Power transmission line safety distance prediction method combining artificial intelligence and power business

Abstract

The invention discloses a power transmission line safety distance prediction method combining artificial intelligence and power business, and the method comprises the steps: 1, carrying out the detection of a scene image through an artificial intelligence detection model, recognizing and marking all types of known defect hidden troubles in the scene image; 2, on the basis of a monocular vision distance calculation algorithm and power transmission service actual information, combining target hidden danger actual information, and calculating actual distance information between a hidden danger point or a target object and a power transmission tower; and 3, carrying out distance-based early warning and pushing on the defect by combining scene sensing equipment and a distance early warning guide rule. The method is suitable for early warning of external damage hidden danger of the power transmission line in advance, and monitoring, early warning and emergency disposal of the threat condition of the tower at the defect distance are achieved by carrying scene sensing equipment on line equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring of power transmission lines, and in particular to a method for predicting safe distances of power transmission lines by combining artificial intelligence with power services. Background Art

[0002] Illegal construction, illegal construction operations, and unplanned and seasonal construction within transmission line protection zones pose significant threats to the safe and stable operation of transmission lines. Due to the lack of safety awareness among drivers of special vehicles such as cranes, grouting trucks, pump trucks, pile drivers, excavators, and dump trucks, coupled with improper handling during large-scale construction, personal injuries and transmission line tripping are common, posing a significant threat to personal safety and the power grid. These accidents can cause power outages at the very least, while at the worst, they can cause severe damage to lines and other facilities, widespread power outages, and even fatalities. Currently, intelligent image recognition technology solely based on construction features, construction vehicles, and other objects lacks the spatial and temporal accuracy required for power generation. Therefore, further improvements are needed to improve the early warning and detection system for defects and hidden dangers that can cause external damage to transmission lines. This requires integrating specialized knowledge of transmission line operations with artificial intelligence image recognition technology. This approach, combining AI with power transmission operations, can shift early warning systems for transmission line defects and hidden dangers from empirical to standardized, quantitative approaches. Doing a good job in preventing defects and early warning of power equipment is related to the reliable supply of electricity and the safe operation of power equipment, and is also related to the smooth implementation of emergency rescue and post-disaster rescue and production resumption work. Summary of the Invention

[0003] The purpose of the present invention is to provide a transmission line safety distance prediction method that combines artificial intelligence with power business. By combining artificial intelligence, transmission business and defect and hidden danger information, the threat level of defects or external force damage to the transmission line body and line channel environment to the tower is calculated, so as to realize the threat of defects and hidden dangers of transmission lines and provide early warning, thereby improving the efficiency of emergency response.

[0004] The technical solution adopted in the present invention is:

[0005] The transmission line safety distance prediction method combining artificial intelligence with power business specifically includes the following steps:

[0006] Step 1: Use the artificial intelligence detection model to detect the scene image, identify various types of known defects and hidden dangers in the scene image, and mark them;

[0007] Step 2: Based on the monocular vision distance calculation algorithm and the actual information of the power transmission business, combined with the actual information of the target hidden danger, the actual distance between the hidden danger point or target object and the transmission tower is calculated;

[0008] Step 3: Combine scene perception equipment with distance warning guidelines to provide distance-based early warning alerts and push notifications for defects.

[0009] Furthermore, the identification of various defects in step 1 includes the following steps:

[0010] Step 1-1: Identify the types of defects that require alarms, use actual images of the power transmission scene to mark the defects, generate annotated images, and use an artificial intelligence deep learning algorithm to generate a negative model using a sufficient number of annotated images;

[0011] Step 1-2: Use the generated negative model to detect the scene image and generate an AI detection result image, which contains the required information on each defect type;

[0012] Furthermore, step 2 specifically includes the following steps:

[0013] Step 2-1: Acquire actual information about power transmission scenarios and collect statistics on actual defect types;

[0014] Step 2-2: Construct a transmission line distance prediction mathematical model by integrating the monocular vision distance calculation principle, actual business information, and actual defect information;

[0015] Step 2-3: Calculate the distance between each defect and the transmission tower in the transmission scenario based on the transmission line distance prediction mathematical model;

[0016] Furthermore, in step 2-1, the actual information acquisition process of the power transmission scenario business is as follows: the coordinates of the transmission line equipment coordinate database host are compared to determine the front and rear towers corresponding to the equipment and their related attribute information (tower height, span, etc.), and the actual information of various defects is counted and recorded.

[0017] Furthermore, the specific steps for creating the transmission line distance prediction mathematical model in step 2-2 are as follows:

[0018] Step 2-2-1: Based on the monocular imaging principle, two sets of corresponding distance formulas are generated when the defect position changes;

[0019] H / h1=D1 / f (1)

[0020] H / h2=D2 / f (2),

[0021] Where H is the height of the defect object, h1 and h2 are the imaging heights of the defect object at different distances; D1 and D2 are the different distances at which the defect object is imaged twice, and f is the distance between the image plane and the camera at the moment of imaging;

[0022] In step 2-2-2, one of the imaging distances D1 of the defective object is selected as the range distance, and the other imaging distance D2 is recorded as the distance between the actual position of the defect and the monitoring position, and D2 = D1*h1 / h2 (3).

[0023] In step 2-2-3, based on the proportional relationship between the height of the defective object and the transmission tower on the span plane, H / H′=h1 / h′, i.e., h1=H*h′ / H′, the distance formula for defective object distance monitoring is obtained by combining formulas (1) and (2):

[0024] D2=D1*H*h′ / (h2*H′) (5)

[0025] Among them, H′ is the height of the tower, and h′ is the height of the tower in the image.

[0026] Specifically, according to the monocular imaging principle, H / h=D / f, where H is the height of the defect object, h is the height of the defect imaged in the image, D is the distance between the defect and the monitoring device, and f is the distance between the image plane and the camera at the moment of imaging. Therefore, when the defect position changes, two sets of corresponding distance formulas H / h1=D1 / f(1) and H / h2=D2 / f(2) will be generated. Combining (1)(2), the relationship between the two different defect distances and imaging heights is as follows: h2 / h1=D1 / D2, where h1 and h2 are the imaging heights of the defect object at different distances; D1 and D2 are the different distances of the defect object at two different imaging times. If D1 is the gear spacing, and D2 is the distance between the actual position of the defect and the monitoring position, then D2=D1*h1 / h2(3). The height between the defect object and the transmission tower on the span plane satisfies the mathematical formula: H / H′=h / h′, where H is the actual height of the defect object, H′ is the tower height, h is the height of the defect in the image, and h′ is the height of the tower in the image. Therefore, h=H*h′ / H′(4), where h in (4) corresponds to h1 in (3). Therefore, combining (3) and (4), the distance formula for defect object distance monitoring is D2=D1*H*h′ / (h2*H′)(5).

[0027] Furthermore, in step 2-3, the distance between each defect and the transmission tower in the transmission scenario is calculated based on the transmission line distance prediction mathematical model: according to the statistical defect type and size information, the mathematical model (5) is introduced to calculate the distance information of each defect model from the tower.

[0028] Furthermore, step 3 specifically includes the following steps:

[0029] Step 3-1: Determine the threat level of the line defects based on the distances of each transmission line defect and the transmission line defect distance warning guideline and send a warning;

[0030] Step 3-2: Combine the transmission line defect threat prediction model with the defect distance warning and diagnosis model to form an application scenario for line defect threats, and integrate and apply it in the monitoring platform.

[0031] Further, in step 3-1, the distance of each defect on the transmission line and the warning guidelines for the distance of transmission line defects: Based on the predicted defect distance D, divide the warning guidelines: when 0 < D ≤ 15m, it is classified as a high threat; when 15 < D ≤ 20m, it is classified as a medium threat; when D > 20m, it is classified as a low threat.

[0032] The present invention adopts the above technical solutions. By collecting corresponding defect samples to train the detection model, the defect positions in the image are identified, and the detection results of the artificial intelligence model are utilized; the method of monocular vision distance calculation is used for distance calculation, combined with the actual information of the transmission line scene and the actual information of the scene defects, the actual distance between the line tower and the hidden danger point of foreign objects is calculated, and early warning and alarm processing are carried out for the threatening defects by setting distance warning guidelines. The present invention is applicable to the early warning of the harm caused by external damage hidden dangers of transmission lines. By installing scene perception devices (such as video and image hosts) on line equipment, the monitoring, early warning and emergency disposal of the threat situation of the defect distance from the tower are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The following further describes the present invention in detail with reference to the drawings and specific embodiments;

[0034] Figure 1 It is a schematic diagram of the principle of the mathematical model for predicting the distance of the transmission line of the present invention;

[0035] Figure 2 It is a schematic diagram of the process for predicting the distance of transmission line defects by combining artificial intelligence and power services of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

[0037] As Figures 1 to 2 shown in one of them, the present invention discloses a method for predicting the safe distance of a transmission line by combining artificial intelligence and power services, which specifically includes the following steps:

[0038] Step 1: Use an artificial intelligence detection model to detect the scene image, identify various known types of defect hidden dangers in the scene image and mark them;

[0039] Step 2: Based on the monocular vision distance calculation algorithm and the actual information of the power transmission service, combined with the actual information of the target hidden danger, calculate the actual distance information between the hidden danger point or the target object and the transmission tower.

[0040] Step 3: Combine scene perception equipment with distance warning guidelines to provide distance-based early warning alerts and push notifications for defects.

[0041] Furthermore, the identification of various defects in step 1 includes the following steps:

[0042] Step 1-1: Identify the types of defects that require alarms, use actual images of the power transmission scene to mark the defects, generate annotated images, and use an artificial intelligence deep learning algorithm to generate a negative model using a sufficient number of annotated images;

[0043] Step 1-2: Use the generated negative model to detect the scene image and generate an AI detection result image, which contains the required information on each defect type;

[0044] Furthermore, step 2 specifically includes the following steps:

[0045] Step 2-1: Acquire actual information about power transmission scenarios and collect statistics on actual defect types;

[0046] Step 2-2: Build a transmission line distance prediction mathematical model by integrating the monocular vision distance calculation principle, actual business information, and actual defect information;

[0047] Step 2-3: Calculate the distance between each defect and the transmission tower in the transmission scenario based on the transmission line distance prediction mathematical model;

[0048] Furthermore, in step 2-1, the actual information acquisition process of the power transmission scenario business is as follows: the coordinates of the transmission line equipment coordinate database host are compared to determine the front and rear towers corresponding to the equipment and their related attribute information (tower height, span, etc.), and the actual information of various defects is counted and recorded.

[0049] Furthermore, the specific steps for creating the transmission line distance prediction mathematical model in step 2-2 are as follows:

[0050] Step 2-2-1: Based on the monocular imaging principle, two sets of corresponding distance formulas are generated when the defect position changes;

[0051] H / h1=D1 / f (1)

[0052] H / h2=D2 / f (2),

[0053] Where H is the height of the defect object, h1 and h2 are the imaging heights of the defect object at different distances; D1 and D2 are the different distances at which the defect object is imaged twice, and f is the distance between the image plane and the camera at the moment of imaging;

[0054] In step 2-2-2, one of the imaging distances D1 of the defective object is selected as the range distance, and the other imaging distance D2 is recorded as the distance between the actual position of the defect and the monitoring position, and D2 = D1*h1 / h2 (3).

[0055] In step 2-2-3, based on the proportional relationship between the height of the defective object and the transmission tower on the span plane, H / H′=h1 / h′, i.e., h1=H*h′ / H′, the distance formula for defective object distance monitoring is obtained by combining formulas (1) and (2):

[0056] D2=D1*H*h′ / (h2*H′) (5)

[0057] Among them, H′ is the height of the tower, and h′ is the height of the tower in the image.

[0058] Specifically, according to the monocular imaging principle, H / h=D / f, where H is the height of the defect object, h is the height of the defect imaged in the image, D is the distance between the defect and the monitoring device, and f is the distance between the image plane and the camera at the moment of imaging. Therefore, when the defect position changes, two sets of corresponding distance formulas H / h1=D1 / f(1) and H / h2=D2 / f(2) will be generated. Combining (1)(2), the relationship between the two different defect distances and imaging heights is as follows: h2 / h1=D1 / D2, where h1 and h2 are the imaging heights of the defect object at different distances; D1 and D2 are the different distances of the defect object at two different imaging times. If D1 is the gear spacing, and D2 is the distance between the actual position of the defect and the monitoring position, then D2=D1*h1 / h2(3). The height between the defect object and the transmission tower on the span plane satisfies the mathematical formula: H / H′=h / h′, where H is the actual height of the defect object, H′ is the tower height, h is the height of the defect in the image, and h′ is the height of the tower in the image. Therefore, h=H*h′ / H′(4), where h in (4) corresponds to h1 in (3). Therefore, combining (3) and (4), the distance formula for defect object distance monitoring is D2=D1*H*h′ / (h2*H′)(5).

[0059] Furthermore, in step 2-3, the distance between each defect and the transmission tower in the transmission scenario is calculated based on the transmission line distance prediction mathematical model: according to the statistical defect type and size information, the mathematical model (5) is introduced to calculate the distance information of each defect model from the tower.

[0060] Furthermore, step 3 specifically includes the following steps:

[0061] Step 3-1: Determine the threat level of the line defects based on the distances of each transmission line defect and the transmission line defect distance warning guideline and send a warning;

[0062] Step 3-2: Combine the transmission line defect threat prediction model with the defect distance early warning diagnosis model to form an application scenario for line defect threats and integrate and apply it in the monitoring platform.

[0063] Furthermore, for each defect distance of the transmission line in Step 3-1 and the transmission line defect distance early warning guidelines: Based on the predicted defect distance D, divide the early warning guidelines: when 0 < D ≤ 15m, it is classified as a high threat; when 15 < D ≤ 20m, it is classified as a medium threat; when D > 20m, it is classified as a low threat.

[0064] The present invention adopts the above technical solutions. By collecting corresponding defect samples to train the detection model, the defect position in the image is identified, and the detection results of the artificial intelligence model are utilized; the method of monocular vision distance calculation is used for distance calculation, combined with the actual information of the transmission line scene and the actual information of the scene defects, to calculate the actual distance between the line tower and the hidden danger point of foreign objects, and early warning and alarm processing are carried out for the threatening defects by setting distance warning guidelines. The present invention is applicable to the early warning of the hazards caused by external damage hidden dangers of transmission lines. By installing scene perception devices (such as video and image hosts) on line equipment, the monitoring, early warning and emergency disposal of the threat situation of the defect distance from the tower are realized.

[0065] Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. Generally, the components of the embodiments of the present application described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application is not intended to limit the scope of the present application claimed, but merely represents the selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

Claims

1. A transmission line safety distance prediction method combining artificial intelligence with power business, characterized by: It includes the following steps: Step 1: Use the artificial intelligence detection model to detect the scene image, identify various types of known defects and hidden dangers in the scene image, and mark them; Step 2: Based on the monocular vision distance calculation algorithm and the actual information of the power transmission business, combined with the actual information of the target hidden danger, the actual distance between the hidden danger point or target object and the transmission tower is calculated; Step 3: Combine scene perception equipment with distance warning guidelines to provide distance-based early warning alerts and push notifications for defects.

2. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 1 is characterized by: Step 1 specifically includes the following steps: Step 1-1: Identify the types of defects that require alarms, use actual images of the power transmission scene to annotate the defects and generate annotated images. Based on the number of annotated images, use an artificial intelligence deep learning algorithm to generate a negative model. Step 1-2: Use the generated negative model to detect the scene image and generate an artificial intelligence detection result image, which contains the required information on each defect type.

3. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 1 is characterized by: Step 2 specifically includes the following steps: Step 2-1: Obtain actual information about power transmission scenarios and collect statistics on actual defect types; Step 2-2: Construct a transmission line distance prediction mathematical model by integrating the monocular vision distance calculation principle, actual business information, and actual defect information; Step 2-3: Calculate the distance between each defect and the transmission tower in the transmission scenario based on the transmission line distance prediction mathematical model.

4. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 3 is characterized by: The actual information acquisition process of the power transmission scenario in step 2-1 is as follows: the coordinates of the transmission line equipment are compared with the host coordinates in the database, the front and rear towers and tower attribute information corresponding to the equipment are determined, and the actual information of various defects is counted and recorded.

5. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 3 is characterized by: The specific steps for creating the transmission line distance prediction mathematical model in step 2-2 are as follows: Step 2-2-1, based on the principle of monocular imaging, two sets of corresponding distance formulas are generated when the defect position changes; H / h1=D1 / f (1) H / h2=D2 / f (2), Where H is the height of the defect object, h1 and h2 are the imaging heights of the defect object at different distances; D1 and D2 are the different distances at which the defect object is imaged twice, and f is the distance between the image plane and the camera at the moment of imaging; Step 2-2-2, select one of the imaging distances D1 of the defect object and record it as the range distance, and the other imaging distance D2 is recorded as the distance between the actual position of the defect and the monitoring position, then D2 = D1*h1 / h2(3); In step 2-2-3, based on the proportional relationship between the height of the defective object and the transmission tower on the span plane, H / H′=h1 / h′, i.e., h1=H*h′ / H′, the distance formula for defective object distance monitoring is obtained by combining formulas (1) and (2): D2=D1*H*h′ / (h2*H′) (5) Among them, H′ is the height of the tower, and h′ is the height of the tower in the image.

6. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 5 is characterized by: In step 2-3, the statistical defect type and size information is substituted into the distance formula for defect object distance monitoring to calculate the distance information of each defect model from the tower.

7. The method for predicting safe distance of power transmission lines combining artificial intelligence with power services according to claim 1 is characterized by: Step 3 specifically includes the following steps: Step 3-1: Determine the threat level of the line defects based on the distances of each transmission line defect and the transmission line defect distance warning guideline and send a warning; Step 3-2: Combine the transmission line defect threat prediction model with the defect distance warning and diagnosis model to form an application scenario for line defect threats, and integrate and apply it in the monitoring platform.

8. The method for predicting safe distances of power transmission lines combining artificial intelligence with power services according to claim 7 is characterized by: For each defect distance of the transmission line in Step 3-1 and the transmission line defect distance warning guidelines: Based on the predicted defect distance D, divide the warning guidelines as follows: When 0 < D ≤ 15m, it is classified as a high threat; when 15 < D ≤ 20m, it is classified as a medium threat; when D > 20m, it is classified as a low threat.

Images