An intelligent path planning method for power inspection drones

Abstract

The present invention discloses a method for intelligent planning of the path of an electric power inspection drone, and relates to the technical field of intelligent path planning. The present invention comprises S10: constructing a three-dimensional inspection model of an inspection area; S20: comprehensively analyzing the inspection difficulty of the drone in each segmented inspection area; S30: preliminarily planning the inspection path of the drone in the inspection area; S40: planning and analyzing the inspection path of the drone at each damaged position; S50: predicting the real-time yaw adjustment angle of the drone; and S60: judging whether the drone is navigating according to the planned inspection path. The present invention enables the drone to preferentially perform electric power inspection operations on the segmented inspection areas with more serious damage, which is conducive to reducing the probability of safety accidents. At the same time, it can increase the inspection time of the drone on the damaged position, avoid the drone from simply inspecting the damaged position due to insufficient power, and improve the inspection effect of the drone on the electric power operation.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent path planning, and in particular to an intelligent path planning method for an electric power inspection unmanned aerial vehicle. Background Art

[0002] In the power system, the operation and maintenance of the transmission system is an important task. In traditional technology, the operation and maintenance of the power system is achieved through manual inspections, which has low maintenance efficiency. For some remote and rugged places, the safety factor is low. With the rapid development of drone technology, the use of drones instead of manual power inspections has become a trend.

[0003] In the prior art, in the process of planning the inspection path of the drone, the inspection path of the drone is determined only through the real-time inspection images collected by the drone. The damaged locations in the power system are not inspected carefully, resulting in differences between the inspection results and the actual results, reducing the system's inspection accuracy for the power system, and making it impossible to effectively analyze the return time of the drone's inspection.

[0004] Therefore, people urgently need an intelligent path planning method for power inspection drones to solve the above technical problems. Summary of the invention

[0005] The purpose of the present invention is to provide a method for intelligent path planning of a power inspection drone to solve the problems raised in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for intelligent path planning of a power inspection drone, the method comprising:

[0007] S10: according to the power inspection task of the drone, the inspection area of ​​the drone is determined, the inspection area is segmented according to the erection distribution of power towers in the inspection area and the line location information of the transmission line, and the inspection three-dimensional model of the inspection area is constructed according to the historical image information taken by the drone in the inspection area;

[0008] S20: Comprehensively analyze the inspection difficulty of the UAV in each segmented inspection area;

[0009] S30: Preliminary planning of the inspection path of the UAV within the inspection area based on the correlation coefficient between two adjacent segmented inspection areas;

[0010] S40: The UAV performs power inspection operations in the inspection area according to the initially planned inspection path, and plans and analyzes the inspection path of the UAV at each damaged location according to the flight turning angle of the UAV at each damaged location;

[0011] S50: construct a relationship model between the direction angle and the route deflection angle based on the historical navigation data of the UAV, and predict the real-time yaw adjustment angle of the UAV in combination with the acquired real-time wind direction data;

[0012] S60: judging whether the drone is navigating according to the planned inspection route according to the real-time remaining power of the drone.

[0013] Further, the S10 includes:

[0014] S101: using each power tower erected in the inspection area as a boundary mark of the inspection area, dividing the inspection area, each divided inspection area includes a power tower and a transmission line erected between two adjacent power towers;

[0015] S102: Based on the historical stereoscopic images taken by the drone in the inspection area most recently, a three-dimensional model of the inspection area is constructed. According to the maintenance and processing of each inspection target in the inspection area by the power inspection and maintenance personnel, the historical abnormal inspection images collected by the drone in each divided inspection area most recently are screened and processed. Based on the retained historical abnormal inspection images, the damage conditions of the power towers, transmission lines and insulator strings at the corresponding damaged positions are simulated in the three-dimensional model to obtain the inspection three-dimensional model of the inspection area.

[0016] Furthermore, the specific method of S20 for comprehensively analyzing the inspection difficulty of the drone in each segmented inspection area is as follows:

[0017] The split inspection areas in the inspection area are numbered, and the numbering result is: j=1, 2, ..., n, where n represents the total number of split inspection areas;

[0018] In the inspection three-dimensional model, the longest damaged length of the insulator string damaged in the split inspection area numbered j and the length L of the transmission line damaged in the split inspection area numbered j are calculated. j According to the information obtained, the inspection difficulty coefficient of the UAV on the damaged insulator string in the jth segmented inspection area is calculated. And the inspection difficulty coefficient of the UAV on the damaged transmission line in the jth segmented inspection area Perform calculations;

[0019] In the inspection 3D model, the inspection angle and inspection length of each damaged position of the power tower erected by the UAV in the segmented inspection area numbered j are obtained, and the inspection angle of the corresponding damaged position is put into the set M, M = {θ1,θ2,…,θ g}, put the inspection length of the corresponding damaged position into the set N, N = {γ1,γ2,…,γg},according to The inspection difficulty coefficient of the UAV on the damaged power tower in the jth segmented inspection area is calculated, where p = 1, 2, ..., g, represents the number of each damaged position in the power tower erected in the segmented inspection area numbered j, g represents the total number of damaged positions, and γ p represents the inspection length corresponding to the damaged position numbered p, j represents the shooting range of the drone when the drone keeps a safe distance from the transmission line, θ p Indicates the inspection angle corresponding to the damaged position numbered p, π = 180°;

[0020] according to The inspection difficulty coefficient of the UAV in the segmented inspection area numbered j is predicted, where α, β, and ε all represent proportional coefficients, and α+β+ε=1.

[0021] Further, the S20 includes:

[0022] pass The inspection difficulty coefficient of the UAV on the damaged insulator string in the jth segmented inspection area is calculated, where i = 1, 2, ..., m, representing the number value corresponding to each damaged insulator string in the jth segmented inspection area, m represents the total number of damaged insulator strings, and d i represents the longest damaged length of the damaged insulator string numbered i, and r represents d i The maximum circumference of the insulator element, e = 2.71, roundup (value, number of decimal places) means the rounding up function;

[0023] pass The inspection difficulty coefficient of the UAV on the damaged transmission line in the jth segmented inspection area is calculated.

[0024] Furthermore, the specific method of performing preliminary planning of the inspection path of the UAV in the inspection area in S30 is:

[0025] In the inspection three-dimensional model, the three-dimensional space coordinate system is constructed with the inspection starting point of the inspection area as the coordinate origin. According to the three-dimensional distance formula, the shortest distance value U between the segmented inspection area numbered j and the inspection starting point is calculated. j Perform calculations;

[0026] The correlation coefficient between two adjacent split inspection areas is predicted. Based on the prediction results, each split inspection area is classified and processed. j The inspection path of the UAV in the inspection area is preliminarily planned.

[0027] Furthermore, according to model W j→j+1=1-(|τ j -τ j+1 | / X) predicts the correlation coefficient between two adjacent segmented inspection areas, where τ j+1 represents the inspection difficulty coefficient of the UAV in the segmented inspection area numbered j+1, X=max{τ1,τ2,…,τ n}, max indicates the maximum value symbol, W j→j+1 represents the correlation coefficient between the split inspection area numbered j and the split inspection area numbered j+1;

[0028] If W j→j+1 ≥Y, then put numbers j and j+1 into set H. If W j→j+1 <Y, then put the number j into the set G, 0.4≤Y≤0.6;

[0029] In the set H, according to U j The inspection paths E1 of the UAVs for each segmented inspection area are determined in order from small to large. In the set G, according to U j The inspection path E2 of the UAV for each divided inspection area is determined in order from large to small. Based on the determination result, the inspection path of the UAV in the inspection area is preliminarily planned. The preliminarily planned inspection path of the UAV in the inspection area is: hangar location → E1 → E2.

[0030] Further, the S40 includes:

[0031] S401: In the initially planned inspection path, the drone has no inspection operations in the inspection path from the hangar position to E1 and the inspection path from E1 to E2, and the drone has inspection operations in E1 and E2;

[0032] The specific inspection path of the drone from the hangar position to E1 is: the path where the connecting line segment between the hangar position and the point in E1 closest to the hangar position is located;

[0033] The method for determining the specific inspection path of the UAV in E1 and E2 is as follows: on any side of each inspection target, the position point with the closest safe distance to each inspection target is determined, and the path of the connecting line segment formed by the determined position points is the specific inspection path of the UAV in E1 and E2;

[0034] The specific inspection path of the drone from E1 to E2 is: the path where the connecting line segment between the point farthest from the hangar in E1 and the point farthest from the hangar in E2 is located;

[0035] S402: When the drone reaches the boundary of the damaged location, the drone determines the flight turning angle of the drone according to the inspection angle corresponding to the damaged location, where the flight turning angle = the inspection angle. Based on the determination result, a new navigation track of the drone at the damaged location is determined;

[0036] S403: The newly added navigation trajectory is: starting from the boundary of the damaged position, the UAV rotates around the damaged position in the extension direction of the damaged position according to the determined flight turning angle. When the rotation angle of the UAV is the flight turning angle, the UAV returns to the starting point according to the rotation trajectory.

[0037] Further, the S50 includes:

[0038] S501: Determine the preset route of the drone based on the newly added inspection paths of the drone at each damaged location analyzed in the planning, and the specific inspection paths of the drone in the inspection area preliminarily planned;

[0039] According to the historical navigation data of the UAV, the navigation data refers to the wind direction angle f that the UAV is subjected to at time t. t , and the UAV at the wind direction angle f t The route deflection angle v generated by the action t , construct a relationship model between the wind direction angle F and the route deflection angle V, the relationship model is: F = k*V, where k represents the weight coefficient of the relationship model;

[0040] S502: Collect the real-time wind direction data at the location of the UAV, input the collected real-time wind direction data into the relationship model F=k*V, and obtain the real-time route deflection angle of the UAV. The real-time yaw adjustment angle of the UAV=F, and the real-time yaw adjustment direction of the UAV depends on the direction of the UAV nose compared to the preset route.

[0041] Furthermore, the specific method of determining whether the UAV is navigating according to the planned inspection trajectory in S60 is as follows:

[0042] The remaining power D of the drone at time t t According to the power consumption mileage curve of the drone, the remaining power of the drone is D t The corresponding cruising range is H t According to the three-dimensional distance formula, the shortest distance value E between the segmented inspection area numbered j and the hangar position is obtained. j Perform calculations;

[0043] According to the preset route of the UAV, the mileage T that the UAV has traveled when completing the power inspection task is determined. According to the execution degree zt of the inspection task by the UAV at time t, the mileage z to be inspected by the UAV is determined. t*T to obtain;

[0044] If H t -z t *Tc*maxE j > 0, the UAV will navigate according to the planned inspection path at time t. If H t -z t *Tc*maxE j ≤0, the UAV returns to the hangar at time t, where 1.2<c<1.5.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The present invention plans and analyzes the inspection path of the UAV by predicting the inspection difficulty coefficient of the UAV in each segmented inspection area, so that the UAV prioritizes the power inspection operations in the segmented inspection areas with more serious damage, which is beneficial to reducing the probability of safety accidents. At the same time, it can increase the inspection time of the UAV on the damaged location, avoid the UAV performing simple inspections on the damaged location due to insufficient power, and further improve the inspection effect of the UAV on the power operation.

[0047] 2. When inspecting the damaged location, the present invention determines the flight turning angle of the UAV at the damaged location according to the damage range of the damaged location, thereby ensuring that the UAV can effectively perform the power inspection operation at the damaged location, which is beneficial for the UAV to discover new damaged locations and further improves the power inspection accuracy of the UAV.

[0048] 3. The present invention draws a power consumption mileage curve of the UAV and the degree to which the UAV performs power inspection tasks, and conducts real-time and effective analysis of the inspection return time of the UAV, thereby avoiding the situation where the UAV cannot return due to insufficient power, and further improving the system's intelligent planning effect on the inspection path. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the workflow of an intelligent path planning method for a power inspection drone according to the present invention;

[0050] Figure 2 A schematic diagram of segmenting inspection areas for an intelligent path planning method for power inspection drones according to the present invention;

[0051] Figure 3 A schematic diagram of an insulator string of an intelligent path planning method for a power inspection drone according to the present invention;

[0052] Figure 4 A schematic diagram of an insulator element of an intelligent path planning method for a power inspection drone according to the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] like Figure 1 , Figure 2 and Figure 3 As shown, the present invention provides the following technical solution, a method for intelligently planning the path of a power inspection drone, the method comprising:

[0055] S10: according to the power inspection task of the drone, the inspection area of ​​the drone is determined, the inspection area is segmented according to the erection distribution of power towers in the inspection area and the line location information of the transmission line, and the inspection three-dimensional model of the inspection area is constructed according to the historical image information taken by the drone in the inspection area;

[0056] S10 includes:

[0057] S101: using each power tower erected in the inspection area as a boundary mark of the inspection area, dividing the inspection area, each divided inspection area includes a power tower and a transmission line erected between two adjacent power towers;

[0058] S102: Based on the historical stereoscopic images taken by the drone in the inspection area most recently, a three-dimensional model of the inspection area is constructed. The stereoscopic image refers to an image that can reflect the positional relationship and proportional size between the transmission lines, power towers and insulator strings in the inspection area. According to the maintenance and processing of each inspection target in the inspection area by the power inspection and maintenance personnel, the inspection targets include power towers, transmission lines and insulator strings. The historical abnormal inspection images collected by the drone in each divided inspection area most recently are screened and processed. If the power inspection and maintenance personnel perform maintenance on the inspection target at the corresponding damaged position, the corresponding damaged position is If the inspection target at the position is not damaged, the historical abnormal inspection images collected by the drone at the corresponding damaged position are screened out. Otherwise, the historical abnormal inspection images collected by the drone at the corresponding damaged position are retained. Based on the retained historical abnormal inspection images, the damage conditions of the power tower, transmission line and insulator string at the corresponding damaged position are simulated in the three-dimensional model to obtain the inspection three-dimensional model of the inspection area. The historical abnormal inspection images include: images of the transmission line at each damaged position collected by the drone, images of the power tower at each damaged position collected by the drone and images of the insulator string at each damaged position collected by the drone;

[0059] S20: Comprehensively analyze the inspection difficulty of the drone in each segmented inspection area. The specific method is as follows:

[0060] The split inspection areas in the inspection area are numbered, and the numbering result is: j=1, 2, ..., n, where n represents the total number of split inspection areas;

[0061] In the inspection three-dimensional model, the longest damaged length of the insulator string damaged in the split inspection area numbered j and the length L of the transmission line damaged in the split inspection area numbered j are calculated. j To obtain;

[0062] pass The inspection difficulty coefficient of the UAV on the damaged insulator string in the jth segmented inspection area is calculated, where i = 1, 2, ..., m, representing the number value corresponding to each damaged insulator string in the jth segmented inspection area, m represents the total number of damaged insulator strings, and d i represents the longest damaged length of the damaged insulator string numbered i, and r represents d i The maximum circumference of the insulator element, e = 2.71, roundup (value, number of decimal places) means the rounding up function;

[0063] pass Calculate the inspection difficulty coefficient of the UAV on the damaged power transmission line in the jth segmented inspection area;

[0064] In the inspection 3D model, the inspection angle and inspection length of each damaged position of the power tower erected by the drone in the segmented inspection area numbered j are obtained. The inspection length refers to the longest damaged length value of the damaged position. When the damaged position cannot be fully observed in a certain plane, When the damage position can be fully observed in a certain plane, the inspection angle = 0, and the inspection angle corresponding to the damage position is put into the set M, M = {θ1,θ2,…,θ g}, put the inspection length of the corresponding damaged position into the set N, N = {γ1,γ2,…,γ g},according to The inspection difficulty coefficient of the UAV on the damaged power tower in the jth segmented inspection area is calculated, where p = 1, 2, ..., g, represents the number of each damaged position in the power tower erected in the segmented inspection area numbered j, g represents the total number of damaged positions, and γ p represents the inspection length corresponding to the damaged position numbered p, h represents the shooting range of the drone when the drone keeps a safe distance from the transmission line, θ p Indicates the inspection angle corresponding to the damaged position numbered p, π = 180°;

[0065] according to The inspection difficulty coefficient of the UAV in the segmented inspection area numbered j is predicted, where α, β, and ε all represent proportional coefficients, and α+β+ε=1.

[0066] By predicting the inspection difficulty coefficient of the UAV in the segmented inspection area numbered j, the inspection path of the UAV is planned and analyzed, so that the UAV will give priority to the power inspection operation in the segmented inspection area with more serious damage, which is conducive to reducing the probability of safety accidents and improving the power inspection efficiency of the UAV;

[0067] S30: Based on the correlation coefficient between two adjacent segmented inspection areas, a preliminary inspection path of the UAV within the inspection area is planned. The specific method is as follows:

[0068] In the inspection 3D model, the 3D space coordinate system is constructed with the inspection starting point of the inspection area as the coordinate origin. The inspection starting point refers to the position of the inspection area closest to the drone hangar. According to the 3D distance formula, the shortest distance value U between the segmented inspection area numbered j and the inspection starting point is calculated. j Perform calculations;

[0069] The correlation coefficient between two adjacent split inspection areas is predicted. The specific prediction formula is:

[0070] W j→j+1 =1-(|τ j -τ j+1 | / X);

[0071] Among them, τ j+1 represents the inspection difficulty coefficient of the UAV in the segmented inspection area numbered j+1, X=max{τ1,τ2,…,τ n}, max indicates the maximum value symbol, W j→j+1 represents the correlation coefficient between the split inspection area numbered j and the split inspection area numbered j+1;

[0072] If W j→j+1 ≥Y, then put numbers j and j+1 into set H. If W j→j+1 <Y, then put the number j into the set G, 0.4≤Y≤0.6;

[0073] In the set H, according to U j The inspection paths E1 of the UAVs for each segmented inspection area are determined in order from small to large. In the set G, according to U jThe inspection paths E2 of the UAVs for each segmented inspection area are determined in descending order. Based on the determination results, the inspection paths of the UAVs in the inspection areas are preliminarily planned. The preliminarily planned inspection paths of the UAVs in the inspection areas are: hangar position → E1 → E2, where the hangar position refers to the location of the hangar compartment where the UAVs are stored;

[0074] S40: The UAV performs power inspection operations in the inspection area according to the initially planned inspection path, and plans and analyzes the inspection path of the UAV at each damaged location according to the flight turning angle of the UAV at each damaged location;

[0075] S40 includes:

[0076] S401: In the initially planned inspection path, the drone has no inspection operations in the inspection path from the hangar position to E1 and the inspection path from E1 to E2, and the drone has inspection operations in E1 and E2;

[0077] The specific inspection path of the drone from the hangar position to E1 is: the path where the connecting line segment between the hangar position and the point in E1 closest to the hangar position is located;

[0078] The method for determining the specific inspection path of the UAV in E1 and E2 is as follows: on any side of each inspection target, the position point with the closest safe distance to each inspection target is determined, and the path of the connecting line segment formed by the determined position points is the specific inspection path of the UAV in E1 and E2;

[0079] The specific inspection path of the drone from E1 to E2 is: the path where the connecting line segment between the point farthest from the hangar in E1 and the point farthest from the hangar in E2 is located;

[0080] S402: When the UAV navigates to the boundary of the damaged location, the UAV determines the flight turning angle of the UAV according to the inspection angle corresponding to the damaged location, where the flight turning angle = the inspection angle. Based on the determination result, a new navigation track of the UAV at the damaged location is determined, where the flight turning angle refers to the angle at which the UAV rotates around the damaged location;

[0081] S403: A new navigation trajectory is added: starting from the boundary of the damaged position, the UAV rotates around the damaged position in the extension direction of the damaged position according to the determined flight turning angle. When the rotation angle of the UAV is the flight turning angle, the UAV returns to the starting position according to the rotation trajectory;

[0082] In the process of determining the specific navigation trajectory of the UAV, the specific navigation trajectory of the UAV is determined according to whether the UAV needs to perform power inspection operations in the corresponding inspection section and whether the damaged location needs to be inspected, which is conducive to improving the inspection efficiency and inspection accuracy of the UAV;

[0083] S50: construct a relationship model between the direction angle and the route deflection angle based on the historical navigation data of the UAV, and predict the real-time yaw adjustment angle of the UAV in combination with the acquired real-time wind direction data;

[0084] S50 includes:

[0085] S501: Determine the preset route of the drone based on the newly added inspection paths of the drone at each damaged location analyzed in the planning, and the specific inspection paths of the drone in the inspection area preliminarily planned;

[0086] According to the historical navigation data of the UAV, the navigation data refers to the wind direction angle f that the UAV is subjected to at time t. t , and the UAV at the wind direction angle f t The route deflection angle v generated by the action t , construct a three-dimensional coordinate system with the nose vertex of the drone as the coordinate origin, take the nose direction of the drone as the positive direction of the x-axis, take the horizontal right direction of the drone body as the positive direction of the y-axis, and take the vertical downward direction of the drone body as the positive direction of the z-axis. The determined preset route is expressed in the earth coordinate system. The earth coordinate system refers to a right-handed rectangular coordinate system fixed to the earth with the center of the earth as the origin and one axis coinciding with the polar axis of the earth. If the origin of the three-dimensional coordinate system coincides with any point on the preset route, the deflection angle of the drone's route at this moment is 0. If the origin of the three-dimensional coordinate system does not coincide with any point on the preset route, the origin of the three-dimensional coordinate system constructed at the previous moment is described in the three-dimensional coordinate system constructed at this moment, and the description point is recorded as A. In the preset route, the target navigation position point B of the drone at this moment is searched. The target navigation position point refers to the point where the nose vertex of the drone coincides with the preset route at this moment when the drone is not yawed. The vector is calculated according to the three-dimensional vector angle calculation formula. With vector The angle between them is calculated, and the calculated angle is the deflection angle of the UAV's route at this moment, where O represents the origin of the three-dimensional coordinate system constructed at this moment, and the wind direction angle is the angle between the wind direction and the positive direction of the x-axis. A relationship model between the wind direction angle F and the route deflection angle V is constructed, and the relationship model is: F = k*V, where k represents the weight coefficient of the relationship model;

[0087] S502: collect the real-time wind direction data at the location of the UAV, input the collected real-time wind direction data into the relationship model F=k*V, obtain the real-time route deflection angle of the UAV, the real-time yaw adjustment angle of the UAV=F, the real-time yaw adjustment direction of the UAV depends on the direction of the UAV nose compared to the preset route, if the UAV nose is located on the right side of the preset route, the yaw adjustment direction of the UAV is to the left, if the UAV nose is located on the left side of the preset route, the yaw adjustment direction of the UAV is to the right;

[0088] By predicting the yaw adjustment direction of the UAV, it is possible to avoid the increase in the mileage of the inspection path due to yaw, which is conducive to improving the inspection efficiency of the UAV;

[0089] S60: Based on the real-time remaining power of the drone, determine whether the drone is navigating according to the planned inspection route. The specific method is as follows:

[0090] The remaining power D of the drone at time t t The power consumption mileage curve of the drone is drawn based on the historical power consumption data of the drone. The power consumption data includes the remaining power of the drone and the mileage that the drone can continue to fly with the remaining power. t The corresponding cruising range is H t According to the three-dimensional distance formula, the shortest distance value E between the segmented inspection area numbered j and the hangar position is obtained. j Perform calculations;

[0091] According to the preset route of the UAV, the mileage T that the UAV travels when completing the power inspection task is determined, and the execution degree z of the inspection task by the UAV at time t is determined. t , the mileage z of the drone to be inspected t *T to obtain;

[0092] If H t -z t *Tc*maxE j > 0, the UAV will navigate according to the planned inspection path at time t. If H t -z t *Tc*maxE j ≤0, the UAV returns to the hangar at time t, where 1.2<c<1.5.

[0093] Example 1: Assume that one insulator string is damaged in the first split inspection area, and three locations in the insulator string are damaged, and the lengths of the three damaged locations are S1, S2, and S3, respectively. Figure 3As shown, let S1 = 8 cm, S2 = 6 cm, S3 = 2 cm, then the longest damaged length of the insulator string is d1 = 8 cm, let the maximum circumference value of the insulator element where S1 is located r = 20 cm, the maximum circumference value is Figure 3 The outer diameter circumference of the middle insulator element ST, the inspection difficulty coefficient of the UAV on the damaged insulator string in the first split inspection area is:

[0094]

[0095] The inspection difficulty coefficient of the UAV on the damaged insulator string in the first divided inspection area is 0.63.

[0096] Example 2: Assume that there are 4 split inspection areas in the inspection area, τ1 = 0.1, U1 = 2 km, τ2 = 0.5, U2 = 5 km, τ3 = 0.6, U3 = 7 km, τ4 = 0.4, U4 = 10 km, Y = 0.5, then:

[0097] X=max{τ1,τ2,τ3,τ4}=τ3=0.6;

[0098] W 1→2 =1-(|τ1-τ2| / X)=0.33;

[0099] w 2→3 =1-(|τ2-τ3| / X)=0.83;

[0100] W 3→4 =1-(|τ3-τ4| / X)=0.67;

[0101] Because W 1→2 <Y,W 2→3 >Y,W 3→4＞ Y, so put number 1 into set G, and put numbers 2, 3, and 4 into set H;

[0102] In the set H, according to the arrangement order of U2=5km<U3=7km<U4=10km, the inspection path of the UAV for each segmented inspection area is determined, and the inspection path is: segmented inspection area numbered 2→segmented inspection area numbered 3→segmented inspection area numbered 4;

[0103] It can be seen that the initially planned inspection path of the UAV in the inspection area is: hangar location → split inspection area numbered 2 → split inspection area numbered 3 → split inspection area numbered 4 → split inspection area numbered 1.

[0104] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for intelligent path planning of a power inspection drone, characterized in that: The method comprises: S10: according to the power inspection task of the drone, the inspection area of ​​the drone is determined, the inspection area is segmented according to the erection distribution of power towers in the inspection area and the line location information of the transmission line, and the inspection three-dimensional model of the inspection area is constructed according to the historical image information taken by the drone in the inspection area; S20: Comprehensively analyze the inspection difficulty of the UAV in each segmented inspection area; S30: Preliminary planning of the inspection path of the UAV within the inspection area based on the correlation coefficient between two adjacent segmented inspection areas; S40: The UAV performs power inspection operations in the inspection area according to the initially planned inspection path, and plans and analyzes the inspection path of the UAV at each damaged location according to the flight turning angle of the UAV at each damaged location; S50: construct a relationship model between the direction angle and the route deflection angle based on the historical navigation data of the UAV, and predict the real-time yaw adjustment angle of the UAV in combination with the acquired real-time wind direction data; S60: judging whether the drone is navigating according to the planned inspection route according to the real-time remaining power of the drone.

2. According to claim 1, a method for intelligent path planning of a power inspection drone is characterized by: The S10 includes: S101: using each power tower erected in the inspection area as a boundary mark of the inspection area, dividing the inspection area, each divided inspection area includes a power tower and a transmission line erected between two adjacent power towers; S102: Based on the historical stereoscopic images taken by the drone in the inspection area most recently, a three-dimensional model of the inspection area is constructed. According to the maintenance and processing of each inspection target in the inspection area by the power inspection and maintenance personnel, the historical abnormal inspection images collected by the drone in each divided inspection area most recently are screened and processed. Based on the retained historical abnormal inspection images, the damage conditions of the power towers, transmission lines and insulator strings at the corresponding damaged positions are simulated in the three-dimensional model to obtain the inspection three-dimensional model of the inspection area.

3. According to claim 2, a method for intelligently planning the path of a power inspection drone is characterized by: The specific method of S20 for comprehensively analyzing the inspection difficulty of the drone in each segmented inspection area is as follows: The split inspection areas in the inspection area are numbered, and the numbering result is: j=1, 2, ..., n, where n represents the total number of split inspection areas; In the inspection three-dimensional model, the longest damaged length of the insulator string damaged in the split inspection area numbered j and the length L of the transmission line damaged in the split inspection area numbered j are calculated. j According to the information obtained, the inspection difficulty coefficient of the UAV on the damaged insulator string in the jth segmented inspection area is calculated. And the inspection difficulty coefficient of the UAV on the damaged transmission line in the jth segmented inspection area Perform calculations; In the inspection 3D model, the inspection angle and inspection length of each damaged position of the power tower erected by the UAV in the segmented inspection area numbered j are obtained, and the inspection angle of the corresponding damaged position is put into the set M, M = {θ1,θ2,…,θ g }, put the inspection length of the corresponding damaged position into the set N, N = {γ1,γ2,…,γ g },according to The inspection difficulty coefficient of the UAV on the damaged power tower in the jth segmented inspection area is calculated, where p = 1, 2, ..., g, represents the number of each damaged position in the power tower erected in the segmented inspection area numbered j, g represents the total number of damaged positions, and γ p represents the inspection length corresponding to the damaged position numbered p, h represents the shooting range of the drone when the drone keeps a safe distance from the transmission line, θ p Indicates the inspection angle corresponding to the damaged position numbered p, π = 180°; according to The inspection difficulty coefficient of the UAV in the segmented inspection area numbered j is predicted, where α, β, and ε all represent proportional coefficients, and α+β+ε=1.

4. The method for intelligent path planning of a power inspection drone according to claim 3 is characterized by: The S20 includes: pass The inspection difficulty coefficient of the UAV on the damaged insulator string in the jth segmented inspection area is calculated, where i = 1, 2, ..., m, representing the number value corresponding to each damaged insulator string in the jth segmented inspection area, m represents the total number of damaged insulator strings, and d i represents the longest damaged length of the damaged insulator string numbered i, and r represents d i The maximum circumference of the insulator element, e = 2.71, roundup (value, number of decimal places) means the rounding up function; pass The inspection difficulty coefficient of the UAV on the damaged transmission line in the jth segmented inspection area is calculated.

5. The method for intelligent path planning of a power inspection drone according to claim 4 is characterized in that: The S30 includes: In the inspection three-dimensional model, the three-dimensional space coordinate system is constructed with the inspection starting point of the inspection area as the coordinate origin. According to the three-dimensional distance formula, the shortest distance value U between the segmented inspection area numbered j and the inspection starting point is calculated. j Perform calculations; The correlation coefficient between two adjacent split inspection areas is predicted. Based on the prediction results, each split inspection area is classified and processed. j The inspection path of the UAV in the inspection area is preliminarily planned.

6. The method for intelligent path planning of a power inspection drone according to claim 5 is characterized by: The specific method of performing preliminary planning of the inspection path of the UAV in the inspection area in S30 is: According to the model W j→j+1 =1-(|τ j -τ j+1 | / X) predicts the correlation coefficient between two adjacent segmented inspection areas, where τ j+1 represents the inspection difficulty coefficient of the UAV in the segmented inspection area numbered j+1, X=max{τ1,τ2,…,τ n }, max indicates the maximum value symbol, W j→j+1 represents the correlation coefficient between the split inspection area numbered j and the split inspection area numbered j+1; If W j→j+1 ≥Y, then put numbers j and j+1 into set H. If W j→j+1 <Y, then put the number j into the set G, 0.4≤Y≤0.6; In the set H, according to U j The inspection paths E1 of the UAVs for each segmented inspection area are determined in order from small to large. In the set G, according to U j The inspection path E2 of the UAV for each divided inspection area is determined in order from large to small. Based on the determination result, the inspection path of the UAV in the inspection area is preliminarily planned. The preliminarily planned inspection path of the UAV in the inspection area is: hangar location → E1 → E2.

7. The method for intelligent path planning of a power inspection drone according to claim 6 is characterized by: The S40 includes: S401: In the initially planned inspection path, the drone has no inspection operations in the inspection path from the hangar position to E1 and the inspection path from E1 to E2, and the drone has inspection operations in E1 and E2; The specific inspection path of the drone from the hangar position to E1 is: the path where the connecting line segment between the hangar position and the point in E1 closest to the hangar position is located; The method for determining the specific inspection path of the UAV in E1 and E2 is as follows: on any side of each inspection target, the position point with the closest safe distance to each inspection target is determined, and the path of the connecting line segment formed by the determined position points is the specific inspection path of the UAV in E1 and E2; The specific inspection path of the drone from E1 to E2 is: the path where the connecting line segment between the point farthest from the hangar in E1 and the point farthest from the hangar in E2 is located; S402: When the drone reaches the boundary of the damaged location, the drone determines the flight turning angle of the drone according to the inspection angle corresponding to the damaged location, where the flight turning angle = the inspection angle. Based on the determination result, a new navigation track of the drone at the damaged location is determined; S403: The newly added navigation trajectory is: starting from the boundary of the damaged position, the UAV rotates around the damaged position in the extension direction of the damaged position according to the determined flight turning angle. When the rotation angle of the UAV is the flight turning angle, the UAV returns to the starting point according to the rotation trajectory.

8. The method for intelligent path planning of a power inspection drone according to claim 7 is characterized by: The S50 includes: S501: Determine the preset route of the drone based on the newly added inspection paths of the drone at each damaged location analyzed in the planning, and the specific inspection paths of the drone in the inspection area preliminarily planned; According to the historical navigation data of the UAV, the navigation data refers to the wind direction angle f that the UAV is subjected to at time t. t , and the UAV is at wind direction angle f t The route deflection angle v generated by the action t , construct a relationship model between the wind direction angle F and the route deflection angle V, the relationship model is: F = k*V, where k represents the weight coefficient of the relationship model; S502: Collect the real-time wind direction data at the location of the UAV, input the collected real-time wind direction data into the relationship model F=k*V, and obtain the real-time route deflection angle of the UAV. The real-time yaw adjustment angle of the UAV=F, and the real-time yaw adjustment direction of the UAV depends on the direction of the UAV nose compared to the preset route.

9. The method for intelligent path planning of a power inspection drone according to claim 6, characterized in that: The specific method of determining whether the UAV is navigating according to the planned inspection track in S60 is as follows: The remaining power D of the drone at time t t According to the power consumption mileage curve of the drone, the remaining power of the drone is D t The corresponding cruising range is H t According to the three-dimensional distance formula, the shortest distance value E between the segmented inspection area numbered j and the hangar position is obtained. j Perform calculations; According to the preset route of the UAV, the mileage T that the UAV travels when completing the power inspection task is determined, and the execution degree z of the inspection task by the UAV at time t is determined. t , the mileage z of the drone to be inspected t *T to obtain; If H t -z t *Tc*maxE j > 0, the UAV will navigate according to the planned inspection path at time t. If H t -z t *Tc*maxE j ≤0, the UAV returns to the hangar at time t, where 1.2<c<1.5.

Images