Electromagnetic calculation method and system for hypersonic flight vehicle

Abstract

The invention relates to the field of electromagnetic property calculation of hypersonic flight vehicles, in particular to an electromagnetic calculation method and system for a hypersonic flight vehicle. The intelligent agent system is designed to intelligently classify the flow field calculation of the hypersonic flight vehicle, and the output key physical quantity is collected, reconstructed, backed up and subjected to high-fidelity line extraction, so that the problems of tedious data integration and low calculation precision in the electromagnetic calculation of the hypersonic flight vehicle at present are solved, and support is provided for subsequent ground analysis.

Description

Technical Field

[0001] This application relates to the field of electromagnetic characteristic calculation of hypersonic vehicles, and more specifically, to electromagnetic calculation methods and systems for hypersonic vehicles. Background Technology

[0002] To study the electromagnetic properties of aircraft under high-speed extreme conditions, researchers both domestically and internationally generally rely on computational fluid dynamics (CFD) and computational electromagnetics (CEA) methods for numerical research. On the one hand, with the widespread adoption of commercial software such as ANSYS and FASTRAN, numerical simulation has become inexpensive and fast; on the other hand, numerical simulation can compensate for configuration conditions that are difficult to achieve in wind tunnel tests and reveal complex mechanisms. However, current calculations still have certain limitations: for example, hypersonic targets involve multi-physics coupling, requiring the combined use of CFD and CEA methods, which, along with data input compatibility issues, makes information integration and analysis very cumbersome; at the same time, due to the uneven distribution of the plasma sheath on the aircraft surface, the traditional WKB method approximates the surface sheath as a metal layer with different dielectric constants when calculating the electromagnetic properties of the target, and its calculation accuracy is often determined by the surface layering strategy, with simplified surface layering leading to a decrease in calculation accuracy; furthermore, flow field calculations require solving different Navier-Stokes equations according to different flight speeds, and their depth depends on human experience.

[0003] Although advancements in supercomputing capabilities have made computer numerical simulations crucial for solving challenging engineering problems and conducting complex scientific research, the following issues still exist: (1) Cumbersome information integration. At present, the calculation of the flow field environment of hypersonic targets relies heavily on commercial software such as ANSYS and FASTRAN, while electromagnetic calculations widely use commercial software such as Foke. Due to the need for interdisciplinary integration and multi-physics coupling, researchers often use fluid dynamics methods to calculate the flow field environment and perform electromagnetic calculations based on the environmental parameters. Due to interdisciplinary interaction, data compatibility problems arise (such as mesh generation and information import). At present, the phenomenon of cumbersome information integration is common.

[0004] (2) Reliance on human experience. Flow field environment calculation requires the selection of flow field control equations based on different limiting factors such as flight altitude, speed, and atmospheric conditions. Its level of intelligence is low, and it relies on human experience for selection and debugging.

[0005] (3) Low computational accuracy. Traditional electromagnetic calculations treat the plasma sheath on the surface of the aircraft as a layered structure, approximating each layer as a stack of metal scattering layers with different dielectric constants. This layered approximation disrupts the spatial continuity of plasma physical parameters (such as electron density), simplifying them into discrete step functions. Consequently, it cannot accurately describe the changes in physical quantity gradients in key regions such as before and after the shock wave and within the boundary layer. Due to this fundamental error in the traditional "layered approximation," the subsequent computational accuracy decreases.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] The purpose of this application is to provide a hypersonic flight electromagnetic calculation method and system with high computational accuracy, good information integration and analysis, and a high degree of intelligence. By intelligently classifying the flow field calculation of hypersonic vehicles and collecting, calculating, backing up, and extracting key physical quantities, this method aims to solve the problems of cumbersome data integration and low computational accuracy in current hypersonic vehicle electromagnetic calculations, thus providing support for subsequent ground analysis.

[0008] Firstly, this application provides an electromagnetic calculation method for hypersonic vehicles, the technical solution of which is as follows: A NS control equation is selected from a pre-built rule base based on user input parameters; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed. The flow field data is calculated based on the Navier-Stokes governing equations and collected; the flow field data includes physical and chemical quantities. The flow field data is reconstructed into electromagnetic parameters required for electromagnetic calculation, generating a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant, and conductivity; The surface feature lines of the aircraft are determined, electromagnetic input parameters are extracted from the reconstructed electromagnetic parameter set, and one-dimensional spatial sequence data distributed along the feature lines are generated. Electromagnetic environment calculations are performed based on the one-dimensional spatial sequence data; the electromagnetic environment calculations include scattering characteristic calculations and transmission characteristic calculations.

[0009] Furthermore, the calculation of flow field data based on the NS governing equations includes calculations of physical quantities and chemical quantities; The physical quantity calculation includes: using the finite element method to discretize and solve the NS control equations on a spatial grid to calculate the flow field temperature and pressure; In three-dimensional Cartesian coordinates, the NS governing equations are expressed as:

[0010] Where Q is a conserved quantity, and E, F, and G are diffusion terms in the x, y, and z directions, respectively. V F V G V Let S be the convection term in three directions, and S be the source vector.

[0011] Furthermore, the chemical quantity calculation includes: if the user-input parameters meet the chemical reaction conditions, then solving the chemical reaction transport equation and reaction kinetic equation in the Park reaction model or Gupta reaction model according to the actual environmental conditions, and calculating the chemical variables; the chemical variables include electron density and reaction rate. In this example, the chemical calculation relies on existing chemical reaction models, such as the Park model and the Gupta model, whose calculation equations include the reactions of dozens of air components.

[0012] Furthermore, the step of reconstructing the flow field data into electromagnetic parameters required for electromagnetic calculation, and generating a reconstructed electromagnetic parameter set, includes: Plasma frequency: ; Where, n e where e is the electron density, e is the electron charge, and m is the electron density. e Where ε is the electron mass, and ε0 is the dielectric constant; Using an empirical formula based on gas density and temperature: Calculate the collision frequency between electrons and neutral particles; where For collision frequency, Where T is the total number of particles and T is the temperature; Using a collision-based plasma Drude model: Used to calculate the relative permittivity; , used to calculate relative conductivity; where The dielectric constant in vacuum is . The relative permittivity, and These are the real and imaginary parts of the relative permittivity, respectively. The incident wave frequency, For collision frequency, relative conductivity and , respectively, represent the real and imaginary parts of the relative conductivity; i is the imaginary part of the complex function. The relative permittivity, is the vacuum dielectric constant.

[0013] A set of reconstructed electromagnetic parameters is generated based on the plasma frequency, collision frequency, dielectric constant, and conductivity, and stored in the log library.

[0014] Furthermore, the process of determining the surface feature lines of the aircraft, extracting electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generating one-dimensional spatial sequence data distributed along the feature lines includes: Identify at least one feature line on the surface of the aircraft and generate a corresponding feature line point set; associate and locate the feature points in the feature line point set with the mesh vertices on the surface of the aircraft. Based on the physical quantities at the associated grid vertices, the physical quantity values ​​of each point on the feature line of the aircraft surface are obtained by interpolation calculation; Generate a one-dimensional data sequence in which the physical quantities are continuously varied along the feature lines on the surface of the aircraft.

[0015] Furthermore, determining at least one feature line on the surface of the aircraft includes: extracting at least one of the following from the apex of the nose cone to the tip of the tail wing, a continuous curve on the plane of symmetry, a nose stagnation line, and a wing edge line based on the geometric features of the aircraft. The step of associating and locating the feature points in the feature line point set with the mesh vertices on the aircraft surface includes: emitting a ray from the feature point along the approximate normal direction of its mesh cell, calculating the intersection of the ray with the mesh on the aircraft surface, and determining the nearest surface mesh vertex.

[0016] Furthermore, the calculation of the scattering characteristics includes: When electromagnetic waves irradiate a target, the scattered electric field on the target surface is obtained using the physical optics PO method. E s :

[0017] In the formula, K is the incident wave beam, r is the field point position vector, r^ is the scattered wave unit vector, n^ is the external unit normal vector of the surface element, Z0 is the free space wave impedance, r' is the source point position vector, and E and H are the total electric field and magnetic field of the boundary, respectively.

[0018] Furthermore, the transmission characteristic calculation includes: The propagation characteristics of electromagnetic waves in a plasma sheath were calculated using the WKB method; the propagation characteristics include signal attenuation, transmission coefficient, and reflection coefficient.

[0019] Secondly, this application also provides an electromagnetic computing system for hypersonic vehicles based on intelligent agents, including an intelligent agent subsystem, a computing subsystem, and a post-processing subsystem; the intelligent agent subsystem includes an intelligent classification module, a pre-built strategy library, a data collection module, a data integration module, a log library, and a data extraction module; the computing subsystem includes a flow field computing module and an electromagnetic computing module; the post-processing subsystem includes a data receiving module and a visualization processing module; wherein: The intelligent classification module is used to automatically identify, evaluate, and classify user input parameters, and select the NS control equation based on a pre-built strategy library; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed. A pre-built strategy library is used to store rule sets for the intelligent classification module to call; The flow field calculation module is used to perform computational fluid dynamics simulation based on the selection results of the intelligent classification module to obtain flow field data; the flow field data includes physical quantities and chemical quantities. The data integration module is used to reconstruct the flow field data into electromagnetic parameters required for electromagnetic calculation, and generate a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant and conductivity; The data extraction module is used to determine the feature lines on the surface of the aircraft, extract electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generate one-dimensional spatial sequence data distributed along the feature lines. The electromagnetic calculation module is used to perform electromagnetic environment calculations based on the one-dimensional spatial sequence data; the electromagnetic environment calculations include scattering characteristic calculations and transmission characteristic calculations. The visualization module is used to visualize the electromagnetic environment calculation results.

[0020] Furthermore, the system is deployed on a hardware platform that includes a high-performance computing server, a large-capacity storage device, and a high-speed internal bus, and the modules interact with each other through the high-speed internal bus; The system includes a data collection module for capturing and converting flow field calculation results; a log library for storing, backing up, and managing all intermediate and final data; and a data receiving module for receiving and backing up electromagnetic calculation results.

[0021] Compared with existing technologies, the electromagnetic calculation method and system for hypersonic vehicles provided by this invention, through its complete technical solution, brings the following significant beneficial effects: 1. Improved accuracy and physical fidelity of electromagnetic calculations: The core innovation of this application lies in the use of "data line extraction" technology. This technology directly extracts a one-dimensional continuous sequence of electromagnetic parameters from the integrated and reconstructed 3D dataset along the surface contour or feature lines of the aircraft through linear interpolation sampling. This method completely abandons the simplified model in the traditional WKB method that approximates the plasma sheath as a finite number of discrete dielectric constant layers. By fully preserving the spatial gradient information and continuity of physical parameters (such as electron density and plasma frequency) along the surface, it fundamentally avoids model errors introduced by coarse layering. Compared to traditional layering strategies, this invention can obtain more accurate calculation results of electromagnetic scattering and transmission characteristics at the same grid density, especially reflecting the electromagnetic effects of key regions such as shock waves and boundary layers more realistically.

[0022] 2. This application achieves efficient and automated integration of interdisciplinary data, reducing manual intervention and the tediousness of integration. It constructs a fully automated data processing pipeline through an "intelligent classification module," a "data integration module," and a "log library." The intelligent classification module automatically identifies input parameters and selects governing equations; the data integration module automatically reconstructs the flow field data into the input parameters required for electromagnetic calculations using built-in physical algorithms (such as electron density and plasma frequency calculation formulas), and stores them in association. This solution solves the information integration problem caused by the incompatibility of data formats and grid systems between CFD and CEM. The system achieves full-link automation from flow field calculation to electromagnetic parameter generation. Researchers no longer need to manually perform tedious data format conversion, transmission, and preprocessing, greatly improving work efficiency, reducing the risk of errors due to human error, and enabling rapid, multi-condition numerical analysis.

[0023] 3. It lowers the barrier to entry for algorithm use and reduces reliance on specific human experience, thereby improving the level of intelligence. The intelligent agent subsystem of this application (especially the intelligent classification module) integrates a "pre-made strategy library" formed by domain expert knowledge, which can automatically and intelligently select the most suitable NS control equations and chemical reaction models based on the input flight conditions. This design enables researchers without a strong background in fluid mechanics to quickly and accurately initiate complex multiphysics simulation calculations.

[0024] 4. A complete, efficient, and traceable simulation analysis ecosystem has been constructed. The technical solution presented in this application is a complete system integrating intelligent classification, flow field calculation, data integration, line extraction, electromagnetic calculation, and visualization. The system's designed "log library" records all intermediate data and process logs from initial decision-making to the final result. This ecosystem ensures high reproducibility and traceability of the calculation process. The result of any calculation can be fully traced, facilitating error analysis, scheme optimization, and knowledge accumulation. Simultaneously, the modules are connected via a high-speed bus and accelerated using hardware such as GPUs, maintaining overall computational efficiency while ensuring high accuracy, thus meeting the needs of large-scale, high-precision simulations in engineering practice. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart illustrating the steps of the electromagnetic calculation method for hypersonic vehicles disclosed in an embodiment of the present invention. Figure 2 This is a schematic diagram of the electromagnetic computing system structure of a hypersonic vehicle based on an intelligent agent, as disclosed in an embodiment of the present invention. Figure 3 This is an operation flowchart of the electromagnetic computing system for hypersonic vehicles based on intelligent agents disclosed in an embodiment of the present invention. Detailed Implementation

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which these embodiments belong; the terminology used herein and in the specification of the application is for the purpose of describing particular embodiments only and is not intended to limit these embodiments; the terms "comprising" and "having," and any variations thereof, in the specification of these embodiments and the foregoing drawings, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification of these embodiments and the foregoing drawings are used to distinguish different objects, not to describe a particular order.

[0028] The implementation details of the technical solution in this embodiment are described in detail below: This application proposes an electromagnetic calculation method for hypersonic vehicles, such as... Figure 1 As shown, the method includes: S101, Select an NS control equation from the pre-made rule base according to the user input parameters; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed.

[0029] Specifically, in this embodiment, based on user input parameters, automated identification and classification of the input parameters are required. Intelligent classification is a preprocessing step in this embodiment, executed by an intelligent agent system, to achieve automated identification and classification of input parameters, reducing manual intervention. This includes the following sub-steps: (1) Format Classification After receiving the user's input model parameters, the intelligent agent first determines the format of the input model by parsing the file header, recognizing the file extension, and detecting content features. This determines whether the input model is three-dimensional or two-dimensional, and then selects the appropriate calculation format.

[0030] (2) Input parameter evaluation The intelligent agent evaluates the parsed parameters, including atmospheric environmental parameters, flight altitude, flight speed, and chemical reaction models. The evaluation process checks the completeness and rationality of the parameters. For example, atmospheric environmental parameters are used to assess whether the fluid is compressible, flight speed is used to assess the presence of a plasma sheath, and chemical reaction models are used to assess whether the fluid is in thermochemical non-equilibrium flow.

[0031] (3) Selection of governing equations Based on the evaluation results, the intelligent agent will intelligently select applicable Navier-Stokes equations from a pre-built rule base. The selection process comprehensively considers the flow state and thermodynamic processes to achieve adaptive matching of equation types, thereby improving computational efficiency and accuracy. For example, for an aircraft flying at an altitude of 61 km and a speed of Mach 23, the intelligent agent will select three-dimensional compressible Navier-Stokes governing equations, a 7-component chemical reaction model, and a viscous solid wall to solve for the flow field environment.

[0032] S102, Calculate the flow field data based on the NS governing equations and collect the flow field data; the flow field data includes physical quantities and chemical quantities; Furthermore, the calculation of flow field data based on the NS governing equations includes physical quantity calculation and chemical quantity calculation; wherein, the physical quantity calculation includes: using the finite element method to discretize and solve the NS governing equations on a spatial grid to calculate the flow field temperature and pressure.

[0033] Furthermore, the chemical quantity calculation includes: if the user input parameters meet the chemical reaction conditions, then solving the chemical reaction transport equation and the reaction kinetic equation to calculate the chemical variables; the chemical variables include electron density and reaction rate.

[0034] Specifically, in this embodiment, the flow field environment calculation is a prerequisite for the electromagnetic calculation and is performed by the flow field calculation module. It mainly obtains the flow field data by solving the Navier-Stokes equations selected by the intelligent agent. The specific sub-steps are as follows: (1) Calculation of physical processes (calculation of physical quantities) The flow field calculation module uses the finite element method to discretize and solve the Navier-Stokes equations on a spatial grid, calculating physical quantities such as temperature and pressure in the flow field. In three-dimensional Cartesian coordinates, the three-dimensional conserved form of the compressible Navier-Stokes equations can be expressed as:

[0035] Where Q is a conserved quantity, and E, F, and G are diffusion terms in the x, y, and z directions, respectively. V F V G V Let S be the convection term in three directions, and S be the source vector. Its expression is as follows:

[0036] In the above formula, u, v, and w are the velocities in the x, y, and z directions, respectively, and p, ρ, ρ S These represent the gas pressure, density, and density of component s, respectively; ρ s For fluid density, The energy density of the turbulent flow; , , The momentum densities in the x, y, and z directions are respectively; Total energy density; Let x be the species mass flux in the x-direction; Let x be the turbulent energy flux in the x-direction; The energy flux in the x-direction; , , The momentum flux in the x, y, and z directions along the X direction; Let x be the total enthalpy flux in the x-direction.

[0037] During the calculation, the calculation module iteratively solves the problem based on the boundary conditions until the physical quantities converge. This step reveals the flow characteristics of the aircraft surface, such as the location of the shock wave and the distribution of the boundary layer, providing a qualitative distribution reference for subsequent electromagnetic calculations.

[0038] (1) Chemical process calculation (chemical quantity calculation) Depending on the flight altitude and speed, if the input parameters satisfy the chemical reaction conditions, the flow field calculation module will additionally solve the chemical reaction transport equations and reaction kinetic equations, calculating chemical variables such as electron density and reaction rate. The module is compatible with various chemical reaction models, such as Gupta_7 / 11, Park97, and Dunn Kang chemical reaction models.

[0039] During the calculation, the calculation module iteratively solves the problem based on the chemical reaction conditions until the chemical variables converge. This step reveals the chemical reaction characteristics on the surface of the spacecraft, such as the concentration distribution of substances, providing a quantitative reference for subsequent electromagnetic calculations.

[0040] S103, the flow field data is reconstructed into electromagnetic parameters required for electromagnetic calculation, generating a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant and conductivity.

[0041] Prior to S103, the method further includes a flow field data collection step. This flow field data collection step is performed by a data collection module, enabling comprehensive acquisition and storage of the flow field calculation results. Specifically, it includes the following sub-steps: (1) Data collection: The module connects to the CFD solver through an interface to read the flow field output data, including grid node temperature, pressure, velocity, etc.

[0042] (2) Format conversion: In order to be compatible with the format requirements of various post-processing software, this module has a built-in format conversion program that can convert the data format into the required format according to the user's needs, such as ASCII, CGNS, Binary and other formats.

[0043] Furthermore, S103 is the data integration and reconstruction step in this embodiment. The data integration and reconstruction step converts the flow field data into the input parameters required for electromagnetic calculations, resolving the multiphysics data compatibility issue. The specific process is as follows: The module incorporates physical algorithms to reconstruct and calculate the output flow field data. For example, it can deduce electron density from temperature and pressure based on chemical reaction concentration.

[0044] Using the formula:

[0045] Derive the plasma frequency, where n e where e is the electron density, e is the electron charge, and m is the electron density. e Where ε is the electron mass, and ε0 is the dielectric constant; The electron-neutral particle collision frequency was calculated using an empirical formula based on gas density and temperature.

[0046] The reconstructed parameters include plasma frequency, collision frequency, dielectric constant, and conductivity. These parameters will be stored in the log database to enable data backup and fast retrieval.

[0047] Among them, the empirical formula based on gas density and temperature is used: Calculate the collision frequency between electrons and neutral particles; where For collision frequency, Where T is the total number of particles and T is the temperature; Using a collision-based plasma Drude model: Used to calculate the relative permittivity; , used to calculate relative conductivity; where The dielectric constant in vacuum is . The relative permittivity, and These are the real and imaginary parts of the relative permittivity, respectively. The incident wave frequency, For collision frequency, relative conductivity and , respectively, represent the real and imaginary parts of the relative conductivity; i is the imaginary part of the complex function. The relative permittivity, is the vacuum dielectric constant.

[0048] S104, based on the determined feature lines of the aircraft surface, extract electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generate one-dimensional spatial sequence data distributed along the feature lines.

[0049] Furthermore, the step of determining the feature lines on the aircraft surface, extracting electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generating one-dimensional spatial sequence data distributed along the feature lines includes: determining at least one feature line on the aircraft surface and generating a corresponding feature line point set; associating and locating the feature points in the feature line point set with the grid vertices on the aircraft surface; obtaining the physical quantity values ​​of each point on the feature line on the aircraft surface through interpolation calculation based on the physical quantities at the associated grid vertices; and generating a one-dimensional data sequence in which the physical quantity values ​​continuously change along the feature lines on the aircraft surface.

[0050] Furthermore, determining at least one feature line on the surface of the aircraft includes: extracting at least one of the following from the nose cone apex to the tail tip on the plane of symmetry: a continuous curve, a nose stagnation line, and a wing edge line, based on the geometric features of the aircraft; associating and locating the feature points in the feature line point set with the grid vertices on the surface of the aircraft includes: emitting a ray from the feature point along the approximate normal direction of its grid cell, calculating the intersection of the ray with the grid on the surface of the aircraft, and determining the nearest surface grid vertex.

[0051] Specifically, in this embodiment, S104 is the data extraction step.

[0052] The data extraction step is performed by the data extraction module, which focuses on extracting key electromagnetic input parameters from the integrated and reconstructed dataset with high fidelity. Specifically, this includes: (1) Determining surface feature lines: The data extraction module first establishes feature lines based on the aircraft's geometric features, such as extracting continuous curves from the tip of the nose cone to the tip of the tail fin, nose stationary point lines, wing edge lines, or user-specified paths on the plane of symmetry. Then, it outputs a set of feature line points L. i (i=1, 2, 3, ..., n) (2) Mesh Association: For the extracted feature lines, this module uses the ray projection method to map the L-axis of each target on the feature lines. i The location association with the target surface mesh is specifically implemented as follows: A ray is emitted from the target point along the approximate line of its corresponding mesh cell; the intersection of this ray with all surface meshes is calculated; and the nearest intersecting cell C with the target is found. j .

[0053] (3) High-fidelity interpolation: Once the data extraction module determines that the target point is within the grid cell, it performs interpolation calculations. Based on the physical quantity (such as electron density) values ​​P(V0), P(V1), and P(V2) at the grid vertices V0, V1, and V2, the physical quantity at the target point can be obtained by linear interpolation: P(L i )=λ0* P(V0)+λ1* P(V1)+λ2* P(V2) (4) One-dimensional sequence generation: all target points L on the feature line i After the above calculations, a set of spatially continuous physical quantity values ​​{P(L)} distributed along the characteristic line is finally obtained. i The one-dimensional data sequence is stored in the form of data pairs, which fully describes the continuous distribution of physical quantities along the characteristic line and is transmitted to the electromagnetic calculation module.

[0054] Because "line extraction" provides high-fidelity boundary conditions with spatial continuity, subsequent physical optics and WKB methods can realize their full accuracy potential.

[0055] S105, Electromagnetic environment calculation is performed based on the one-dimensional spatial sequence data; the electromagnetic environment calculation includes scattering characteristic calculation and transmission characteristic calculation.

[0056] Specifically, in this embodiment, S105 is the electromagnetic environment calculation step. The electromagnetic environment calculation step uses the extracted line data to analyze electromagnetic characteristics, including scattering and transmission characteristics. It specifically includes the following sub-steps: (1) Scattering Characteristics Calculation: The module solves Maxwell's equations based on the Physical Optics (PO) method to calculate the radar cross section of the aircraft. This method can perform rapid calculations while maintaining computational accuracy and is compatible with high-frequency electromagnetic wave calculations. When electromagnetic waves irradiate a target, the scattered electric field on the target surface can be obtained using the PO method. E s :

[0057] In the formula, K For the incident wave beam, r The field point position vector, r ^ The unit vector of the scattered wave. n ^ Let be the outer unit normal vector of the surface element. Z 0 represents the free-space wave impedance. r’ The source point position vector, E and H These represent the total electric field and magnetic field at the boundary, respectively. This calculation module takes the extracted surface electromagnetic data as input parameters and outputs the scattered field distribution and the trend of radar cross-section variation with angle and frequency. j is the imaginary unit, which can be understood as i. jk and jkr are multiplications of the aforementioned physical quantities and do not represent individual physical quantities. jk is wavenumber k multiplied by the imaginary unit j, representing the phase change rate; jkr is jk multiplied by the distance r, representing the total phase delay at distance r; jkr^ is jk multiplied by r^, representing the wavenumber multiplied by the unit vector of the scattering direction.

[0058] (2) Transmission characteristics calculation: The module uses the WKB method to calculate the propagation characteristics of electromagnetic waves in the plasma sheath, including signal attenuation, transmission coefficient, reflection coefficient, etc. The transmission characteristics are used to evaluate the performance of the communication link and the signal penetration capability.

[0059] Furthermore, this embodiment also includes a step for visualizing the calculation results. Specifically, the calculation visualization module receives the electromagnetic wave calculation results data and visualizes the binary dataset through a graphical interface and data reports to support user analysis and decision-making. The visualized content includes radar cross-section distribution, transmission attenuation curves, and electromagnetic transmission intensity distribution.

[0060] Secondly, this embodiment also provides an electromagnetic computing system for hypersonic vehicles based on intelligent agents, such as... Figure 2 As shown, the system includes an intelligent agent subsystem, a computing subsystem, and a post-processing subsystem; the intelligent agent subsystem includes an intelligent classification module, a pre-built strategy library, a data collection module, a data integration module, a log library, and a data extraction module; the computing subsystem includes a flow field calculation module and an electromagnetic calculation module; the post-processing subsystem includes a data receiving module and a visualization processing module; wherein: The intelligent classification module is used to automatically identify, evaluate, and classify user input parameters, and select the NS control equation based on a pre-built strategy library; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed. A pre-built strategy library is used to store rule sets for the intelligent classification module to call; The flow field calculation module is used to perform computational fluid dynamics simulation based on the selection results of the intelligent classification module to obtain flow field data; the flow field data includes physical quantities and chemical quantities. The data integration module is used to reconstruct the flow field data into electromagnetic parameters required for electromagnetic calculation, and generate a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant and conductivity; The data extraction module is used to determine the feature lines on the surface of the aircraft, extract electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generate one-dimensional spatial sequence data distributed along the feature lines. The electromagnetic calculation module is used to perform electromagnetic environment calculations based on the one-dimensional spatial sequence data; the electromagnetic environment calculations include scattering characteristic calculations and transmission characteristic calculations. The visualization module is used to visualize the electromagnetic environment calculation results.

[0061] Furthermore, the system is deployed on a hardware platform that includes a high-performance computing server, a large-capacity storage device, and a high-speed internal bus. The various modules interact with each other via the high-speed internal bus. Specifically, the data collection module captures and transforms the flow field calculation results; the log library stores, backs up, and manages all intermediate and final data; and the data receiving module receives and backs up the electromagnetic calculation results.

[0062] like Figure 3 The diagram shown is an operation flowchart of the electromagnetic computing system for hypersonic vehicles based on intelligent agents in this embodiment.

[0063] 1. Intelligent classification module: This module is responsible for automatically identifying, evaluating, and classifying the initial model parameters input by the user, replacing the traditional manual selection process. This module is integrated and deployed on the front-end server, operating by calling a pre-built strategy library. It mainly includes the following functions: (a) Format recognition: Parse the file header, extension and internal data structure of the input file, automatically determine whether the model is in two-dimensional or three-dimensional format, and adapt the data interface required for subsequent calculation processes accordingly.

[0064] (b) Parameter Evaluation: Logical and physical verification is performed on the analyzed parameters to evaluate their completeness and rationality (atmospheric environment, flight speed, flight altitude, chemical reaction indicators). For example, flight speed is used to determine whether a plasma sheath will be generated, and flight altitude and chemical composition are used to determine whether the flow is in a thermochemical non-equilibrium state.

[0065] (c) Selection decision: Based on the evaluation results, query and call the rules in the pre-made strategy library, intelligently select the most suitable NS control equation and chemical reaction model, and send the instructions to the flow field calculation module.

[0066] 2. Flow field calculation module: This module is responsible for performing computational fluid dynamics simulations, accurately solving the governing equations specified by the intelligent classification module to obtain the flow field distribution around the aircraft. This module runs on a high-performance server that handles large-scale numerical calculations and mainly includes the following functions: (a) Physical solution: The NS equations are discretized in space and iterated using the finite element method until the physical quantities of the flow field (such as temperature, pressure, density, and velocity) converge.

[0067] (b) Chemical Solving: When the chemical reaction indicator is activated, this module solves the chemical component transport equations and reaction kinetic equations in parallel, and calculates chemical parameters such as electron density and concentration of each component.

[0068] 3. Data Collection Module: This module is responsible for capturing the raw calculation results comprehensively and without loss from the output of the flow field calculation module. It connects to the flow field calculation module via a high-speed internal bus (such as PCIe) to achieve low-latency throughput of the output data; at the same time, it has multiple built-in data converters that can automatically convert the raw data into the formats required by general or specific post-processing software such as ASCII and CGNS, ensuring downstream compatibility of the data.

[0069] 4. Data Integration Module: This module is responsible for converting flow field data into input parameters that can be directly used for electromagnetic calculations. It relies on the system's central processing unit to call built-in physics algorithms to perform a series of conversion calculations on the collected flow field data. For example, it calculates electron density using the concentration of chemical reaction components combined with temperature and pressure; and it uses the formula relating electron density to plasma frequency.

[0070] The plasma frequency was derived; the collision frequency was derived using the gas density and temperature formulas, etc. At the same time, the reconstructed electromagnetic parameters (dielectric constant, conductivity, etc.) were logically correlated with the original flow field data to form a complete multiphysics dataset.

[0071] 5. Log library: The log library is responsible for storing, backing up, and managing all intermediate and final data. It is mainly implemented by a large-capacity disk array. It not only stores the reconstruction parameters output by the data integration module, but also records the decision logs of the intelligent classification module, the raw data of flow field calculation, the output data of electromagnetic calculation, etc., supporting the rapid retrieval and retrospective analysis of data.

[0072] 6. Data Extraction Module: This module is responsible for accurately extracting electromagnetic parameters from the aircraft surface from the integrated 3D dataset. Its computation relies on the system-integrated graphics processing unit hardware and efficient network topology processing hardware. First, based on the aircraft's geometric model, the module automatically identifies its surface mesh nodes. Simultaneously, along preset surface contour lines or feature lines, it samples and extracts a one-dimensional, continuous sequence of parameters from the 3D dataset using linear interpolation (e.g., the electron density distribution along a generatrix from the aircraft's nose cone to its tail). This "line extraction" method avoids traditional hierarchical approximations and fully preserves the spatial gradient information of the data.

[0073] 7. Electromagnetic Calculation Module: This module is responsible for simulating the interaction between electromagnetic waves and the spacecraft and its plasma sheath based on extracted surface line data. It mainly includes the following functions: (a) Calculation of electromagnetic scattering characteristics: The radar cross section of the aircraft is calculated by solving Maxwell's equations using the physical optics method. The calculation process uses the extracted surface electromagnetic parameters as boundary conditions and outputs scattering field data that vary with angle and frequency.

[0074] (b) Calculation of electromagnetic propagation characteristics: The WKB calculation method is used to analyze the propagation effect of electromagnetic waves in the plasma sheath and calculate the signal attenuation, transmission and reflection coefficients to evaluate the performance of the communication link.

[0075] 8. Data receiving module: This module is responsible for receiving the complete output results from the electromagnetic computing module. It connects to the electromagnetic computing module via a high-speed internal bus (such as PCIe) to achieve low-latency throughput of the output data; at the same time, it connects to the intelligent agent log library to quickly save the received data to the log library for backup.

[0076] 9. Visualization Processing Module: The visualization processing module serves as the interactive interface between the system and the user. It is responsible for converting binary results into intuitive graphics and reports. This module is responsible for receiving data from the data receiving module and generating visualization results such as radar cross section distribution maps, transmission attenuation curves, and electromagnetic field intensity cloud maps. It supports interactive analysis, result export, and report generation for users.

[0077] 10. Pre-built strategy library: As the knowledge support for the intelligent classification module, it stores a set of rules formed by domain expert knowledge and historical computational experience, mainly stored in high-speed memory (such as SSD) for the intelligent classification module to read in real time; the rules in the library exist in the form of "condition-action" pairs. For example, the rule is defined as: "Condition: flight speed > Mach 20, flight altitude > 60km; Action: select the three-dimensional compressible Navier-Stokes equation, select the temperature-based input window, start the 7-component chemical reaction model (such as Gupta_7), and set the viscous solid wall boundary conditions." Furthermore, this embodiment uses the calculation of the electromagnetic characteristics of a classic hypersonic target—the RAMC-II aircraft—at a flight altitude of 60 km and a flight speed of Mach 20 as an example to fully demonstrate the implementation process of this system. The software program of this system is deployed on one or more high-performance computing servers interconnected by a network. These servers are equipped with multi-core central processing units, large-capacity memory, high-speed solid-state drives, and professional graphics processing units. The specific implementation steps are as follows: Step 1: Parameter Input and Intelligent Classification (1) User operation: The user uploads the three-dimensional geometric model mesh file of the RAMC-II aircraft (such as .stl or .obj format) through the system's graphical interface, and enters the flight conditions in the parameter setting interface: flight altitude 60km, flight speed Mach 20 (about 6800m / s), and checks the "Enable chemical reaction" option.

[0078] (2) System execution - intelligent classification: The intelligent classification module automatically parses the uploaded file header, confirms it is in 3D model format, and prepares the 3D calculation process accordingly. Furthermore, the module evaluates the input flight conditions. Based on the high speed of Mach 20 and the altitude of 60 km, it determines the flow to be compressible and that a plasma sheath will inevitably be generated; the activation of the chemical reaction flag indicates the need for thermochemical nonequilibrium calculations.

[0079] Based on the above assessment, the intelligent classification module queries the pre-built strategy library and matches the rule: "If the flight speed is > Mach 5 and chemical reaction is enabled, select the three-dimensional compressible NS control equation and adopt a 7-component air chemical reaction model (such as the Park model), and set viscous, no-slip wall boundary conditions." Subsequently, this decision instruction is sent to the flow field calculation module.

[0080] Step 2: Flow field environment calculation After receiving the instruction from the intelligent classification module, the flow field calculation module starts the computational fluid dynamics solver.

[0081] Physical process calculation: The solver uses the finite volume method to discretize and iterate through 10,000 steps to solve the three-dimensional compressible Navier-Stokes equations, calculating the temperature, pressure, density, and velocity distributions of the flow field. The calculation continues until the residuals of all physical quantities converge to a preset threshold of 10. -5 At this point, the flow field reaches a stable state.

[0082] Chemical process calculation: In parallel, the solver solves the chemical component transport equations based on the specified 7-component chemical reaction model, and calculates the concentration of chemical components such as electron density and ion density at each point in the flow field.

[0083] Step 3: Flow field data collection and format conversion The data collection module tracks the flow field calculation process and, upon convergence, immediately reads all physical quantities (temperature, pressure, etc.) and chemical quantities (electron density, etc.) from the CFD solver via the application programming interface. In this embodiment, the RAMC spacecraft flow field contains 79% nitrogen, 19% oxygen, and 2% other gases, with a stagnation temperature of 7000 K and a stagnation pressure of 9000 Pa.

[0084] To achieve seamless integration with subsequent modules, this module automatically converts the raw binary data into the standard ASCII format, ensuring the integrity of the data structure and ease of subsequent processing.

[0085] Step 4: Data Integration and Restructuring The data integration module reads the flow field data in ASCII format. Then, it calls the built-in physics algorithm for reconstruction calculations: using the electron density *ne* obtained from chemical reaction concentration correlation calculations; and using the frequency formula... The plasma frequency ωp was calculated point by point; the collision frequency between electrons and neutral particles was calculated using empirical formulas based on the gas density and temperature in the flow field. All these reconstructed electromagnetic parameters (plasma frequency, collision frequency, and the derived dielectric constant and conductivity) were associated and packaged with the original flow field data and stored in a log repository for backup and recording.

[0086] In this embodiment, calculations show that the peak electron density of the flow field is 1020 (Ne / m3), the axial plasma frequency is 25GHz-9GHz, and the collision frequency is 0.9GHz-2GHz.

[0087] Step 5: Electromagnetic parameter extraction The data extraction module acquires the 3D dataset integrated by the data integration module. First, it identifies the surface mesh of the RAM-II aircraft's geometric model. Then, along a key feature line from the nose cone apex to the tail (such as generatrices on symmetry planes, vertical wall normals, and stationary point lines), linear interpolation sampling techniques are used to accurately extract the electron density, plasma frequency, and collision frequency data for all nodes on this line. Finally, a one-dimensional line array that maintains spatial continuity is generated and sent to the electromagnetic calculation module for computation. This "line extraction" method avoids the information loss caused by traditional coarse layering and, compared to "3D volume extraction," avoids full 3D field calculations, reducing computational load.

[0088] Step 6: Electromagnetic Environment Calculation (1) Scattering characteristics calculation: The module adopts the physical optics (PO) method. The extracted surface line data is used as electromagnetic boundary conditions. The incident electromagnetic wave frequency (e.g., 10 GHz in the X-band) and incident angle (0° perpendicular to the wall) are set, and Maxwell's equations are solved. The scattered field of the aircraft is obtained by calculation and integration, and then the curve of its radar cross section as a function of observation angle and incident frequency is calculated.

[0089] (2) Transmission characteristics calculation: The module adopts the WKB method. The propagation process of communication electromagnetic waves of continuous frequency (0-10GHz) in the calculated plasma sheath (whose characteristics are described by the extracted parameters) is analyzed, and the attenuation constant, transmission coefficient and reflection coefficient of the signal are calculated to evaluate the severity of the aircraft's "blackout" effect and the performance of the communication link under this flight condition.

[0090] This embodiment calculates that when the incident electromagnetic wave is 10 GHz, the penetration coefficient increases to 80% and the radar cross section decreases by 15 dB.

[0091] Step 7: Visualizing and Outputting Results The visualization module acquires the raw results of electromagnetic calculations from the data receiving module. On the graphics workstation, it renders the binary data into intuitive charts, including: a radar cross-section distribution cloud map of the RAM-II aircraft; a polar coordinate plot of the radar cross-section as a function of azimuth; a curve showing the signal attenuation of electromagnetic waves in the plasma sheath as a function of frequency; and a curve showing the signal transmission / reflection coefficient of electromagnetic waves in the plasma sheath as a function of frequency. Users can interactively view these charts on the graphical interface, zoom, rotate, and analyze them, and export the charts and data reports for final technical reports and performance evaluations.

[0092] Through the seven logically rigorous steps outlined above (steps 1-7), this embodiment fully demonstrates how the system and method, starting from the initial parameter input, automatically completes flow field calculations, data conversion and integration through an intelligent agent, and utilizes high-precision line extraction data as input parameters for electromagnetic simulation, ultimately providing users with intuitive visualization results. The entire process minimizes human intervention, solves the challenge of data integration under multi-physics coupling, and significantly improves computational accuracy by preserving complete surface gradient data, fully demonstrating the superiority and practicality of this invention.

[0093] Compared with existing technologies, the intelligent calculation method and system for the electromagnetic characteristics of hypersonic vehicles based on intelligent agents provided by this invention, through its complete technical solution, brings the following significant beneficial effects: 1. Improved accuracy and physical fidelity of electromagnetic calculations. The core innovation of this invention lies in the use of "data line extraction" technology. It directly extracts a one-dimensional continuous sequence of electromagnetic parameters from the integrated and reconstructed 3D dataset along the surface contour or feature lines of the aircraft through linear interpolation sampling. This method completely abandons the simplified model in the traditional WKB method that approximates the plasma sheath as a finite number of discrete dielectric constant layers. By fully preserving the spatial gradient information and continuity of physical parameters (such as electron density and plasma frequency) along the surface, it fundamentally avoids model errors introduced by coarse layering. Compared to traditional layering strategies, this invention can obtain more accurate electromagnetic scattering and transmission characteristic calculation results at the same grid density, especially reflecting the electromagnetic effects of key regions such as shock waves and boundary layers more realistically.

[0094] 2. This invention achieves efficient and automated integration of interdisciplinary data, reducing manual intervention and the tediousness of integration. It constructs a fully automated data processing pipeline through an "intelligent classification module," a "data integration module," and a "log library." The intelligent classification module automatically identifies input parameters and selects governing equations; the data integration module automatically reconstructs the flow field data into the input parameters required for electromagnetic calculations using built-in physical algorithms (such as electron density and plasma frequency calculation formulas), and stores them in association. This solution solves the information integration problem caused by the incompatibility of data formats and grid systems between CFD and CEM. The system achieves full-link automation from flow field calculation to electromagnetic parameter generation. Researchers no longer need to manually perform tedious data format conversion, transmission, and preprocessing, greatly improving work efficiency, reducing the risk of errors due to human error, and enabling rapid, multi-condition numerical analysis.

[0095] 3. It lowers the barrier to entry for algorithm use and reduces reliance on specific human experience, thereby enhancing the level of intelligence. The intelligent agent subsystem (especially the intelligent classification module) integrates a "pre-built strategy library" formed by domain expert knowledge, enabling it to automatically and intelligently select the most suitable NS control equations and chemical reaction models based on input flight conditions (altitude, speed, chemical reactions, etc.). This design allows researchers without a strong background in fluid mechanics to quickly and accurately initiate complex multiphysics simulation calculations.

[0096] 4. A complete, efficient, and traceable simulation analysis ecosystem has been constructed. The technical solution of this invention is a complete system integrating intelligent classification, flow field calculation, data integration, line extraction, electromagnetic calculation, and visualization. The system's designed "log library" records all intermediate data and process logs from initial decision-making to the final result. This ecosystem ensures high reproducibility and traceability of the calculation process. The result of any calculation can be fully traced, facilitating error analysis, scheme optimization, and knowledge accumulation. Simultaneously, the modules are connected via a high-speed bus and accelerated using hardware such as GPUs (e.g., data extraction and electromagnetic calculation), maintaining high accuracy while preserving overall computational efficiency, thus meeting the needs of large-scale, high-precision simulations in engineering practice.

[0097] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for electromagnetic calculation of hypersonic vehicles, characterized in that, include: A NS control equation is selected from a pre-built rule base based on user input parameters; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed. The flow field data is calculated based on the Navier-Stokes governing equations and collected; the flow field data includes physical and chemical quantities. The flow field data is reconstructed into electromagnetic parameters required for electromagnetic calculation, generating a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant, and conductivity; The surface feature lines of the aircraft are determined, electromagnetic input parameters are extracted from the reconstructed electromagnetic parameter set, and one-dimensional spatial sequence data distributed along the feature lines are generated. Electromagnetic environment calculations are performed based on the one-dimensional spatial sequence data; the electromagnetic environment calculations include scattering characteristic calculations and transmission characteristic calculations.

2. The electromagnetic calculation method for hypersonic vehicles according to claim 1, characterized in that, The calculation of flow field data based on the NS governing equations includes calculations of physical quantities and chemical quantities. The physical quantity calculation includes: using the finite element method to discretize and solve the NS control equations on a spatial grid to calculate the flow field temperature and pressure; In three-dimensional Cartesian coordinates, the NS governing equations are expressed as: Where Q is a conserved quantity, and E, F, and G are diffusion terms in the x, y, and z directions, respectively. V F V G V Let S be the convection term in three directions, and S be the source vector.

3. The electromagnetic calculation method for hypersonic vehicles according to claim 2, characterized in that, The calculation of the chemical quantities includes: If the user-input parameters meet the chemical reaction conditions, the chemical reaction transport equations and reaction kinetic equations in the Park reaction model or Gupta reaction model are solved according to the actual environmental conditions, and the chemical variables are calculated; the chemical variables include electron density and reaction rate.

4. The electromagnetic calculation method for hypersonic vehicles according to claim 1, characterized in that, The step of reconstructing the flow field data into electromagnetic parameters required for electromagnetic calculation, and generating a reconstructed electromagnetic parameter set, includes: Plasma frequency: ; Where, n e where e is the electron density, e is the electron charge, and m is the electron density. e Where ε is the electron mass, and ε0 is the dielectric constant; Using an empirical formula based on gas density and temperature: Calculate the collision frequency between electrons and neutral particles; where For collision frequency, Where T is the total number of particles and T is the temperature; Using a collision-based plasma Drude model: Used to calculate the relative permittivity; , used to calculate relative conductivity; where The dielectric constant in vacuum is . The relative permittivity, and These are the real and imaginary parts of the relative permittivity, respectively. The incident wave frequency, For collision frequency, relative conductivity and , respectively, represent the real and imaginary parts of the relative conductivity; i is the imaginary part of the complex function. The relative permittivity, It is the vacuum dielectric constant; A set of reconstructed electromagnetic parameters is generated based on the plasma frequency, collision frequency, dielectric constant, and conductivity, and stored in the log library.

5. The electromagnetic calculation method for hypersonic vehicles according to any one of claims 1-4, characterized in that, The process of determining the surface feature lines of the aircraft, extracting electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generating one-dimensional spatial sequence data distributed along the feature lines includes: Identify at least one feature line on the surface of the aircraft and generate a corresponding feature line point set; associate and locate the feature points in the feature line point set with the mesh vertices on the surface of the aircraft. Based on the physical quantities at the associated grid vertices, the physical quantity values ​​of each point on the feature line of the aircraft surface are obtained by interpolation calculation; Generate a one-dimensional data sequence in which the physical quantity values ​​change continuously along the feature lines on the surface of the aircraft.

6. The electromagnetic calculation method for hypersonic vehicles according to claim 5, characterized in that, The determination of at least one feature line on the surface of the aircraft includes: extracting at least one of the following based on the geometric features of the aircraft: a continuous curve on the plane of symmetry from the apex of the nose cone to the tip of the tail fin: a nose stagnation line; and a wing edge line. The step of associating and locating the feature points in the feature line point set with the mesh vertices on the aircraft surface includes: emitting a ray from the feature point along the approximate normal direction of its mesh cell, calculating the intersection of the ray with the mesh on the aircraft surface, and determining the nearest surface mesh vertex.

7. The electromagnetic calculation method for hypersonic vehicles according to claim 1, characterized in that, The calculation of the scattering characteristics includes: When electromagnetic waves irradiate a target, the scattered electric field E on the target surface is obtained using the physical optics PO method. s : In the formula, K is the incident wave beam, and r is the field point position vector. The unit vector of the scattered wave. Z0 is the external unit normal vector of the surface element, and Z0 is the free space wave impedance. Let E be the source point location vector, and let H be the total electric field and magnetic field at the boundary, respectively.

8. The electromagnetic calculation method for hypersonic vehicles according to claim 1, characterized in that, The transmission characteristic calculation includes: The propagation characteristics of electromagnetic waves in a plasma sheath were calculated using the WKB method; the propagation characteristics include signal attenuation, transmission coefficient, and reflection coefficient.

9. An electromagnetic computing system for hypersonic vehicles based on intelligent agents, characterized in that, It includes an intelligent agent subsystem, a computing subsystem, and a post-processing subsystem; the intelligent agent subsystem includes an intelligent classification module, a pre-built strategy library, a data collection module, a data integration module, a log library, and a data extraction module; the computing subsystem includes a flow field calculation module and an electromagnetic calculation module; the post-processing subsystem includes a data receiving module and a visualization processing module; wherein: The intelligent classification module is used to automatically identify, evaluate, and classify user input parameters, and select the NS control equation based on a pre-built strategy library; the user input parameters include atmospheric environmental parameters, flight altitude, and flight speed. A pre-built strategy library is used to store rule sets for the intelligent classification module to call; The flow field calculation module is used to perform computational fluid dynamics simulation based on the selection results of the intelligent classification module to obtain flow field data; the flow field data includes physical quantities and chemical quantities. The data integration module is used to reconstruct the flow field data into electromagnetic parameters required for electromagnetic calculation, and generate a reconstructed electromagnetic parameter set; the electromagnetic parameters include plasma frequency, collision frequency, dielectric constant and conductivity; The data extraction module is used to determine the feature lines on the surface of the aircraft, extract electromagnetic input parameters from the reconstructed electromagnetic parameter set, and generate one-dimensional spatial sequence data distributed along the feature lines. The electromagnetic calculation module is used to perform electromagnetic environment calculations based on the one-dimensional spatial sequence data; the electromagnetic environment calculations include scattering characteristic calculations and transmission characteristic calculations. The visualization module is used to visualize the electromagnetic environment calculation results.

10. The electromagnetic computing system for hypersonic vehicles based on intelligent agents according to claim 9, characterized in that, The system is deployed on a hardware platform that includes a high-performance computing server, a large-capacity storage device, and a high-speed internal bus. The modules interact with each other through the high-speed internal bus. The system includes a data collection module for capturing and converting flow field calculation results; a log library for storing, backing up, and managing all intermediate and final data; and a data receiving module for receiving and backing up electromagnetic calculation results.

Images