A method for evaluating the healthy operation of drones based on digital twins

Abstract

The present invention discloses a method for evaluating the healthy operation of an unmanned aerial vehicle (UAV) based on digital twin. First, relevant data of the UAV is obtained to construct a digital twin model of the UAV. Then, the relevant data of the UAV is cross-fused and iteratively fed back to fuse the relevant data of the UAV with the digital twin model of the UAV, and a digital twin health management and control platform for the UAV is constructed. Based on the platform, fault diagnosis and prediction are carried out on the operation process of the UAV, health assessment parameters are output and analyzed, the operation state of the UAV is visualized in the digital twin health management and control platform for the UAV, and decisions are made on the health assessment of the UAV. The method for evaluating the healthy operation of the UAV provided by the present invention truly displays the UAV in the physical environment in a virtual environment, can quickly and intuitively show the healthy operation status of the UAV, so as to make a health assessment in time and reduce corresponding losses. Through data interaction and feedback, the real situation of the UAV is accurately displayed, and the maintenance cost is saved.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV operation safety, and mainly relates to a method for evaluating the healthy operation of UAVs based on digital twins. Background Art

[0002] Currently, the global UAV industry is growing rapidly. Aspects such as the scale of the UAV fleet, UAV participants, and UAV operation volume are all increasing continuously. The rapid development and popularization of UAVs have enabled UAVs to be widely used in fields such as agricultural pest control, land resource planning, emergency rescue, forest fire prevention, smart cities, military politics, and leisure tourism. Different models of UAVs vary in size, performance, tasks, etc., resulting in different faults and health conditions of UAVs. The large-scale development of UAVs has increasingly exposed the problem of UAV operation safety, posing new requirements and challenges for the health management of UAVs, and is also the key to ensuring aviation safety.

[0003] With the deep integration of emerging communication technologies such as artificial intelligence technology, Internet of Things technology, and 5G, it is possible to realize the real-time monitoring of the UAV operation status, timely detect UAV faults, and conduct a more accurate, efficient, and intelligent evaluation of the UAV health condition. At the same time, through digital twin technology, the information of the UAV operation process is fed back to the UAV development process and maintenance process, reducing the full life cycle cost of the UAV while improving the reliability of UAV operation. Therefore, it is of great significance to conduct a health assessment of UAV operation. Summary of the Invention

[0004] Object of the Invention: Aiming at the problems existing in the above background art, the present invention provides a method for evaluating the healthy operation of UAVs based on digital twins, reducing the UAV operation risk, improving the UAV operation safety, so as to improve the healthy operation of UAVs and solve the problem of UAV health assessment.

[0005] Technical Solution: To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for evaluating the healthy operation of UAVs based on digital twins, comprising the following steps:

[0007] Step S1, collect UAV-related data and construct a UAV digital twin model;

[0008] Step S2, cross-fuse the UAV-related data and perform iterative feedback to fuse the UAV-related data with the UAV digital twin model; construct a UAV digital twin health management and control platform based on the fusion result to realize the supervision of the UAV health status;

[0009] Step S3: Based on the UAV digital twin health management and control platform, conduct fault diagnosis and prediction on the UAV operation process, and output health assessment parameters;

[0010] Step S4: Analyze based on the health assessment parameters, visualize the UAV operation status in the UAV digital twin health management and control platform, and make decisions on the UAV health assessment.

[0011] Furthermore, in step S1, the UAV-related data includes operation status data, operation environment data, and UAV geometric and physical behavior rules;

[0012] The operation status data includes UAV component connection status data, UAV activity status data, and UAV operation attitude data;

[0013] The operation environment data includes 5G radio transmission signal data, ground auxiliary equipment information data, meteorological environment data, and geographical environment data;

[0014] The UAV geometric and physical behavior rules include UAV shape parameters, size parameters, performance parameters, material characteristics, and activity range;

[0015] The establishment of the UAV digital twin model includes the UAV physical entity, the UAV virtual entity, and the connection relationship between the UAV physical entity and the virtual entity; First, according to the geometric data of the UAV physical entity, complete the three-dimensional modeling of the UAV geometric model and physical model in CATIA; Then integrate the UAV operation status data, operation environment data, and UAV geometric and physical behavior rule data into the Simulink environment in MATLAB to complete the construction of the UAV behavior model and UAV rule model, and realize the digital mapping between the UAV virtual entity and the physical entity. The data exchange between the UAV physical entity and the virtual entity is completed by using an embedded system and several sensors.

[0016] Furthermore, in step S3, use deep learning technology to conduct fault diagnosis and prediction on the UAV operation process; specifically including:

[0017] Step S3.1: Collect the UAV operation status data and operation environment data as the original input data;

[0018] Step S3.2: Preprocess the original input data and extract features from the preprocessed data;

[0019] Step S3.3: Input the data after feature extraction into the deep learning model for training, conduct fault diagnosis and prediction on the UAV operation process, and finally output health assessment parameters.

[0020] Furthermore, the drone health assessment parameters include: drone fault parameters, drone component reliability information parameters, drone key component stress state parameters and fatigue strength parameters.

[0021] Furthermore, in step S4, the impact of the drone's operating status is analyzed based on the drone health assessment parameters, the drone's operating status is visualized in the drone health management and control platform, and targeted health assessment decisions are made; specifically,

[0022] Conduct drone operational status impact analysis based on deep learning models, including drone fault propagation impact analysis and drone function impact analysis. The drone fault propagation impact analysis mainly focuses on whether it affects other structural components, and the drone function impact analysis mainly focuses on whether it affects the functions necessary for the normal operation of the drone.

[0023] Visualize the drone's operating status. It is a virtual control platform for the drone's digital twin. The drone's digital twin is connected to the platform display through a server to monitor the drone's operating health in real time.

[0024] The health assessment decision shown is to evaluate the health status of the UAV and the UAV mission capability based on the health assessment parameters, and then make the UAV maintenance decision.

[0025] The fault diagnosis parameters, the health assessment parameters, and the health assessment decision data information are fed back to the physical world drone database; the physical world drone feeds back drone-related data to the digital twin drone to continuously iterate and update the digital twin drone-related data and monitor the drone's operating health status. Beneficial effects

[0026] The method for healthy operation of drones provided by the present invention adopts digital twin technology to effectively achieve the coordination of safe operation of drones and cost reduction. Combined with 5G technology, the communication problem of drones is effectively improved. Finite element technology is used, or a drone digital twin model is constructed in a Simulink environment to truly display the drone in a physical environment in a virtual environment. The constructed drone visualization management and control platform can quickly and intuitively display the healthy operation status of the drone, so that health assessments can be made in time and corresponding losses can be reduced. Through the interactive feedback of drone data in the virtual environment and the real environment, the real status of the drone can be accurately and efficiently displayed in the virtual environment, providing support for the development, operation and maintenance of drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of the UAV health operation assessment method based on digital twins provided by the present invention;

[0028] Figure 2It is a schematic diagram for obtaining relevant data of the drone provided by the present invention;

[0029] Figure 3 It is a schematic diagram for constructing a digital twin model of the drone provided by the present invention;

[0030] Figure 4 It is a flow chart for fault diagnosis and prediction provided by the present invention;

[0031] Figure 5 It is a schematic diagram for health assessment decision provided by the present invention. Detailed implementation manners

[0032] The present invention will be further described below with reference to the accompanying drawings. However, this description should not be construed as a limitation of the present invention, but rather as a more detailed description of certain aspects, features and embodiments of the present invention.

[0033] As Figure 1 shown, the steps of the drone health operation assessment method based on digital twin provided by the present invention are as follows:

[0034] Step S1: Collect relevant data of the drone and construct a digital twin model of the drone.

[0035] Specifically, the relevant data of the drone, as Figure 2 shown, includes: operation status data, operation environment data, and drone geometric and physical behavior rules;

[0036] The operation status data includes, but is not limited to, drone component connection status data, drone activity status data, and drone operation attitude data;

[0037] The operation environment data includes, but is not limited to, 5G radio transmission signal data, ground auxiliary equipment information data, meteorological environment data, and geographical environment data.

[0038] The meteorological environment data includes, but is not limited to, wind speed, rain and snow, fog, air density, and atmospheric temperature. The geographical environment data includes data such as the drone flying in mountains, canyons, forests, highways, etc.

[0039] The drone geometric and physical behavior rules include, but are not limited to, drone shape parameters, size parameters, performance parameters, material properties, and activity ranges.

[0040] The drone digital twin model includes a drone physical entity, a drone virtual entity, and the connection relationship between the drone physical entity and the virtual entity. As Figure 3As shown in the figure, first, according to the geometric data of the physical entity of the drone, the three-dimensional modeling of the geometric model and physical model of the drone is completed in CATIA; then, the drone operation state data, operation environment data, and drone geometric and physical behavior rule data are integrated into the Simulink environment in MATLAB to complete the construction of the drone behavior model and the drone rule model, realizing the digital mapping of the drone virtual entity and the physical entity. The data exchange between the drone physical entity and the virtual entity is completed by using an embedded system and several sensors.

[0041] Step S2: Cross-fuse the drone-related data and perform iterative feedback to integrate the drone-related data with the drone digital twin model; build a drone digital twin health management and control platform based on the fusion result to realize the supervision of the drone health status.

[0042] Step S3: Based on the drone digital twin health management and control platform, conduct fault diagnosis and prediction on the drone operation process and output health assessment parameters. Specifically, as Figure 4 shown:

[0043] Step S3.1: Collect the drone operation state data and operation environment data as the original input data;

[0044] Step S3.2: Preprocess the original input data, including but not limited to filtering, noise reduction, and time-frequency conversion. Extract features from the preprocessed data, including but not limited to extracting the average value and standard deviation.

[0045] Step S3.3: Input the data after feature extraction into the deep learning model for training, conduct fault diagnosis and prediction on the drone operation process, and finally output health assessment parameters.

[0046] The drone health assessment parameters include: drone fault parameters, drone component reliability information parameters, stress state parameters of drone key components, and fatigue strength parameters.

[0047] Step S4: Analyze based on the health assessment parameters, visualize the drone operation state in the drone digital twin health management and control platform, and make decisions on the drone health assessment.

[0048] Conduct an analysis of the impact of the drone operation state according to the fault diagnosis model, including the analysis of the impact of drone fault propagation and the analysis of the impact of drone functions. Among them, the analysis of the impact of drone fault propagation mainly involves whether it affects other structural components, and the analysis of the impact of drone functions mainly involves whether it affects the functions necessary for the normal operation of the drone;

[0049] Visualize the operating status of the drone, which is a virtual control platform for the drone digital twin. The drone digital twin can be connected to the platform through a server and then to a display to monitor the operating health status of the drone in real time;

[0050] The health assessment decision shown is to evaluate the health status of the drone and the mission capabilities of the drone based on health assessment parameters, and then make a targeted drone maintenance decision. The fault diagnosis parameters, health assessment parameters, health assessment decision data information, etc. are continuously fed back into the physical world drone database, and the physical world drone data is fed back into the digital twin drone to continuously iterate and update the digital twin drone parameters to help more scientifically monitor the operating health status of the drone.

[0051] This embodiment also provides a drone health operation assessment system based on digital twin, including a status acquisition module, a drone digital twin module, and a health control platform module;

[0052] Status acquisition mainly collects data from sensors distributed on the drone, such as data collected by current sensors, magnetic sensors, tilt sensors, engine intake flow sensors, acceleration sensors, and inertial measurement units, and connects to the server through 5G network communication;

[0053] The drone digital twin is used to determine the actual operating status of the drone based on the collected status data, and is also used to determine the fault diagnosis model and health assessment of the drone;

[0054] The health control platform module is mainly used to continuously feed back the fault diagnosis parameters, the health assessment parameters, the health assessment decision data information, etc. into the physical world drone database, and the physical world drone data is fed back into the digital twin drone to continuously iterate and update the digital twin drone parameters to help more scientifically monitor the operating health status of the drone.

[0055] The above is only the preferred embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for evaluating the healthy operation of an unmanned aerial vehicle based on digital twin, characterized in that, It includes the following steps: Step S1: Collect relevant data of the drone and construct a digital twin model of the drone; The relevant data of the drone includes operating status data, operating environment data, and geometric and physical behavior rules of the drone; The operating status data includes data on the connection status of drone components, data on the activity status of the drone, and data on the operating attitude of the drone; The operating environment data includes 5G radio transmission signal data, ground auxiliary equipment information data, meteorological environment data, and geographical environment data; The geometric and physical behavior rules of the drone include the shape parameters, size parameters, performance parameters, material properties, and activity range of the drone; The establishment of the digital twin model of the drone includes the physical entity of the drone, the virtual entity of the drone, and the connection relationship between the physical entity and the virtual entity of the drone; First, according to the geometric data of the physical entity of the drone, complete the three-dimensional modeling of the geometric model and physical model of the drone in CATIA; Then integrate the operating status data, operating environment data, and geometric and physical behavior rule data of the drone into the Simulink environment in MATLAB to complete the construction of the drone behavior model and the drone rule model, and realize the digital mapping between the virtual entity and the physical entity of the drone; The data exchange between the physical entity and the virtual entity of the drone is completed by using an embedded system and several sensors; Step S2: Cross-fuse and iteratively feedback the relevant data of the drone to integrate the relevant data of the drone with the digital twin model of the drone; Based on the fusion result, construct a digital twin health management and control platform for the drone to realize the supervision of the health status of the drone; Step S3: Conduct fault diagnosis and prediction on the operation process of the drone based on the digital twin health management and control platform for the drone, and output health assessment parameters, specifically including: Step S3.1: Collect the operating status data and operating environment data of the drone as the original input data; Step S3.2: Preprocess the original input data and extract features from the preprocessed data; Step S3.3: Input the data after feature extraction into a deep learning model for training, conduct fault diagnosis and prediction on the operation process of the drone, and finally output health assessment parameters; The health assessment parameters include: drone fault parameters, drone component reliability information parameters, stress state parameters and fatigue strength parameters of key drone components; Step S4: Analyze based on the health assessment parameters, visualize the operating status of the drone in the digital twin health management and control platform for the drone, and make decisions on the health assessment of the drone, specifically as follows: Conduct an impact analysis on the operating status of the drone according to the deep learning model, including an impact analysis of drone fault propagation and an impact analysis of drone functions. Among them, the impact analysis of drone fault propagation includes judging whether it affects other structural components, and the impact analysis of drone functions includes judging whether it affects the functions necessary for the normal operation of the drone; The visualization of the operating status of the drone is a virtual management and control platform for the digital twin of the drone. The digital twin model of the drone is connected to the platform display through a server to monitor the operating health status of the drone in real time; The health assessment decision shown is to evaluate the health status of the UAV and the UAV mission capabilities based on the health assessment parameters, and then make a UAV maintenance decision; The fault diagnosis parameters, the health assessment parameters, and the health assessment decision data information are fed back to the physical world UAV database; the physical world UAV feeds the UAV-related data back to the digital twin UAV to continuously iterate and update the digital twin UAV-related data and monitor the UAV operation health status.

Images