A real-time prediction method for aircraft maneuverability based on digital twin

Abstract

The present application provides a real-time prediction method for aircraft maneuverability based on digital twins, comprising: first, constructing a digital twin model that can accurately characterize the strength and / or stiffness characteristics of the aircraft body; second, ensuring that the digital twin model characterizing the strength and / or stiffness characteristics of the aircraft body is consistent with the digital-physical twin formed by the physical entity of the aircraft in the initial physical state; obtaining the data set generated by the physical twin of the aircraft during the manufacturing stage and assigning it to the digital twin; third, using the digital twin model of the aircraft body to predict the strength and / or stiffness characteristics of the aircraft; fourth, determining the stress and strain parameters of the current aircraft body structure based on the predicted aircraft strength and / or stiffness characteristics, thereby determining the stress and strain level and residual load-bearing capacity of the aircraft body structure; and conducting a safe use condition analysis based on the influence of multi-level and multi-dimensional parameters of the basic parameters of the entire aircraft, the maneuverability parameters of the entire aircraft, and the component-level parameters to determine the aircraft flight capability envelope.

Description

Technical Field

[0001] The present application relates to the field of aircraft design technology, and in particular to a real-time prediction method for aircraft maneuverability based on digital twins. Background Art

[0002] Currently, traditional aircraft design assumes that the flight capability of an aircraft remains constant throughout its service life. However, due to individual differences in aircraft usage, as well as the presence of manufacturing defects, fatigue cracks, corrosion damage, and other issues, each aircraft's airframe structural characteristics are different. Furthermore, as aircraft usage continues to change, changes in the structural strength and stiffness directly lead to changes in the load-bearing performance of the aircraft's airframe platform, which in turn causes changes in the aircraft's flight capability. Failure to predict and control the aircraft's flight capability in real time can lead to mission failures due to these changes, or even catastrophic accidents caused by exceeding flight capabilities.

[0003] In order to achieve real-time prediction of aircraft flight capability, a real-time prediction method for aircraft maneuverability is needed. Summary of the Invention

[0004] The purpose of this application is to provide a real-time prediction method for aircraft maneuverability based on digital twins to solve or alleviate at least one problem in the background technology.

[0005] The technical solution of this application is: a real-time prediction method for aircraft maneuverability based on digital twins, the method comprising:

[0006] First, a digital twin model is constructed that can accurately characterize the strength and / or stiffness characteristics of the aircraft body to simulate the dynamic changes of the aircraft body characteristic parameters during actual use.

[0007] Second, based on the aircraft design information, the digital twin model representing the strength and / or stiffness characteristics of the aircraft body and the digital-physical twin composed of the aircraft physical entity are kept consistent in their initial physical state;

[0008] Acquire the dataset generated by the physical twin of the aircraft during the manufacturing phase and assign the dataset to the digital twin, so that the digital twin model can truly reflect the physical characteristics of the aircraft entity;

[0009] Third, predict aircraft strength and / or stiffness characteristics using a digital twin model of the aircraft body;

[0010] Fourth, determining the stress and strain parameters of the current aircraft airframe structure based on the aircraft strength and / or stiffness characteristics predicted by the digital twin model, and determining the stress and strain levels that meet the aircraft structure failure criteria based on the failure mode of the aircraft airframe structure, thereby determining the residual load-bearing capacity of the aircraft airframe structure;

[0011] Based on the flight load data corresponding to the carrying capacity obtained from the strength analysis, a safe use condition analysis is conducted based on the influence of multi-level and multi-dimensional parameters including the basic parameters of the entire aircraft, the maneuvering parameters of the entire aircraft, and the component-level parameters, and the aircraft's flight capability envelope is ultimately determined.

[0012] Furthermore, when establishing a digital twin model of the entire aircraft body, important areas that affect the overall stiffness and / or strength of the aircraft are refined and modeled, and the boundary conditions of the refined model are kept consistent with the entire aircraft body model.

[0013] Furthermore, the data set includes: empirical data from physical prototypes or experiments, boundaries of key design elements, and uncertainty statistical engineering data.

[0014] Furthermore, in the process of using the digital twin model of the aircraft body to predict the strength and / or stiffness characteristics of the aircraft, the strength / stiffness characteristics of the aircraft during service are predicted in real time through the Bayesian network prediction model.

[0015] Furthermore, the Bayesian network prediction model is used to predict the strength / stiffness characteristics of the aircraft during service in real time, including:

[0016] Before entering service, the Bayesian network prediction model is initially trained based on the simulation prediction results of the digital twin model to form an initial intelligent and rapid prediction model;

[0017] During the service process, based on the real-time measurement data in the aircraft, the Bayesian network prediction model is improved and evolved by integrating simulation data and measured data on the basis of the initial prediction model.

[0018] Furthermore, the basic parameters of the entire aircraft include flight overload, flight weight, flight altitude, and flight speed.

[0019] Furthermore, the whole aircraft maneuvering parameters include three-axis angular velocity and three-axis angular acceleration.

[0020] Furthermore, the component-level parameters include the limited deflection of each control surface.

[0021] The method of the present application adopts digital twin technology to synchronously twin the change process of characteristic parameters that affect the flight capability of the aircraft. By accurately predicting the change trend and range of the characteristic parameters, the flight capability of the aircraft can be evaluated in real time, and the use and maintenance of the aircraft can be guided, which plays an important role in improving the safety and efficiency of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions provided by this application, the following is a brief introduction to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of this application.

[0023] Figure 1 This is a flow chart of the real-time prediction method for aircraft maneuverability of this application.

[0024] Figure 2 Flowchart for predicting strength / stiffness characteristics of typical aircraft structures.

[0025] Figure 3 Flowchart of Bayesian network prediction of strength / stiffness of typical aircraft structures. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the implementation of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below in conjunction with the drawings in the embodiments of this application.

[0027] In order to achieve real-time prediction of aircraft flight capabilities, this application establishes a digital twin model that can fully express the real-time flight capabilities of the aircraft. Based on the high-fidelity simulation characteristics of the digital twin model, it fully utilizes big data information from various stages such as design, manufacturing, use, and maintenance to accurately predict the structural strength / stiffness characteristics of the aircraft and effectively evaluate the flight capabilities of the aircraft.

[0028] like Figure 1 As shown, the real-time prediction method of aircraft maneuverability based on digital twin proposed in this application includes the following steps:

[0029] First, build a digital twin model of the aircraft body.

[0030] Build a digital twin model that can accurately characterize the strength / stiffness characteristics of the aircraft body, so as to simulate the dynamic changes of various characteristic parameters of the aircraft body during actual use, thereby evaluating the impact of these characteristic parameters on the aircraft's flight capability and laying the foundation for accurate evaluation of the aircraft's flight capability.

[0031] The digital twin model of the aircraft body has parametric and multi-scale characteristics. Among them, parameterization refers to the dynamic change process of various characteristic parameters of the aircraft body during actual use. These parameters are variables, thereby realizing real-time changes to the parameters during use; multi-scale refers to the multi-scale digital twin model established to describe the physical characteristics of the body from macro to micro scales. It can not only characterize the macro strength and stiffness characteristics of the aircraft body, but also characterize micro characteristics such as material properties and initial defects.

[0032] Taking a certain type of aircraft as an example, a parametric, multi-scale digital twin model is constructed that can accurately characterize the strength / stiffness characteristics of the aircraft body. The connection area between the wing and the fuselage is an important area that affects the overall stiffness and strength of the aircraft. Therefore, on the basis of establishing a digital twin model of the entire aircraft, this area is refined and modeled, and the boundary conditions of the refined model are kept consistent with the full-aircraft model. At the same time, a mesoscopic model that can characterize the size of microscopic cracks is established, and the connection bolt hole size, assembly clearance, hole margin, crack size, etc. are parametrically characterized.

[0033] Second, the actual manufacturing, testing, and usage data of the aircraft are interacted and integrated so that the digital twin model can truly reflect the physical characteristics of the aircraft entity.

[0034] The data interaction between the aircraft's digital-physical twin model is an important foundation for solving the problem of synchronous twinning between the aircraft's physical entity and digital model. First, based on a parametric, multi-scale, high-fidelity digital twin model, the overall digital twin benchmark is determined based on aircraft design information to ensure a high degree of consistency between the digital-physical twin in its initial physical state (material properties, geometric dimensions, etc.). Through online digital detection and measurement during the aircraft manufacturing process, the data generated by the physical twin during the manufacturing stage is processed to form an authoritative digital model and data set as a unified data source, which is then transferred to the digital twin in the virtual space to support simulation of all aspects of the entire life cycle. On this basis, a method for the synchronous evolution of digital-physical twins is formed by combining technologies such as intelligent data interaction and multi-scale model linkage.

[0035] To address the intelligent data interaction issues of the digital-physical twin of the aircraft airframe platform, the planning, development, and use of models must first be incorporated into standardized processes to ensure the continuous transfer of models and data throughout the lifecycle. This allows for a comprehensive, end-to-end digital representation of the target entity, supporting consistent analysis and decision-making for this complex organization.

[0036] Secondly, a public and dynamically updated authoritative digital model and database is formed, and its integrity, validity, consistency, timeliness, and accuracy are guaranteed. This ensures that the access, management, analysis, use, and distribution of the latest data information remain authoritative and consistent throughout the entire digital-physical twin throughout its lifecycle, enabling collaborative work with shared knowledge and resources. This authoritative dataset primarily includes: empirical data from physical prototypes or tests, and statistical engineering data on the boundaries and uncertainties of key design elements.

[0037] Finally, the latest data information is associated with each model at each level of the simulation model system in the form of variables, forming an internal update driving mechanism of the twin based on design variables, and further forming this mechanism into an automated process driven by data flow to achieve real-time synchronous update of the internal simulation model system of the digital twin and the physical twin.

[0038] For example, through online digital inspection and measurement during the aircraft manufacturing process, the specific dimensional values ​​of parameters such as bolt hole size, assembly clearance, and hole margin in the model are assigned to the digital twin model, enabling the synchronous evolution of the digital twin model and the physical entity model during the manufacturing phase. At the same time, a sensor system is installed in the aircraft entity to monitor structural characteristic parameters in key areas, including structural strain and crack size. This monitoring data is incorporated into a unified database, and a data association is established between the database and the digital twin model to assign data values ​​to the corresponding parameters, realizing the dynamic evolution of the digital twin model based on measured data.

[0039] Third, use the digital twin of the aircraft body to predict the aircraft strength / stiffness characteristics.

[0040] The prediction of the strength / stiffness characteristics of aircraft structures is the basis for predicting the flight capability of aircraft. The traditional method is to predict the strength / stiffness characteristics based on the simulation results of digital prototypes. However, due to the many factors that affect the strength / stiffness characteristics and the complexity of the structural nonlinear response process of multi-physical field coupling, the simulation model is very large. Faced with the complex and random use environment of aircraft during service, even with the use of high-performance computing platforms, it is still impossible to meet the efficiency requirements of the prediction of the strength / stiffness characteristics of individual aircraft during service by exhaustive simulation of each digital twin. It also makes it impossible to fuse and analyze the simulation data with the sensor data during service, so as to make more accurate predictions for more complex working conditions in the future. In the method provided in this application, the Bayesian network method is used to solve the above problems. The typical aircraft structural strength / stiffness prediction process is shown in Figure 2 Among them, the load history, structure and corrosion environment are inputs, and the life and bearing capacity are outputs. The input goes through a series of processes (anti-corrosion system attenuation, structural corrosion, structural response, damage accumulation, crack initiation / propagation, etc.) to obtain the output, that is, the prediction result.

[0041] like Figure 3 The structural strength / stiffness prediction process shown, the real-time prediction of the strength / stiffness characteristics of the aircraft during service is a directed acyclic process, which is consistent with the research object of the Bayesian network. Therefore, the Bayesian network method is used in this application to solve the problem of real-time prediction of the strength / stiffness characteristics of the aircraft during service. Before service, the Bayesian network model is initially trained based on the simulation prediction results of the digital twin model to form an initial intelligent rapid prediction model, for example, the initial intelligent rapid prediction model of the structural stress and strain of the wing and fuselage area in the above embodiment can be formed; during service, based on the measured load results and the measured data of the sensor, on the basis of the initial prediction model, the simulation data and the measured data are integrated to improve the evolutionary Bayesian network prediction model, improve its prediction accuracy, and perform subsequent strength / stiffness intelligent prediction based on the improved prediction model.

[0042] Fourth, the aircraft's flight capability is predicted based on the aircraft's carrying capacity prediction results.

[0043] Based on the analysis and prediction results of the third step, the structural stress, strain and other parameters of the current aircraft are determined. Based on the structural failure mode analysis, the stress and strain levels that meet the structural failure criteria are determined. These are used as the maximum allowable stress and strain boundary values ​​to determine the residual load-bearing capacity of the structure. Based on the flight load data corresponding to the load-bearing capacity obtained from the strength analysis, a safe operating condition analysis is conducted based on the influence of three levels of multi-dimensional parameters: basic aircraft parameters (flight overload, flight weight, flight altitude, flight speed, etc.), aircraft maneuvering parameters (three-axis angular velocity, three-axis angular acceleration, etc.), and component-level parameters (such as the limit deflection of each control surface). The final result is the aircraft's flight capability envelope.

[0044] The method of the present application adopts digital twin technology to synchronously twin the change process of characteristic parameters that affect the flight capability of the aircraft. By accurately predicting the change trend and range of the characteristic parameters, the flight capability of the aircraft can be evaluated in real time, and the use and maintenance of the aircraft can be guided, which plays an important role in improving the safety and efficiency of the aircraft.

[0045] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A real-time prediction method for aircraft maneuverability based on digital twins, characterized in that: The method comprises: First, a digital twin model is constructed that can accurately characterize the strength and / or stiffness characteristics of the aircraft body to simulate the dynamic changes of the aircraft body characteristic parameters during actual use. Second, based on the aircraft design information, the digital twin model representing the strength and / or stiffness characteristics of the aircraft body and the digital-physical twin composed of the aircraft physical entity are kept consistent in their initial physical state; Acquire the dataset generated by the physical twin of the aircraft during the manufacturing phase and assign the dataset to the digital twin, so that the digital twin model can truly reflect the physical characteristics of the aircraft entity; Third, predict aircraft strength and / or stiffness characteristics using a digital twin model of the aircraft body; Fourth, determining the stress and strain parameters of the current aircraft airframe structure based on the aircraft strength and / or stiffness characteristics predicted by the digital twin model, and determining the stress and strain levels that meet the aircraft structure failure criteria based on the failure mode of the aircraft airframe structure, thereby determining the residual load-bearing capacity of the aircraft airframe structure; Based on the flight load data corresponding to the carrying capacity obtained from the strength analysis, a safe use condition analysis is conducted based on the influence of multi-level and multi-dimensional parameters including the basic parameters of the entire aircraft, the maneuvering parameters of the entire aircraft, and the component-level parameters, and the aircraft's flight capability envelope is ultimately determined.

2. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: When establishing a digital twin model of the entire aircraft body, important areas that affect the overall stiffness and / or strength of the aircraft are refined and modeled, and the boundary conditions of the refined model are kept consistent with the full aircraft body model.

3. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: The data set includes: empirical data from physical prototypes or experiments, boundaries of key design elements, and uncertainty statistical engineering data.

4. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: In the process of using the digital twin model of the aircraft body to predict the strength and / or stiffness characteristics of the aircraft, the Bayesian network prediction model is used to perform real-time prediction of the strength / stiffness characteristics of the aircraft during its service.

5. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 4, characterized in that: Real-time prediction of aircraft strength / stiffness characteristics during service is performed using a Bayesian network prediction model, including: Before entering service, the Bayesian network prediction model is initially trained based on the simulation prediction results of the digital twin model to form an initial intelligent and rapid prediction model; During the service process, based on the real-time measurement data in the aircraft, the Bayesian network prediction model is improved and evolved by integrating simulation data and measured data on the basis of the initial prediction model.

6. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: The basic parameters of the entire aircraft include flight overload, flight weight, flight altitude, and flight speed.

7. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: The full aircraft maneuvering parameters include three-axis angular velocity and three-axis angular acceleration.

8. The method for real-time prediction of aircraft maneuverability based on digital twins according to claim 1, characterized in that: The component-level parameters include the limit deflection of each control surface.

Images