, then the area of ​​the substation is a non-safe area, and the non-safe areas of the substation are screened out, and then an early warning is issued.

[0063] The present invention can timely discover potential safety hazards and avoid equipment failures or accidents by real-time monitoring and analyzing the status parameters of equipment in various areas of the substation, thereby significantly improving the overall safety of the substation. Timely warning and intervention can significantly reduce the risk of accidents in the substation and ensure the stable operation of the substation.

[0064] Database: used to store the historical inspection path diagrams of the inspection robot in each area of ​​the substation, store the standard images of each device in each area of ​​the substation, store the standard color component values ​​of each pixel in the standard image of each device, and store the predefined key features of each obstacle.

[0065] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A substation intelligent patrol robot based on machine vision, characterized by: include: Area division module: used to divide the substation into various areas based on the preset area division rules to obtain various areas of the substation; Priority analysis module: used to extract the historical inspection path diagram of the inspection robot in each area of ​​the substation from the database and the collected soil moisture, total area and soil area of ​​each area of ​​the substation, analyze the priority index of each area of ​​the substation, and then set the inspection path of the inspection robot according to the priority index of each area of ​​the substation; Equipment status analysis module: used to collect images of each device in each area of ​​the substation and analyze the damage index of each device in each area of ​​the substation gj and corrosion index gj , where g represents the number corresponding to the g-th area of ​​the substation, g = 1, 2, 3, ... G, and j represents the number corresponding to the j-th device, j = 1, 2, 3, ... J; Equipment temperature anomaly determination module: used to collect thermal imaging images of each device in each area of ​​the substation and analyze the temperature anomaly index of each device in each area of ​​the substation gj , determine whether the temperature of each device in each area of ​​the substation is abnormal, screen out the devices with abnormal temperatures in each area of ​​the substation, and then determine whether there are obstacles in the ventilation openings of the devices with abnormal temperatures in each area of ​​the substation, and issue an early warning; Safety warning module: used to analyze the safety index of each area of ​​the substation based on the damage index, corrosion index and temperature anomaly index of each equipment in each area of ​​the substation, determine whether each area of ​​the substation is safe, and then issue an early warning; Database: used to store the historical inspection path diagrams of the inspection robot in each area of ​​the substation, store the standard images of each device in each area of ​​the substation, store the standard color component values ​​of each pixel in the standard image of each device, and store the predefined key features of each obstacle.

2. According to claim 1, a substation intelligent inspection robot based on machine vision is characterized in that: The priority index of each area of ​​the analysis substation includes: According to the historical inspection path diagram of the inspection robot in each area of ​​the substation, the number of inspections of the inspection robot in each area of ​​the substation within the set monitoring time period is counted, and the number is divided by the monitoring time corresponding to the set monitoring time period to obtain the historical inspection frequency of the inspection robot in each area of ​​the substation within the set monitoring time period, which is recorded as frequency g ; The soil moisture in each area of ​​the substation is collected through humidity sensors and recorded as humidity g , analyze the soil moisture index midity of each area of ​​the substation g , where humidity′ represents the permissible soil moisture in the area of ​​the substation; By measuring and collecting the soil area of ​​each area of ​​the substation and the total area of ​​the substation, the soil area of ​​each area of ​​the substation is divided by the total area of ​​the substation to obtain the soil area ratio of each area of ​​the substation, which is recorded as Propor g ; Analyze the priority index of each area of ​​the substation g , Wherein ι1, ι2 and ι3 represent the weights corresponding to the soil moisture index, soil area ratio and historical inspection frequency of the substation area, respectively, and ι1+ι2+ι3=1.

3. The substation intelligent patrol robot based on machine vision according to claim 1, characterized in that: The specific process of setting the inspection path of the inspection robot is as follows: The priority indexes of the various areas of the substation are sorted in descending order to obtain the sorted areas of the substation, and the inspection paths of the inspection robots are set based on the sorted areas of the substation.

4. The intelligent substation inspection robot based on machine vision according to claim 1 is characterized in that: The analysis of the damage index of each device in each area of ​​the substation includes: A1. Extracting standard images of each device in each area of ​​the substation from the database, and obtaining feature points in the images of each device in each area of ​​the substation and feature points in the standard images of each device through a feature extraction algorithm; A2. Match each feature point in the image of each device in each area of ​​the substation with each feature point in the standard image of each device to obtain the feature point in the device standard image corresponding to each feature point in the image of each device in each area of ​​the substation, and then obtain the coordinates of each feature point in the image of each device in each area of ​​the substation and the coordinates of the feature point in the device standard image corresponding to it by image processing technology, substitute it into the Euclidean distance formula to obtain the displacement between each feature point in the image of each device in each area of ​​the substation and the feature point in the device standard image corresponding to it, select the maximum displacement as the displacement of the feature point in the image of each device in each area of ​​the substation and its standard image, recorded as displace gj ; A3. Analyze the damage index of each equipment in each area of ​​the substation gj .

5. The substation intelligent patrol robot based on machine vision according to claim 1 is characterized by: The analysis of the corrosion index of each device in each area of ​​the substation includes: The color component values ​​of each pixel in the image of each device in each area of ​​the substation are obtained through the OpenCV library. The color component values ​​include hue component value, saturation component value, and brightness component value, which are respectively denoted as H gji , S gji and L gji , where i represents the number corresponding to the i-th pixel, i=1,2,3,...I; Extract the hue component value H from the standard color component value of each pixel in the standard image of each device from the database j ' i , saturation component value S j ' i , brightness component value L j ' i ; Analyze the corrosion index of each equipment in each area of ​​the substation gj , Wherein φ1, φ2 and φ3 represent the deviation coefficient of the hue component value, the deviation coefficient of the saturation component value and the deviation coefficient of the brightness component value of the pixel point in the image of the substation equipment respectively, and φ1+φ2+φ3=1.

6. The substation intelligent inspection robot based on machine vision according to claim 1, characterized in that: The analysis of the temperature anomaly index of each device in each area of ​​the substation includes: The thermal imaging images of each device in each area of ​​the substation are collected by a thermal imager, and then the highest temperature of the surface of each device in each area of ​​the substation is extracted by image processing technology, which is used as the surface temperature of each device in each area of ​​the substation, recorded as htem gj ; Analyze the temperature anomaly index tem of each device in each area of ​​the substation gj , where tem′ j Indicates the reference temperature of the j-th device in the area of ​​the set substation.

7. The substation intelligent inspection robot based on machine vision according to claim 1 is characterized by: The specific process of judging whether the temperature of each device in each area of ​​the substation is abnormal is as follows: The temperature anomaly index of each device in each area of ​​the substation is compared with the preset temperature anomaly index threshold. If the temperature anomaly index of a device in a certain area of ​​the substation is greater than the preset temperature anomaly index threshold, the device in the area of ​​the substation is abnormal, and the temperature abnormal devices in each area of ​​the substation are screened out.

8. The intelligent substation inspection robot based on machine vision according to claim 7 is characterized in that: The determining whether there are obstacles at the ventilation openings of each temperature abnormality equipment in each area of ​​the substation includes: Screening out images of devices with abnormal temperatures in each area of ​​the substation from images of devices in each area of ​​the substation; The key features of the vent area in the image of each temperature abnormal device in each area of ​​the substation are obtained by feature extraction technology; Extract the predefined key features of each obstacle from the database, compare the key features of the vent area in the image of each temperature abnormal device in each area of ​​the substation with the predefined key features of each obstacle, and obtain the obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation; The obstacle similarity of the vent area in the image of each temperature abnormal device in each area of ​​the substation is compared with the preset obstacle similarity threshold. If the obstacle similarity of the vent area in the image of a certain temperature abnormal device in a certain area of ​​the substation is greater than the preset obstacle similarity threshold, there is an obstacle at the vent of the temperature abnormal device in the area of ​​the substation. Similarly, it is judged whether there is an obstacle at the vent of each temperature abnormal device in each area of ​​the substation.

9. The intelligent substation inspection robot based on machine vision according to claim 1, characterized in that: The specific formula for analyzing the safety index of each area of ​​the substation is: Where ε1, ε2 and ε3 represent the weights corresponding to the damage degree index, corrosion degree index and temperature anomaly degree index of the equipment in the substation area, respectively, and ε1+ε2+ε3=1, where e represents a natural constant, and J represents the total number of equipment.

10. The intelligent substation inspection robot based on machine vision according to claim 1, characterized in that: The specific process of judging whether each area of ​​the substation is safe is as follows: The safety index of each area of ​​the substation is compared with the preset safety index threshold. If the safety index of an area of ​​the substation is less than the preset safety index threshold, the area of ​​the substation is a non-safe area. The non-safe areas of the substation are screened out and an early warning is issued.

Images