CN112650218B

CN112650218B - Transformer substation inspection route planning method and device based on collision detection

Info

Publication number: CN112650218B
Application number: CN202011401642.4A
Authority: CN
Inventors: 汪杨凯; 曾宏宇; 许涛; 杨冰; 李非; 韩继东; 赵然; 许悦; 张勇; 李云越; 潘屹峰; 柳红凯; 王丹
Original assignee: Maintenance Branch of State Grid Hubei Electric Power Co Ltd
Current assignee: Maintenance Branch of State Grid Hubei Electric Power Co Ltd
Priority date: 2020-12-04
Filing date: 2020-12-04
Publication date: 2023-05-16
Anticipated expiration: 2040-12-04
Also published as: CN112650218A

Abstract

The invention discloses a transformer substation inspection route planning method and device based on collision detection, wherein an aerial photographing point is set by adopting a mode of binding a power equipment station account and a point cloud model, and after collision detection is carried out on the aerial photographing point, an absolute safe transformer substation inspection route is planned, so that the safety and reliability of unmanned aerial vehicle inspection of a transformer substation are ensured, and the working efficiency of an electric power inspection operation system is improved.

Description

Transformer substation inspection route planning method and device based on collision detection

Technical Field

The invention relates to the technical field of electric power inspection, in particular to a substation inspection route planning method and device based on collision detection.

Background

The power system plays an extremely critical role in the development of modern economic society, and the safe and reliable operation of the power system is a long-term and careless daily power work. The transformer substation is an indispensable important link in the power system, and whether the equipment in the transformer substation runs safely directly relates to whether the whole power grid can run normally or not, so that the transformer substation plays a significant role in the safe and economic operation of the power grid. With the rapid development of the electric power technology in China, the scale, capacity and voltage level of the transformer substation are higher and higher, higher requirements are also provided for safe and stable operation of the transformer substation, and timely checking of the operation condition of electric equipment in the transformer substation becomes an important guarantee measure for safe and reliable operation of the transformer substation.

With the deep application of unmanned aerial vehicle inspection technology in electric power system, unmanned aerial vehicle has the characteristics of higher flying height, wider field of vision, with its application in the transformer substation inspection can inspect the people and patrol the place that the robot can't patrol, eliminates inspection dead angle and blind area to can in time discover transformer substation equipment defect hidden danger. It is the advantage that unmanned aerial vehicle patrols and examines at the transformer substation, makes transformer substation unmanned aerial vehicle patrols and examines to become a trend.

By adopting the unmanned aerial vehicle inspection mode, the operation condition of the transformer substation can be accurately and comprehensively inspected, the inspection efficiency and quality of the transformer substation are greatly improved, the workload of operation and maintenance personnel is reduced, and the inspection operation cost is reduced.

However, the feasibility of the unmanned aerial vehicle applied to the substation inspection field is preliminarily verified at present, but the unmanned aerial vehicle mainly stays in a long-distance and high-altitude winding flight stage, so that the safety risk problem of the station inspection passing through dense equipment flight is avoided, and the difficulty of popularization of unmanned aerial vehicle inspection in the substation is essentially overcome. Meanwhile, the unmanned aerial vehicle can trigger the unmanned aerial vehicle obstacle avoidance mode when in short-distance inspection in a station because the unmanned aerial vehicle generally has the vision obstacle avoidance function, so that the unmanned aerial vehicle cannot fly, and therefore, in order to inspect substation equipment in short distance, the unmanned aerial vehicle vision obstacle avoidance function needs to be turned off, the phenomenon that the unmanned aerial vehicle collides with surrounding equipment is most likely to occur, and accidents such as unmanned aerial vehicle crash and power equipment damage are caused.

Disclosure of Invention

Aiming at the problems in the background art, the transformer substation inspection route planning method and device based on collision detection are provided, an aerial photographing point can be set according to the binding mode of a power equipment ledger and a point cloud model, and after the aerial photographing point is subjected to collision detection, an absolute safe transformer substation inspection route is planned, so that when an unmanned aerial vehicle vision obstacle avoidance function is closed, the unmanned aerial vehicle can also shuttle in transformer substation equipment in a short distance, an unmanned aerial vehicle inspection task is completed according to a given path, and the safety and reliability of transformer substation unmanned aerial vehicle inspection are guaranteed.

The invention discloses a substation inspection route planning method based on collision detection, which comprises the following steps:

s1, collecting high-precision laser point cloud data of substation power equipment;

s2, denoising the point cloud data to form a point cloud model;

s3, manufacturing a transformer substation power equipment ledger, and importing the ledger into three-dimensional route planning software;

s4, carrying out association binding on the power equipment in the ledger and the point cloud model;

s5, setting aerial photographing point parameter information when photographing each electric power device in three-dimensional route planning software;

s6, adjusting aerial photographing points according to the space distances and the minimum safe distances of the power equipment with different voltage levels;

s7, setting a patrol entrance node according to the point cloud model of the minimum patrol unit divided by the standing book;

s8, setting road crossing points according to the point cloud model, and constructing a global map of the transformer substation;

s9, packaging the point cloud data, the aerial photographing point parameter information, the inspection entrance node, the road intersection point and the global map to form an input data packet of a three-dimensional route planning algorithm;

s10, generating a modularized inspection route interface according to the standing book and the input data packet, selecting an option control in the inspection route interface by a user, and automatically generating an inspection route.

The point cloud data is a data type obtained through a 3D scanner. The scanning data is recorded in the form of points, each point contains three-dimensional coordinates, and can also contain color information (R, G, B) or object reflecting surface intensity, and the intensity information is obtained by the echo intensity collected by the laser scanner receiving device and is related to the surface material, roughness and incident angle direction of the target, the emission energy of the instrument and the laser wavelength.

The denoising processing is a processing method commonly used in point cloud data processing, and the denoising processing commonly used at present comprises bilateral filtering, gaussian filtering, box-division denoising, KD-Tree isolated forest, straight-through filtering, random sampling consistency filtering and the like, wherein the acquired point cloud data density is irregular, outliers are required to be removed due to the problems of shielding and the like, a large amount of data are required to be downsampled, noise data are required to be removed and the like.

The standing book is the standing book in the aspect of safety management of the power equipment, and is a data record reflecting the overall situation of the power equipment and the workflow of the transformer substation. The transformer substation equipment ledger comprises specific information such as a primary equipment list, a secondary equipment list, names, models, capacities, voltages, standard codes, phase numbers, manufacturers, tapping positions and the like of all equipment.

According to the invention, an aerial photographing point is set by adopting a binding mode of the power equipment standing book and the point cloud model, and after collision detection is carried out on the aerial photographing point, an absolute safe substation inspection route is planned, so that the security and reliability of the substation unmanned aerial vehicle inspection are ensured, and the working efficiency of the power inspection operation system is improved.

Further, the high-precision laser point cloud data is real-color rendered point cloud data which is acquired by adopting a ground laser radar scanning system and has coordinate precision within 10cm and density of 500 points per square meter and is based on a WGS84 coordinate system.

Further, the step of denoising the high-precision laser point cloud data to form a point cloud model includes:

removing secondary echoes, air floaters and bird outlier cloud noise points in the point cloud data;

the point cloud data of the target object is automatically denoised through a denoising algorithm, so that the point cloud of the target object can be distinguished by naked eyes, no shielding exists, and no scattered disorder points exist around the point cloud;

and removing foreign matter point clouds affecting modeling accuracy in the point cloud model space.

Specifically, the step of manufacturing a transformer substation power equipment ledger and importing the ledger into three-dimensional route planning software comprises the following steps:

constructing five-level electric power equipment ledger files, wherein the first level is the name of a transformer substation, the second level is the names of main transformer areas and equipment areas of the transformer substation, the third level is the names of the main transformer areas and the equipment areas according to interval subdivision, the fourth level is the names of equipment parts needing inspection shooting according to the direction and the height on the basis of the third level;

and importing the power equipment ledger file into three-dimensional route planning software to form a five-level ledger tree list.

Further, the step of performing association binding on the power equipment in the ledger and the point cloud model includes:

locking a point cloud area needing route planning;

selecting equipment components needing to be inspected and shot in the point cloud area, and associating with the point cloud model;

and binding the account information of the equipment component.

Further, the step of setting the parameter information of the aerial photographing points when photographing each electric device in the three-dimensional route planning software comprises the following steps:

picking up a target point, and checking the distribution condition of power equipment around the target point by rotating the point cloud view;

selecting a shooting view angle with the least shielding object, and setting parameter information of an aerial shooting point when shooting a target point; the parameter information comprises a holder angle, a machine head orientation and a distance between an aerial photographing point and a target point;

and the three-dimensional route planning software longitudinally stretches the normal direction space corresponding to the shooting visual angle to generate a frame selection area according to the shooting rule of the RTK unmanned aerial vehicle by taking the target point as the center, and sets a plurality of aerial shooting points for the same equipment, so that the frame selection area shoots the whole appearance of the equipment.

Further, the step of adjusting the aerial photographing point according to the spatial distance and the minimum safe distance of the power equipment with different voltage levels comprises:

for a 220kv transformer substation, the space distances between aerial photographing points at a low-altitude vertical plane, a hollow horizontal plane, a high-altitude lightning rod, a main transformer area and special parts and target points are within the range of 2-3 meters; the minimum safe distance between the aerial photographing point and the nearest barrier is 1.3 meters;

for a 500kv transformer station, the space distances between aerial photographing points at a low-altitude vertical plane, a hollow horizontal plane, a high-altitude lightning rod, a main transformer area and special parts and target points are within the range of 3-5 meters; the minimum safe distance between the aerial photographing point and the nearest barrier is 2.2 meters;

for a 1000kv transformer station, the space distances between aerial photographing points at a low-altitude vertical plane, a hollow horizontal plane, a high-altitude lightning rod, a main transformer area and special parts and target points are within the range of 4-5 meters; the minimum safe distance between the aerial photographing point and the nearest barrier is 2.8 meters;

setting aerial photographing points according to the space distances between the aerial photographing points of the substations with different voltage levels and the target points, searching the point cloud of the obstacle nearest to the aerial photographing points by adopting an octree neighbor searching algorithm, calculating whether the distance between the aerial photographing points and the nearest obstacle is greater than the minimum safe distance, and reminding and adjusting the aerial photographing points smaller than the minimum safe distance;

the octree neighbor searching algorithm takes an aerial photographing point as a center, and takes 4 meters as a side length to construct a space cube to surround all point clouds taking the aerial photographing point as the center;

dividing the space cube into 8 sub cubes step by step towards the central position until the number of point clouds in the sub cubes closest to the central point is less than ten;

calculating the Euclidean distance between each point cloud and the aerial photographing point in the sub-square;

the Euclidean distance calculating method comprises the following steps:

wherein the three-dimensional coordinates of the aerial photographing point p are (p 1, p2, p 3), the three-dimensional coordinates of the point cloud q are (q 1, q2, q 3), and d is the Euclidean distance between the aerial photographing point p and the point cloud q;

comparing the smallest Euclidean distance with the smallest safe distance; if the Euclidean distance is greater than the minimum safety distance, judging the safety of the aerial photographing point; if the Euclidean distance is smaller than the minimum safety distance, judging that the aerial photographing point is unsafe, reminding a user, adjusting the aerial photographing point, and then judging whether the new aerial photographing point is safe or not by using an octree neighbor searching algorithm again.

Further, according to the point cloud model of the minimum inspection unit divided by the standing book, an inspection entrance node is set, and the method comprises the following steps:

setting a minimum inspection unit in a power equipment ledger, wherein the minimum inspection unit is a route which can inspect all target points in a frame and ensure that the residual electric quantity can return;

and setting a patrol entrance node at the point cloud of the patrol unit according to the point cloud model corresponding to the minimum patrol unit division, wherein the distance between the entrance node and the nearest power equipment is greater than the safety distance.

The entrance node guarantees unmanned aerial vehicle flight safety, and in actual inspection process, unmanned aerial vehicle flies into inspection unit area from the entrance node and goes to inspection equipment.

Further, the road crossing point is a plurality of marking points arranged at the center and corners of the road crossing; the global map of the transformer substation is an air route map with longitude and latitude and elevation information of point cloud data;

the input data packet is a json format file;

and the modularized routing inspection route interface receives a selection instruction of a user for the routing inspection option control, generates an optimal routing inspection route according to a route planning algorithm, and marks and previews the optimal routing inspection route on a point cloud.

The road intersection is set as a marking point for assisting route planning, and a passage is displayed at the road intersection in the global map for the unmanned aerial vehicle to fly. The optimal inspection route of the unmanned aerial vehicle is that in the detection process of a transformer substation, the optimal unmanned aerial vehicle flight route is obtained, the unmanned aerial vehicle takes off from a preset starting point, passes through a target point as many times as possible, and finally reaches the terminal point to finish the inspection task. The routing inspection route planning principle is as follows: shortest route principle and safe flight principle.

The invention also provides a substation inspection route planning device based on collision detection, which comprises:

a device for acquiring high-precision laser point cloud data of substation power equipment;

a device for denoising the high-precision laser point cloud data to form a point cloud model;

the device is used for manufacturing a transformer substation power equipment ledger and importing the ledger into three-dimensional route planning software;

means for performing an association binding of the power devices in the ledger with the point cloud model;

the device is used for setting aerial photographing point parameter information when photographing all the electric power equipment in the three-dimensional route planning software;

means for adjusting the distance of the aerial photo point from the power device according to the minimum safe distance of the power device at different voltage levels;

means for setting a patrol entry node according to the point cloud model of the ledger partition minimum patrol unit;

setting a road intersection point according to the point cloud model, and constructing a global map of the transformer substation;

the device is used for packaging the point cloud data, the aerial photographing point parameter information, the inspection entrance node, the road intersection and the global map to form an input data packet of a three-dimensional route planning algorithm;

and the device is used for generating a modularized inspection route interface according to the standing book and the input data packet, selecting an option control in the inspection route interface by a user and automatically generating an inspection route.

In order that the invention may be more clearly understood, specific embodiments thereof will be described below with reference to the accompanying drawings.

Drawings

Fig. 1 is a flowchart of a substation inspection route planning method based on collision detection according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of cube segmentation in an octree neighbor search algorithm according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of an octree structure in an octree neighbor search algorithm according to an embodiment of the present invention.

Detailed Description

s2, denoising the point cloud data to form a point cloud model;

According to the invention, the point cloud model is built by collecting the point cloud data of the power equipment of the transformer substation, the equipment account files are bound, so that the inspection objects and the inspection data are in one-to-one correspondence and are automatically associated, the aerial photographing points are set to photograph the power equipment of different levels, the whole global map of the transformer substation is built, the optimal inspection route is planned by the three-dimensional route planning software, the security and reliability of unmanned aerial vehicle inspection of the transformer substation are ensured, and the working efficiency of the power inspection operation system is improved.

The unmanned aerial vehicle inspection is mainly stopped at a long-distance and high-altitude winding stage, so that the safety risk problem of the station inspection passing through dense equipment for flying is avoided, and the difficulty of popularization of unmanned aerial vehicle inspection in a transformer substation is not basically overcome. Meanwhile, the unmanned aerial vehicle can trigger the unmanned aerial vehicle obstacle avoidance mode when in short-distance inspection in a station because the unmanned aerial vehicle generally has the vision obstacle avoidance function, so that the unmanned aerial vehicle cannot fly, and therefore, in order to inspect substation equipment in short distance, the unmanned aerial vehicle vision obstacle avoidance function needs to be turned off, the phenomenon that the unmanned aerial vehicle collides with surrounding equipment is most likely to occur, and accidents such as unmanned aerial vehicle crash and power equipment damage are caused. The invention provides a method and a device for planning a substation inspection route based on collision detection, which can set aerial photo points according to the binding mode of a power equipment station account and a point cloud model, and plan an absolute safe substation inspection route after collision detection of the aerial photo points, so that when an unmanned aerial vehicle vision obstacle avoidance function is closed, an unmanned aerial vehicle can also shuttle in substation equipment in a short distance, unmanned aerial vehicle inspection tasks are completed according to a set path, and the safety and reliability of unmanned aerial vehicle inspection of the substation are ensured

The substation inspection route planning method based on collision detection of the invention is described in one embodiment.

In this embodiment:

the acquisition of the point cloud data is to acquire high-precision laser point cloud data of low, medium and high-rise equipment of a transformer substation by adopting a ground laser radar scanning system, a WGS84 coordinate system is adopted to ensure the quality of the point cloud data, the coordinate precision is controlled within 10cm, the point cloud density is 500 points per square meter, the point cloud adopts true color rendering, all the equipment and meters are completely visible, no component loss exists, and the point cloud data format is a las format.

The collected original point cloud data is influenced by the power equipment and the external environment, comprises a large number of hash points, isolated points and the like, and influences the planning of a route. Denoising the acquired original point cloud data, and removing secondary echoes, air floaters and bird outlier point cloud noise points in the point cloud data; the point cloud data of the target object is automatically denoised through a denoising algorithm, so that the point cloud of the target object can be distinguished by naked eyes, no shielding exists, and no scattered disorder points exist around the point cloud; and removing foreign matter point clouds affecting modeling accuracy in the point cloud model space.

Before the obtained point cloud model is associated with the standing book, a standing book file corresponding to the relationship of the transformer substation equipment is required to be constructed, so that the equipment inspection unit and the shooting points can be conveniently searched and managed during planning of a transformer substation route.

In this embodiment, the power equipment ledger file includes five levels, one level is a transformer substation name, two levels are a transformer substation main transformer area and an equipment area name, three levels are main transformer area subdivision and equipment area subdivision names according to intervals, four levels are subdivided according to directions and heights on the basis of three levels, such as a height direction, a middle-high layer, a low layer, and the like, and five levels are equipment component names required to be inspected and shot. The binding of the standing book and the setting of the aerial photo points are carried out on fifth-level equipment.

In this embodiment, the csv form of the ledger is adopted, and in other alternative embodiments, the ledger file may be in txt form or json form.

And storing the manufactured standing book file, and importing the standing book file of the power equipment in three-dimensional route planning software through an importing standing book function to form a five-level standing book tree list after importing.

In order to facilitate the association of the point cloud model and the ledger, the point cloud area needing route planning is firstly locked, equipment components for inspection shooting in a five-level ledger tree list are selected, the equipment components are associated with the high-precision three-dimensional laser point cloud model of the transformer substation, and ledger information of the equipment components is bound. The inspection data and the inspection equipment objects are in one-to-one correspondence and are automatically associated.

the three-dimensional route planning software generates a frame selection area by taking a target point as a center and longitudinally stretching a normal direction space corresponding to a shooting visual angle according to shooting rules of the RTK unmanned aerial vehicle, wherein the frame selection area represents a visual field range which can be shot by an unmanned aerial vehicle camera under the parameter settings in actual shooting, namely, the frame selection area is large, the shooting range is large, a user can conveniently conduct route planning and presetting, and under the condition of large equipment size, a plurality of aerial shooting points are arranged on the same equipment, so that the frame selection area shoots the equipment overall view. The method is beneficial to a user to analyze whether the current aerial photographing point setting is comprehensive in photographing the same equipment, redundant aerial photographing points are not arranged, planning efficiency is improved, data analysis efficiency of photographed pictures is improved, and excessive redundant photographed pictures are avoided.

The aerial photographing points are adjusted according to the space distances and the minimum safety distances of the power equipment with different voltage levels, and for a 220kv transformer substation, the space distances between the aerial photographing points and target points at a low-altitude vertical plane, a hollow horizontal plane, a high-altitude lightning rod, a main transformer area and a special part are within the range of 2-3 meters; the minimum safe distance between the aerial photographing point and the nearest barrier is 1.3 meters;

please refer to fig. 2, which is a block diagram illustrating a square segmentation in an octree neighbor search algorithm according to an embodiment of the present invention.

The simplest method for searching the existing space point cloud is to traverse all point clouds, calculate Euclidean distances between all point clouds and aerial photo points, and then find out the nearest distance and compare with the safe distance. The method is simple but is time consuming for computation of a large number of point clouds. In order to improve the calculation speed, the embodiment adopts a search method based on space division, the method is to divide the space of the point cloud into blocks according to a certain rule, and when space search is performed, the nearest point search is performed in the divided space according to a depth-first strategy. The existing search method based on space division mainly adopts an octree neighbor search method, and the method has the advantages of low time complexity and high calculation speed.

The octree neighbor search algorithm in the embodiment uses an aerial photographing point as a center, and uses 4 meters as a side length to construct a space cube to surround all point clouds which use the aerial photographing point as the center; according to the space size of the point cloud, the space is gradually divided into subspaces with the same size by utilizing a hierarchical structure of the tree, when the space cube of the point cloud is divided, the space cube is divided equally in two in each dimension, namely, the cube is divided by taking 3 vertical middle division planes of the cube, and 8 equal subcubes which are called 8 voxels are obtained. The divided 8 voxels are segmented again by the method to form a spatial tree hierarchy. Each voxel is each node of the tree.

See the octree structure schematic of fig. 3.

Each node is encoded, the purpose of the encoding is to calculate the index number and depth information of the point cloud, and the position of the point cloud is determined according to the index number and the depth information, and several concepts are defined at first:

(1) Ancestor node: the voxel represented at the uppermost layer is at the very top of the tree structure, called ancestor node, i.e., the original square bounding box, from which all nodes are partitioned.

(2) Parent and child nodes: except for ancestor nodes, the partitioned voxels are called parent nodes, and the partitioned 8 voxels are called child nodes.

(3) Black node: the black node indicates that a certain point cloud falls in the space cube corresponding to the node, and the position of the space cube is closer to the aerial point in the 8 cubes, and the black node needs to divide the space downwards continuously for 8 times.

(4) Bai Jiedian: the white node indicates that a certain point cloud does not fall into the space cube corresponding to the node or falls into the space cube corresponding to the node, but the position of the space cube is not closer to the aerial photographing point in the 8 cubes, and the white node does not need to divide 8 spaces continuously.

(5) Index: each node is coded by the user definition, and the codes are used for distinguishing the nodes and have uniqueness. According to the node position of the point cloud in the tree, encoding the first layer father node to all father nodes where the cubes do not need to be continuously segmented to form a group of series in sequence, and using the series as index information of the point cloud.

(6) Depth: and calculating the total number in the number sequence forming the index, and taking the total number as depth, wherein the deeper the depth is, the closer the point cloud is to the aerial photographing point.

And then carrying out space division on each point in the point cloud according to the method, classifying the point cloud into a corresponding subspace, wherein the boundary for ending the division is that 8 split voxel nodes are white, indicating that the cube corresponding to the father node of the layer is the position of the point cloud, finding out all the point clouds with the maximum depth by utilizing the depth information of the index number of the point cloud, realizing the search of nearest neighbor points, calculating the positions of the point clouds from the aerial photographing points, and comparing the nearest distance with the safe distance.

The Euclidean distance calculating method comprises the following steps:

comparing the smallest Euclidean distance with the smallest safe distance; if the Euclidean distance is greater than the minimum safety distance, judging that the aerial photographing point is safe, and not colliding with surrounding equipment during route planning; if the Euclidean distance is smaller than the minimum safety distance, judging that the aerial photographing point is unsafe, after the aerial photographing point is reminded and adjusted by the route planning software, carrying out collision safety inspection again by using an octree neighbor searching algorithm, judging whether the new aerial photographing point is safe or not, and carrying out next operation until all the aerial photographing points pass the safety inspection.

Before setting the inspection entrance node, setting a minimum inspection unit in the power equipment ledger, and performing inspection route planning based on each minimum inspection unit. The minimum inspection unit is a route which can inspect all target points in an overhead period and ensure that the residual electric quantity can return;

and according to the point cloud model corresponding to the minimum inspection unit division, an inspection entrance node is arranged at the point cloud of the inspection unit, and the distance between the entrance node and the nearest power equipment is larger than the safety distance, so that the flight safety of the unmanned aerial vehicle is ensured. In the actual inspection process, the unmanned aerial vehicle needs to fly into the inspection unit area from the entrance node to inspect equipment.

In this embodiment, after the minimum inspection unit and the inspection entrance node are set, a road intersection point is set as an automatic route selection path, where the road intersection point is a plurality of marking points set at the center and corners of the road intersection; and in relation to the calculation of the entry point and the exit point of the inspection unit, a passage is displayed at the road intersection point in the global map for the unmanned aerial vehicle to fly.

And constructing a global map for the whole transformer substation according to the longitude and latitude, the elevation and other information of the point cloud data, and taking the global map as one of basic data of route planning, so that the information of the map on the planned route is used for guiding the unmanned aerial vehicle to fly when the unmanned aerial vehicle is used for inspection, and the safety is ensured. And when the global map is constructed, the planned route safety distances are synchronously set, so that all routes and surrounding equipment keep a certain safety distance.

Packaging the point cloud data, the aerial photographing point parameter information, the inspection entrance node, the road intersection point and the global map to form an input data packet of a three-dimensional route planning algorithm;

the input data packet is a json format file;

and the modularized routing inspection route interface receives a selection instruction of a user for the routing inspection option control, generates an optimal routing inspection route according to a route planning algorithm, and marks and previews the optimal routing inspection route on a point cloud. The distribution of the airlines and surrounding devices in all directions can be checked through view rotation on the point cloud. The routing inspection route planning principle is as follows: shortest route principle and safe flight principle.

According to the method, according to the requirements of the inspection task of the substation power equipment and the setting of the inspection scheme of the unmanned aerial vehicle, how to construct an absolute safe substation inspection route is studied, the space distance between the aerial photographing points and the equipment and the distance between the aerial photographing points and the nearest obstacle calculated by the octree neighbor search algorithm are calculated through the substation power equipment with different voltage levels, and the aerial photographing points in the planning route are adjusted, so that the unmanned aerial vehicle can closely shuttle in the substation equipment when the inspection task is executed, and the inspection task is completed according to a set path.

In traditional unmanned aerial vehicle routing planning, basically all are long-distance high altitude around the way of flying, have not flexibility, also lead to the data of patrolling and examining inadequately accurately easily. Meanwhile, due to the vision obstacle avoidance function of the unmanned aerial vehicle, the obstacle avoidance mode is frequently triggered by mistake when the power equipment is inspected in a short distance, so that the unmanned aerial vehicle stops the flight task, the work of the inspection task is wasted, the phenomenon that the unmanned aerial vehicle collides with the peripheral power equipment easily occurs when the vision obstacle avoidance function is turned off, and the inspection work safety is extremely unfavorable.

Compared with the prior art, the invention can plan an absolute safe substation inspection route, thereby ensuring that the unmanned aerial vehicle can also shuttle in the substation equipment in a short distance when the unmanned aerial vehicle vision obstacle avoidance function is closed, reducing the workload of inspection personnel, greatly improving the efficiency, shortening the planning time of the inspection shooting path and further promoting the development of the unmanned aerial vehicle substation inspection application.

The present invention is not limited to the above-described embodiments, and if various modifications or variations of the present invention are not departing from the spirit and scope of the present invention, the present invention also includes such modifications and variations provided they fall within the scope of the claims and the equivalents thereof.

Claims

1. A substation inspection route planning method based on collision detection comprises the following steps:

collecting high-precision laser point cloud data of substation power equipment;

denoising the point cloud data to form a point cloud model; removing secondary echoes, air floaters and bird outlier cloud noise points in the point cloud data; the point cloud data of the target object is automatically denoised through a denoising algorithm, so that the point cloud of the target object can be distinguished by naked eyes, no shielding exists, and no scattered disorder points exist around the point cloud; removing foreign matter point clouds affecting modeling accuracy in a point cloud model space;

manufacturing a transformer substation power equipment ledger, and importing the ledger into three-dimensional route planning software; the method comprises the steps of constructing five levels of electric equipment ledger files, wherein one level is a transformer substation name, the second level is a transformer substation main transformer area and an equipment area name, the third level is a main transformer area subdivision and an equipment area subdivision name according to intervals, the fourth level is subdivided according to directions and heights on the basis of the third level, and the fifth level is an equipment part name needing inspection shooting; importing the power equipment ledger file into three-dimensional route planning software to form a five-level ledger tree list;

carrying out association binding on the power equipment in the standing book and the point cloud model;

setting aerial photographing point parameter information when photographing each power device in three-dimensional route planning software; the method comprises the steps of picking up a target point, and checking distribution conditions of power equipment around the target point through a rotating point cloud view; selecting a shooting view angle with the least shielding object, and setting parameter information of an aerial shooting point when shooting a target point; the parameter information comprises a holder angle, a machine head orientation and a distance between an aerial photographing point and a target point; the three-dimensional route planning software longitudinally stretches the normal direction space corresponding to the shooting visual angle to generate a frame selection area according to the shooting rule of the RTK unmanned aerial vehicle by taking the target point as the center, and sets a plurality of aerial shooting points for the same equipment to enable the frame selection area to shoot the whole appearance of the equipment;

adjusting aerial photographing points according to the space distances and the minimum safe distances of the power equipment with different voltage levels;

setting a patrol entrance node according to the point cloud model of the minimum patrol unit divided by the standing book; the method comprises the steps that a minimum inspection unit is arranged in a power equipment ledger, and the minimum inspection unit is a route which can inspect all target points in an overhead period and ensure that the residual electric quantity can return; according to the point cloud model corresponding to the minimum inspection unit division, an inspection entrance node is arranged at the point cloud of the inspection unit, and the distance between the entrance node and the nearest power equipment is larger than the safety distance;

and generating a modularized inspection route interface according to the standing book and the input data packet, selecting an option control in the inspection route interface by a user, and automatically generating an inspection route.

2. The substation inspection route planning method based on collision detection according to claim 1, wherein the high-precision laser point cloud data is real-color rendered point cloud data which is acquired by adopting a ground laser radar scanning system, has coordinate precision within 10cm and density of 500 points per square meter and is based on a WGS84 coordinate system.

3. The substation inspection route planning method based on collision detection according to claim 1, wherein the step of associatively binding the power equipment in the ledger with the point cloud model comprises:

locking a point cloud area needing route planning;

and binding the account information of the equipment component.

4. The method for planning a patrol route of a transformer substation based on collision detection according to claim 1, wherein the step of adjusting the aerial photographing point according to the spatial distance and the minimum safe distance of the power equipment of different voltage classes comprises:

the Euclidean distance calculating method comprises the following steps:

wherein the three-dimensional coordinates of the aerial points p are (p 1, p2,p 3), the three-dimensional coordinates of the point cloud q are (q 1, q2, q 3), and d is the Euclidean distance between the aerial photographing point p and the point cloud q;

5. The substation inspection route planning method based on collision detection according to claim 1, wherein the road intersection is a plurality of marking points arranged at the road intersection center and corners; the global map of the transformer substation is an air route map with longitude and latitude and elevation information of point cloud data;

the input data packet is a json format file;

6. A substation inspection route planning device based on collision detection, comprising:

a device for denoising the high-precision laser point cloud data to form a point cloud model; removing secondary echoes, air floaters and bird outlier cloud noise points in the point cloud data; the point cloud data of the target object is automatically denoised through a denoising algorithm, so that the point cloud of the target object can be distinguished by naked eyes, no shielding exists, and no scattered disorder points exist around the point cloud; removing foreign matter point clouds affecting modeling accuracy in a point cloud model space;

the device is used for manufacturing a transformer substation power equipment ledger and importing the ledger into three-dimensional route planning software; the method comprises the steps of constructing five levels of electric equipment ledger files, wherein one level is a transformer substation name, the second level is a transformer substation main transformer area and an equipment area name, the third level is a main transformer area subdivision and an equipment area subdivision name according to intervals, the fourth level is subdivided according to directions and heights on the basis of the third level, and the fifth level is an equipment part name needing inspection shooting; importing the power equipment ledger file into three-dimensional route planning software to form a five-level ledger tree list;

the device is used for setting aerial photographing point parameter information when photographing all the electric power equipment in the three-dimensional route planning software; the method comprises the steps of picking up a target point, and checking distribution conditions of power equipment around the target point through a rotating point cloud view; selecting a shooting view angle with the least shielding object, and setting parameter information of an aerial shooting point when shooting a target point; the parameter information comprises a holder angle, a machine head orientation and a distance between an aerial photographing point and a target point; the three-dimensional route planning software longitudinally stretches the normal direction space corresponding to the shooting visual angle to generate a frame selection area according to the shooting rule of the RTK unmanned aerial vehicle by taking the target point as the center, and sets a plurality of aerial shooting points for the same equipment to enable the frame selection area to shoot the whole appearance of the equipment;

means for setting a patrol entry node according to the point cloud model of the ledger partition minimum patrol unit; the method comprises the steps that a minimum inspection unit is arranged in a power equipment ledger, and the minimum inspection unit is a route which can inspect all target points in an overhead period and ensure that the residual electric quantity can return; according to the point cloud model corresponding to the minimum inspection unit division, an inspection entrance node is arranged at the point cloud of the inspection unit, and the distance between the entrance node and the nearest power equipment is larger than the safety distance;