Power transmission channel hidden danger evaluation method

Abstract

The invention relates to the technical field of power transmission lines, in particular to a power transmission channel hidden danger evaluation method which comprises the steps of S10, determining the position of a power transmission channel in a task area, and detecting the surrounding environment of the power transmission channel to obtain basic information; s20, identifying a plurality of hidden danger areas according to the basic information, and obtaining hidden danger points from the hidden danger areas; s30, obtaining correlation factors of the hidden danger points according to the hidden danger areas and the corresponding basic information; s40, setting an initial weight for each hidden danger area according to the correlation factors; calculating the weight ratio of the correlation factors of the hidden danger points and scoring; s50, calculating a comprehensive risk score Rk according to scores G1-Gh of all hidden danger points in the hidden danger area; s60, grading the hidden danger area; combining the basic information of the corresponding hidden danger area and the distribution of the power transmission channels to generate a power transmission channel hidden danger evaluation report of the task period; the objective of the invention is to more accurately judge the hidden danger of the power transmission channel.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission lines, and in particular to a method for evaluating hidden dangers in power transmission channels. Background Art

[0002] In modern society, the power system is an important part of urban infrastructure, and its stable operation plays a vital role in ensuring the normal operation of society. In recent years, with the continuous adjustment of energy structure and the rapid development of the power market, the safety and stability of transmission channels have become more critical. However, it cannot be ignored that due to various factors such as changes in natural conditions and human operational errors, there are many hidden dangers in our transmission channels, and the distribution of various hidden dangers has obvious regional characteristics and environmental associations. If they are not identified and resolved in time, they will pose a threat to the normal operation of the entire power system, and thus affect the stable power supply of society. Therefore, a comprehensive and scientific assessment of the hidden dangers in the transmission channel, timely discovery of hidden dangers and formulation of targeted solutions are important means to ensure the safe operation of the transmission channel.

[0003] Existing technologies lack in-depth analysis of the correlation between potential hazards and the operating environment of transmission channels during potential hazard assessment, and are unable to effectively quantify the specific impact of potential hazards on the safety of transmission channels. Moreover, inspection plans are often based on fixed cycles or lines, and are unable to conduct focused inspections and rapid responses to high-risk potential hazard areas, affecting the efficiency and accuracy of operation and maintenance work. Summary of the invention

[0004] The present invention provides a method for evaluating hidden dangers in power transmission channels, which can effectively and comprehensively consider the hidden danger factors of power transmission channels, quantify the impact of hidden dangers on power transmission channels, improve the level of identification of hidden danger points, and improve the efficiency and accuracy of operation and maintenance work.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: The steps include: S10: at intervals of a manually set task cycle, determining the location of the power transmission channel from the task area, and detecting the surrounding environment of the power transmission channel to obtain basic information; S20: Based on the basic information, multiple potential hazard areas that affect the safe operation of the power transmission channel are identified, and potential hazard points are obtained from the potential hazard areas; S30: Obtaining a correlation factor of a hidden danger point in the hidden danger area according to the hidden danger area and its corresponding basic information; S40: according to the correlation factor of the hidden danger point, an initial weight is set for each hidden danger area; the weight ratio of the correlation factor of the hidden danger point is calculated and scored by using the principal component analysis method; The score of a potential hazard point in a potential hazard area is calculated as: Gh=W1·f1+W2·f2+……Wn·fn; Among them, Gh is the score of hidden danger point h; fn is the score of hidden danger point h on the nth correlation factor, obtained by principal component analysis; Wn is the weight ratio of the nth correlation factor of the hth hidden danger point in the hidden danger area; S50: Calculate the comprehensive risk score Rk of each hidden danger area according to the scores G1~Gh of all hidden danger points in the hidden danger area; S60: Compare the comprehensive risk score Rk with the manually set threshold, and classify the hidden danger areas corresponding to each comprehensive risk score Rk; combine the basic information of the corresponding hidden danger areas and the distribution of transmission channels to generate a transmission channel hidden danger evaluation report for this task cycle.

[0006] Furthermore, in step S10, the location of the power transmission channel is determined as follows: Establish a table to distinguish between clouds, land and water; Take aerial photos of the mission area where the power transmission channel is located at multiple times; collect surface and air information data in the mission area and process them into environmental data; Extract visible light band and infrared band from environmental data; From the environmental data, combined with the images taken at high altitude, the reflectivity of each pixel in the image in the visible light band and the temperature in the infrared band are extracted; Corresponding to the classification table, each pixel is classified into cloud layer, land, and water surface to form a classification result; In the pixel points corresponding to the land, shape feature extraction is first performed to extract the target area containing the shape target features of the tower and the conductor; Obtain the target areas of the images at multiple times, and match the target areas belonging to the same tower and conductor; Combine the target areas on the images at multiple times to perform edge detection and shadow feature extraction to specifically identify the position and shape of the tower and conductor. Then the texture features are extracted to detect whether it is a pole tower and a conductor, and determine the location of the transmission channel.

[0007] Furthermore, step S10 also includes calculating a similarity value S between every two pixels in the image, and verifying the distinction result according to the value of S: From the environmental data, the characteristic elements of each pixel point X are extracted, including the reflectivity Rr of the red light band, the reflectivity Rg of the green light band, the reflectivity Rb of the blue light band, the reflectivity R of the near infrared band, and the temperature T of the far infrared band, to form the characteristic vector of each pixel point: X1=[Rr1, Rg1, Rb1, R1, T1], X2=[Rr2, Rg2, Rb2, R2, T2], ..., Xn=[Rrn, Rgn, Rbn, Rn, Tn], where n is the total number of pixels; Use the feature vector to calculate the similarity value S between each two pixel feature vectors; If the similarity value S of two pixels is less than the artificially set threshold, the two pixels are recorded as the same category, and a total of three categories are divided; By judging whether the representative points in each category are clouds, land or water, each category is judged as clouds, land or water; thereby verifying whether it is consistent with the distinction result, and if it is consistent, proceed to the subsequent steps.

[0008] Furthermore, in step S10, the calculation process of the similarity value S is: Take pixel point Xi = [Rri, Rgi, Rbi, Ri, Ti] and pixel point Xj = [Rrj, Rgj, Rbj, Rj, Tj] as an example: Calculate first ; Xik represents one of [Rri, Rgi, Rbi, Ri, Ti], and Xjk represents one of [Rrj, Rgj, Rbj, Rj, Tj]; Then calculate the standard deviation σ of each corresponding feature between the feature vectors of two pixels k , and calculate Md=σ1+σ2+σ3+σ4+σ5; When k is 1 to 5, the standard deviation σ is calculated by taking values ​​from left to right in the eigenvector; Finally, the similarity value S=We·Ed+Wm·Md is calculated, where We and Wm are artificially set weight coefficients and are constants.

[0009] Furthermore, in step S40, the weight ratio is calculated as: Wn=An / (A1+……+An), Among them, An is the eigenvalue corresponding to the nth correlation factor obtained in the principal component analysis method; n is the total number of eigenvalues; Wn is the weight ratio of the nth correlation factor of the hth hidden danger point in the hidden danger area.

[0010] Furthermore, the comprehensive risk score Rk of each potential hazard area is calculated as: Rk=(G1+……+Gh) / mk; Among them, Gh is the score of hidden danger point h in the hidden danger area; mk is the total number of hidden danger points in the hidden danger area.

[0011] Furthermore, step S60 further includes: Set the dynamic warning threshold Tt of the comprehensive risk score Rk; The comprehensive score change rate in the dynamic monitoring potential hazard area is ΔRk=[Rk(t)-Rk(t-1)]×100% / Rk(t-1); Among them, Rk(t) is the comprehensive risk score at time t; When Rk≥Tt or ΔRk≥ΔTt, the warning signal is triggered, a warning report is generated, and auxiliary processing opinions are generated according to the warning signal level.

[0012] Furthermore, in step S60, the dynamic warning threshold Tt is set to: Tt = Tb·(1+α·Hk); Among them, Tb is the basic threshold; α is the adjustment factor, which is a man-made constant; Hk is the historical accident rate of the hidden danger area.

[0013] Furthermore, in step S60, it also includes: By prioritizing inspections of potential danger areas; Prioritize the identified hidden danger areas according to the comprehensive risk score Rk, the number of hidden danger points mk and the historical accident rate Hk; According to the priority sorting results, formulate inspection plans for each potential risk area; operation and maintenance personnel conduct inspections according to the inspection plans.

[0014] Furthermore, in step S60, it also includes: Combine the risk point scores and correlation factors to identify the key risk points that have the greatest impact on the comprehensive risk score Rk; Generate risk change trend charts for key hidden danger points separately, set higher inspection frequency and processing priority, and formulate targeted processing plans based on the characteristics of the hidden danger points.

[0015] The technical solution of the present invention can achieve the following technical effects: The present invention comprehensively analyzes the environmental data around the transmission channel, introduces the correlation factors of the hidden danger points and the principal component analysis method, scientifically calculates the weight ratio and score of the hidden danger points, and accurately identifies the hidden danger areas and hidden danger points that affect the safety of the transmission channel; and adjusts the risk assessment results in real time in combination with the dynamic early warning mechanism, so that the risk assessment results are more objective and timely; by comparing the comprehensive risk score of the hidden danger area with the preset threshold, the risk level of the hidden danger area is divided, providing an effective means for quantitatively evaluating the safety status of the hidden danger area; by prioritizing the hidden danger areas, the pertinence and efficiency of the inspection are further improved, and the operation and maintenance resource allocation can be optimized to the maximum extent, reducing the time cost of inspection and hidden danger handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of the steps of the method for evaluating hidden dangers of power transmission channels in the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] In the description of the present invention, it should be noted that the orientations or positional relationships indicated by “center”, “up”, “down”, “left”, “right”, “vertical”, “horizontal”, “inside” and “outside”, etc., are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0020] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0021] A method for evaluating hidden dangers in power transmission channels, such as Figure 1 As shown, the following steps are included: S10: at intervals of a manually set task cycle, the location of the power transmission channel is determined from within the task area, and the surrounding environment of the power transmission channel is detected to obtain basic information; S20: Based on the basic information, multiple potential hazard areas that affect the safe operation of the power transmission channel are identified, and potential hazard points are obtained from the potential hazard areas; S30: Obtaining a correlation factor of a hidden danger point in the hidden danger area according to the hidden danger area and its corresponding basic information; S40: according to the correlation factor of the hidden danger point, an initial weight is set for each hidden danger area; the weight ratio of the correlation factor of the hidden danger point is calculated and scored by using the principal component analysis method; The score of a potential hazard point in a potential hazard area is calculated as: Gh=W1·f1+W2·f2+……Wn·fn; The weight score is obtained by adding the product of the weight ratio of all correlation factors related to the potential danger point h in a potential danger area and the score to represent the impact of this potential danger point h on the transmission channel. The higher the weight score, the greater the impact. Wherein, Gh is the score of the potential danger point h; fn is the score of the potential danger point h on the nth correlation factor, which is obtained by multiplying the standardized eigenvector of the potential danger point h in the principal component analysis method with the eigenvector corresponding to the correlation factor in the covariance; Wn is the weight ratio of the nth correlation factor of the hth potential danger point in the potential danger area; S50: Calculate the comprehensive risk score Rk of each hidden danger area according to the scores G1~Gh of all hidden danger points in the hidden danger area; S60: Compare the comprehensive risk score Rk with the manually set threshold, and classify the hidden danger areas corresponding to each comprehensive risk score Rk; combine the basic information of the corresponding hidden danger areas and the distribution of transmission channels to generate a transmission channel hidden danger evaluation report for this task cycle.

[0022] The specific working principle of the present invention is as follows: By regularly using a variety of detection technologies, such as staff inspections, drones, high-altitude photography, satellite images and other methods, combined with visible light sensors, infrared sensors and microwave sensors, basic information is collected from the mission area where the transmission channel is located, including information on the terrain, landforms, vegetation, climate and other information around the transmission line; by combining image processing and big data analysis technology, by identifying high-risk areas around the transmission line (such as near tall trees, buildings or areas prone to geological disasters), potential danger points are extracted from them, including greenhouses, dust-proof nets, ground films, color steel tiles, trees and bamboo forests, ponds, construction areas, and piles under the line; the correlation factors of each potential danger point are analyzed, and the correlation factors include the type of potential danger, the size of the area, the degree of damage, the destructive intensity of the potential danger point, the distance from the potential danger to the line protection area, wind force and direction data and other factors that affect the harm of the potential danger to the transmission channel; because each potential danger has its own area, there will be various factors in this area, that is, correlation factors that affect the harm of the potential danger to the transmission channel, so the potential danger points and factors are summarized into potential danger areas; According to the correlation factor of the hidden danger point, an initial weight is set for each hidden danger area in the panoramic image; when the hidden danger point is a greenhouse, ground film or other floating objects, its hidden danger risk intensity is higher; the closer the hidden danger point is to the transmission channel, the higher its hidden danger risk intensity is; the larger the area of ​​the hidden danger point is and the higher the percentage of the damaged area is, the higher its hidden danger risk intensity is; when the hidden danger point is in the wind direction, its hidden danger risk intensity is higher; The principal component analysis method is used to calculate and score the weight ratio. Each column corresponds to a data type, and each row corresponds to a correlation factor of a hidden danger area. The covariance matrix is ​​constructed, and the eigenvalues ​​and eigenvectors of the covariance matrix are decomposed. According to the order of eigenvalue size, the eigenvalue with the largest eigenvalue is selected as the main correlation factor of the transmission channel. The weight ratio of each correlation factor is calculated based on the eigenvalue, and finally the weight score is calculated. The weight score represents the size of the factor in the risk impact of the hidden danger area on the transmission channel. The larger the weight score, the greater the impact of the factor. Finally, a comprehensive score is given to the hidden danger points in the hidden danger area, and the comprehensive risk score Rk of the area is calculated. Rk is compared with the set threshold, and the hidden danger area is divided into high-risk, medium-risk and low-risk levels. Combined with the distribution of transmission channels and hidden danger information, a hidden danger evaluation report is generated to provide a basis for inspection and treatment.

[0023] High-altitude photography (such as photography by drones, satellites or airplanes) can cover a large mission area in a short time; and high-altitude photography combined with imaging technologies such as visible light, infrared and multi-spectral can more accurately distinguish between clouds, land, water, poles, wires and other details, which is faster and more comprehensive than ground detection methods. It is especially suitable for targets such as transmission channels that are widely distributed and have long lines. It can capture images and data of the entire mission area at one time, and can easily cope with complex terrain (such as mountains, river valleys, forests, etc.), is not restricted by ground traffic conditions, and does not require manual time-consuming section-by-section inspections, which greatly improves detection efficiency. Therefore, in step S10 of this method, high-altitude photography is used to determine the location of the transmission channel. When using high-altitude photography to determine the transmission channel, it is necessary to first distinguish between clouds, land, and water, and then further determine the location of the transmission channel in the area belonging to the land, specifically: A cloud, land, and water surface distinction table is established, and the reflectivity and temperature data corresponding to the cloud, land, and water surface are stored in the distinction table.

[0024] Take aerial photos of the mission area where the power transmission channel is located at multiple times; collect surface and air information data in the mission area and process them into environmental data.

[0025] When making the distinction, we can base it on the physical properties of the clouds, such as temperature, thickness, and optical properties such as reflectivity and transmittance. Reflectivity refers to the ability of clouds to reflect sunlight, and the color brightness of the clouds can be analyzed through image analysis. Usually white indicates high reflectivity, while gray or blue indicates lower reflectivity. Transmittance refers to the degree to which light penetrates the clouds, and it can be estimated by analyzing the relationship between the intensity of light received on the ground and the transparency of the atmosphere. The identified clouds can then be further classified, for example: high clouds, including cirrus, cirrostratus, and cirrocumulus clouds; middle clouds, including high-level clouds and altocumulus clouds; low clouds, including stratus clouds and stratocumulus clouds; and straight-spreading clouds, including cumulus clouds and cumulonimbus clouds. This not only helps find the land, but also helps classify the clouds, which can help better cope with climate change.

[0026] The classification table distinguishes clouds, land, and water surface based on reflectivity and temperature; clouds have a higher reflectivity in the visible light band, while land and water have lower reflectivity; clouds have a lower temperature in the infrared band, while land and water have higher temperature; the classification table can be based on: clouds have high reflectivity and low temperature, land has moderate reflectivity and high temperature, and water has low reflectivity and high temperature.

[0027] By collecting surface and air information data within the mission area, we can conduct preliminary quality checks, eliminate abnormal or damaged data, ensure the validity of the data, and process it into environmental data.

[0028] Visible light band and infrared band are extracted from environmental data.

[0029] From the environmental data, combined with the images taken at high altitudes, the reflectivity of each pixel in the image in the visible light band and the temperature in the infrared band are extracted.

[0030] The corresponding differentiation table divides each pixel into clouds, land, and water to form a differentiation result; in the hidden danger monitoring of transmission lines, it is possible to clearly distinguish which parts are clouds rather than ground features, so as to avoid false alarms and missed alarms, and to remove the moments when clouds seriously block hidden dangers and transmission channels through screening; the identification and classification of clouds is helpful for meteorological research, especially for short-term weather forecasts and long-term climate change research; understanding the dynamic changes of clouds can assist in predicting extreme weather events that may affect the safety of transmission channels.

[0031] In the pixel points corresponding to the land, shape feature extraction is first performed to extract the target area containing the shape target features of the tower and the conductor; The target areas of the images at multiple times are obtained, and the target areas belonging to the same tower and conductor are matched.

[0032] Combining the target areas on the images at multiple times, and then performing edge detection and shadow feature extraction will be faster and reduce the amount of calculation; edge detection can identify the boundaries of objects by detecting sharp changes in pixel values ​​in the image, and shadow features use the position and shape of shadows to assist in identifying targets. Poles usually cast shadows on the ground, and by identifying these shadows, the poles can be better located; specifically identifying the position and shape of poles and conductors is a key step in identifying the contours of objects in the image.

[0033] Then the texture features are extracted to detect whether it is a pole tower and a conductor and determine the location of the transmission channel. The texture feature line calculates the grayscale co-occurrence matrix, normalizes the number of co-occurrences of specific grayscale pixels into probability, and then extracts a variety of statistics from the matrix, including energy, entropy, contrast, and homogeneity, to describe the characteristics of the texture. It can also combine local binary patterns, LBP histograms, structural tensors, etc. to extract texture features more comprehensively and meticulously, thereby improving the detection accuracy of pole towers and conductors.

[0034] Step S10 also includes calculating the similarity value S between every two pixels in the image, and verifying the distinction result according to the value of S: From the environmental data, the characteristic elements of each pixel point X are extracted, including the reflectivity Rr of the red light band, the reflectivity Rg of the green light band, the reflectivity Rb of the blue light band, the reflectivity R of the near infrared band, and the temperature T of the far infrared band, to form the characteristic vector of each pixel point: X1=[Rr1, Rg1, Rb1, R1, T1], X2=[Rr2, Rg2, Rb2, R2, T2], ..., Xn=[Rrn, Rgn, Rbn, Rn, Tn], where n is the total number of pixels; If the similarity value S of two pixels is less than the artificially set threshold, it means that the two pixels are very similar, and the two pixels are recorded as the same category. A total of three categories are divided, namely clouds, land and water surface, but it is not known what each category corresponds to, and subsequent judgment is needed.

[0035] By judging whether the representative point in each category is cloud, land or water, each category is judged as cloud, land or water; thereby verifying whether it is consistent with the distinction result, and if it is consistent, proceeding to the subsequent steps; the representative point here can be the pixel with the lowest temperature in each category, the pixel at the center of the area composed of each category, or the pixel with the highest frequency in each category.

[0036] In step S10, the calculation process of the similarity value S is: Take pixel point Xi = [Rri, Rgi, Rbi, Ri, Ti] and pixel point Xj = [Rrj, Rgj, Rbj, Rj, Tj] as an example: Calculate first ; The feature vectors of two pixels can be understood as two coordinate points in multidimensional space. First, the distance between the two pixels is calculated. The smaller the distance, the smaller the similarity value, and the more similar they are; Xik represents one of [Rri, Rgi, Rbi, Ri, Ti], and Xjk represents one of [Rrj, Rgj, Rbj, Rj, Tj]; during calculation, these two are selected correspondingly, and the same characteristic element is selected for subtraction; Then calculate the standard deviation σ of each corresponding feature between the feature vectors of two pixels k , the standard deviation σ of each corresponding feature k The difference between the two eigenvectors in each axis can be understood, and the sum of the differences in each axis Md=σ1+σ2+σ3+σ4+σ5 can be calculated; When k is 1 to 5, the standard deviation σ is calculated by taking values ​​from left to right in the eigenvector; Finally, the similarity value S=We·Ed+Wm·Md is calculated, where We and Wm are artificially set weight coefficients and are constants; by comprehensively calculating the similarity value from two aspects, the similarity value can more accurately classify the pixels into three categories. Using the feature vector composed of elements extracted from multiple bands to calculate the similarity value S between the feature vectors of each two pixels, the similarity value S can be calculated more comprehensively and accurately, reducing errors, classifying pixels more accurately, and improving the accuracy of verification.

[0037] In step S40, the weight ratio is calculated as: Wn=An / (A1+…+An), this formula ensures that the sum of all weights is 1, and can more intuitively and uniformly compare the weight ratio of each correlation factor in the hidden danger area; Among them, An is the eigenvalue corresponding to the nth correlation factor obtained in the principal component analysis method; n is the total number of eigenvalues; Wn is the weight ratio of the nth correlation factor of the hth hidden danger point in the hidden danger area.

[0038] The comprehensive risk score Rk of each potential hazard area is calculated as follows: Rk=(G1+……+Gh) / mk; Among them, Gh is the score of hidden danger point h in the hidden danger area; mk is the total number of hidden danger points in the hidden danger area; the risks of all hidden danger points in the hidden danger area are scored, and the comprehensive score Rk is used to quantify the risk situation of the entire area, avoiding scoring bias, especially when the number of hidden danger points is different, so that the risk scores of different hidden danger areas are comparable, providing a basis for subsequent hidden danger level classification and treatment; the comprehensive score is calculated by averaging to avoid misjudgment of the risk of the entire area due to the extreme value of a single hidden danger point.

[0039] Step S60 also includes: Set the dynamic warning threshold Tt of the comprehensive risk score Rk; The comprehensive score change rate in the dynamic monitoring potential hazard area is ΔRk=[Rk(t)-Rk(t-1)]×100% / Rk(t-1); Among them, Rk(t) is the comprehensive risk score at time t, that is, the comprehensive risk score of the current task cycle, and Rk(t-1) is the comprehensive risk score of the previous task cycle; When Rk≥Tt (the comprehensive risk score reaches or exceeds the threshold) or ΔRk≥ΔTt (the risk change rate reaches or exceeds the threshold), the warning signal is triggered, a warning report is generated, and auxiliary processing opinions are generated according to the warning signal level; in addition to the static threshold Tt, the score change rate ΔRk is also dynamically monitored, which can more sensitively capture risk trends and abnormal growth of risks in potential danger areas, avoid the lag caused by relying solely on a single scoring threshold, and provide support for dynamic risk management of transmission channels.

[0040] In step S60, the dynamic warning threshold Tt is set to: Tt = Tb·(1+α·Hk); Among them, Tb is the basic threshold, which can be set by referring to expert opinions and statistics of historical hidden danger events; α is the adjustment factor, which is a man-made constant; Hk is the historical accident rate of the hidden danger area, which can be obtained by dividing the total number of accidents in the hidden danger area within the set time period by the product of the time period length and the area of ​​the hidden danger area.

[0041] In step S60, the method preferably further includes: By prioritizing inspections of potential danger areas; Prioritize the identified hidden danger areas according to the comprehensive risk score Rk, the number of hidden danger points mk and the historical accident rate Hk; According to the priority sorting results, formulate inspection plans for each potential risk area; operation and maintenance personnel conduct inspections according to the inspection plans; Inspection priority sorting is to optimize the order and method of inspection resource allocation based on risk level, which is used for the execution of actual inspection plan and to guide the dispatch of inspection teams. Risk level is only a single indicator and is not equivalent to inspection priority. For example, if there are two high-risk areas, one with 10 potential hazards and the other with only 1 potential hazard, it is obvious that the area with more potential hazards needs a higher priority. Or in areas with a history of frequent accidents, even if the current risk score is low, it is necessary to increase the priority for preventive inspections. Through priority sorting, limited resources can be invested in areas with the greatest impact on overall safety, rather than simply processing them one by one according to risk level. The processing order of potential hazard areas can be adjusted in real time based on new monitoring data (such as changes in environmental conditions such as wind speed and rainfall).

[0042] In step S60, the method preferably further includes: Combine the risk point scores and correlation factors to identify the key risk points that have the greatest impact on the comprehensive risk score Rk, so as to focus on the key risk points for governance; Generate risk change trend charts for key hidden danger points separately to help determine the degree of deterioration and time nodes of hidden dangers, set higher inspection frequencies and processing priorities, and formulate targeted processing plans based on the characteristics of hidden danger points to improve resource utilization efficiency, ensure that the most urgent hidden danger points are resolved first, prevent hidden danger points from becoming key factors leading to systemic accidents, and further ensure the overall safety of transmission channels.

[0043] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for evaluating hidden dangers of power transmission channels, characterized in that: The steps include: S10: at intervals of a manually set task cycle, the location of the power transmission channel is determined from within the task area, and the surrounding environment of the power transmission channel is detected to obtain basic information; S20: Based on the basic information, multiple potential hazard areas that affect the safe operation of the power transmission channel are identified, and potential hazard points are obtained from the potential hazard areas; S30: Obtaining a correlation factor of a hidden danger point in the hidden danger area according to the hidden danger area and its corresponding basic information; S40: according to the correlation factor of the hidden danger point, an initial weight is set for each hidden danger area; the weight ratio of the correlation factor of the hidden danger point is calculated and scored by using the principal component analysis method; The score of a potential hazard point in a potential hazard area is calculated as: Gh=W1·f1+W2·f2+……Wn·fn; Among them, Gh is the score of hidden danger point h; fn is the score of hidden danger point h on the nth correlation factor, obtained by principal component analysis; Wn is the weight ratio of the nth correlation factor of the hth hidden danger point in the hidden danger area; S50: Calculate the comprehensive risk score Rk of each hidden danger area according to the scores G1~Gh of all hidden danger points in the hidden danger area; S60: Compare the comprehensive risk score Rk with the manually set threshold, and classify the hidden danger areas corresponding to each comprehensive risk score Rk; combine the basic information of the corresponding hidden danger areas and the distribution of transmission channels to generate a transmission channel hidden danger evaluation report for this task cycle.

2. The method for evaluating hidden dangers of power transmission channels according to claim 1, characterized in that: In step S10, the location of the power transmission channel is determined as follows: Establish a table to distinguish between clouds, land and water; Take aerial photos of the mission area where the power transmission channel is located at multiple times; collect surface and air information data in the mission area and process them into environmental data; Extract visible light band and infrared band from environmental data; From the environmental data, combined with the images taken at high altitude, the reflectivity of each pixel in the image in the visible light band and the temperature in the infrared band are extracted; Corresponding to the classification table, each pixel is classified into cloud, land, and water surface; In the pixel points corresponding to the land, shape feature extraction is first performed to extract the target area containing the shape target features of the tower and the conductor; Obtain the target areas of the images at multiple times, and match the target areas belonging to the same tower and conductor; Combine the target areas on the images at multiple times to perform edge detection and shadow feature extraction to specifically identify the position and shape of the tower and conductor. Then the texture features are extracted to detect whether it is a pole tower and a conductor, and determine the location of the transmission channel.

3. The method for evaluating hidden dangers of power transmission channels according to claim 2, characterized in that: Step S10 also includes calculating the similarity value S between every two pixels in the image, and verifying the distinction result according to the value of S: From the environmental data, the characteristic elements of each pixel point X are extracted, including the reflectivity Rr of the red light band, the reflectivity Rg of the green light band, the reflectivity Rb of the blue light band, the reflectivity R of the near infrared band, and the temperature T of the far infrared band, to form the characteristic vector of each pixel point: X1=[Rr1, Rg1, Rb1, R1, T1], X2=[Rr2, Rg2, Rb2, R2, T2], ……， Xn=[Rrn, Rgn, Rbn, Rn, Tn], where n is the total number of pixels; Use the feature vector to calculate the similarity value S between each two pixel feature vectors; If the similarity value S of two pixels is less than the artificially set threshold, the two pixels are recorded as the same category, and a total of three categories are divided; By judging whether the representative points in each category are clouds, land or water, each category is judged as clouds, land or water; thereby verifying whether it is consistent with the distinction result, and if so, proceed to the subsequent steps.

4. The method for evaluating hidden dangers of power transmission channels according to claim 3, characterized in that: In step S10, the calculation process of the similarity value S is: Take pixel point Xi = [Rri, Rgi, Rbi, Ri, Ti] and pixel point Xj = [Rrj, Rgj, Rbj, Rj, Tj] as an example: Calculate first ; Xik represents one of [Rri, Rgi, Rbi, Ri, Ti], and Xjk represents one of [Rrj, Rgj, Rbj, Rj, Tj]; Then calculate the standard deviation σ of each corresponding feature between the feature vectors of two pixels k , and calculate Md=σ1+σ2+σ3+σ4+σ5; When k is 1 to 5, the standard deviation σ is calculated by taking values ​​from left to right in the eigenvector; Finally, the similarity value S=We·Ed+Wm·Md is calculated, where We and Wm are artificially set weight coefficients and are constants.

5. The method for evaluating hidden dangers of power transmission channels according to claim 1, characterized in that: In step S40, the weight ratio is calculated as: Wn=An / (A1+……+An), Among them, An is the eigenvalue corresponding to the nth correlation factor obtained in the principal component analysis method; n is the total number of eigenvalues; Wn is the weight ratio of the nth correlation factor of the hth hidden danger point in the hidden danger area.

6. The method for evaluating hidden dangers of power transmission channels according to claim 1, characterized in that: The comprehensive risk score Rk of each potential hazard area is calculated as follows: Rk=(G1+……+Gh) / mk; Among them, Gh is the score of hidden danger point h in the hidden danger area; mk is the total number of hidden danger points in the hidden danger area.

7. The method for evaluating hidden dangers of power transmission channels according to claim 1, characterized in that: Step S60 also includes: Set the dynamic warning threshold Tt of the comprehensive risk score Rk; The comprehensive score change rate in the dynamic monitoring potential hazard area is ΔRk=[Rk(t)-Rk(t-1)]×100% / Rk(t-1); Among them, Rk(t) is the comprehensive risk score at time t; When Rk≥Tt or ΔRk≥ΔTt, the warning signal is triggered, a warning report is generated, and auxiliary processing opinions are generated according to the warning signal level.

8. The method for evaluating hidden dangers of power transmission channels according to claim 7, characterized in that: In step S60, the dynamic warning threshold Tt is set as: Tt = Tb·(1+α·Hk); Among them, Tb is the basic threshold; α is the adjustment factor, which is a man-made constant; Hk is the historical accident rate of the hidden danger area.

9. The method for evaluating hidden dangers of power transmission channels according to claim 8, characterized in that: In step S60, it also includes: By prioritizing inspections of potential danger areas; Prioritize the identified hidden danger areas according to the comprehensive risk score Rk, the number of hidden danger points mk and the historical accident rate Hk; According to the priority sorting results, formulate inspection plans for each potential risk area; operation and maintenance personnel conduct inspections according to the inspection plans.

10. The method for evaluating hidden dangers of power transmission channels according to claim 9, characterized in that: In step S60, it also includes: Combine the risk point scores and correlation factors to identify the key risk points that have the greatest impact on the comprehensive risk score Rk; Generate risk change trend charts for key hidden danger points separately, set higher inspection frequency and processing priority, and formulate targeted processing plans based on the characteristics of the hidden danger points.

Images