Unmanned aerial vehicle digital twin modeling method based on real-time data driving

By constructing a digital twin model of the main wing spars of a drone and combining it with Ansys and Unity3D for simulation and visualization, the problem of the single drone modeling method was solved, and real-time monitoring and management of the drone status was realized, reducing the equipment damage rate.

CN116150886BActive Publication Date: 2026-02-27XIDIAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310141233.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-20
Publication Date
2026-02-27
Estimated Expiration
2043-02-20

AI Technical Summary

Technical Problem

Traditional UAV modeling methods are limited and cannot achieve multi-dimensional mapping of UAV digital models. This results in the inability to transmit real-time data of physical entities to the model in a timely manner, and a lack of interaction and integration between the model and the data.

Method used

A real-time data-driven digital twin modeling method for UAVs was adopted to construct a digital twin model of the wing main beam from four dimensions: geometric model, physical model, behavioral model, and rule model. The model was verified by Ansys simulation software, visualized by virtual animation using Unity3D, and stored and persisted in real time using a MySQL database.

Benefits of technology

It enables real-time visualization of the deformation attitude of UAV wings, reduces the damage rate of UAV equipment, and improves the real-time performance and accuracy of UAV status monitoring and management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116150886B_ABST
    Figure CN116150886B_ABST
Patent Text Reader

Abstract

The application discloses a kind of unmanned aerial vehicle digital twin modeling methods based on real-time data driving, it includes: from the four dimensions of geometry model, physical model, behavior model and rule model, construct wing girder digital twin model;The wing girder digital twin model is simulated, and the physical space real-time state of unmanned aerial vehicle is mapped to virtual space;Real-time display the deformation posture of unmanned aerial vehicle wing using virtual animation visualization method, to provide visual basis for operator to know the health status of unmanned aerial vehicle.The application uses digital twin technology to monitor the state of unmanned aerial vehicle, simulates the working state of physical entity in virtual space, assists technical personnel to make decisions, reduces the probability of occurrence of various accidents of unmanned aerial vehicle, with real-time, high fidelity and high inheritance.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of modeling, in particular to a real-time data-driven unmanned aerial vehicle (UAV) digital twin modeling method. BACKGROUND

[0002] An unmanned aerial vehicle, referred to as a UAV for short, is an unmanned aircraft that is controlled by using radio remote control equipment and self-provided program control devices or is completely or intermittently autonomously operated by a vehicle-mounted computer. Compared with manned aircraft, the unmanned aerial vehicle has the advantages of small size, low cost, convenient use, low requirement for the operating environment and strong battlefield survival capability. After years of research and practice of unmanned aerial vehicles, the unmanned aerial vehicle has gradually matured and is widely used in fields such as police, energy, land resources, water conservancy, agriculture, forestry, electricity, business, medical treatment, disaster prevention and reduction, entertainment and logistics.

[0003] From the currently disclosed data, the research on the unmanned aerial vehicle digital twin modeling method is relatively insufficient. For example, Su Ben designed and developed a UAV digital simulation GUI interactive environment and simulated the take-off and landing and waypoint flight of the UAV, simplifying the UAV modeling process, but the reliability of the simulation data is not high. Ji Guang et al. took the flight process of a quadrotor UAV as the task requirement, proposed a quadrotor UAV flight process simulation research based on digital twin technology, and fused and constructed a digital twin model of the quadrotor UAV from the aspects of geometry, physics, behavior and rules, verifying the interaction between the virtual and the real, the accuracy of the model and the feasibility of the system, but there are difficulties in the data fusion process. Liu Yan et al. designed a UAV hardware-in-the-loop flight simulation platform based on digital twin technology to improve the research and development efficiency of flight control algorithms and reduce the research and development cost. The platform meets the real-time and stability requirements of flight simulation and has important application value for subsequent control algorithm optimization and complex application scenario flight tests. Shi Baiyu et al. took a quadrotor UAV as the research object based on the Unity physical engine, established a mathematical model of the UAV operation mode and conducted simulation model research. The model meets the reality of simulation and simplifies the modeling process, but does not consider the physical properties of motors, electronic speed controllers and batteries, and the aerodynamic resistance of propellers. The dSPACE real-time simulation system is a control system development and semi-physical simulation software and hardware work platform based on MATLAB / Simulink developed by the German dSPACE company, which realizes complete seamless connection with MATLAB / Simulink / RTW. Domestic universities and relevant research units have done a lot of research and development work on UAV real-time simulation systems based on dSPACE and other real-time simulation devices, but such platforms are very expensive and have a long development cycle.

[0004] In the actual working process, the unmanned aerial vehicle will inevitably produce structural deformation and even damage. In order to reduce the damage rate of the unmanned aerial vehicle, a mapping system is needed to reflect the equipment state of the unmanned aerial vehicle in real time. The unmanned aerial vehicle state monitoring needs to have real-time performance, so as to display the state to the ground technical personnel in the most intuitive visual way within the shortest data signal delay range, and ensure the effectiveness of the ground-air information interaction. The traditional unmanned aerial vehicle modeling method is single, cannot realize unmanned aerial vehicle digital model mapping from multiple dimensions, and causes that the real-time data of the physical entity cannot be transmitted to the model in time, and the interactive fusion of the model and the data is lacked. SUMMARY

[0005] In order to solve the problem that the traditional unmanned aerial vehicle modeling method is single, cannot realize unmanned aerial vehicle digital model mapping from multiple dimensions, and causes that the real-time data of the physical entity cannot be transmitted to the model in time, and the interactive fusion of the model and the data is lacked, the present application provides a kind of unmanned aerial vehicle digital twin modeling method based on real-time data driving to solve the above technical problems.

[0006] The application discloses a kind of unmanned aerial vehicle digital twin modeling method based on real-time data driving, it includes the following steps:

[0007] Step 1: construct wing spar digital twin model from the four dimensions of geometric model, physical model, behavior model and rule model;

[0008] Step 2: simulate the wing spar digital twin model, and map the real-time state of the physical space of the unmanned aerial vehicle to the virtual space;

[0009] Step 3: real-time display the deformation posture of the unmanned aerial vehicle wing using virtual animation visualization method, to provide visual basis for operator to know the health status of the unmanned aerial vehicle.

[0010] Further, in step 1:

[0011] In the geometric model, Solidworks three-dimensional drawing software is used to draw the geometric model;Based on the symmetry characteristics of the wing, only half of the wing spar model is constructed.

[0012] Further, in step 1:

[0013] In the physical model, the unmanned aerial vehicle adopts traditional aerodynamic layout, and when analyzing the load and internal force of the wing, the external load it bears is air dynamic force, wing structure mass force and other components and external concentrated load;

[0014] The air dynamic force load is a distributed load, and the air load acts on the entire wing surface, forming the lift and drag of the entire wing surface, wherein the lift part is the main external load of the wing surface;

[0015] The mass force of the wing structure itself consists of the gravity and inertial force of the wing structure. The two act on the wing structure entity, and their distribution range depends on the mass distribution law of the structure itself. The distribution law of the mass force is approximately proportional to the chord length.

[0016] The loads on other components fixedly connected to the wing are transmitted to the wing in the form of concentrated forces through connectors; wherein, the other components include solar panels, landing gear, and external stores;

[0017] During the transfer of load from the wing to the fuselage, the internal forces generated inside the wing include: shear force S. n S h Bending moment B n B h and torque B t S n ≥S h B n Greater than or equal to B h The structure's width and moment of inertia are very large within the chord plane; therefore, the approximate analysis only considers S. n B n and B t Abbreviated as S, B and B t ;

[0018] The shear force inside the wing is expressed by the following formula:

[0019]

[0020] Where z is the distance integrated from the wingtip towards the wing root, and q = q a -q c q a For aerodynamic distributed load; q c For the mass force distribution load of the wing structure, P i Concentrated loads transmitted from the engine or other components;

[0021] The bending moment inside the wing is expressed by the following formula:

[0022]

[0023] The airfoil torque caused by distributed load is expressed by the following formula:

[0024]

[0025] Where, m t =q a e+q c d is usually taken as the axis of the rigid center as the relative axis, e is the distance between the center line and the relative axis, and d is the distance between the center line and the relative axis.

[0026] Further, in the step 1:

[0027] In the behavior model, the attitude, position and speed in the attitude controller, position controller and speed controller are analyzed respectively, and the analyzed attitude, position and speed are compared with the actual measured signal values of the sensors for error analysis. Each type of controller is analyzed and corrected, and the central executive mechanism adjusts the wing attitude to ensure the flight control accuracy of the unmanned aerial vehicle.

[0028] Further, in the step 1:

[0029] In the rule model, after receiving the task, the unmanned aerial vehicle first rises in height, and then goes to the task area after reaching the set height. During the lifting and flat flying stage, the structural stability of the machine body needs to be judged to determine whether the flight attitude needs to be adjusted or the task plan needs to be changed. If the structure is stable, it is also necessary to judge whether the sunlight radiation in the flight area is sufficient, that is, whether enough energy can be obtained to support the task execution. If the sunlight radiation in the task area is insufficient, the route needs to be adjusted to obtain sufficient energy reserves. If the structure is stable and the sunlight radiation in the flight area is sufficient, the task will be continued.

[0030] Further, the simulation of the wing main beam digital twin model comprises:

[0031] The constructed three-dimensional model of the wing main beam is imported into the Ansys simulation software, various information is added according to the performance parameters of the unmanned aerial vehicle, a virtual simulation environment is constructed, and a simulation experiment is performed to determine whether the error between the stress of the model main beam and the stress measured by the physical entity sensor is small enough; wherein, the various information includes material information and load information.

[0032] Further, the simulation of the wing main beam digital twin model specifically comprises:

[0033] Step 21: Preprocessing stage, using the PCP module of workbench for preprocessing modeling;

[0034] Step 22: Post-processing stage, using the Static Structural module for mechanical analysis;

[0035] Step 23: Analyze the twin model according to the results of the deformation cloud diagram obtained in step 22 to determine whether the error between the stress of the main beam and the stress measured by the physical entity sensor is small enough. If the error is small enough, it indicates that the twin model can replace the physical entity.

[0036] Further, the step 21 comprises:

[0037] Carbon fiber / epoxy composite material is used, and material information is added according to its performance parameters to establish an equivalent thin-walled cylindrical model.

[0038] Setting constraints and dividing the grid, including setting the layer angle: setting according to the specified angle; defining the direction selection set; layering the carbon fiber according to the previously set carbon fiber direction, which is along the axial direction of the cylinder;

[0039] The step 22 comprises:

[0040] The model is regarded as a cantilever beam, one end of which is constrained, and then a load is applied, the wind load being treated as a static load, and a static force analysis is adopted; the wind load belongs to a surface load, and in the modeling process, the equivalent force of the wind load is calculated and applied at the pre-set force point of the structure; finally, the overall strain nephogram of the thin-walled cylinder is solved, and the deformation nephogram at the position of the strain gauge is drawn.

[0041] Further, the virtual animation visualization method comprises:

[0042] Step 31: Maya three-dimensional model construction: model construction is carried out by using Maya software, the model is converted into an.fbx file format for use in a Unity 3D environment; a wing deformation posture representation method based on bone displacement control is designed, bones are added at the positions where strain sensors are installed on the wing, sensor data is mapped to bone displacement to reflect the overall deformation of the wing; wherein the sensor data is stored and read in real time by a MySQL database;

[0043] Step 32: model node loading in Unity 3D: the.fbx model file is loaded into the Models folder of the Unity 3D scene; after the wing model file is loaded into the scene, the displacement of each bone is controlled by a c# script language;

[0044] Step 33: implementation of wing deformation synchronous display function: in actual wing posture changes, the bones of the wing main beam are bound through parent-child relationship; that is, the child object moves with the parent object, that is, the child object inherits the parent object, but the movement of the child object itself will not affect the movement state of the parent object, that is, the parent object will not inherit the movement of the child object; according to the symmetry of the wing, the middle bone is set as the parent object, and the parent-child level extends to both sides in turn;

[0045] Step 34: implementation of wing stress nephogram rendering effect: the wing main beam is rendered by using a Shader, the stress value in the database is read to define the shader value parameter, and when the Unity 3D performs shader rendering, the stress value is read in real time to perform color rendering.

[0046] Further, the data involved in the application of the steps 1 to 3 are all stored and persisted in real time by the MySQL database; in the MySQL database, each function corresponds to a primary table and a secondary table, the primary table is used for real-time data reading, only one row, the data is deleted after reading, and the data is persisted in the secondary table, and the data reading and display are realized by using echarts.

[0047] Due to the adoption of the above technical solutions, the present application has the following advantages:

[0048] 1. The present application constructs a digital twin model of the unmanned aerial vehicle from four dimensions of geometric model, physical model, behavior model and rule model, has better twin performance; in the aspect of visualization, the present application uses Unity3D engine, and designs a virtual animation visualization method, which can display the deformation posture of the unmanned aerial vehicle wing in real time, and provides direct and effective visual basis for the health state of the unmanned aerial vehicle for the operator; the data involved in the application are all stored and persisted in real time by the MySQL database, and have real-time performance.

[0049] 2. The present application realizes the mapping of the key parts of the unmanned aerial vehicle, does not need to model the whole machine, can reflect the equipment state of the unmanned aerial vehicle, and realizes real-time state monitoring and management, thereby reducing the equipment damage rate of the unmanned aerial vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced, and obviously, the drawings in the following description are only some embodiments described in the embodiments of the present application, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0051] Figure 1 It is a schematic diagram of a real-time data driven unmanned aerial vehicle digital twin modeling method of the embodiment of the present application;

[0052] Figure 2 It is a schematic diagram of the wing posture closed loop control loop of the embodiment of the present application;

[0053] Figure 3 It is a rule model flowchart of the embodiment of the present application;

[0054] Figure 4 It is a visualization technology implementation flowchart of the embodiment of the present application. DETAILED DESCRIPTION

[0055] The present application is further illustrated with reference to the accompanying drawings and examples. It will be apparent, however, that the described examples are only a small number of the examples of the present application, and are not all-inclusive of all examples of the present application. All other examples that can be derived by a person of ordinary skill in the art from the present disclosure, should be considered as falling within the scope of the present application.

[0056] With the development of science and technology, we are facing high-difficulty, high-risk and high-content tasks that humans cannot handle. Unmanned aerial vehicles (UAVs) have emerged as the times require. They replace manned aircraft to perform these tasks. They can perfectly utilize cutting-edge technologies such as artificial intelligence, signal processing and autonomous driving. Due to their small size, unmanned driving, long range and other advantages, they are widely used in natural environment investigation, popular science research, agriculture, maintenance of national sovereignty and public health safety and many other aspects.

[0057] However, due to the long flight time of UAVs, the changeable operation environment and other factors, structural deformation and even damage will inevitably occur during operation. Therefore, it is urgent to develop a mapping system to reflect the equipment state of the UAV in real time. The present application constructs a digital twin model of the UAV, simulates the working state of the physical entity in the virtual space, can realize the state monitoring, life prediction and other functions of the UAV, reduces the occurrence rate of various accidents of the UAV, and has far-reaching significance for the state monitoring and management of the UAV.

[0058] Referring to Figure 1 , the present application provides an embodiment of a UAV digital twin modeling method based on real-time data driving, which first constructs a wing spar digital twin model from four dimensions of geometric model, physical model, behavior model and rule model. Then, the Ansys software is used to simulate the three-dimensional twin model, and the real-time state of the physical space of the UAV is mapped to the virtual space. Finally, a virtual animation visualization method is designed by using the Unity3D software, which can display the deformation posture of the UAV wing in real time, and provide direct and effective visualization basis for the health state of the UAV for the operator. The data involved in the application are completed by means of the MySQL database.

[0059] (1) Constructing a wing spar digital twin model

[0060] Since the wing spar of the UAV can well reflect its deformation state, the present embodiment constructs a digital twin model for the wing spar of the UAV, and constructs a wing spar digital twin model from four dimensions of geometric model, physical model, behavior model and rule model.

[0061] (1a) Geometric model

[0062] Since the digital twin seeks precision in the construction of the geometric model, it requires that the appearance, size, and material type be consistent with the physical entity. The geometric model is designed and drawn with the aid of Solidworks three-dimensional drawing software. Due to the characteristics of large wingspan and large aspect ratio of the UAV wing, from the accuracy and effectiveness of model analysis, based on the symmetry characteristics of the wing, this embodiment only constructs a half-wing spar model.

[0063] (1b) Physical model

[0064] When analyzing the load and internal force of the UAV wing, the external load acting on the wing is the air force, the wing structure mass force, and the concentrated load of other components and external attachments.

[0065] Air force:

[0066] The air force load is a distributed load, and the air load acts on the entire wing surface, forming the lift and drag of the entire wing surface, of which the lift part is the main external load of the wing surface.

[0067] Wing structure mass force:

[0068] The mass force of the wing structure itself is composed of the gravity and inertia force of the wing structure, which acts on the wing surface structure entity, and the distribution range depends on the mass distribution law of the structure itself. The mass force distribution law is approximately proportional to the chord length.

[0069] Wing structure mass force and other components and external attachments:

[0070] The load acting on the wing fixedly connected with other components such as solar panels, landing gear, and external attachments will be transmitted to the wing in the form of concentrated force through the connecting piece.

[0071] Wing internal force:

[0072] During the transmission of the load on the wing to the fuselage, the internal forces caused in the wing include shear force S n , S h , bending moment B n , B h , and torque B t . Generally, S n ≥ S h , B n is greater than or equal to B h . Since the structure width and moment of inertia in the chord plane are large, the approximate analysis usually only considers S n , B n , and B t , which are simply written as S, B, and B t .

[0073] The shear force inside the wing is represented by the following equation:

[0074]

[0075] where z is the distance integrated from the wing tip to the wing root, q = q a -q c , q a is the aerodynamic distributed load; q c is the wing structure mass force distributed load, P i is the concentrated load from the engine or other components;

[0076] The bending moment inside the wing is represented by the following equation:

[0077]

[0078] The wing section torsion caused by the distributed load is represented by the following equation:

[0079]

[0080] where m t = q a e + q c d, the centroid axis is usually taken as the relative axis, e is the distance between the center of pressure line and the relative axis, and d is the distance between the center of gravity line and the relative axis.

[0081] (1c) Behavioral model

[0082] Figure 2 The unmanned aerial vehicle attitude closed-loop control loop schematic diagram is shown, the input is the unmanned aerial vehicle set flight attitude, position and speed signal value, including wing deformation, flight pitch angle, height and speed signal, etc., by comparing the error with the sensor measured signal value, the behavior is analyzed and corrected by each type of controller, and the wing attitude is adjusted by the central executive mechanism, which ensures the flight control accuracy of the unmanned aerial vehicle.

[0083] (1d) Rule model

[0084] Figure 3 The unmanned aerial vehicle rule model flow chart is shown, after receiving the task, the unmanned aerial vehicle first lifts the height, and then goes to the task area after reaching the set height, in the lifting and flat flying stage, it is necessary to judge the structural stability of the fuselage, to decide whether to adjust the flight attitude or change the task plan, if the structure is stable, it is also necessary to judge whether the sunlight radiation in the flight area is sufficient, that is, whether it can obtain enough energy to support the task execution, if the sunlight radiation in the task area is insufficient, it is also necessary to adjust the route to obtain enough energy reserve, if the above conditions are met, the task will be continued.

[0085] (2) Simulate the wing spar digital twin model

[0086] The wing spar three-dimensional model constructed above is imported into the Ansys simulation software, material information, load information, etc. are added according to the performance parameters of the unmanned aerial vehicle, a virtual simulation environment is constructed, simulation experiments are carried out, and whether the error between the model spar stress and the stress measured by the physical entity sensor is small enough is judged. The specific steps are as follows:

[0087] (2a) In the pre-processing stage, PCP (Pre) module of workbench is used for pre-processing modeling. First, a certain type of carbon fiber / epoxy composite material is used, and material information is added according to its performance parameters to establish an equivalent thin-walled cylindrical model. Then, the constraint conditions are set and the mesh is divided, including setting the layer angle: setting according to the specified angle; defining the direction selection set; layering the carbon fiber set in advance, here 8 layers are layered, and the final carbon fiber direction: along the axial direction of the cylinder.

[0088] (2b) In the post-processing stage, Static Structural module is used for mechanical analysis. First, the model is regarded as a cantilever beam, and one end is constrained, then the load is applied, the wind load is treated as static load, and static analysis is used. The wind load belongs to surface load, for simplification, the equivalent force of the wind load is calculated and applied at the pre-set stress point of the structure. Finally, the total strain cloud of the thin-walled cylinder is drawn, and the deformation cloud of the strain gauge position is taken.

[0089] (2c) According to the results of the deformation cloud, the twin model is analyzed to judge whether the error between the main beam stress and the stress measured by the physical entity sensor is small enough. If the error is small enough, it means that the twin model can replace the physical entity to complete the state monitoring, life prediction and other functions of the unmanned aerial vehicle.

[0090] (3) Wing deformation state visualization

[0091] This embodiment uses Unity 3D engine to design a virtual animation visualization method, which can display the deformation posture of the unmanned aerial vehicle wing in real time, and provide direct and effective visualization basis for the health status of the unmanned aerial vehicle for the operator, Figure 4 The visualization technical roadmap is shown.

[0092] (3a) Maya three-dimensional model construction: this embodiment uses Maya software to construct the corresponding model, and converts the model into.fbx file format for use in Unity 3D environment. A wing deformation posture representation method based on bone displacement control is designed, bones are added at the strain sensor installation position of the wing, sensor data is read and stored in real time by means of MySQL database, sensor data is mapped to bone displacement, and the overall deformation of the wing is reflected by this method.

[0093] (3b) Model node loading in Unity3D: Load the.fbx model file into the Models folder of the Unity3D scene. After the wing model file is loaded into the scene, the displacement of each bone is controlled by the c# scripting language.

[0094] (3c) Implementation of wing deformation synchronization display function: In the actual wing attitude change, the bones of the wing main beam are bound through the parent-child relationship. The child object follows the parent object, that is, the child object inherits the parent object, but the motion of the child object itself will not affect the motion state of the parent object, that is, the parent object will not inherit the motion of the child object. According to the symmetry of the wing, the middle bone is set as the parent object, and the parent-child level is extended to both sides.

[0095] (3d) Implementation of wing stress cloud rendering effect: The wing main beam is rendered and designed using Shader, and the shader value parameter is defined by reading the stress value in the database. When Unity3D performs shader rendering, it will perform color rendering according to the real-time stress value. By applying this method, the visualization of the stress state of the wing main beam is realized.

[0096] (4) Real-time data reading technology

[0097] The data involved in the application of the embodiment is completed by means of the MySQL database. Each function in the database corresponds to one main table and one secondary table, a total of two tables. The main table is used for real-time data reading, and there is only one row. The data is deleted after reading, and at the same time, the data is persisted into the secondary table. The reading and display of data are realized by using echarts.

[0098] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application. Any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A method for digital twin modeling of unmanned aerial vehicles (UAVs) based on real-time data-driven approaches, characterized in that: Includes the following steps: Step 1: Construct a digital twin model of the wing main sparsity from four dimensions: geometric model, physical model, behavioral model, and rule model; Step 2: Simulate the digital twin model of the wing main sparsity to map the real-time physical space state of the UAV to the virtual space; Step 3: Use virtual animation visualization methods to display the deformation attitude of the drone's wings in real time, so as to provide operators with a visual basis for knowing the health status of the drone; In step 1: In the physical model, the UAV adopts a traditional aerodynamic layout. When analyzing the load and internal force of the wing, the external loads it experiences are aerodynamic forces, wing structural mass forces and other components, and concentrated external loads. Aerodynamic loads are distributed loads, acting on the entire wing surface, forming lift and drag across the entire wing surface. The lift component is the most significant external load on the wing surface. The mass force of the wing structure itself consists of the gravity and inertial force of the wing structure. The two act on the wing structure entity, and their distribution range depends on the mass distribution law of the structure itself. The distribution law of the mass force is approximately proportional to the chord length. The loads on other components fixedly connected to the wing are transmitted to the wing in the form of concentrated forces through connectors; wherein, the other components include solar panels, landing gear, and external stores; During the transfer of load from the wing to the fuselage, the internal forces generated inside the wing include: shear force S. n S h Bending moment B n B h and torque B t S n ≥S h B n Greater than or equal to B h The structure's width and moment of inertia are very large within the chord plane; therefore, the approximate analysis only considers S. n B n and B t Abbreviated as S, B and B t ; The shear force inside the wing is expressed by the following formula: Where z is the distance integraled from the wingtip to the wing root, and q = q a -q c q a For aerodynamic distributed load; q c For the mass force distribution load of the wing structure, Concentrated loads transmitted from the engine or other components; The bending moment inside the wing is expressed by the following formula: The airfoil torque caused by distributed load is expressed by the following formula: in, m t = q a e + q c d Typically, the axis of rigidity is taken as the relative axis. e The distance between the center line and the relative axis. d This is the distance between the centroid line and the relative axis.

2. The method according to claim 1, characterized in that, In step 1: In the geometric model, Solidworks 3D drawing software was used to draw the geometric model; based on the symmetry characteristics of the wing, only half of the wing main spars model was constructed.

3. The method according to claim 1, characterized in that, In step 1: In the behavior model, the attitude, position, and velocity of the attitude controller, position controller, and velocity controller are analyzed respectively. The analyzed attitude, position, and velocity are compared with the measured signal values ​​of the sensors for error analysis. Each type of controller performs behavior analysis and correction. The central actuator adjusts the wing attitude to ensure the flight control accuracy of the UAV.

4. The method according to claim 1, characterized in that, In step 1: In the rule model, after receiving the mission, the UAV first raises its altitude and then proceeds to the mission area after reaching the set altitude. During the raising and level flight phase, the stability of the fuselage structure needs to be assessed to determine whether the flight attitude needs to be adjusted or the mission plan needs to be changed. If the structure is stable, it is necessary to determine whether the solar radiation in the flight area is sufficient, that is, whether enough energy can be obtained to support the mission. If the solar radiation in the mission area is insufficient, the route needs to be adjusted to obtain sufficient energy reserves. If the structure is stable and the solar radiation in the flight area is sufficient, the mission can continue.

5. The method according to claim 1, characterized in that, The simulation of the digital twin model of the wing main sparsity includes: The constructed 3D model of the wing main beam is imported into Ansys simulation software. Various information is added according to the performance parameters of the UAV to build a virtual simulation environment. Simulation experiments are conducted on it to determine whether the error between the stress of the model main beam and the stress measured by the physical sensor is small enough. The various information includes material information and load information.

6. The method according to claim 1, characterized in that, The simulation of the digital twin model of the wing main sparsity specifically includes: Step 21: Preprocessing stage, using the PCP module of Workbench for preprocessing modeling; Step 22: In the post-processing stage, mechanical analysis is performed using the Static Structural module; Step 23: Analyze the twin model based on the deformation cloud map obtained in Step 22 to determine whether the error between the main beam stress and the stress measured by the physical entity sensor is small enough. If the error is small enough, it indicates that the twin model replaces the physical entity.

7. The method according to claim 6, characterized in that, Step 21 includes: Using carbon fiber / epoxy composite material, material information was added according to its performance parameters to establish an equivalent thin-walled cylindrical model. Set constraints and mesh, including setting the layup angle: set it according to a specified angle; define the direction selection set; lay up the pre-defined carbon fibers, with the carbon fiber direction along the axial direction of the cylinder; Step 22 includes: The model is treated as a cantilever beam with one end constrained, and then loads are applied. Wind loads are treated as static loads and static analysis is used. Wind loads are surface loads, and the equivalent force of wind loads is calculated during the modeling process and applied to the pre-defined stress points of the structure. Finally, the model is solved, the overall strain contour map of the thin-walled cylinder is drawn, and the deformation contour map of the strain gauge positions is taken.

8. The method according to claim 1, characterized in that, The virtual animation visualization method includes: Step 31: Maya 3D Model Construction: The model is constructed using Maya software and converted into .fbx file format for use in the Unity 3D environment; a wing deformation attitude representation method based on skeleton displacement control is designed, adding skeletons at the installation locations of the strain sensors on the wing, mapping the sensor data to skeleton displacement to reflect the overall deformation of the wing; wherein, the sensor data is transferred and retrieved in real time through a MySQL database; Step 32: Loading model nodes in Unity3D: Load the .fbx model file into the Models folder of the Unity3D scene. After the wing model file is loaded into the scene, control the displacement of each bone using the C# scripting language. Step 33: Implementation of the synchronized display function for wing deformation: In the actual wing attitude change, the bones of the wing main spars are bound together through parent-child relationships; that is, the child object follows the movement of the parent object, that is, the child object inherits the movement of the parent object, but the movement of the child object itself will not affect the movement state of the parent object, that is, the parent object will not inherit the movement of the child object; according to the symmetrical characteristics of the wing, the middle bone is set as the parent object, and the parent-child level extends to both sides in sequence. Step 34: Rendering effect of wing stress cloud map: Use shaders to render the main wing spars. Define shader numerical parameters by reading stress values ​​from the database. When Unity3D performs shader rendering, it performs shading rendering based on the stress values ​​read in real time.

9. The method according to claim 1, characterized in that, Also includes: In steps 1 to 3, the data involved in the application is transferred and persisted in real time using a MySQL database. Each function in the MySQL database corresponds to a main table and a secondary table. The main table is used for real-time data reading and contains only one row. The data is deleted after reading and the data is persisted into the secondary table. Data reading and display are implemented using echarts.