CN115454132A - Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station - Google Patents
Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station Download PDFInfo
- Publication number
- CN115454132A CN115454132A CN202211210434.5A CN202211210434A CN115454132A CN 115454132 A CN115454132 A CN 115454132A CN 202211210434 A CN202211210434 A CN 202211210434A CN 115454132 A CN115454132 A CN 115454132A
- Authority
- CN
- China
- Prior art keywords
- path
- unmanned aerial
- aerial vehicle
- target
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000009194 climbing Effects 0.000 claims abstract description 11
- 239000003016 pheromone Substances 0.000 claims description 35
- 230000005611 electricity Effects 0.000 claims description 30
- 241000257303 Hymenoptera Species 0.000 claims description 29
- 230000007704 transition Effects 0.000 claims description 21
- 230000011218 segmentation Effects 0.000 claims description 7
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 5
- 238000005192 partition Methods 0.000 claims description 4
- 229910052739 hydrogen Inorganic materials 0.000 claims description 3
- 239000001257 hydrogen Substances 0.000 claims description 3
- 125000004435 hydrogen atom Chemical class [H]* 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 abstract description 19
- 230000001788 irregular Effects 0.000 abstract description 5
- 238000011960 computer-aided design Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 238000000638 solvent extraction Methods 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The unmanned aerial vehicle routing inspection three-dimensional path planning system and method for the mountain photovoltaic power station solve the problem that the mountain photovoltaic power station is poor in shooting quality due to irregular distribution, large distribution area and fluctuation of terrain in an array area of the mountain photovoltaic power station. On the basis of information of a plant area CAD graph and a digital elevation model map DEM, a mountain photovoltaic power station is efficiently clustered and partitioned by improving a density clustering algorithm, the improved clustering algorithm is more prone to bringing latitude and altitude approaching photovoltaic array areas into clusters, turning of an unmanned aerial vehicle is reduced, climbing times are increased, the unmanned aerial vehicle flies along a straight line as far as possible at full speed, a PT camera model of the unmanned aerial vehicle is built, when three-dimensional path planning is carried out, the distance between the camera and the ground is ensured to be constant, the angle is perpendicular to a photovoltaic panel, a power consumption model of the unmanned aerial vehicle is built, the three-dimensional path planning is carried out on the basis of the optimal power consumption path rule of an improved ant colony, and the single routing inspection efficiency of the unmanned aerial vehicle is effectively improved.
Description
Technical Field
The invention relates to the technical field of new energy power generation, in particular to a mountain photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method and system.
Background
As a big energy consumption country, the energy consumption structure of China is accelerating to change to clean and low carbon. With the increasing number of photovoltaic power stations, land resources with flat terrain and better construction conditions are gradually reduced, land which depends on inferior construction conditions of lakes, mountains and the like gradually becomes the land construction resources of the photovoltaic power stations, particularly the construction of the photovoltaic power stations in the mountains is gradually increased in recent years, and the land construction resources become important photovoltaic power station construction resources.
However, most mountain photovoltaic power stations are located in mountainous areas, the terrain structures are complex, the array areas are scattered, the block shapes are different, and meanwhile, the slope and height of the mountain are greatly changed. Consequently in the aspect of photovoltaic power plant operation and maintenance patrols and examines, compare in the manual work and patrol and examine, use unmanned aerial vehicle to patrol and examine and be an outstanding alternative, the infrared image of all subassemblies is found out the trouble that exists in the power station through the power station that the analysis unmanned aerial vehicle shot to operation and maintenance personnel.
Under the condition that the cruising ability of the unmanned aerial vehicle and the resolution of the carried infrared camera are determined, the planning of the routing inspection path is the most important factor influencing the automatic routing inspection efficiency of the unmanned aerial vehicle. Suitable flight path not only can save unmanned aerial vehicle's flight time, can reduce the higher action of power consumption such as turn, acceleration and deceleration simultaneously, improves the single flight volume of patrolling and examining, reduces the flight number of times.
At present, most path planning systems are designed for flat ground photovoltaic power stations, and based on the characteristics of high component density, uniform distribution and very regular block shapes, the unmanned aerial vehicle can adopt a simple coverage type to patrol and examine to shoot comprehensively. But for mountain photovoltaic power stations, a targeted unmanned aerial vehicle routing inspection path planning algorithm is also lacked. If a large amount of open spaces are shot by adopting simple coverage type inspection, the flight time is wasted, and simultaneously, due to the large number of components, the complexity of various commonly used problem algorithms of the travelers is high.
Disclosure of Invention
In order to solve the problems, the invention provides a mountain land photovoltaic power station-oriented unmanned aerial vehicle routing inspection three-dimensional path planning method and system, which can realize unmanned aerial vehicle three-dimensional routing inspection path planning for mountain land photovoltaic power stations with scattered array area distribution and different block shapes.
The invention provides a mountain photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method, which comprises the following steps:
s1, inputting design drawing information of factory planning of a mountain photovoltaic power station in a system, and acquiring position information and elevation information of n array area array areas of the photovoltaic power station;
s2, dividing the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm;
s3, based on the optimal power consumption path rule of the improved ant colony, path planning in each area is carried out; and outputting the three-dimensional track of the unmanned aerial vehicle, wherein n is larger than 1, and m is larger than 1.
Preferably, step S2 uses an improved density clustering algorithm to divide the photovoltaic power plant into m blocks suitable for unmanned aerial vehicle inspection; also comprises the following steps;
s21, obtaining related parameters, wherein the related parameters comprise a field parameter epsilon, minPts
Sample set D = { p = { (p) 1 ,p 2 ,…,p n For any one sample point p, where ε field is defined as:
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
sample set D = { p = { (p) 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel;
where dist (p, q) denotes the sample point p (x) p ,y p ,z p ) And q (x) q ,y q ,z q ) The distance between
For a distance dist (p, q) between arbitrary sample points p and q the formula is
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Wherein x, y and z respectively represent three-dimensional coordinates of the photovoltaic panels, and subscripts are used for distinguishing the two photovoltaic panels. Randomly selecting two photovoltaic panels, and respectively expressing a distance formula between the two photovoltaic panels by using p and q; p is more than or equal to 1, q is more than or equal to 1;
initializing a set of core objectsInitializing cluster number k =0, initializing unvisited sample set Γ = D, clustering
S22: traversing all sample points in the sample set which is not visited to obtain the epsilon field;
if N ε (p) i ) If the point is more than or equal to MinPts, the point is classified into a core object set;
wherein N is ε (p i ) Represents p i The number of sample points contained in the epsilon field of (2), and MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = { o }, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = { o }, update unvisited sample set Γ = Γ - { o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After generation, update cluster partition C = { C = { C = } 1 ,C 2 ,…,C n H, updating the core object set omega to omega-C k Step S23 is executed. Otherwise, updating the core object set omega to omega-C k (ii) a Wherein k is greater than or equal to 1;
s26: and taking out one core object from the target cluster core object queue, updating a target cluster sample set, updating an unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
Preferably, the step S3 performs path planning inside each area based on the optimal power consumption path rule for improving ant colony, and further includes the following steps:
s31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions;
s32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; and obtaining an aerial route point set according to photogrammetric constraint conditions, namely that the vertical distance between the camera and the shot slope is constant, and the shooting direction of the camera is vertical to the shot slope.
Step S31, establishing a camera model, and obtaining an aerial route point set according to a photogrammetry constraint condition, wherein the aerial route point set is marked as N; further comprising the steps of:
s311: fitting the discrete digital elevation model by using a B spline surface to obtain a fitted surface S formula (3);
wherein, B ij Is a control point set which is a position coordinate of the photovoltaic panel; n is a radical of hydrogen i,p (x) And N j,q (y) is a B-spline surface basis function;
s312: establishing a camera model, defining the constant offset distance d between the position of each aerial route point and the target surface, respectively recording the horizontal rotation angle and the vertical rotation angle of the PT camera as P and T, and recording the horizontal overlapping rate and the vertical overlapping rate as sigma 0 And σ 1 ;
The waypoints are obtained as follows:
the camera rotation angles are as follows:
wherein the content of the first and second substances,representing the coordinates of the three-dimensional waypoints in the air,andrespectively representing coordinates and normal vectors on the ground curved surface which are in one-to-one correspondence with the aerial three-dimensional waypoints; i =1,2,3, \ 8230;, N 0 ,j=1,2,3,…,N 1 ;T i,j Indicating the vertical angle of rotation P of the camera i,j (ii) a Representing the horizontal rotation angle of the camera;
N 0 and N 1 The following conditions are satisfied:
l and W respectively represent the length and width of the projection on the ground taken by the camera, N 0 、N 1 Is an intermediate parameter defined to simplify the expression.
Preferably, step S32: performing path planning inside each area based on the optimal power consumption path rule of the improved ant colony; obtaining an aerial route point set for photogrammetry constraint conditions, namely that the vertical distance between a camera and a shot slope is constant, and the shooting direction of the camera is vertical to the shot slope; also comprises the following steps:
s321: initializing ant colony algorithm parameters alpha and beta, wherein alpha and beta respectively represent the size of heuristic factors and the size of expected heuristic factors, M represents the number of colonies, rho represents an information volatilization factor, and setting an upper limit of iteration times;
s322: obtaining the transfer probability of ants from the target alpha, beta, node i to node j at the moment tSelecting a next target point by adopting a roulette mode and moving the target point; s323: judging whether the target ants traverse all paths or not;
selecting a next target point and moving the target point by adopting a roulette mode as shown in the following formula;
wherein j ∈ allowed k Indicating that the next target point j belongs to the set of waypoints allowed to arrive;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to arrive in the formula (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of the path points allowed to reach, the transition probability is equal to 0;
wherein, the first and the second end of the pipe are connected with each other,heuristic information indicative of the path transitions,indicating the concentration of pheromone on the path from node i to node j at time t,the amount of information arriving at the target point s from point i is shown,friend arriving at the target point s from point i with a desired value, allowed k Representing a set of waypoints allowed to be reached; the path traversed by the ant is the waypoint generated in step S312
Preferably, step S32: performing path planning inside each area based on the optimal power consumption path rule of the improved ant colony; obtaining an aerial route point set for photogrammetry constraint conditions, namely that the vertical distance between a camera and a shot slope is constant, and the shooting direction of the camera is vertical to the shot slope; also comprises the following steps:
s323: judging whether the target ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal or not;
s325: acquiring the electricity consumption cost required under all paths, and selecting the path with the optimal electricity consumption cost as the current optimal path;
the electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively the power consumption coefficient of the unmanned aerial vehicle during horizontal flight, climbing flight and landing flight, d x Is the lateral distance of the target viewpoint from the next viewpoint, d y Is the longitudinal distance of the target viewpoint from the next viewpoint, d z The vertical height distance from the target viewpoint to the next viewpoint; selecting a path with the optimal electricity consumption cost as a current optimal path; the minimum electricity consumption cost is optimal;
s326: updating the concentration of pheromones on the path;
wherein, in the formula (10), τ ij (t + n) represents the pheromone concentration on paths i to j after the time t + n is updated;
representing the total amount of pheromones left on paths between nodes after all ants traverse the paths after one round of iteration;
ρ is a volatilization coefficient of pheromone concentration;
and adding 1 to the iteration times, judging whether all search rounds are completed or not, and outputting a three-dimensional optimal path of the unmanned aerial vehicle, otherwise, returning to S322, wherein the output optimal path is the three-dimensional trajectory path of the unmanned aerial vehicle with high pheromone concentration after iteration.
The invention provides a planning method for routing inspection three-dimensional paths of unmanned aerial vehicles for mountain photovoltaic power stations, which is used for obtaining three-dimensional routing inspection paths of unmanned aerial vehicles aiming at the characteristics of scattered distribution and different block shapes of mountain photovoltaic power station array areas on the basis of information of a plant area CAD (computer-aided design) diagram and a Digital Elevation Model (DEM). By improving a density clustering algorithm, the mountain land photovoltaic power station is subjected to efficient clustering and partitioning, and the characteristics that the mountain land photovoltaic power station array area is irregular in distribution, large in distribution area and limited in unmanned aerial vehicle operation time are effectively solved; the improved clustering algorithm is more prone to bringing the photovoltaic array area with the latitude and the elevation close to the cluster, and fully considers the characteristic that the photovoltaic panels are arranged in the east-west direction, so that the turning and climbing times of the unmanned aerial vehicle are reduced, the unmanned aerial vehicle flies at full speed along a straight line as far as possible, and the inspection efficiency is improved. Through establishing unmanned aerial vehicle PT camera model, when carrying out three-dimensional path planning, ensure that camera and ground distance are invariable, the angle is perpendicular with the photovoltaic panel, has effectively solved mountain region photovoltaic power plant because the not good condition of shooting quality that relief caused. By establishing the power consumption model of the unmanned aerial vehicle and based on the optimal power consumption path rule of the improved ant colony, three-dimensional path planning is carried out, and the single-time routing inspection efficiency of the unmanned aerial vehicle is effectively improved.
The invention also provides a mountain land photovoltaic power station-oriented unmanned aerial vehicle routing inspection three-dimensional path planning system, which comprises a photovoltaic power station information acquisition module, an area division module and a path planning module;
the system comprises a photovoltaic power station information acquisition module, a power distribution module and a power distribution module, wherein the photovoltaic power station information acquisition module is used for acquiring design drawing information of factory planning of a mountain photovoltaic power station and acquiring position information and elevation information of n array area array areas of the photovoltaic power station;
the region segmentation module is used for segmenting the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm;
the route planning module is used for planning the route in each area by improving the optimal power consumption route rule of the ant colony; wherein n is greater than 1 and m is greater than 1.
The region segmentation module is used for segmenting the photovoltaic power station into m unmanned aerial vehicle patrol blocks by using an improved density clustering algorithm, and is obtained by the following steps:
s21, obtaining related parameters, wherein the related parameters comprise a field parameter epsilon, minPts and a sample set D = { p = 1 ,p 2 ,…,p n For any one sample point p, where ε field is defined as:
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
sample set D = { p = { (p) 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel;
where dist (p, q) denotes the sample point p (x) p ,y p ,z p ) And q (x) q ,y q ,z q ) The distance between
For a distance dist (p, q) between arbitrary sample points p and q the formula is
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Wherein x, y and z respectively represent three-dimensional coordinates of the photovoltaic panels, and subscripts are used for distinguishing the two photovoltaic panels. Randomly selecting two photovoltaic panels, and respectively expressing a distance formula between the two photovoltaic panels by using p and q; p is more than or equal to 1, q is more than or equal to 1;
initializing a set of core objectsInitializing cluster number k =0, initializing unvisited sample set Γ = D, clustering
S22: traversing all sample points in the sample set which is not visited to obtain the epsilon field;
if N is present ε (p i ) If the point is more than or equal to MinPts, the point is classified into a core object set;
wherein, N ε (p i ) Denotes p i The number of sample points contained in the epsilon field of (a), and MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
for any sample point p its epsilon field is denoted as N ε (p i );
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = o, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = o, update unvisited sample set Γ = Γ - { o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After generation, update cluster partition C = { C = { C = } 1 ,C 2 ,...,C n H, updating the core object set omega to C k Step S23 is executed. Otherwise, updating the core object set omega to omega-C k (ii) a Wherein k is greater than or equal to 1;
s26: and taking out one core object from the target cluster core object queue, updating the target cluster sample set, updating the unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
The route planning module is used for planning the route in each area by improving the optimal power consumption route rule of the ant colony and is obtained through the following steps;
s31: establishing a camera model, and obtaining an aerial route point set according to a photogrammetry constraint condition, wherein the aerial route point set is marked as N;
s311: fitting the discrete digital elevation model by using a B spline surface to obtain a fitted surface S formula (3);
wherein, B ij Is a control point set which is a position coordinate of the photovoltaic panel; n is a radical of i,p (x) And N j,q (y) is a B-spline surface basis function;
s312: establishing a camera model, defining the constant offset distance d between the position of each aerial route point and the target surface, respectively recording the horizontal rotation angle and the vertical rotation angle of the PT camera as P and T, and recording the horizontal overlapping rate and the vertical overlapping rate as sigma 0 And σ 1 ;
Waypoints were obtained as follows:
the camera rotation angles are as follows:
wherein, the first and the second end of the pipe are connected with each other,representing the coordinates of three-dimensional waypoints in the air,andrespectively representing coordinates and normal vectors on the ground surface which are in one-to-one correspondence with the aerial three-dimensional waypoints; i =1,2,3, \8230;, N0, j =1,2,3, \8230;, N 1 ;T i,j Indicating the vertical rotation angle of the camera; p is i,j Indicating the horizontal rotation angle of the camera;
N 0 and N 1 The following conditions are satisfied:
s32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; and obtaining an aerial route point set according to photogrammetric constraint conditions, namely that the vertical distance between the camera and the shot slope is constant, and the shooting direction of the camera is vertical to the shot slope. The method comprises the following steps of;
the method comprises the following steps of obtaining an air route point set:
s321: initializing ant colony algorithm parameters alpha and beta, wherein alpha and beta respectively represent the size of heuristic factors and the size of expected heuristic factors, M represents the number of colonies, rho represents an information volatilization factor, and setting iterationThe upper limit of the generation number; s322: obtaining the transition probability of ants from the target alpha, the target beta, the node i to the node j at the moment tSelecting a next target point by adopting a roulette mode and moving the target point; s323: judging whether the target ants traverse all paths or not;
selecting a next target point and moving the target point by adopting a roulette mode;
wherein j ∈ allowed k Indicating that the next target point j belongs to the set of allowed path points;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to be reached in the equation (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of the path points allowed to reach, the transition probability is equal to 0;
wherein, the first and the second end of the pipe are connected with each other,heuristic information indicative of the path transitions,indicating the concentration of pheromones on the path from node i to node j at time t,representing the amount of information arriving at the target point s from point i,indicating the arrival at the target point s of the desired value from the point i,allowed k Representing a set of waypoints allowed to be reached;
s323: judging whether the target ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal or not;
s325: acquiring the electricity consumption cost required under all paths, and selecting the path with the optimal electricity consumption cost as the current optimal path;
the electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively the power consumption coefficient of the unmanned aerial vehicle during horizontal flight, climbing flight and landing flight, d x Is the lateral distance of the target viewpoint from the next viewpoint, d y Is the longitudinal distance of the target viewpoint from the next viewpoint, d z The vertical height distance from the target viewpoint to the next viewpoint; selecting a path with the optimal electricity consumption cost as a current optimal path; the minimum electricity consumption cost is optimal;
s326: updating the concentration of pheromones on the path;
expressed by the formula (10), the pheromone concentrations on paths i to j after t + n are updated;
friend all ants leave the total amount of pheromones on the paths between nodes after traversing the paths after one round of iteration; ρ is a volatilization coefficient of pheromone concentration;
and adding 1 to the iteration times, judging whether all search rounds are completed or not, and outputting a three-dimensional optimal path of the unmanned aerial vehicle, otherwise, returning to S322, wherein the output optimal path is the three-dimensional trajectory path of the unmanned aerial vehicle with high pheromone concentration after iteration.
The unmanned aerial vehicle inspection three-dimensional path planning system comprises a photovoltaic power station information acquisition module, an area division module and a path planning module, acquires information based on a factory area CAD graph and a digital elevation model DEM (digital elevation model) graph, and obtains unmanned aerial vehicle three-dimensional inspection paths aiming at the characteristics of scattered distribution and different block shapes of an array area of a mountain photovoltaic power station. By improving a density clustering algorithm, the mountain land photovoltaic power station is subjected to efficient clustering and partitioning, and the characteristics that the mountain land photovoltaic power station array area is irregular in distribution, large in distribution area and limited in unmanned aerial vehicle operation time are effectively solved; the improved clustering algorithm is more prone to bringing the photovoltaic array area with the latitude and the elevation close into a cluster, and fully considers the characteristic that the photovoltaic panels are arranged in the east-west direction, so that the turning and climbing times of the unmanned aerial vehicle are reduced, the unmanned aerial vehicle flies at full speed along a straight line as far as possible, and the inspection efficiency is improved. Through establishing unmanned aerial vehicle PT camera model, when carrying out three-dimensional path planning, ensure that camera and ground distance are invariable, the angle is perpendicular with the photovoltaic panel, has effectively solved mountain region photovoltaic power plant because the not good condition of shooting quality that relief caused. By establishing the power consumption model of the unmanned aerial vehicle and based on the optimal power consumption path rule of the improved ant colony, three-dimensional path planning is carried out, and the single-time routing inspection efficiency of the unmanned aerial vehicle is effectively improved.
Drawings
FIG. 1 is a flow chart of a mountain photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method;
FIG. 2 is a diagram of a model of an unmanned aerial vehicle carrying camera for the unmanned aerial vehicle inspection three-dimensional path planning method for the mountain photovoltaic power station;
FIG. 3 is a schematic diagram of route points of an unmanned aerial vehicle inspection three-dimensional path planning method for a mountain photovoltaic power station;
FIG. 4 is a block diagram of a mountain-land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method;
fig. 5 is a schematic diagram of a mountain-land-oriented unmanned aerial vehicle inspection three-dimensional path planning method for a photovoltaic power station, which performs path planning for each block.
Detailed Description
The invention will be described in further detail below with reference to the drawings and specific examples, which are intended to be illustrative and not to be construed as limiting the invention.
The embodiment of the invention provides a mountain land photovoltaic power station-oriented unmanned aerial vehicle routing inspection three-dimensional path planning method, which specifically comprises the following steps of:
s1: inputting a CAD (computer-aided design) drawing of a factory planning of a mountain photovoltaic power station in a system, and acquiring position information and elevation information of each array area of the photovoltaic power station;
s2: dividing the power station into a plurality of blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm; as shown in fig. 4.
S3: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; as shown in fig. 5.
In one embodiment, the power station is divided into a plurality of blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm, and the method comprises the following steps:
s21, obtaining related parameters, wherein the related parameters comprise the field parameters epsilon =16 and MinPts =5, and the specific numerical values can be adjusted according to the actual situation. Sample set D = { p = 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel; wherein, the first and the second end of the pipe are connected with each other,
for any sample point, the distance dist (p, q) between p and q is defined as
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Wherein x, y and z respectively represent three-dimensional coordinates of the photovoltaic panels, and subscripts are used for distinguishing the two photovoltaic panels. Randomly selecting two photovoltaic panels, and respectively expressing a distance formula between the two photovoltaic panels by using p and q; p is more than or equal to 1, q is more than or equal to 1;
x, y and z respectively represent coordinate neighborhoods of each photovoltaic panel and refer to areas around one point, and the epsilon area of any sample point p is defined as
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
Wherein N is ε (p i ) Represents p i The epsilon field of (c) contains the number of sample points,
MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
the epsilon field for any sample point p is denoted as N ε (pi);
Defining a set of core objectsInitializing cluster number k =0, initializing unaccessed sample set Γ = D, clustering
S22: and traversing all sample points in the unvisited sample set to obtain the epsilon area. If N is present ε (p i ) If the point is more than or equal to MinPts, the point is classified into a core object set;
s23: if the core object set is an empty set, ending the algorithm, otherwise executing S24;
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = o, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = { o }, update unvisited sample set Γ = Γ - { o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After the generation is finished, updating cluster division C = { C = { C = 1 ,C 2 ,...,C n H, updating the core object set omega to omega-C k Step S23 is executed. Otherwise, updating the core object set omega to omega-C k ;
S26: and taking out one core object from the target cluster core object queue, updating the target cluster sample set, updating the unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
In one embodiment, planning the path inside each region includes the following steps:
s31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions; as shown in fig. 3.
S32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony;
for photogrammetric constraint conditions, namely that the vertical distance between a camera and a shot slope is constant, and the shooting direction of the camera is vertical to the shot slope, obtaining an air route point set, and specifically comprising the following steps:
s311: fitting the discrete DEM model by using a B-spline surface, and obtaining a fitted surface S according to the following formula:
wherein, B ij Is a set of control points, N i,p (u) and N j,q (v) B spline surface basis function;
and (3) taking a 3-order B spline basis function, namely p = q =3, substituting the 3-order B spline basis function into a position coordinate xy of the photovoltaic panel, and respectively recurrently obtaining:
S312: referring to fig. 2, a camera model is established. Recording the constant offset distance between the position of each air route point and the target surface as d, P represents the horizontal rotation angle of the camera and T represents the vertical rotation angle, and the horizontal overlapping rate and the vertical overlapping rate are respectively recorded as sigma 0 And σ 1 . The waypoint and the camera rotation angle can be obtained by:
wherein the content of the first and second substances,representing the coordinates of the three-dimensional waypoints in the air,andand respectively representing the coordinates on the ground surface and the normal vector thereof, which are in one-to-one correspondence with the aerial three-dimensional waypoints. i =1,2,3, ·, N 0 ,j=1,2,3,··,N 1 。N 0 And N 1 The following conditions are satisfied:
l and W respectively represent the length and width of the projection on the ground photographed by the camera, N 0 、N 1 Is an intermediate parameter defined to simplify the expression.
Based on the optimal power consumption path rule of the improved ant colony, path planning in each area is carried out, and the specific steps are as follows:
s321: and initializing ant colony algorithm related parameters. Alpha and beta respectively represent the size of a heuristic factor and the size of an expected heuristic factor, M represents a group number, rho represents an information volatilization factor, and an upper limit of iteration times is set;
s322: obtaining the transition probability of ants from the current node i to the node j at the moment tAs shown in the following formula, and select the next target point and move it by roulette.
Wherein j ∈ allowed k Indicating that the next target point j belongs to the set of waypoints allowed to arrive;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to arrive in the formula (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of the path points allowed to reach, the transition probability is equal to 0;
wherein, the first and the second end of the pipe are connected with each other,heuristic information indicative of the path transitions,indicating the concentration of pheromone on the path from node i to node j at time t,representing the amount of information arriving at the target point s from point i,indicating the arrival of the desired value, allowed, at the target point s from point i k Representing the set of allowed waypoints, the path traversed by the ant being the waypoint generated in step 312
S323: judging whether the current ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal;
s325: and acquiring the electricity consumption cost required under the path, and selecting the path with the optimal electricity consumption cost as the current optimal path. The electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively the power consumption coefficient of the unmanned aerial vehicle during horizontal flight, climbing flight and landing flight, d x Is the lateral distance of the current viewpoint from the next viewpoint, d y Is the longitudinal distance from the current viewpoint to the next viewpoint, d z Is the vertical height distance of the front viewpoint from the next viewpoint.
Selecting a path with the optimal electricity consumption cost as a current optimal path; the electricity consumption cost is optimal at a minimum.
S326: updating pheromone concentration on the path:
wherein, in the formula (10), τ ij (t + n) represents the pheromone concentration on paths i to j after the time t + n is updated;
friend all ants traverse the path after one round of iteration and then leave the total amount of pheromones on the paths between the nodes;
ρ represents a volatilization coefficient of pheromone concentration;
and adding 1 to the iteration times, judging whether all search rounds are finished, and outputting the optimal path, otherwise, returning to the S322.
The invention provides a mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method which is used for obtaining unmanned aerial vehicle three-dimensional inspection paths aiming at the characteristics of scattered distribution and different block shapes of mountain land photovoltaic power station array areas on the basis of information of a factory area CAD graph and a Digital Elevation Model (DEM). By improving a density clustering algorithm, the mountain land photovoltaic power station is efficiently clustered and partitioned, and the characteristics that the mountain land photovoltaic power station array area is irregular in distribution, large in distribution area and limited in unmanned aerial vehicle operation time are effectively solved; the improved clustering algorithm is more prone to bringing the photovoltaic array area with the latitude and the elevation close into a cluster, and fully considers the characteristic that the photovoltaic panels are arranged in the east-west direction, so that the turning and climbing times of the unmanned aerial vehicle are reduced, the unmanned aerial vehicle flies at full speed along a straight line as far as possible, and the inspection efficiency is improved. Through establishing unmanned aerial vehicle PT camera model, when carrying out three-dimensional path planning, ensure that camera and ground distance are invariable, the angle is perpendicular with the photovoltaic panel, has effectively solved mountain region photovoltaic power plant because the not good condition of shooting quality that relief caused. By establishing the power consumption model of the unmanned aerial vehicle and based on the optimal power consumption path rule of the improved ant colony, three-dimensional path planning is carried out, and the single routing inspection efficiency of the unmanned aerial vehicle is effectively improved.
The embodiment of the invention also provides a mountain land photovoltaic power station-oriented unmanned aerial vehicle routing inspection three-dimensional path planning system, which comprises a photovoltaic power station information acquisition module, an area division module and a path planning module;
the system comprises a photovoltaic power station information acquisition module, a power distribution module and a power distribution module, wherein the photovoltaic power station information acquisition module is used for acquiring design drawing information of mountain photovoltaic power station factory planning and acquiring position information and elevation information of n array area array areas of the photovoltaic power station;
the region segmentation module is used for segmenting the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm;
the route planning module is used for planning the route in each area by improving the optimal power consumption route rule of the ant colony; wherein n is greater than 1 and m is greater than 1.
In one embodiment, the region segmentation module, which segments the photovoltaic power station into m unmanned aerial vehicle patrol blocks by using an improved density clustering algorithm, is obtained by the following steps:
s21, obtaining related parameters, wherein the related parameters comprise a field parameter epsilon, minPts and a sample set D = { p = 1 ,p 2 ,…,p n For any one sample point p, where ε field is defined as:
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
sample set D = { p = { (p) 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel;
where dist (p, q) denotes the sample point p (x) p ,y p ,z p ) And q (x) q ,y q ,z q ) The distance between
For a distance dist (p, q) between arbitrary sample points p and q the formula is
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Wherein x, y and z respectively represent three-dimensional coordinates of the photovoltaic panels, and subscripts are used for distinguishing the two photovoltaic panels. Randomly selecting two photovoltaic panels, and respectively representing a distance formula between the two photovoltaic panels by p and q; p is more than or equal to 1, q is more than or equal to 1:
initializing a set of core objectsInitializing cluster number k =0, initializing unaccessed sample set Γ = D, clustering
S22: traversing all sample points in the sample set which is not visited to obtain the epsilon field;
if N is present ε (p i ) If the point is more than or equal to MinPts, the point is classified into a core object set;
wherein N is ε (p i ) Represents p i The epsilon field of (c) contains the number of sample points,
MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
the epsilon field for any sample point p is denoted as N ε (p i );
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = o, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = { o }, update unvisited sample set Γ = Γ - (o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After the generation is finished, updating cluster division C = { C = { C = 1 ,C 2 ,…,C n H, updating the core object set omega to omega-C k Step S23 is executed. Otherwise, updating the core object set omega to omega-C k ;
S26: and taking out one core object from the target cluster core object queue, updating the target cluster sample set, updating the unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
In one embodiment, the path planning module is configured to perform path planning inside each area by improving an optimal power consumption path rule of an ant colony, and the path planning module is obtained through the following steps;
s31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions;
s311: fitting the discrete digital elevation model by using a B spline surface to obtain a fitted surface S formula (3);
wherein, B ij Is a control point set which is a position coordinate of the photovoltaic panel; n is a radical of hydrogen i,p (x) And N j,q (y) is a B-spline surface basis function;
s312: establishing a camera model, defining the constant offset distance d between the position of each aerial route point and the target surface, and defining the horizontal rotation angle P and the vertical rotation angle T of the camera, and recording the horizontal overlapping rate and the vertical overlapping rate as sigma 0 And σ 1 ;
Waypoints were obtained as follows:
the camera rotation angles are as follows:
wherein, the first and the second end of the pipe are connected with each other,representing the coordinates of three-dimensional waypoints in the air,andrespectively representing coordinates and normal vectors on the ground curved surface which are in one-to-one correspondence with the aerial three-dimensional waypoints; i =1,2,3, \ 8230;, N 0 ,j=1,2,3,…,N 1 ;T i,j Indicating the vertical rotation angle of the camera; p i,j Indicating the horizontal rotation angle of the camera;
N 0 and N 1 The following conditions are satisfied:
s32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; and obtaining an aerial route point set for photogrammetry constraint conditions, namely that the vertical distance between the camera and the shot slope is constant, and the shooting direction of the camera is vertical to the shot slope. The method comprises the following steps;
the method comprises the following steps of obtaining an air route point set:
s321: initializing ant colony algorithm parameters alpha and beta, wherein alpha and beta respectively represent the size of a heuristic factor and the size of an expected heuristic factor, M represents the number of colonies, rho represents an information volatilization factor, and an upper limit of iteration times is set; s322: obtaining the transition probability of ants from the target alpha, the target beta, the node i to the node j at the moment tSelecting a next target point by adopting a roulette mode and moving the target point; s323: judging whether the target ants traverse all paths or not;
selecting a next target point and moving the target point by adopting a roulette mode as shown in the following formula;
wherein j ∈ allowed k Indicating that the next target point j belongs to the set of allowed path points;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to be reached in the equation (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of the path points allowed to reach, the transition probability is equal to 0;
heuristic information indicative of the transition of the path,indicating the concentration of pheromones on the path from node i to node j at time t,representing the amount of information arriving at the target point s from point i,indicating the arrival of the desired value, allowed, at the target point s from point i k Representing a set of waypoints allowed to be reached;
s323: judging whether the target ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal;
s325: acquiring the electricity consumption cost required under all paths, and selecting the path with the optimal electricity consumption cost as the current optimal path;
the electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively the power consumption coefficient of the unmanned aerial vehicle during horizontal flight, climbing flight and landing flight, d x Is the lateral distance of the target viewpoint from the next viewpoint, d y Is the longitudinal distance of the target viewpoint from the next viewpoint, d z The vertical height distance from the target viewpoint to the next viewpoint; selecting a path with the optimal electricity consumption cost as a current optimal path; the minimum electricity consumption cost is optimal;
s326: updating the concentration of pheromones on the path;
expressed by the formula (10), the pheromone concentrations on paths i to j after t + n are updated;
representing the total amount of pheromones left on paths between nodes after all ants traverse the paths after one round of iteration; ρ is a volatilization coefficient of pheromone concentration;
adding 1 to the iteration times, judging whether all search rounds are finished or not, and outputting an optimal path, otherwise, returning to the S322; and the output optimal path is a path with high pheromone concentration after iteration is selected.
The unmanned aerial vehicle routing inspection three-dimensional path planning system comprises a photovoltaic power station information acquisition module, an area segmentation module and a path planning module, and acquires information based on a plant area CAD (computer-aided design) diagram and a digital elevation model DEM (digital elevation model) diagram to obtain unmanned aerial vehicle three-dimensional routing inspection paths aiming at the characteristics of scattered distribution and various block shapes of an array area of the mountain photovoltaic power station. By improving a density clustering algorithm, the mountain land photovoltaic power station is subjected to efficient clustering and partitioning, and the characteristics that the mountain land photovoltaic power station array area is irregular in distribution, large in distribution area and limited in unmanned aerial vehicle operation time are effectively solved; the improved clustering algorithm is more prone to bringing the photovoltaic array area with the latitude and the elevation close into a cluster, and fully considers the characteristic that the photovoltaic panels are arranged in the east-west direction, so that the turning and climbing times of the unmanned aerial vehicle are reduced, the unmanned aerial vehicle flies at full speed along a straight line as far as possible, and the inspection efficiency is improved. Through establishing unmanned aerial vehicle PT camera model, when carrying out three-dimensional path planning, ensure that camera and ground distance are invariable, the angle is perpendicular with the photovoltaic panel, has effectively solved mountain region photovoltaic power plant because the not good condition of shooting quality that relief caused. By establishing the power consumption model of the unmanned aerial vehicle and based on the optimal power consumption path rule of the improved ant colony, three-dimensional path planning is carried out, and the single routing inspection efficiency of the unmanned aerial vehicle is effectively improved.
It should be noted that those skilled in the art can change, modify, replace and modify the above embodiments without departing from the principle and spirit of the invention, and still fall into the protection scope of the invention.
Claims (10)
1. An unmanned aerial vehicle inspection three-dimensional path planning method for a mountain land photovoltaic power station is characterized by comprising the following steps of,
s1, inputting design drawing information planned in a mountain region photovoltaic power station factory area in a system, and acquiring position information and elevation information of n array areas of the photovoltaic power station;
s2, dividing the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm;
s3, planning paths in each region based on the optimal power consumption path rule of the improved ant colony; outputting a three-dimensional track of the unmanned aerial vehicle; wherein n is greater than 1 and m is greater than 1.
2. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method according to claim 1, wherein S2 is used for dividing the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm; also comprises the following steps;
s21, obtaining related parameters, wherein the related parameters comprise a field parameter epsilon, minPts,
ε represents the size of the region taken, minPts represents the number of sample points in the core object set that contain the least, and sample set D = { p = 1 ,p 2 ,…,p n For any one sample point p, where ε field is defined as:
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
sample set D = { p = 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel;
where dist (p, q) denotes the sample point p (x) p ,y p ,z p ) And q (x) q ,y q ,z q ) The distance between
For a distance dist (p, q) between arbitrary sample points p and q the formula is
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Wherein x, y and z respectively represent three-dimensional coordinates of the photovoltaic panels, and subscripts are used for distinguishing the two photovoltaic panels. Randomly selecting two photovoltaic panels, and respectively representing a distance formula between the two photovoltaic panels by p and q; p is more than or equal to 1, q is more than or equal to 1;
initializing a set of core objectsInitializing cluster number k =0, initializing unvisited sample set Γ = D, clustering
S22: traversing all sample points in the sample set which is not visited to obtain the epsilon field;
if N is present ε (p i )≥MinPts,
Wherein N is ε (p i ) Denotes p i The epsilon field of (c) contains the number of sample points,
MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
the epsilon field for any sample point p is denoted as N ε (p i );
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = o, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = { o }, update unvisited sample set Γ = Γ - { o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After generation, update cluster partition C = { C = { C = } 1 ,C 2 ,…,C n H, updating the core object set omega to C k Executing the step S23, otherwise, updating the core object set omega to omega-C k (ii) a Wherein k is greater than or equal to 1;
s26: and taking out one core object from the target cluster core object queue, updating the target cluster sample set, updating the unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
3. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method according to claim 1, wherein S3 performs path planning inside each area based on an ant colony-improving optimal power consumption path rule, and further comprises the following steps:
s31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions;
s32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; and for the photogrammetry constraint condition, the vertical distance between the camera and the shot slope is constant, and the shooting direction of the camera is vertical to the shot slope, so that the aerial route point set is obtained.
4. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method according to claim 3, wherein the step S31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions; further comprising the steps of:
s311: fitting the discrete digital elevation model by using a B spline surface to obtain a fitted curved surface S (x, y) formula (3);
wherein, B ij Is a control point set which is a position coordinate of the photovoltaic panel; n is a radical of hydrogen i,p (x) And N j,q (y) is a B-spline surface basis function;
s312: establishing a camera model, defining the constant offset distance d between the position of each aerial route point and the target surface, the horizontal rotation angle P and the vertical rotation angle T of the camera, and the horizontal overlapping rate sigma 0 And vertical overlap ratio σ 1 ;
The waypoints are obtained as follows:
the camera horizontal rotation angle is as follows:
the vertical rotation angle of the camera is as follows:
wherein the content of the first and second substances,representing the coordinates of the three-dimensional waypoints in the air,andrespectively representing coordinates and normal vectors on the ground curved surface which are in one-to-one correspondence with the aerial three-dimensional waypoints;
I=1,2,3,…,N 0 ,j=1,2,3,…,N 1 ;
T i,j indicating the vertical rotation angle of the camera;
P i,j representing the horizontal rotation angle of the camera;
N 0 and N 1 The following conditions are satisfied:
l and W respectively represent the length and width of the projection on the ground taken by the camera, N 0 、N 1 Is an intermediate parameter for simplifying the definition of the expression.
5. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method according to claim 3, wherein S32: performing path planning inside each area based on the optimal power consumption path rule of the improved ant colony; obtaining an aerial route point set for photogrammetry constraint conditions, namely that the vertical distance between a camera and a shot slope is constant, and the shooting direction of the camera is vertical to the shot slope; also comprises the following steps:
s321: initializing ant colony algorithm parameters alpha and beta, wherein alpha and beta respectively represent the size of heuristic factors and the size of expected heuristic factors, M represents the number of colonies, rho represents an information volatilization factor, and setting an upper limit of iteration times;
s322: obtaining the transfer probability of ants from the target alpha, beta, node i to node j at the moment tSelecting a next target point by adopting a roulette mode and moving the target point; s323: judging whether the target ants traverse all paths or not;
selecting a next target point and moving the target point by adopting a roulette mode as shown in the following formula;
wherein j ∈ allowed k Indicating that the next target point j belongs to the set of allowed path points;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to be reached in the equation (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of path points allowed to reach, the transition probability is equal to 0;
wherein the content of the first and second substances,heuristic information indicative of the path transitions,indicating the concentration of pheromone on the path from node i to node j at time t,representing the amount of information arriving at the target point s from point i,indicating the arrival of the desired value, allowed, at the target point s from point i k Representing a set of waypoints allowed to be reached; the path traversed by the ant is the waypoint generated in step 312
6. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning method according to claim 5, wherein S32: performing path planning inside each area based on the optimal power consumption path rule of the improved ant colony; obtaining an aerial route point set for photogrammetry constraint conditions, namely that the vertical distance between a camera and a shot slope is constant, and the shooting direction of the camera is vertical to the shot slope; also comprises the following steps:
s323: judging whether the target ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal or not;
s325: acquiring the electricity consumption cost required under all paths, and selecting the path with the optimal electricity consumption cost as the current optimal path;
the electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively for the unmanned plane to carry out horizontal flight, climbing flight and landingPower consumption coefficient in flight;
d x the transverse distance from the target viewpoint to the next viewpoint;
d y the longitudinal distance from the target viewpoint to the next viewpoint;
d z the vertical height distance from the target viewpoint to the next viewpoint;
selecting a path with the optimal electricity consumption cost as a current optimal path; the minimum electricity consumption cost is optimal;
s326: updating the concentration of pheromones on the path;
wherein, in the formula (10), τ ij (t + n) represents the pheromone concentration on paths i to j after the time t + n is updated;
representing the total amount of pheromones left on paths between nodes after all ants traverse the paths after one round of iteration;
ρ is a volatilization coefficient of pheromone concentration;
and adding 1 to the iteration times, judging whether all the search rounds are completed or not, and outputting a three-dimensional optimal path of the unmanned aerial vehicle, otherwise, returning to S322, wherein the output optimal path is the three-dimensional track path of the unmanned aerial vehicle with high pheromone concentration after iteration.
7. An unmanned aerial vehicle inspection three-dimensional path planning system for a mountain photovoltaic power station is characterized by comprising a photovoltaic power station information acquisition module, an area division module and a path planning module;
the photovoltaic power station information acquisition module is used for acquiring design drawing information of mountain photovoltaic power station plant planning and acquiring position information and elevation information of n array area array areas of the photovoltaic power station;
the region division module is used for dividing the photovoltaic power station into m blocks suitable for unmanned aerial vehicle inspection by using an improved density clustering algorithm;
the path planning module is used for planning paths in each area by improving the optimal power consumption path rule of the ant colony; wherein n is greater than 1 and m is greater than 1.
8. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning system according to claim 7, wherein the region segmentation module segments the photovoltaic power station into m unmanned aerial vehicle inspection blocks by using an improved density clustering algorithm, and the segmentation is obtained by the following steps:
s21, obtaining related parameters, wherein the related parameters comprise a field parameter epsilon, minPts and a sample set D = { p = 1 ,p 2 ,…,p n For any one sample point p, where ε field is defined as:
N ε (p)={q∈D|dist(p,q)≤ε}; (1)
sample set D = { p = 1 ,p 2 ,…,p n The elements in the sample set are coordinates of each photovoltaic panel;
where dist (p, q) denotes the sample point p (x) p ,y p ,z p ) And q (x) q ,y q ,z q ) A distance therebetween
For any sample point, the distance dist (p, q) between p and q is formulated as
dist(p,q)=|x p -x q |+3×|y p -y q |+2×|z p -z q |; (2)
Initializing a set of core objectsInitializing cluster number k =0, initializing unvisited sample set Γ = D, clustering
S22: traversing all sample points in the sample set which is not visited to obtain the epsilon field;
if N is present ε (p i ) If the point is more than or equal to MinPts, the point is classified into a core object set;
wherein, N ε (p i ) Denotes p i The number of sample points contained in the epsilon field of (a), and MinPts represents the number of sample points contained in the core object set at least, and then the points are classified into the core object set;
s24: randomly selecting a core object o from the core object set omega, and initializing a target cluster core object queue omega cur = o, initialization class number k n+1 =k n +1, initializing the target cluster sample set Ω k = o, update unvisited sample set Γ = Γ - { o };
s25: if the target cluster core object queue is an empty setThen the target cluster C k After generation, update cluster partition C = { C = { C = } 1 ,C 2 ,…,C n H, updating the core object set omega to omega-C k Step S23 is executed. Otherwise, updating the core object set omega to omega-C k (ii) a Wherein k is more than or equal to 1;
s26: and taking out one core object from the target cluster core object queue, updating the target cluster sample set, updating the unaccessed sample set, updating the target cluster core object queue, and executing the step S25.
9. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning system according to claim 7, wherein the path planning module is configured to perform path planning inside each area by improving an ant colony optimal power consumption path rule, and is obtained through the following steps:
s31: establishing a camera model, and acquiring an aerial route point set, which is recorded as N, according to photogrammetric constraint conditions;
s311: fitting the discrete digital elevation model by using a B spline surface to obtain a fitted surface S formula (3);
wherein, B ij Is a control point set which is a position coordinate of the photovoltaic panel; n is a radical of i,p (x) And N j,q (y) is a B-spline surface basis function;
s312: establishing a camera model, defining the constant offset distance between the position of each aerial route point and the target surface as d, respectively recording the horizontal rotation angle and the vertical rotation angle of the PT camera as P and T, and recording the horizontal overlapping rate and the vertical overlapping rate as sigma 0 And σ 1 ;
The waypoints are obtained as follows:
the camera rotation angles are as follows:
wherein the content of the first and second substances,representing the coordinates of the three-dimensional waypoints in the air,andrespectively representing coordinates and normal vectors on the ground curved surface which are in one-to-one correspondence with the aerial three-dimensional waypoints; i =1,2,3, \ 8230;, N 0 ,j=1,2,3,…,N 1 ;T i,j Indicating the vertical rotation angle of the camera; p is i,j Indicating the horizontal rotation angle of the camera;
N 0 and N 1 The following conditions are satisfied:
s32: performing path planning inside each region based on the optimal power consumption path rule of the improved ant colony; and obtaining an aerial route point set according to photogrammetric constraint conditions, namely that the vertical distance between the camera and the shot slope is constant, and the shooting direction of the camera is vertical to the shot slope.
10. The mountain land photovoltaic power station-oriented unmanned aerial vehicle inspection three-dimensional path planning system according to claim 9, wherein the obtaining of the aerial waypoint set is obtained by:
s321: initializing ant colony algorithm parameters alpha and beta, wherein alpha and beta respectively represent the size of a heuristic factor and the size of an expected heuristic factor, M represents the number of colonies, rho represents an information volatilization factor, and an upper limit of iteration times is set; s322: obtaining the transfer probability of ants from the target alpha, beta, node i to node j at the moment tSelecting a next target point by adopting a roulette mode and moving the target point; s323: judging whether the target ants traverse all paths or not;
selecting a next target point and moving the target point by adopting a roulette mode;
wherein j ∈ allowed k Indicating that the next target point j belongs to the set of waypoints allowed to arrive;
otherwise denotes when node j does not belong to the set of allowed path points;
when the next target point j belongs to the set of path points allowed to arrive in the formula (9), the method is usedCalculating the transition probability by a formula;
when the node j does not belong to the set of path points allowed to reach, the transition probability is equal to 0;
wherein the content of the first and second substances,heuristic information indicative of the path transitions,indicating the concentration of pheromones on the path from node i to node j at time t,representing the amount of information arriving at the target point s from point i,indicating the arrival at the target point s of the desired value, allowed, from the point i k Representing a set of waypoints allowed to be reached;
s323: judging whether the target ants traverse all paths or not;
s324: adding one to the ant number, and judging whether all ants finish path traversal or not;
s325: acquiring the electricity consumption cost required under all paths, and selecting the path with the optimal electricity consumption cost as the current optimal path;
the electricity consumption cost is obtained by the following formula:
wherein, an unmanned aerial vehicle power consumption model is established, w f 、w u 、w d Respectively the power consumption coefficient of the unmanned aerial vehicle during horizontal flight, climbing flight and landing flight, d x Is the lateral distance of the target viewpoint from the next viewpoint, d y Is the longitudinal distance of the target viewpoint from the next viewpoint, d z The vertical height distance from the target viewpoint to the next viewpoint; selecting a path with the optimal electricity consumption cost as a current optimal path; the minimum electricity consumption cost is optimal;
s326: updating the concentration of pheromones on the path;
expressed by the formula (10), the pheromone concentrations on paths i to j after t + n are updated;
representing the total amount of pheromones left on paths between nodes after all ants traverse the paths after one round of iteration; ρ represents a volatilization coefficient of pheromone concentration;
and adding 1 to the iteration times, judging whether all the search rounds are completed or not, and outputting a three-dimensional optimal path of the unmanned aerial vehicle, otherwise, returning to S322, wherein the output optimal path is the three-dimensional track path of the unmanned aerial vehicle with high pheromone concentration after iteration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211210434.5A CN115454132A (en) | 2022-09-30 | 2022-09-30 | Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211210434.5A CN115454132A (en) | 2022-09-30 | 2022-09-30 | Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115454132A true CN115454132A (en) | 2022-12-09 |
Family
ID=84309109
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211210434.5A Pending CN115454132A (en) | 2022-09-30 | 2022-09-30 | Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115454132A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116388691A (en) * | 2023-04-07 | 2023-07-04 | 宁夏百川电力股份有限公司 | Intelligent inspection method and system for photovoltaic power generation field |
CN116703043A (en) * | 2023-08-09 | 2023-09-05 | 华北电力大学 | Unmanned aerial vehicle inspection point planning method and device and electronic equipment |
-
2022
- 2022-09-30 CN CN202211210434.5A patent/CN115454132A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116388691A (en) * | 2023-04-07 | 2023-07-04 | 宁夏百川电力股份有限公司 | Intelligent inspection method and system for photovoltaic power generation field |
CN116388691B (en) * | 2023-04-07 | 2023-10-20 | 宁夏百川电力股份有限公司 | Intelligent inspection method and system for photovoltaic power generation field |
CN116703043A (en) * | 2023-08-09 | 2023-09-05 | 华北电力大学 | Unmanned aerial vehicle inspection point planning method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111695776B (en) | Unmanned aerial vehicle cluster distributed online cooperative area reconnaissance method and device | |
CN108229719B (en) | Multi-objective optimization method and device for unmanned aerial vehicle formation task allocation and flight path planning | |
CN115454132A (en) | Unmanned aerial vehicle inspection three-dimensional path planning method and system for mountain photovoltaic power station | |
CN110428111B (en) | UAV/UGV (unmanned aerial vehicle/user generated Union vector) cooperative long-time multitask operation trajectory planning method | |
CN107807665B (en) | Unmanned aerial vehicle formation detection task cooperative allocation method and device | |
CN108958285B (en) | Efficient multi-unmanned aerial vehicle collaborative track planning method based on decomposition idea | |
CN113342046B (en) | Power transmission line unmanned aerial vehicle routing inspection path optimization method based on ant colony algorithm | |
CN110262548B (en) | Unmanned aerial vehicle track planning method considering arrival time constraint | |
CN105679102B (en) | A kind of national flight flow spatial and temporal distributions prediction deduction system and method | |
CN106295141B (en) | A plurality of unmanned plane determining method of path and device for reconstructing three-dimensional model | |
CN111609864A (en) | Multi-policeman cooperative trapping task allocation and path planning method under road network constraint | |
CN109871031A (en) | A kind of method for planning track of fixed-wing unmanned plane | |
CN115903879A (en) | Unmanned aerial vehicle track planning method based on terrain data interpolation technology | |
CN116257082B (en) | Distributed active cooperative detection method for multiple unmanned aerial vehicles | |
CN112327939B (en) | Collaborative path planning method for high-rise fire-fighting multiple unmanned aerial vehicles in city block environment | |
CN111121784B (en) | Unmanned reconnaissance aircraft route planning method | |
CN115060263A (en) | Flight path planning method considering low-altitude wind and energy consumption of unmanned aerial vehicle | |
CN115185303A (en) | Unmanned aerial vehicle patrol path planning method for national parks and natural protected areas | |
CN115935610A (en) | Method and system for optimizing routing inspection strategy of unmanned aerial vehicle on overhead line | |
CN110986954A (en) | Military transport plane route planning method based on wolf optimization algorithm | |
CN113220008B (en) | Collaborative dynamic path planning method for multi-Mars aircraft | |
CN116518982B (en) | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method | |
CN115390584B (en) | Multi-machine collaborative searching method | |
CN116772848A (en) | Green real-time planning method for four-dimensional flight track of aircraft terminal area | |
Liu et al. | UAV Routine Optimization and Obstacle Avoidance Based on ACO for Transmission Line Inspection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |