Power line inspection system and method for unmanned aerial vehicle

Abstract

The invention discloses a power line inspection system and method for an unmanned aerial vehicle. The system comprises a central control module, a data acquisition module and an inspection planning module, relates to the technical field of power line inspection, and solves the technical problem that the efficiency of inspecting a power line by an unmanned aerial vehicle in the prior art is not high. The method comprises the following steps: acquiring historical icing information of a power line; dividing the power line into a plurality of inspection sections based on the iron tower distribution information; calculating a fault risk coefficient of each inspection section based on the historical icing information and the tree feature information; obtaining a fault risk level of each inspection section based on the fault risk coefficient; and determining the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level. The fault risk coefficient of the high-risk inspection section is substituted into the formula to calculate the inspection frequency, and the inspection frequency can be dynamically adjusted along with the change of the fault risk of the inspection section, so that the inspection efficiency of the unmanned aerial vehicle on the power line can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power line inspection, and specifically relates to a power line inspection system and method for an unmanned aerial vehicle. Background Art

[0002] Electricity is the foundation of modern society. Ensuring the safe operation of the power system is crucial to economic development and social stability. Power facilities, such as transmission lines and substations, require regular inspections to prevent failures and accidents. With the rapid development of drone technology, drones have significantly improved in terms of flight stability, endurance, and payload capacity, making drones have broad application potential in the field of power inspection.

[0003] The prior art uses drones to shoot power lines to obtain inspection videos, which are then uploaded to a remote server. The remote server then performs a series of processing on the inspection videos to obtain the fault status of the power lines. However, the prior art lacks specificity in the inspection of power lines by drones, resulting in low efficiency in the inspection of power lines by drones.

[0004] The present invention provides a power line inspection system and method for an unmanned aerial vehicle to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a power line inspection system and method for unmanned aerial vehicles, which are used to solve the technical problem that the prior art unmanned aerial vehicles lack specificity in power line inspection, resulting in low efficiency in power line inspection by unmanned aerial vehicles.

[0006] To achieve the above-mentioned object, a first aspect of the present invention provides a power line inspection system for a drone, comprising: a central control module, and a data acquisition module and an inspection planning module connected thereto;

[0007] The data acquisition module is used to obtain the historical icing information of the power line and the tower distribution information; the original video of the power line is shot and pre-processed to obtain the inspection video; the historical icing information includes the annual average number of icing times and the annual average icing duration; the tower distribution information includes the total number of towers and the distance between adjacent towers;

[0008] The central control module is used to extract tree feature information from the inspection video based on the image recognition algorithm; divide the power line into several inspection sections based on the tower distribution information; calculate the fault risk coefficient of each inspection section based on the historical ice coverage information and the tree feature information; obtain the fault risk level of each inspection section based on the fault risk coefficient; wherein the fault risk level includes high risk and low risk;

[0009] The inspection planning module is used to determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level; and to plan the inspection route of the UAV based on the inspection frequency of each inspection section to obtain a plurality of inspection routes.

[0010] Preferably, the extracting tree feature information from the inspection video based on the image recognition algorithm includes:

[0011] A1: Extract inspection videos of power lines;

[0012] A2: Obtain video frames from the inspection video to obtain several inspection images;

[0013] A3: Preprocess the inspection image based on the image processing algorithm to obtain the image to be tested;

[0014] A4: Based on the image recognition algorithm, feature recognition is performed on the image to be tested to obtain tree feature information; wherein the tree feature information includes the main tree types, the average tree height and the average distance between the trees and the power lines.

[0015] Preferably, the method of dividing the power line into several inspection sections based on tower distribution information includes:

[0016] B1: Extract the total number of towers and the distance between adjacent towers from the tower distribution information to obtain the round-trip mileage of the drone;

[0017] B2: Starting from one end of the power line, select several consecutive towers and compile them into the inspection set, so that the sum of the distances between the towers in a single inspection set does not exceed the round-trip mileage of the drone;

[0018] B3: Mark the first tower in the inspection set as the starting point, mark the last tower in the inspection set as the return point, and mark the part of the power line between the starting point and the return point as the inspection section; wherein the number of inspection sets corresponds to the number of inspection sections;

[0019] B4: Mark the corresponding inspection segments as i according to the generation order of the inspection sets; wherein i=1, 2, ..., n, and n is the total number of inspection segments.

[0020] The present invention obtains the distribution information of iron towers in the power line, starts from one end of the power line, selects a number of iron towers with continuous positions in sequence and compiles them into an inspection set, and the sum of the distances between the iron towers in a single inspection set does not exceed the round-trip mileage of the unmanned aerial vehicle, which can ensure that the unmanned aerial vehicle will not run out of power when inspecting a single inspection section, which is beneficial to improving the reliability of the unmanned aerial vehicle's inspection of the power line; the power lines of the first iron tower and the last iron tower in the inspection set are marked as inspection sections, so that the power line is divided into a number of continuous inspection sections; it is convenient for the unmanned aerial vehicle to conduct key inspections on the inspection sections in the power line that are prone to failure, which is beneficial to improving the efficiency of the unmanned aerial vehicle's inspection of the power line.

[0021] It should be noted that two adjacent inspection sets are continuous, for example, the last tower in the first inspection set and the first tower in the second inspection set are the same tower.

[0022] Preferably, the calculation of the failure risk coefficient of each inspection section based on the historical ice coverage information and tree feature information includes:

[0023] C1: Extract historical ice cover information and tree feature information;

[0024] C2: Calculate the icing risk coefficient of each inspection section based on historical icing information;

[0025] C3: Calculate the tree leaning risk coefficient of each inspection section based on tree characteristic information;

[0026] C4: Calculate the fault risk coefficient GZXi of the inspection section i by the formula GZXi=α×FBXi+β×SKXi; where FBXi is the icing risk coefficient of the inspection section i, SKXi is the tree leaning risk coefficient of the inspection section i, and α and β are proportional coefficients greater than 0.

[0027] The present invention calculates the icing risk coefficient and the tree leaning risk coefficient of each inspection section respectively, and then multiplies the icing risk coefficient and the tree leaning risk coefficient by their respective proportional coefficients to obtain the fault risk coefficient of the inspection section. This comprehensively considers the influence of line icing and surrounding trees on the safe operation of the power line, facilitates the subsequent determination of the fault risk level of each inspection section according to the fault risk coefficient, and is beneficial to improving the efficiency of UAV inspection of power lines.

[0028] Preferably, the calculating of the icing risk coefficient of each inspection section based on historical icing information includes:

[0029] Extract the historical icing information of inspection section i; calculate the icing risk coefficient FBXi of inspection section i by the formula FBXi=a1×PFCi+a2×ln PFSi; where PFCi is the annual average number of icing times of inspection section i, and PFSi is the annual average icing duration of inspection section i; a1 and a2 are proportional coefficients greater than 0.

[0030] Preferably, the tree leaning risk coefficient of each inspection section is calculated based on tree characteristic information, including:

[0031] Extract the tree feature information of inspection section i; through the formula Calculate the tree leaning risk coefficient SKXi of inspection section i; where SMXi is the influence coefficient corresponding to the main tree type in inspection section i, SMHi is the average tree height of inspection section i, SMLi is the average distance between the tree and inspection section i; b1 and b2 are proportional coefficients greater than 0.

[0032] It should be noted that the tree-leaning risk factor is used to describe the risk level of contact between power lines and surrounding trees.

[0033] Preferably, the obtaining of the fault risk level of each inspection section based on the fault risk coefficient includes:

[0034] Extract the fault risk coefficient of each inspection section; determine whether the fault risk coefficient is greater than a preset risk coefficient threshold; if yes, mark the fault risk level of the corresponding inspection section as high risk; if no, mark the fault risk level of the corresponding inspection section as low risk.

[0035] Preferably, the determining the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level includes:

[0036] T1: Extract the fault risk coefficient and fault risk level of each inspection section;

[0037] T2: Determine whether the fault risk level is high risk;

[0038] If yes, the inspection frequency XJPi of inspection section i is calculated by the formula XJPi=c×GZXi; where c is an influence coefficient greater than 0;

[0039] If not, set the inspection frequency of the corresponding inspection segment to once a quarter.

[0040] It should be noted that the unit of inspection frequency is times / quarter; when the calculated inspection frequency is not an integer, the actual drone inspection frequency is obtained by rounding up.

[0041] The present invention determines the corresponding inspection frequency according to the fault risk level of the inspection section, and calculates the UAV inspection frequency by substituting the fault risk coefficient of the inspection section with a high fault risk level into the formula. The UAV inspection frequency can be dynamically adjusted as the fault risk of the inspection section changes, so that the UAV can inspect the power line more specifically, which is beneficial to improving the efficiency of the UAV inspection of the power line.

[0042] Preferably, the planning of the inspection route of the drone based on the inspection frequency of each inspection section includes:

[0043] The inspection frequency of each inspection section is extracted, and the inspection sections with equal and adjacent inspection frequencies are classified into the same inspection route; when the inspection frequency of an inspection section is not equal to the inspection frequency of an adjacent inspection section, a separate inspection route is planned.

[0044] A second aspect of the present invention provides a power line inspection method for a drone, comprising:

[0045] S1: Obtain historical ice coverage information of power lines; shoot original videos of power lines and perform preprocessing to obtain inspection videos;

[0046] S2: Extract tree feature information from inspection videos based on image recognition algorithms;

[0047] S3: Divide the power line into several inspection sections based on the tower distribution information;

[0048] S4: Calculate the failure risk coefficient of each inspection section based on historical ice coverage information and tree characteristic information;

[0049] S5: Obtaining the fault risk level of each inspection section based on the fault risk coefficient;

[0050] S6: Determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level;

[0051] S7: Planning the inspection routes of the UAV based on the inspection frequencies of each inspection section to obtain a number of inspection routes.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention obtains the distribution information of iron towers in the power line, starts from one end of the power line, selects several iron towers with continuous positions in sequence and compiles them into the inspection set, and the sum of the distances between the iron towers in a single inspection set does not exceed the round-trip mileage of the drone, which can ensure that the drone will not run out of power when inspecting a single inspection section, which is beneficial to improving the reliability of the drone's inspection of the power line; the power line of the first iron tower and the part of the last iron tower in the inspection set are marked as inspection sections, so that the power line is divided into several continuous inspection sections; it is convenient for the drone to focus on the inspection section in the power line that is prone to failure, which is beneficial to improving the efficiency of the drone's inspection of the power line.

[0054] 2. The present invention calculates the icing risk coefficient and tree leaning risk coefficient of each inspection section respectively, and then multiplies the icing risk coefficient and tree leaning risk coefficient by their respective proportional coefficients to obtain the fault risk coefficient of the inspection section. This comprehensively considers the impact of line icing and surrounding trees on the safe operation of power lines, facilitates the subsequent determination of the fault risk level of each inspection section according to the fault risk coefficient, and is conducive to improving the efficiency of drone inspections of power lines.

[0055] 3. The present invention determines the corresponding inspection frequency according to the fault risk level of the inspection section, and calculates the UAV inspection frequency by substituting the fault risk coefficient of the inspection section with a high fault risk level into the formula. The UAV inspection frequency can be dynamically adjusted as the fault risk of the inspection section changes, so that the UAV can inspect the power line more specifically, which is beneficial to improving the efficiency of the UAV inspection of the power line. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] 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 of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 It is an overall flow chart of the power line inspection method for drones in the present invention;

[0058] Figure 2 It is a schematic diagram of the principle of the power line inspection system for UAV in the present invention;

[0059] Figure 3 The flowchart of the present invention is to divide the power line into several inspection sections. DETAILED DESCRIPTION

[0060] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0061] See also Figure 1-Figure 3 , a first aspect of the present invention provides a power line inspection system for a drone, comprising: a central control module, and a data acquisition module and an inspection planning module connected thereto;

[0062] Data acquisition module: used to obtain historical icing information of power lines and tower distribution information; shoot the original video of the power lines and pre-process it to obtain the inspection video; the historical icing information includes the annual average number of icing times and the annual average icing duration; the tower distribution information includes the total number of towers and the distance between adjacent towers;

[0063] Central control module: used to extract tree feature information from inspection videos based on image recognition algorithms; divide power lines into several inspection sections based on tower distribution information; calculate the fault risk coefficient of each inspection section based on historical ice coverage information and tree feature information; obtain the fault risk level of each inspection section based on the fault risk coefficient; where the fault risk level includes high risk and low risk;

[0064] Inspection planning module: used to determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and fault risk level; plan the inspection route of the drone based on the inspection frequency of each inspection section to obtain several inspection routes.

[0065] In this embodiment, tree feature information is extracted from the inspection video based on an image recognition algorithm, including:

[0066] A1: Extract inspection videos of power lines;

[0067] A2: Obtain video frames from the inspection video to obtain several inspection images;

[0068] A3: Preprocess the inspection image based on the image processing algorithm to obtain the image to be tested;

[0069] A4: Based on the image recognition algorithm, feature recognition is performed on the image to be tested to obtain tree feature information; wherein the tree feature information includes the main tree types, the average tree height and the average distance between the trees and the power lines.

[0070] In this embodiment, the power line is divided into several inspection sections based on the tower distribution information, including:

[0071] B1: Extract the total number of towers and the distance between adjacent towers from the tower distribution information to obtain the round-trip mileage of the drone;

[0072] B2: Starting from one end of the power line, select several consecutive towers and compile them into the inspection set, so that the sum of the distances between the towers in a single inspection set does not exceed the round-trip mileage of the drone;

[0073] B3: Mark the first tower in the inspection set as the starting point, mark the last tower in the inspection set as the return point, and mark the part of the power line between the starting point and the return point as the inspection section; wherein the number of inspection sets corresponds to the number of inspection sections;

[0074] B4: Mark the corresponding inspection segments as i according to the generation order of the inspection sets; wherein i=1, 2, ..., n, and n is the total number of inspection segments.

[0075] For example, assume that the total length of the power line is 50 kilometers, the total number of towers is 320, and the round-trip mileage of the drone is 8 kilometers; set the distance between the first 25 consecutive towers to 7.8 kilometers, and the distance between the first 26 consecutive towers to 8.3 kilometers, and add the first 25 consecutive towers to the inspection set; mark the first tower in the inspection set as the starting point, mark the last tower in the inspection set as the return point, and mark the part of the power line between the starting point and the return point as inspection section 1.

[0076] The present invention obtains the distribution information of iron towers in the power line, starts from one end of the power line, selects a number of iron towers with continuous positions in sequence and compiles them into an inspection set, and the sum of the distances between the iron towers in a single inspection set does not exceed the round-trip mileage of the unmanned aerial vehicle, which can ensure that the unmanned aerial vehicle will not run out of power when inspecting a single inspection section, which is beneficial to improving the reliability of the unmanned aerial vehicle's inspection of the power line; the power lines of the first iron tower and the last iron tower in the inspection set are marked as inspection sections, so that the power line is divided into a number of continuous inspection sections; it is convenient for the unmanned aerial vehicle to conduct key inspections on the inspection sections in the power line that are prone to failure, which is beneficial to improving the efficiency of the unmanned aerial vehicle's inspection of the power line.

[0077] It should be noted that two adjacent inspection sets are continuous. For example, the last tower in the first inspection set and the first tower in the second inspection set are the same tower, thereby ensuring that some towers will not be missed when the drone inspects the power lines.

[0078] In this embodiment, the failure risk coefficient of each inspection section is calculated based on the historical ice coverage information and tree feature information, including:

[0079] C1: Extract historical ice cover information and tree feature information;

[0080] C2: Calculate the icing risk coefficient of each inspection section based on historical icing information;

[0081] C3: Calculate the tree leaning risk coefficient of each inspection section based on tree characteristic information;

[0082] C4: The fault risk coefficient GZXi of the inspection section i is calculated by the formula GZXi=α×FBXi+β×SKXi; wherein FBXi is the icing risk coefficient of the inspection section i, SKXi is the tree leaning risk coefficient of the inspection section i, α and β are proportional coefficients greater than 0, and the specific values ​​of α and β are determined according to the experience of experts in related fields.

[0083] Exemplarily, the proportional coefficients α=1.2 and β=1.5 are set; the icing risk coefficient FBX1 of inspection section 1 is set to 52.44, and the tree leaning risk coefficient SKX1 of inspection section i is set to 48.09; and the failure risk coefficient GZX1=135.063 of inspection section 1 is calculated by the formula.

[0084] The present invention calculates the icing risk coefficient and the tree leaning risk coefficient of each inspection section respectively, and then multiplies the icing risk coefficient and the tree leaning risk coefficient by their respective proportional coefficients to obtain the fault risk coefficient of the inspection section. This comprehensively considers the influence of line icing and surrounding trees on the safe operation of the power line, facilitates the subsequent determination of the fault risk level of each inspection section according to the fault risk coefficient, and is beneficial to improving the efficiency of UAV inspection of power lines.

[0085] In this embodiment, the icing risk coefficient of each inspection section is calculated based on historical icing information, including:

[0086] Extract the historical icing information of inspection section i; calculate the icing risk coefficient FBXi of inspection section i by the formula FBXi=a1×PFCi+a2×ln PFSi; where PFCi is the annual average number of icing times of inspection section i, and PFSi is the annual average icing duration of inspection section i; a1 and a2 are proportional coefficients greater than 0.

[0087] Exemplarily, the proportional coefficients a1=2 and a2=8 are set; the annual average number of icing times of inspection section 1 PFC1=12 times, and the annual average icing duration of inspection section 1 PFS1=35 days; the icing risk coefficient FBX1≈52.44 of inspection section 1 is calculated by the formula.

[0088] It should be noted that the specific values ​​of the proportional coefficients a1 and a2 are determined by the geographical region. For example, the values ​​of a1 and a2 may be appropriately increased in high-latitude regions.

[0089] In this embodiment, the tree leaning risk coefficient of each inspection section is calculated based on the tree feature information, including:

[0090] Extract the tree feature information of inspection section i; through the formula Calculate the tree leaning risk coefficient SKXi of inspection section i; where SMXi is the influence coefficient corresponding to the main tree type in inspection section i, SMHi is the average tree height of inspection section i, SMLi is the average distance between the tree and inspection section i; b1 and b2 are proportional coefficients greater than 0.

[0091] Exemplarily, the proportional coefficients b1=15 and b2=0.1 are set; the influence coefficient corresponding to the main tree types in inspection section 1 is SMX1=2, the average tree height of inspection section 1 is SMH1=7.6m, and the average distance between trees and inspection section 1 is SML1=9.4m; the tree leaning risk coefficient of inspection section 1 is calculated by the formula as SKX1≈48.09.

[0092] It should be noted that the tree-leaning risk coefficient is used to describe the risk level of mutual contact between power lines and surrounding trees. The larger the tree-leaning risk coefficient, the greater the risk level of mutual contact between power lines and surrounding trees. The influence coefficient corresponding to the main tree types on the inspection section can be appropriately increased for trees with faster growth rates.

[0093] In this embodiment, the fault risk level of each inspection section is obtained based on the fault risk coefficient, including:

[0094] Extract the fault risk coefficient of each inspection section; determine whether the fault risk coefficient is greater than a preset risk coefficient threshold; if yes, mark the fault risk level of the corresponding inspection section as high risk; if no, mark the fault risk level of the corresponding inspection section as low risk.

[0095] Exemplarily, the fault risk coefficient GZX1 of inspection section 1 is set to 135.063, and the preset risk coefficient threshold is 100; since the fault risk coefficient GZX1 = 135.063 is greater than the preset risk coefficient threshold, the fault risk level of inspection section 1 is marked as high risk.

[0096] In this embodiment, the inspection frequency corresponding to each inspection section is determined based on the fault risk coefficient and the fault risk level, including:

[0097] T1: Extract the fault risk coefficient and fault risk level of each inspection section;

[0098] T2: Determine whether the fault risk level is high risk;

[0099] If yes, the inspection frequency XJPi of inspection section i is calculated by the formula XJPi=c×GZXi; where c is an influence coefficient greater than 0;

[0100] If not, set the inspection frequency of the corresponding inspection segment to once a quarter.

[0101] It should be noted that when the calculated inspection frequency is not an integer, the actual drone inspection frequency is obtained by rounding up.

[0102] Exemplarily, c=0.1 is set; the fault risk level of inspection section 1 is marked as high risk, and the fault risk coefficient of inspection section 1 GZX1=135.063; the inspection frequency XJP1=13.5063 of inspection section 1 is calculated by the formula, and the actual drone inspection frequency is rounded up to 14 times / quarter.

[0103] The present invention determines the corresponding inspection frequency according to the fault risk level of the inspection section, and calculates the UAV inspection frequency by substituting the fault risk coefficient of the inspection section with a high fault risk level into the formula. The UAV inspection frequency can be dynamically adjusted as the fault risk of the inspection section changes, so that the UAV can inspect the power line more specifically, which is beneficial to improving the efficiency of the UAV inspection of the power line.

[0104] In this embodiment, the inspection route of the drone is planned based on the inspection frequency of each inspection section, including:

[0105] The inspection frequency of each inspection section is extracted, and the inspection sections with equal and adjacent inspection frequencies are classified into the same inspection route; when the inspection frequency of an inspection section is not equal to the inspection frequency of an adjacent inspection section, a separate inspection route is planned.

[0106] A second aspect of the present invention provides a power line inspection method for a drone, comprising:

[0107] S1: Obtain historical ice coverage information of power lines; shoot original videos of power lines and perform preprocessing to obtain inspection videos;

[0108] S2: Extract tree feature information from inspection videos based on image recognition algorithms;

[0109] S3: Divide the power line into several inspection sections based on the tower distribution information;

[0110] S4: Calculate the failure risk coefficient of each inspection section based on historical ice coverage information and tree characteristic information;

[0111] S5: Obtaining the fault risk level of each inspection section based on the fault risk coefficient;

[0112] S6: Determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level;

[0113] S7: Planning the inspection routes of the UAV based on the inspection frequencies of each inspection section to obtain a number of inspection routes.

[0114] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0115] Working principle of the present invention:

[0116] The present invention obtains historical icing information of power lines; shoots original videos of power lines and performs preprocessing to obtain inspection videos; extracts tree feature information from the inspection videos based on an image recognition algorithm; divides the power lines into a number of inspection sections based on tower distribution information; calculates the fault risk coefficient of each inspection section based on historical icing information and tree feature information; obtains the fault risk level of each inspection section based on the fault risk coefficient; determines the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level; and plans the inspection routes of drones based on the inspection frequencies of each inspection section to obtain a number of inspection routes.

[0117] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A power line inspection system for drones, comprising: The central control module, and the data acquisition module and inspection planning module connected thereto are characterized in that: The data acquisition module is used to obtain historical icing information and tower distribution information of power lines; shoot the original video of the power line and perform preprocessing to obtain the inspection video; wherein the historical icing information includes the annual average number of icing times and the annual average icing duration; the tower distribution information includes the total number of towers and the distance between adjacent towers; The central control module is used to extract tree feature information from the inspection video based on the image recognition algorithm; divide the power line into several inspection sections based on the tower distribution information; calculate the fault risk coefficient of each inspection section based on the historical ice coverage information and the tree feature information; obtain the fault risk level of each inspection section based on the fault risk coefficient; wherein the fault risk level includes high risk and low risk; The inspection planning module is used to determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level; and to plan the inspection route of the UAV based on the inspection frequency of each inspection section to obtain a plurality of inspection routes.

2. A power line inspection system for drones according to claim 1, characterized in that: The method of extracting tree feature information from inspection videos based on an image recognition algorithm includes: A1: Extract inspection videos of power lines; A2: Obtain video frames from the inspection video to obtain several inspection images; A3: Preprocess the inspection image based on the image processing algorithm to obtain the image to be tested; A4: Based on the image recognition algorithm, feature recognition is performed on the image to be tested to obtain tree feature information; wherein the tree feature information includes the main tree types, the average tree height and the average distance between the trees and the power lines.

3. The power line inspection system for unmanned aerial vehicles according to claim 1, characterized in that: The power line is divided into several inspection sections based on the tower distribution information, including: B1: Extract the total number of towers and the distance between adjacent towers from the tower distribution information to obtain the round-trip mileage of the drone; B2: Starting from one end of the power line, select several consecutive towers and compile them into the inspection set, so that the sum of the distances between the towers in a single inspection set does not exceed the round-trip mileage of the drone; B3: Mark the first tower in the inspection set as the starting point, mark the last tower in the inspection set as the return point, and mark the part of the power line between the starting point and the return point as the inspection section; wherein the number of inspection sets corresponds to the number of inspection sections; B4: Mark the corresponding inspection segments as i according to the generation order of the inspection sets; wherein i=1, 2, ..., n, and n is the total number of inspection segments.

4. The power line inspection system for unmanned aerial vehicles according to claim 3, characterized in that: The calculation of the failure risk coefficient of each inspection section based on the historical ice coverage information and tree characteristic information includes: C1: Extract historical ice cover information and tree feature information; C2: Calculate the icing risk coefficient of each inspection section based on historical icing information; C3: Calculate the tree leaning risk coefficient of each inspection section based on tree characteristic information; C4: The fault risk coefficient of each inspection section is obtained by adding the products of the icing risk coefficient and the tree leaning risk coefficient multiplied by their respective proportional coefficients.

5. The power line inspection system for unmanned aerial vehicles according to claim 4, characterized in that: The calculation of the icing risk coefficient of each inspection section based on historical icing information includes: Extract the historical icing information of inspection section i; calculate the icing risk coefficient FBXi of inspection section i by the formula FBXi=a1×PFSi+a2×ln PFSi; where PFCi is the annual average number of icing times of inspection section i, PFSi is the annual average icing duration of inspection section i; a1 and a2 are proportional coefficients greater than 0.

6. The power line inspection system for unmanned aerial vehicles according to claim 4, characterized in that: The calculation of the tree leaning risk coefficient of each inspection section based on the tree characteristic information includes: Extract the tree feature information of inspection section i; through the formula Calculate the tree leaning risk coefficient SKXi of inspection section i; where SMXi is the influence coefficient corresponding to the main tree type in inspection section i, SMHi is the average tree height of inspection section i, SMLi is the average distance between the tree and inspection section i; b1 and b2 are proportional coefficients greater than 0.

7. The power line inspection system for unmanned aerial vehicles according to claim 1, characterized in that: The obtaining of the fault risk level of each inspection section based on the fault risk coefficient includes: Extract the fault risk coefficient of each inspection section; determine whether the fault risk coefficient is greater than a preset risk coefficient threshold; if yes, mark the fault risk level of the corresponding inspection section as high risk; if no, mark the fault risk level of the corresponding inspection section as low risk.

8. The power line inspection system for unmanned aerial vehicles according to claim 4, characterized in that: The determining of the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level includes: T1: Extract the fault risk coefficient and fault risk level of each inspection section; T2: Determine whether the fault risk level is high risk; If yes, then the inspection frequency corresponding to the inspection section is obtained by calculating the product of the fault risk coefficient and the influence coefficient; If not, set the inspection frequency of the corresponding inspection segment to once a quarter.

9. The power line inspection system for unmanned aerial vehicles according to claim 1, characterized in that: The inspection route of the drone is planned based on the inspection frequency of each inspection section, including: The inspection frequency of each inspection section is extracted, and the inspection sections with equal and adjacent inspection frequencies are classified into the same inspection route; when the inspection frequency of an inspection section is not equal to the inspection frequency of an adjacent inspection section, a separate inspection route is planned.

10. A power line inspection method for drones, based on the operation of a power line inspection system for drones according to any one of claims 1 to 9, characterized in that: include: S1: Obtain historical ice coverage information of power lines; shoot original videos of power lines and perform preprocessing to obtain inspection videos; S2: Extract tree feature information from inspection videos based on image recognition algorithms; S3: Divide the power line into several inspection sections based on the tower distribution information; S4: Calculate the failure risk coefficient of each inspection section based on historical ice coverage information and tree characteristic information; S5: Obtaining the fault risk level of each inspection section based on the fault risk coefficient; S6: Determine the inspection frequency corresponding to each inspection section based on the fault risk coefficient and the fault risk level; S7: Planning the inspection routes of the UAV based on the inspection frequencies of each inspection section to obtain a number of inspection routes.

Images