A method and system for simulating the flight of unmanned aerial vehicles (UAVs) in substations

Abstract

This invention discloses a method and system for simulating the flight of a UAV in a substation, comprising: digitally reconstructing collected substation geographic data and equipment data to generate a three-dimensional terrain model; simulating the flight characteristics of the UAV in the substation using the three-dimensional terrain model, and performing physical motion simulation analysis of the UAV's motion from UAV motion analysis, collision detection, and UAV attitude calculation; and planning the UAV's inspection trajectory based on the physical motion simulation analysis, generating the UAV's flight path and conducting simulated flight to realize the simulation of UAV flight in the substation. This invention realistically recreates the substation environment through high-precision three-dimensional modeling technology, enabling real-time data transmission and display, providing operators with a highly realistic training platform; simultaneously, the built-in dynamic defect library supports flexible setting of common equipment defects, ensuring the diversity of training content, and the full-scenario simulation assessment helps operators comprehensively improve their UAV inspection capabilities.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) simulation technology, and in particular to a method and system for simulating the flight of UAVs in substations. Background Technology

[0002] With the continuous advancement of smart grid construction, drone technology is increasingly being applied in substation inspection. Leveraging its advantages of efficiency, safety, and flexibility, drones have become an important supplement to traditional manual inspections. They can penetrate deep into substations to observe equipment up close, promptly identifying potential faults and significantly improving inspection efficiency and quality. However, existing substation drone inspection systems still face some significant challenges in practical application, particularly in operator training. Due to a lack of effective training simulation tools, newly hired operators often require extensive hands-on experience to become proficient in drone operation. This not only increases training costs but also raises the risk of accidents caused by improper operation. Furthermore, operators may struggle to accurately identify and assess equipment defects when facing complex equipment and challenging environments due to insufficient experience, impacting the effectiveness of inspection work.

[0003] Current technologies lack the ability to accurately recreate 3D scenes and simulate defects, which directly hinders the development of training simulation tools. On the one hand, substations have complex internal structures and a wide variety of equipment, making high-precision 3D modeling difficult. Existing modeling techniques often fall short of ideal results, especially in detail rendering and dynamic change simulation. On the other hand, equipment defects are diverse, ranging from minor corrosion to severe damage. Each defect manifests differently and has varying degrees of impact. Existing simulation technologies struggle to comprehensively and realistically reproduce these defects, leading to significant discrepancies between the operator's experience in the simulation environment and actual conditions, thus affecting training effectiveness.

[0004] Therefore, developing a flight simulation method for UAV inspection of substations is particularly important. This method should be able to achieve high-precision 3D reconstruction of the internal and external environment of the substation, including various equipment and their layout, ensuring a high degree of realism in the simulated environment. Simultaneously, it needs to possess powerful defect simulation capabilities, capable of simulating different types and degrees of equipment defects, providing operators with a rich variety of training cases. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, this invention provides a method and system for simulating the flight of UAVs in substations. Through functions such as 3D modeling, real-time data transmission from UAVs, and defect identification, it comprehensively simulates the UAV inspection scenario in substations. The system adopts the UE4 engine, which can realistically reproduce the equipment structure and environment of a 110kV substation. At the same time, it has a built-in function to automatically generate inspection reports. Operators can use the simulation platform to conduct UAV operation training and defect detection assessment, further improving the inspection effect and efficiency of UAVs.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] In a first aspect, the present invention provides a method for simulating the flight of a UAV in a substation, comprising: digitally reconstructing collected geographical data and equipment data of the substation to generate a three-dimensional terrain model; using the three-dimensional terrain model to simulate the flight characteristics of the UAV in the substation, and performing physical motion simulation analysis of the UAV's motion from UAV motion analysis, collision detection and UAV attitude calculation; and planning the UAV's patrol trajectory based on the physical motion simulation analysis, generating the UAV's flight path and performing simulated flight to realize the simulation of the flight of the UAV in the substation.

[0009] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the step of digitally reconstructing the collected substation geographic data and equipment data to generate a three-dimensional terrain model includes:

[0010] The raw point cloud data of the substation's precise geography and equipment is collected using UAV lidar. Different categories of candidate point clouds are labeled using clustering methods to separate ground feature point clouds from non-ground feature point clouds. The raw point cloud data is then subjected to noise filtering and enhancement.

[0011] By analyzing the three-dimensional coordinates and morphological features of point cloud data, the orientation of the terrain and the distribution of ground features can be determined.

[0012] The selected point cloud data, combined with its spatial location, is used to create a 3D model of the power lines and the entire Earth surface.

[0013] In 3D terrain modeling, a 3D terrain model is generated based on triangulation and Poisson reconstruction algorithms, and the 3D terrain model is then smoothed.

[0014] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the step of simulating the flight characteristics of the substation UAV includes:

[0015] Based on the aforementioned 3D terrain model, the flight characteristics are simulated using the Airsim+Pixhawk flight control hardware platform to simulate the real physical state of the drone's flight, and a real drone remote controller is used for control.

[0016] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the UAV motion analysis is a physical motion simulation analysis of the UAV's motion based on the force analysis of the UAV, the UAV's acceleration, and the UAV's speed, position, angular velocity, and direction.

[0017] The force analysis of the UAV includes:

[0018] Linear drag and angular velocity drag are important factors that change the motion state of a drone during its movement. The direction of linear drag is opposite to the velocity vector v.

[0019] Assuming that the partial angular velocity of the rigid body surface ds is ω, then ds is subjected to two types of resistance torques: friction torque, which is the torque generated by the tangential component of the velocity, and torque generated by shear stress. The linear resistance on ds is obtained according to the linear resistance equation, and the resistance torque on the surface of the UAV is calculated by integration.

[0020] This allows us to obtain the total force and torque F acting on the drone. net for:

[0021] F net =∑ i F i +F d

[0022] Among them, F i F is the thrust at the four vertices relative to the drone's center of gravity. d For linear resistance;

[0023] The drone acceleration includes: calculating the total linear and angular acceleration of the drone based on the total force and torque acting on it; combining the calculated linear and angular accelerations with the gravitational acceleration g to obtain the actual acceleration a of the drone.

[0024]

[0025] Where m represents the mass of the drone;

[0026] The speed, position, angular velocity, and direction of the drone include:

[0027] At time t+1, the velocity V is updated by integrating the velocity from the previous time t. t+1 and position p t+1At time t+1, the angular velocity ω is updated by integrating the angular velocity from the previous time t. t+1 ;

[0028] Calculate the angle axis pair (a) dt ,u), where a dt It is the angular velocity of rotation around the unit vector u, and the change in direction is represented by the change of the quaternion over time dt.

[0029] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the physical motion simulation analysis of the UAV's motion from the perspective of collision detection includes:

[0030] In the collision detection system, each entity has a collision attribute, which determines whether the entity collides with other entities. If a collision or overlap event occurs during the rendering interval, the collision detection system calls the collision response, which returns the collision location, collision type, and input depth. The system then calculates the returned data and modifies the drone's motion state based on the calculation results.

[0031] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the physical motion simulation analysis of the UAV's motion from both aspects of UAV attitude estimation and attitude control includes:

[0032] The UAV attitude in the geographic coordinate system is stored using quaternions. When input into the UAV control algorithm, it is converted into Euler angles. The pitch angle (θ), yaw angle (ψ), and roll angle (φ) are used to represent the UAV attitude relative to the ground. The pitch angle represents the UAV's rotation around the x-axis, with the UAV nose pointing upwards being positive; the yaw angle represents the UAV's rotation around the y-axis, with the UAV nose yawing to the right being positive; and the roll angle represents the UAV's rotation around the z-axis, with the UAV nose rolling to the right being positive.

[0033] The UAV attitude estimation uses IMU sensor data and employs the Mahony complementary filtering algorithm combined with a PID feedback controller to compensate for and correct gyroscope errors.

[0034] The attitude control of the UAV adopts a nested PI-PID controller, which controls the pitch angle, yaw angle and roll angle by adjusting the angular velocity of the propeller.

[0035] As a preferred embodiment of the substation UAV flight simulation method described in this invention, the UAV inspection trajectory planning includes:

[0036] Within the inspection area, individuals in the population are transformed into waypoints, and each waypoint is represented by its distance and orientation angle.

[0037] pass Describe the length of the flight path, where P i D(P) represents the waypoint. i ,P i+1 The distance between two consecutive waypoints is represented by the distance between the waypoints. The penalty design is based on the safe distance from the obstacle. The smoothness of the flight is measured by calculating the change in the heading angle between consecutive waypoints.

[0038] The fitness function F, which combines the path length, obstacle avoidance performance, and smoothness, is expressed as:

[0039]

[0040] Among them, w i Here are the weight values ​​for each reference, where s represents the smoothness of flight and p represents the obstacle avoidance capability of flight.

[0041] Suppose the set of points that the drone needs to inspect is X = {X1, X2, ..., X...} n}, where X n This represents an inspection point. The planned path during the inspection process visits all inspection points sequentially and returns to the starting point.

[0042] Perform a full permutation of all inspection points (n-1)!, exhaustively list all inspection paths, calculate the inspection process and total distance to return to the origin for each path, and select the path with the shortest distance as the optimal inspection path.

[0043] Secondly, the present invention provides a substation unmanned aerial vehicle (UAV) flight simulation system, comprising:

[0044] The terrain model generation module is used to digitally reconstruct a three-dimensional terrain model from the collected substation geographic data and equipment data.

[0045] The simulation analysis module is used to simulate the flight characteristics of the substation UAV using the three-dimensional terrain model, and to perform physical motion simulation analysis of the UAV's motion from the perspectives of UAV motion analysis, collision detection and UAV attitude calculation.

[0046] The flight simulation module is used to plan the flight path of UAV inspection based on physical motion simulation analysis, generate the flight path of UAV and conduct simulated flight to realize the simulation of UAV flight in substations.

[0047] Thirdly, the present invention provides an electronic device, comprising:

[0048] Memory and processor;

[0049] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the substation UAV flight simulation method are implemented.

[0050] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the substation UAV flight simulation method.

[0051] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a method and system for simulating the flight of unmanned aerial vehicles (UAVs) in substations. Through high-precision 3D modeling technology, it realistically recreates the substation environment. The simulation system can achieve real-time data transmission and display, providing operators with a highly realistic training platform. Inspection reports can be automatically generated in the simulation environment, significantly improving inspection efficiency. Simultaneously, the built-in dynamic defect library supports flexible setting of common equipment defects and allows batch import / export, ensuring the diversity and relevance of training content. Full-scenario simulation assessment helps operators comprehensively improve their UAV inspection capabilities, ensuring efficient and accurate task completion in actual work. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a schematic diagram of the overall process logic of the substation UAV flight simulation method according to an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the geographic coordinate system and the body coordinate system of the substation UAV flight simulation method according to an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of the electrical equipment model of the substation UAV flight simulation method according to an embodiment of the present invention;

[0056] Figure 4 This is a three-dimensional scene diagram of the substation UAV flight simulation method according to an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram of the horizontal figure-eight training of the substation UAV flight simulation method according to an embodiment of the present invention;

[0058] Figure 6 This is a schematic diagram of single-channel hovering training for a substation UAV flight simulation method according to an embodiment of the present invention;

[0059] Figure 7 This is a schematic diagram of dual-channel hovering training for a substation UAV flight simulation method according to an embodiment of the present invention;

[0060] Figure 8 This is a schematic diagram of the AOPA assessment of the substation UAV flight simulation method according to an embodiment of the present invention;

[0061] Figure 9 This is a schematic diagram of a UAV substation inspection defect simulation method according to an embodiment of the present invention. Detailed Implementation

[0062] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0063] Example 1

[0064] Reference Figures 1-9 As one embodiment of the present invention, a method for simulating the flight of a substation unmanned aerial vehicle (UAV) is provided, such as... Figure 1 The specific steps shown are as follows:

[0065] S100: Digitally reconstruct the collected substation geographic data and equipment data to generate a three-dimensional terrain model;

[0066] S200: Uses a 3D terrain model to simulate the flight characteristics of UAVs in substations, and performs physical motion simulation analysis of UAV motion from UAV motion analysis, collision detection and UAV attitude calculation.

[0067] S300: Based on physical motion simulation analysis, it plans the flight path of UAV inspection, generates the flight route of UAV and conducts simulated flight to realize the simulation of UAV flight in substations.

[0068] It should be noted that this invention provides a method and system for simulating the flight of UAVs in substations. Through high-precision 3D modeling technology, it realistically recreates the substation environment. The simulation system can achieve real-time data transmission and display, providing operators with a highly realistic training platform. Inspection reports can be automatically generated in the simulation environment, significantly improving inspection efficiency. Simultaneously, the built-in dynamic defect library supports flexible settings for common equipment defects and allows batch import / export, ensuring the diversity and relevance of training content. Full-scenario simulation assessment helps operators comprehensively improve their UAV inspection capabilities, ensuring efficient and accurate task completion in actual work.

[0069] It should be noted that the precise geographic and equipment data of the substation pre-collected by the UAV LiDAR is digitally reconstructed using image processing tools (such as 3ds Max). Airborne LiDAR involves mounting a LiDAR system on a flight platform to acquire spatial information of ground features through laser scanning, achieving rapid acquisition of high-density, high-precision three-dimensional spatial information of ground features. The series of data obtained by the LiDAR system is called a "point cloud," which is a set of discrete points with spatial three-dimensional coordinate information. Based on the point cloud data, a three-dimensional model of the ground features around the transmission line can be constructed. The processing of laser point cloud data includes point cloud classification, noise reduction filtering, target extraction, and three-dimensional model reconstruction.

[0070] In this embodiment of the application, the above step S100, which involves digitally reconstructing the collected substation geographic data and equipment data to generate a three-dimensional terrain model, includes:

[0071] Using UAV lidar, raw point cloud data of the substation's precise geography and equipment is collected, encompassing all objects in the geographic environment, such as the ground surface, vegetation, buildings, and roads. Clustering methods are used to label candidate points of different categories, achieving separation between ground feature point clouds and non-ground feature point clouds. The formula for the layering threshold is:

[0072]

[0073] Among them, H max and H min These are the maximum and minimum height values ​​in the point cloud data, respectively, and n is the number of layers used to classify different types of ground features;

[0074] Furthermore, geographic environment modeling requires noise filtering of point clouds before data extraction. For power line extraction, noise mainly comes from sky specks or ground reflections. Noise filtering can eliminate these irrelevant data, thereby enhancing the clarity of power line points. In geographic environment modeling, noise mainly comes from occlusion by ground objects or equipment errors. A method similar to Moving Least Squares (MLS) can be used to process noise.

[0075] Where, f(x) i ,y i ) represents the interpolated fitted surface, and p′ represents the filtered points;

[0076] Furthermore, by analyzing the 3D coordinates and morphological features of point cloud data, the orientation of the terrain and the distribution of ground features can be determined. Delaunay triangulation is a common method for processing this type of data. Triangulation can be used to refine the modeling of the land surface, especially for complex terrains (such as mountains and rivers).

[0077]

[0078] Furthermore, using the selected point cloud data and their spatial locations, a 3D model of the power lines and the entire surface is created. Natural neighbor interpolation is employed, and the point cloud is interpolated using a Voronoi diagram structure.

[0079]

[0080] Among them, w i (P) is the weight of the interpolation point relative to its natural neighbors, z i It is the height value of the known neighboring points.

[0081] Furthermore, such as Figure 4 The diagram shows how to generate a 3D terrain model based on triangulation and Poisson reconstruction algorithms in 3D terrain modeling.

[0082] ΔΦ=▽·V

[0083] Where Φ is the 3D shape function of the point cloud generation, and V is the normal vector field;

[0084] Furthermore, the existing 3D model was optimized using 3DS Max software, primarily to reduce surface roughness, improve visual appeal, and refine details. Based on the circuit design drawings, typical tower designs were modeled using 3DS Max software, with details reproduced at a 1:1 scale. Figure 3 As shown.

[0085] It should be noted that step S100 above generates a three-dimensional terrain model by digitally reconstructing the collected substation geographic data and equipment data. This provides a highly realistic and detailed environmental basis for subsequent UAV flight characteristic simulation and trajectory planning, ensuring the accuracy and reliability of the simulation. At the same time, the construction of this three-dimensional model helps to fully understand and evaluate the actual situation of the substation, laying a solid foundation for efficient and safe UAV inspection operations.

[0086] In this embodiment of the application, step S200 includes the following sub-steps B1-B2;

[0087] In B1: The flight characteristics of UAVs in substations are simulated using a 3D terrain model;

[0088] In B2: Physical motion simulation analysis of UAV motion is performed from UAV motion analysis, collision detection, and UAV attitude calculation;

[0089] Specifically, the substation UAV flight characteristic simulation in sub-step B1 includes using the Airsim+Pixhawk flight control hardware platform based on a 3D terrain model to simulate the real physical state of the UAV flight and using a real UAV remote controller for control.

[0090] It should be noted that a realistic 3D model is created using actual drone dimensions, CAD drawings, and prototype photos to reproduce the details of the drone. This model supports mainstream drone models, including the Mavic 2, M210 RTK, Phantom 4, M600, and M300. Based on the 3D model, flight characteristics are simulated using the AirSim platform. AirSim is a cross-platform simulator for drones and other automated mobile devices built on Unreal Engine 4 (UE4). It supports realistic physical and visual simulation of the flight controller. This simulator creates a highly realistic virtual environment, simulating shadows, reflections, and other factors that can interfere with the real world. This embodiment uses AirSim + Pixhawk flight controller semi-physical flight characteristic simulation to simulate the real physical state of drone flight and uses a real drone remote controller for control, closely approximating the feel of operating a real drone. Flight controls cover commonly used quadcopter, hexcopter, and octocopter models.

[0091] It should be noted that the establishment of the UAV model fully considers the forces and motion analysis of the UAV during its movement. This embodiment selects a multi-rotor aircraft as the research and modeling object. The physical characteristics of the UAV mainly involve parameters such as mass, linearity, inertia, coefficient of friction, elasticity, and angular drag coefficient. These parameters are the basis for the force, motion analysis, and attitude estimation of the UAV. In Unreal Engine, the actuator is the source of the forces and torques acting on the UAV. To simplify the calculation, the UAV is abstracted as a cylinder with four vertices and a mass of m, with each vertex at a distance X from its center of mass. i i = {1, 2, 3, 4}, the input angular velocity control signal is u i For i = {1, 2, 3, 4}, the forces and torques acting on the drone are:

[0092] F i =C T σω max D4 u i

[0093]

[0094] Among them, C T and C pow σ represents the thrust coefficient and power coefficient, which are determined by the material of the UAV propeller. σ is the air density, D is the propeller length, and ω is the propeller length. max It is the maximum angular velocity of the propeller per minute;

[0095] Besides the drone's own propulsion force, its movement is also affected by various physical quantities. Gravity, air density, magnetic field, lighting, and visibility all have a significant impact on drone movement. However, this module only models gravity, the Earth's magnetic field, air pressure, and air density. Lighting and visibility do not exert any force on the drone, so they will be set in the UE4 engine and are not included in the drone modeling.

[0096] ① Gravity

[0097] When a drone flies at low altitude, its altitude relative to the Earth's radius is negligible; therefore, the drone's gravitational acceleration is a constant g during low-altitude flight. const =9.8m / s 2 When a drone flies at high altitudes or at relatively high elevations (defined as one-thousandth of the Earth's radius), the change in gravity caused by the drone's altitude is not negligible, and the acceleration due to gravity is:

[0098]

[0099] Among them, R e =6000km represents the Earth's radius, because To reduce computational complexity and eliminate square terms, approximate calculations were performed on quadratic terms.

[0100] ② Magnetic force

[0101] In real-world drones, gyroscopes and accelerometers alone are insufficient to accurately calculate the drone's attitude, including yaw, pitch, and roll angles. Magnetometers, a crucial sensor on drones, are used in the inertial navigation unit, working alongside accelerometers and gyroscopes to determine the drone's heading and velocity vectors relative to the Earth. Therefore, magnetism is a significant factor influencing drone flight. Because modeling the Earth's magnetic field is extremely complex, only approximate values ​​are used in the model. The magnetic field strength is determined by the drone's location, including longitude, latitude, and altitude.

[0102] Given longitude Calculate geomagnetic latitude θ using latitude θ and altitude h. m The formula is shown below. Where θ0 represents the latitude of the North Magnetic Pole in the actual magnetic field. Indicates the longitude of the North Magnetic Pole.

[0103]

[0104] The formula for calculating the total magnetic field strength |B| is shown below. Where represents the average value of the magnetic field at the magnetic equator on the Earth's surface.

[0105]

[0106] The values ​​of the horizontal, vertical, latitude, and longitudinal components of the magnetic field vector were obtained respectively.

[0107] ③ Air pressure and air density

[0108] Air pressure and density have a significant impact on the lift and drag of drones during flight. Air pressure and density change dramatically with altitude, making their simulation crucial. The model uses the US Standard Atmosphere Model at altitudes below 51 km, and the pressure P and temperature T are calculated as shown in the following equations:

[0109]

[0110] Among them, sea level atmospheric pressure P SL =1.033×10 4 kg / m 2 In an atmospheric model, under pressure P and temperature T, the formula for calculating air density is:

[0111]

[0112] Specifically, in sub-step B2 above, the motion of the UAV is analyzed by physical motion simulation, which includes force analysis of the UAV, acceleration of the UAV, and the speed, position, angular velocity and direction of the UAV.

[0113] Specifically, the force analysis of the drone includes:

[0114] Linear drag and angular velocity drag are important factors that change the motion state of a drone during its movement. The direction of linear drag is opposite to the velocity vector v, and the calculation formula is:

[0115]

[0116] Among them, C lin σ is the linear air drag coefficient, A is the cross-sectional area of ​​the UAV, and σ is the air density.

[0117] Angular drag is more complex than linear drag. Assuming the partial angular velocity of the rigid body surface ds is ω, then ds experiences two types of drag torques: frictional torque (generated by the tangential component of the velocity) and torque generated by shear stress. Since shear stress is an internal force acting on the UAV and is independent of its motion, we will primarily analyze the frictional torque. The linear velocity of ds is r. ds ×ω, according to the linear resistance equation, the linear resistance on ds is:

[0118]

[0119] The direction of dF and r ds Conversely, ×ω is calculated by integration to determine the drag torque acting on the surface of the drone:

[0120] τ d =∫r ds ×dF

[0121] In addition to drag and drag torque, the thrust F at the four vertices relative to the drone's center of gravity... i and thrust τ i It is also an important parameter for calculating the motion state of the drone, and thus obtaining the total force and torque F acting on the drone. net for:

[0122] F net =∑ i F i +F d

[0123] Among them, F i F is the thrust at the four vertices relative to the drone's center of gravity. d For linear resistance;

[0124] Specifically, the analysis of drone acceleration includes:

[0125] The total linear acceleration and angular acceleration of the UAV are calculated based on the total forces and torques acting on it. The angular acceleration is given by Euler's rotation equations:

[0126] α=I -1 (τ net -(ω×(Iω))

[0127] Where I is the inertia tensor;

[0128] According to Newton's second law, the linear acceleration can be calculated. By combining the calculated linear acceleration, angular acceleration, and gravitational acceleration g, the actual acceleration 'a' of the UAV can be obtained as follows:

[0129]

[0130] Where m represents the mass of the drone;

[0131] Specifically, the analysis of the drone's speed, position, angular velocity, and direction includes:

[0132] At time t+1, the velocity V is updated by integrating the velocity from the previous time t. t+1 and position p t+1 :

[0133] V t+1 =V t +dt·a t / 2

[0134] p t+1 =p t +dt·v t+1

[0135] At time t+1, the angular velocity ω is updated by integrating the angular velocity from the previous time t. t+1 :

[0136] ω t+1 =ω t +dt·a t / 2

[0137] Orientation updates differ from position updates in that they are not linear. If the orientation is updated using a rotation matrix for each update, slow drift will occur, requiring extensive computation to determine the drone's true orientation. The model transforms the orientation by representing rotation using quaternions. Quaternions only require the rotation axis and rotation angle to represent the rotation state. First, the angle-axis pair (a...) is calculated. dt ,u), where a dt It is the angular velocity of rotation around the unit vector u. Then...

[0138] a t+1 =|ω|·dt

[0139] u=ω / |ω|

[0140] The change in direction is represented by the change of quaternion over time dt:

[0141] q t+1 =q t ·q dt

[0142] It should be noted that, based on the force analysis of the UAV, six parameters of the UAV's motion state can be obtained: position, direction, linear velocity, linear acceleration, angular velocity, and angular acceleration. Applying this physical model to the physics engine can present a motion state in the system that is similar to that of a real UAV.

[0143] Specifically, in sub-step B2 above, the physical motion simulation analysis of the UAV's motion based on collision detection includes each entity in the collision detection system having a collision attribute. The collision attribute determines whether the entity collides with other entities. If a collision or overlap event occurs during the rendering interval, the collision detection system calls the collision response, returning the collision position, collision type, and input depth. By calculating the returned data, the motion state of the UAV is modified according to the calculation results.

[0144] It should be noted that Unreal Engine provides a collision detection system for different types of collisions to simulate drone flight by analyzing and calculating the drone's flight state. This system mainly consists of collision detection, collision response, and trajectory response. During the rendering interval, if a collision or overlap event occurs, the Unreal Engine collision detection system will invoke the collision response, returning the collision location, collision type, and input depth. By calculating the returned data, the drone's motion state is modified according to the calculation results.

[0145] It should be noted that in Unreal Engine, each entity has a Collision property, which determines whether the entity collides with other entities. There are three main Collision properties: No Collision (the entity will not collide with other entities), No Physical Collision (the entity can only be used for raycasting, sweeping, and overlapping), and Collision Enabled (the entity can be used for physics simulation and collision detection). Because forces are mutual, a collision will only occur if both or more colliding entities have the Collision property; otherwise, the entities will not collide, and the visual effect will be that two entities overlap or pass through each other during movement.

[0146] It should be noted that in Unreal Engine, each entity with collision properties has its own channel type. Collisions between different entities are distinguished by these collision channels. Commonly used collision object channels include WorldDynamic, Worldstatic, Pawn, and Character. The WorldDynamic channel type is often used for "moving" objects, such as drones and birds; while the Worldstatic channel type is often used for "stationary" objects, such as the ground and trees. Entities can be configured to collide with other entities of different channel types using collision presets. For example, if the ground entity has a Worldstatic channel and its collision preset is BlockAll (collides with all objects), and the drone is selected in its collision response list, then when the drone comes into contact with the ground, a collision will occur, and the drone's motion state will be changed according to the collision response, returning the collision result.

[0147] Specifically, the physical motion simulation analysis of the UAV's motion based on the UAV attitude calculation in sub-step B2 above includes analysis from two aspects: UAV attitude estimation and UAV attitude control.

[0148] It should be noted that quaternions are used to store the UAV attitude in the geographic coordinate system. When input into the UAV control algorithm, it is converted into Euler angles, and the pitch angle (θ), yaw angle (ψ), and roll angle (φ) are used to represent the UAV attitude relative to the ground, such as... Figure 2 As shown, the pitch angle represents the drone's rotation around the x-axis, with the drone's nose pointing upwards being positive; the yaw angle represents the drone's rotation around the y-axis, with the drone's nose yawing to the right being positive; and the roll angle represents the drone's rotation around the z-axis, with the drone's nose rolling to the right being positive.

[0149] Specifically, the UAV attitude estimation uses IMU sensor data, employs the Mahony complementary filtering algorithm combined with a PID feedback controller to compensate for and correct gyroscope errors, and uses quaternions to describe the attitude.

[0150] q = a + bi + cj + dk

[0151] Where a, b, c, d are real numbers, and i, j, k are mutually orthogonal unit vectors. The quaternions are initialized as follows:

[0152] q0 = 1, q1 = 0, q2 = 0, q3 = 0

[0153] The acceleration data is normalized to obtain the angular velocities ω of the three axes of the sensor. x ,ω y ,ω z acceleration a x ,a y ,a z The acceleration is normalized.

[0154]

[0155] Error vector calculation is performed based on accelerometer and gyroscope data, and the error is calculated through vector product:

[0156] e = a norm ×g

[0157] Where g is the geographic gravity vector transformed into the body coordinate system.

[0158] Update the quaternion using the processed gyroscope data:

[0159]

[0160] Among them, h T=0.001 represents the time step, and Ω represents the angular velocity matrix.

[0161] Specifically, the UAV attitude control employs a nested PI-PID controller, which adjusts the propeller angular velocity to control the pitch, yaw, and roll angles. The UAV's state equation can be expressed as:

[0162]

[0163]

[0164] Where θ is the pitch angle, ψ is the yaw angle, φ is the roll angle, and τ is the pitch angle. θ ,τ ψ ,τ φ Let u be the stress torque, and u be the thrust.

[0165] The PID controller is responsible for the smooth control of the UAV's attitude. It is used to adjust the rate of change of attitude angle, and adjusts the system gain by comparing the filtered attitude angle with the desired angle.

[0166] Furthermore, since the drone operates in a simulated training environment, it lacks real sensors and therefore cannot acquire drone flight status data. Thus, it is necessary to model the sensors within the simulated environment to obtain the flight status and data from the simulated training environment.

[0167] The sensor module mainly consists of five parts: GPS, barometer, magnetometer, accelerometer, and gyroscope. Since sensor data acquisition in actual training involves certain errors, in order to make the sensor data more closely resemble reality, the modeling process not only models the sensor data but also simulates the sensor errors.

[0168] GPS: Provides location data, incorporating horizontal and vertical errors during simulation.

[0169] Barometer: Measures atmospheric pressure based on Torricelli's principle and converts it into altitude.

[0170] Magnetometer: measures magnetic field components and is used for attitude calculation.

[0171] Accelerometers and gyroscopes: Core components of the IMU, responsible for calculating the linear acceleration and angular velocity of the UAV, and the output is used for attitude estimation.

[0172] GPS error simulation is mainly used for horizontal and vertical errors, and the errors gradually decrease over time.

[0173] The barometer simulates the relationship between air pressure and altitude. The air pressure value changes with altitude, and the calculation formula is as follows:

[0174]

[0175] Where p0, L, h, T0, g, and R are the standard atmospheric pressure at sea level, temperature gradient, altitude, reference temperature, gravitational acceleration, and gas constant, respectively.

[0176] Furthermore, the system supports multiple deployment methods and can run on Windows systems:

[0177] Supports video streaming deployment: UE pixel streaming technology is used to deploy on the teacher's computer (high-performance computer), execute system logic and render each frame, supporting a teaching method where the teacher operates the computer and students observe and learn.

[0178] Supports independent deployment: It supports independent deployment for teachers and students, and can run without the need for third-party support.

[0179] It should be noted that the above step S200 can not only predict and avoid potential flight risks in advance, but also optimize the flight path and operation strategy of the UAV, ensuring the flight safety and operational efficiency of the UAV in complex environments. At the same time, through precise physical simulation analysis, it can effectively reduce the number of field tests, save costs, and accelerate the development and deployment process of the UAV inspection system.

[0180] In this embodiment, step S300, based on physical motion simulation analysis, plans the UAV inspection trajectory, generates the UAV flight path, and performs simulated flight to realize the simulation of UAV flight in the substation, including:

[0181] Because substations have high voltage levels and numerous energized devices, the electric field environment for live-line work is more complex than that of conventional transmission lines. Therefore, the substation drone simulation inspection module was developed according to the standards for real substation drone simulation inspections. Autonomous drone inspections require first setting flight areas for the substation based on voltage levels, and then sequentially inspecting the substation structure, lightning rods, HGIS equipment, etc. Safe take-off and landing positions are set for each area, and flight routes are clearly defined. During drone flight, it is important to avoid exceeding the performance specifications. The flight speed during inspection should not exceed 10 m / s. Flight over densely populated equipment areas such as the low-voltage side of the main transformer should be avoided. If it is necessary to cross the lead wire, the crossing method should be adopted, and the clearance distance between the drone and the uppermost lead wire should not be less than 10 m. The drone should not hover directly above electrical equipment for an extended period of time. The main purpose is to combine drone flight parameters, camera parameters, 3D terrain data, and tower coordinate data to calculate the coordinates of the waypoints, aircraft attitude, and gimbal angle, etc., to provide single-point, single-device, and global waypoint corrections at all levels. It has spatial collision detection and automatic avoidance capabilities, and provides standard, open-format flight path file outputs to provide strategic guidance for actual flight.

[0182] Specifically, within the inspection area, individuals in the population are transformed into waypoints, each waypoint represented by its distance and orientation angle:

[0183] Chromosome={(r1,θ1),(r2,θ2),...,(r n ,θ n )}

[0184] Where, r i θ is the distance between the current waypoint and the reference point. i It is the polar angle relative to the reference point. Polar coordinate system encoding has unique advantages in handling angle changes and rotation problems in trajectory planning. When a UAV flies around an obstacle, the trajectory direction can be easily changed by adjusting the angle θ, thereby avoiding the obstacle.

[0185] The design of the fitness function should comprehensively consider various factors in UAV trajectory planning to ensure that the path optimizes the objective as much as possible while meeting mission requirements. The fitness function should be designed by combining the planning mission requirements with consideration of path length, obstacle avoidance performance, and smoothness. Describe the length of the flight path, where P i D(P) represents the waypoint. i ,P i+1 The distance between two consecutive waypoints is represented by the distance between the waypoints. The penalty design is based on the safe distance from the obstacle. The smoothness of the flight is measured by calculating the change in the heading angle between consecutive waypoints.

[0186] It should be noted that the formula for the penalty design is expressed as follows:

[0187]

[0188] It should be noted that the fitness value is low when the flight path changes frequently. Where, θ i This represents the direction angle of the i-th track point.

[0189] The fitness function F, which combines path length, obstacle avoidance performance, and smoothness, is expressed as:

[0190]

[0191] Among them, w i Here are the weight values ​​for each reference, where s represents the smoothness of flight and p represents the obstacle avoidance capability of flight.

[0192] Furthermore, assume that the set of points that the drone needs to inspect is X = {X1, X2, ..., X...} n}, where X n This represents an inspection point. The inspection process plans a path to visit all inspection points sequentially and return to the starting point. Calculate the X values ​​for two inspection points.i ,X j The Euclidean distance between them is:

[0193]

[0194] Where x i y i Indicates inspection point X i coordinate.

[0195] Perform a full permutation (n-1!) of all inspection points, exhaustively list all inspection paths, and calculate the inspection process and the total distance D returning to the origin for each path. Xi =d 12 +d 23 ,...,d n1 The path with the shortest distance is selected as the optimal inspection path.

[0196] After generating the flight path, parameters such as the drone's flight altitude, flight speed, flight distance, and gimbal angle can be set in the simulation scenario. The drone will then perform simulated flight within the scenario according to the planned flight path, ensuring the accuracy of the planned flight path.

[0197] It should be noted that the above step S300 not only ensures the safety and rationality of the flight route and identifies and resolves potential flight obstacles and risks in advance, but also optimizes the flight path, improves the efficiency and quality of the inspection, and allows for the rehearsal of the UAV's operation procedures through simulated flight, helping operators become familiar with the flight mission, reducing errors and uncertainties in actual operation, and comprehensively improving the safety and reliability of UAV inspection operations.

[0198] Furthermore, this embodiment also includes basic drone flight skills training, covering fundamental knowledge of drone operation, flight skills, laws and regulations, safe operation, and practical applications. It designs AOPA flight skills such as drone takeoff, landing, hovering, figure-eight flight, and quadruple path maneuvers. This meets the training needs of students, from basic flight skills training to advanced flight skills training and AOPA exam preparation. Students can complete their training on the job, improving training efficiency while significantly reducing drone crashes caused by training errors.

[0199] like Figure 5 The image shows a horizontal figure-eight flight: Horizontal figure-eight flight is a complex maneuver that requires precise control of the drone's speed, direction, and turning angle. This flight mode requires the drone to make continuous turns, forming two intersecting arcs, with the overall path resembling the number "8".

[0200] like Figure 6The image shows single-channel hovering: GPS flight mode / attitude stabilization mode, with switchable pitch, aileron, and single-channel four-sided hovering (supports one-click switching between tail-to-tail flight training, head-to-head flight training, left-side flight training, and right-side flight training).

[0201] like Figure 7 The image shows dual-channel hovering: GPS flight mode / attitude stabilization mode, with switchable pitch, aileron, and dual-channel four-sided hovering (supports one-click switching between tail-to-tail flight training, head-to-head flight training, left-side flight training, and right-side flight training).

[0202] like Figure 8 The image shows AOPA: It has AOPA flight skills such as drone take-off, landing, hovering, figure-eight flight, and quadruple path, supports flight training track display and hiding functions and can be cleared with one click, and has the ability to display training results such as flight training time and number of successes.

[0203] This embodiment also includes the function of enabling drones to conduct intelligent inspections and identify defects in substation equipment, such as... Figure 9 The operator manipulates a simulated drone to identify various equipment defects within a substation using cameras and sensors, completing training for drone-based equipment defect inspections. Simultaneously, by equipping the drone with infrared and visible light detection devices to capture equipment inspection images, a defect image recognition model identifies existing defects. Deep learning algorithms classify the detected defects, and the system automatically generates inspection reports based on the inspection results, along with new training models and plans. Operators can also manually select defect modes for training. The built-in dynamic defect library includes common defects encountered during power line inspections, such as bird nests, insulator spontaneous explosions, foreign objects on towers, tower corrosion, tilted phase sequence plates, detached phase sequence plates, severely contaminated insulators, hanging objects, detached pins, tilted equipotential rings, displaced vibration dampers, detached vibration dampers, deformed vibration dampers, and loose bolts—common defects in transmission lines. It also includes common substation defects such as loose, detached, and broken bolts on high-altitude equipment like substation structures, lightning rods, busbars, and lead wires, which can randomly appear during inspections, providing operators with a realistic power line inspection scenario.

[0204] As described above, this invention provides a method and system for simulating the flight of unmanned aerial vehicles (UAVs) in substations. Through high-precision 3D modeling technology, it realistically recreates the substation environment. The simulation system can achieve real-time data transmission and display, providing operators with a highly realistic training platform. Inspection reports can be automatically generated in the simulation environment, significantly improving inspection efficiency. Simultaneously, the built-in dynamic defect library supports flexible settings for common equipment defects and allows batch import / export, ensuring the diversity and relevance of training content. Full-scenario simulation assessment helps operators comprehensively improve their UAV inspection capabilities, ensuring efficient and accurate task completion in actual work.

[0205] Example 2

[0206] This embodiment provides a substation UAV flight simulation system, including:

[0207] The terrain model generation module is used to digitally reconstruct a three-dimensional terrain model from the collected substation geographic data and equipment data.

[0208] The simulation analysis module is used to simulate the flight characteristics of UAVs in substations using a 3D terrain model, and to perform physical motion simulation analysis of UAV motion from UAV motion analysis, collision detection and UAV attitude calculation.

[0209] The flight simulation module is used to plan the flight path of UAV inspection based on physical motion simulation analysis, generate the flight path of UAV and conduct simulated flight to realize the simulation of UAV flight in substations.

[0210] It should be noted that the technical solution of the substation drone flight simulation system is based on the same concept as the technical solution of the substation drone flight simulation method described above. For details not described in detail in the technical solution of the substation drone flight simulation system in this embodiment, please refer to the description of the technical solution of the substation drone flight simulation method described above.

[0211] The above-mentioned unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or they can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0212] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of this computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for simulating the flight of a substation drone. The display screen of the computer device can be a liquid crystal display screen or an e-ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0213] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method proposed in the above embodiments.

[0214] The storage medium proposed in this embodiment belongs to the same inventive concept as the method proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0215] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0216] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0217] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0218] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0219] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0220] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0221] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0222] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for simulating the flight of a UAV in a substation, characterized in that, include: The collected geographic data and equipment data of the substation are digitally reconstructed to generate a three-dimensional terrain model; The flight characteristics of the UAV in the substation were simulated using the three-dimensional terrain model, and the physical motion simulation analysis of the UAV was performed from the perspectives of UAV motion analysis, collision detection and UAV attitude calculation. Based on the physical motion simulation analysis, UAV patrol trajectory planning is performed, UAV flight routes are generated and simulated flight is conducted to realize the simulation of UAV flight in substations; Physical motion simulation analysis of drone motion from a collision detection perspective includes: In the collision detection system, each entity has a collision attribute, which determines whether the entity collides with other entities. If a collision or overlap event occurs during the rendering interval, the collision detection system calls the collision response, returns the collision position, collision type, and input depth, and modifies the motion state of the UAV based on the calculation results by calculating the returned data. The planning of drone patrol routes includes: Within the inspection area, individuals in the population are transformed into waypoints, and each waypoint is represented by its distance and orientation angle. pass Describe the length of the flight path, where P i D(P) represents the waypoint. i ,P i+1 The distance between two consecutive waypoints is represented by the distance between the waypoints. The penalty design is based on the safe distance from the obstacle. The smoothness of the flight is measured by calculating the change in the heading angle between consecutive waypoints. The fitness function F, which combines the path length, obstacle avoidance performance, and smoothness, is expressed as: Among them, w i Here are the weight values ​​for each reference, where s represents the smoothness of flight and p represents the obstacle avoidance capability of flight. Suppose the set of points that the drone needs to inspect is X = {X1, X2, ..., X...} n }, where X n This represents an inspection point. The planned path during the inspection process visits all inspection points sequentially and returns to the starting point. Perform a full permutation of all inspection points (n-1)!, exhaustively list all inspection paths, calculate the inspection process and total distance to return to the origin for each path, and select the path with the shortest distance as the optimal inspection path.

2. The substation UAV flight simulation method as described in claim 1, characterized in that, The process of digitally reconstructing the collected substation geographic data and equipment data to generate a three-dimensional terrain model includes: The raw point cloud data of the substation's precise geography and equipment is collected using UAV lidar. Different categories of candidate point clouds are labeled using clustering methods to separate ground feature point clouds from non-ground feature point clouds. The raw point cloud data is then subjected to noise filtering and enhancement. By analyzing the three-dimensional coordinates and morphological features of point cloud data, the orientation of the terrain and the distribution of ground features can be determined. The selected point cloud data, combined with its spatial location, is used to create a 3D model of the power lines and the entire Earth surface. In 3D terrain modeling, a 3D terrain model is generated based on triangulation and Poisson reconstruction algorithms, and the 3D terrain model is then smoothed.

3. A system applying the substation UAV flight simulation method as described in any one of claims 1 to 2, characterized in that, include: The terrain model generation module is used to digitally reconstruct a three-dimensional terrain model from the collected substation geographic data and equipment data. The simulation analysis module is used to simulate the flight characteristics of the substation UAV using the three-dimensional terrain model, and to perform physical motion simulation analysis of the UAV's motion from the perspectives of UAV motion analysis, collision detection and UAV attitude calculation. The flight simulation module is used to plan the flight path of UAV inspection based on physical motion simulation analysis, generate the flight path of UAV and conduct simulated flight to realize the simulation of UAV flight in substations.

4. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the substation UAV flight simulation method according to any one of claims 1 to 2.

5. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the substation UAV flight simulation method according to any one of claims 1 to 2.

Images