Beam decomposition method for electromagnetic acceleration calculation of electrically large target
By employing a conical beam decomposition and regional parallel computing method, the problems of high resource consumption and low accuracy in electromagnetic scattering calculations of electrically large targets are solved, achieving efficient and accurate electromagnetic simulation and supporting large-scale parallel computing.
Patent Information
- Application Number
- CN202511524769.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-24
- Publication Date
- 2026-02-10
AI Technical Summary
Existing technologies consume huge amounts of computational resources and have low computational efficiency when processing electromagnetic scattering calculations for electrically large targets. High-frequency approximation methods lead to a decrease in accuracy, while domain decomposition methods have complex calculations due to coupling between sub-regions and are difficult to achieve independent parallel computing, thus failing to effectively utilize high-performance computing resources.
A cone beam decomposition and regional parallel computing method is adopted. The sub-beam center is determined by the golden spiral method, the amplitude attenuation threshold of the Gaussian beam edge is limited, and high-fidelity beam decomposition is performed by local convolution and phase cloning operations to ensure that each sub-beam is calculated independently.
It significantly improves computational efficiency, suppresses edge diffraction effects, enhances computational accuracy and numerical stability, and possesses good scalability, supporting large-scale distributed computing.
Smart Images
Figure CN121503002A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of computational electromagnetics and high-performance electromagnetic simulation technology, and in particular to a beam decomposition method for electromagnetic acceleration computation of electrically large targets. Background Technology
[0002] In the field of computational electromagnetics and target characteristic simulation, the calculation of electromagnetic scattering of electrically large targets is a core challenge, and the results are crucial for the stealth design, target identification, and electromagnetic compatibility assessment of platforms such as aircraft and ships. Since the target size is much larger than the wavelength, directly using full-wave numerical methods (such as the method of moments and the finite element method) results in enormous computational costs, and is even difficult to implement.
[0003] Traditional acceleration algorithms (such as high-frequency approximation methods) often suffer from accuracy degradation when dealing with complex structures due to factors such as edge diffraction and multiple scattering, while pure domain decomposition methods face the problems of complex computational coupling between sub-regions and significant truncation errors. The trade-off between computational efficiency and accuracy is particularly prominent when high-fidelity simulation of full-space scattering characteristics is required. Summary of the Invention
[0004] Given the limitations of existing technologies, such as truncation edge effects and global computational resource bottlenecks, which hinder efficient handling of electrically large targets, this invention provides a beam decomposition and region-parallel computing method for calculating the electromagnetic scattering characteristics of electrically large targets (such as aircraft, ships, or vehicles). Through a conical beam decomposition and region mapping mechanism, it achieves distributed parallel computing for complex scattering problems, significantly improving computational efficiency while maintaining computational accuracy. Compared to traditional full-wave algorithms, this method achieves breakthroughs in computational scale and scalability, providing a new technical path for the electromagnetic characteristic analysis of electrically large platforms.
[0005] This invention provides a beam decomposition method for electromagnetic acceleration calculations of electrically large targets, comprising:
[0006] Step 1: Obtain the sampling data file and convert the relevant parameter information of the antenna pattern uniform sampling grid representation into a data organization format that can be queried and processed;
[0007] Step 2: Use the golden spiral method to uniformly sample the sphere and determine the center of the sub-beam;
[0008] Step 3: By limiting the amplitude attenuation threshold at the edge of the Gaussian beam and calculating the beam variance, the shape and width of the Gaussian beam to be used are controlled as a constraint.
[0009] Step 4: Use local convolution, amplitude alignment and phase cloning operations to complete high-fidelity beam decomposition of the radiation pattern, ensuring that each sub-beam strictly inherits the original field characteristics.
[0010] In some embodiments of the present invention, the step of acquiring the sampling data file and converting the relevant parameter information read from the uniform sampling grid representation of the antenna pattern into a data organization format capable of query processing includes:
[0011] Import the uniformly sampled gridded data file of the m-row n-column antenna pattern with equal spacing Δd;
[0012] When reading the antenna pattern using a uniformly sampled grid representation, different elevation angles θ and azimuth angles... The following belong to θ polarization, Amplitude A and phase information θ in the polarization direction phase And convert it into a data organization format that facilitates subsequent query processing.
[0013] In some embodiments of the present invention, the step of determining the sub-beam center by uniform sampling on a sphere using the golden spiral method includes:
[0014] Using the golden spiral method on a unit sphere S 2 A total of T sampling points are obtained by uniformly sampling at approximately equal solid angles.
[0015] The sampling center obtained from the i-th sampling is denoted as ω. i The solid angle of the center is denoted as Ω. i solid angle Ω i Corresponding unit sphere S 2 The neighboring region on is denoted as O. i Where i = 1, 2, ..., T, and T is a positive integer;
[0016] The obtained series of sampling points are used as the center of the preset sub-beam, and the center coordinates ω are recorded. i (x,y,z).
[0017] In some embodiments of the present invention, controlling the shape and width of the Gaussian beam to be used by limiting the amplitude attenuation threshold at the edge of the Gaussian beam and calculating the beam variance as a constraint includes:
[0018] Each sampling center ω i The corresponding initial beam setting is a Gaussian beam G. i Furthermore, the attenuation threshold variable δ of the Gaussian beam edge amplitude is defined as a constraint condition, and the variance parameter σ is calculated and determined to control the shape and width of the Gaussian beam to be used.
[0019] In some embodiments of the present invention, the high-fidelity beam decomposition of the radiation pattern using local convolution, amplitude alignment, and phase cloning operations, ensuring that each sub-beam strictly inherits the original field characteristics, includes:
[0020] For the radiation pattern already given and imported in step one, use the Gaussian beam G with a defined shape and width. i (x i ,y i ,z i ,σ) around the sub-beam center ω uniformly acquired in step one i The neighborhood O of (x,y,z) i Perform convolution operations.
[0021] In some embodiments of the present invention, the step of performing high-fidelity beam decomposition of the radiation pattern using local convolution, amplitude alignment, and phase cloning operations to ensure that each sub-beam strictly inherits the original field characteristics further includes:
[0022] For each center ω i The main pointer points to μ i Fixed sub-beam G i , for θ polarization and The amplitude of both polarization directions is adjusted and matched to achieve the desired θ polarization at the center of the sub-beam. Polarization gain amplitude A θ , Align with the previously imported original orientation pattern information.
[0023] In some embodiments of the present invention, the step of performing high-fidelity beam decomposition of the radiation pattern using local convolution, amplitude alignment, and phase cloning operations to ensure that each sub-beam strictly inherits the original field characteristics further includes:
[0024] The phase copy of the corresponding position in the original radiation pattern is cloned as the phase θ of the corresponding position in the sub-beam reconstruction beam. phase To be incorporated into subsequent electromagnetic calculations.
[0025] Compared with existing technologies, the beam decomposition method for electromagnetic acceleration computation of electrically large targets provided by this invention effectively solves the problems of traditional full-wave electromagnetic computation methods, which suffer from huge computational resource consumption and memory requirements when processing electrically large targets, resulting in low computational efficiency and difficulty in completing accurate simulation and efficient solution of ultra-electrically large targets within a limited time. It also addresses the problem that existing high-frequency approximation methods (such as physical optics) are prone to non-physical edge diffraction effects at sub-region truncation points, severely disrupting the continuity of the field distribution, leading to a significant decrease in computational accuracy and affecting the reliability of simulation results. Furthermore, it solves the problems of strong electromagnetic coupling between sub-domains in traditional domain decomposition techniques, making it difficult to achieve truly independent computation, lacking a parallelizable and scalable computational architecture, limiting the scalability of parallel scale, and failing to meet the needs of large-scale distributed computing, thus hindering the effective utilization of high-performance computing resources.
[0026] The present invention has the following beneficial effects:
[0027] (1) Significantly improved computational efficiency: By decomposing the original problem into multiple independently computable beam sub-problems and adopting an efficient parallel computing architecture, the computation time for electromagnetic scattering of electrically large targets is reduced by more than an order of magnitude, greatly improving the simulation speed.
[0028] (2) Effective suppression of edge diffraction effect: By introducing a conical beam and strictly controlling the edge attenuation threshold, the non-physical edge scattering caused by truncation in the traditional domain decomposition method is significantly reduced, and the numerical stability and accuracy of the calculation are improved.
[0029] (3) It has good scalability: Based on the beam-region precise mapping mechanism, this method can be naturally extended to parallel processing in ultra-large-scale computing clusters, breaking through the limitations of traditional methods in terms of memory and computing scale, and providing a feasible path for fine electromagnetic modeling of electrically large targets. Attached Figure Description
[0030] Figure 1 A flowchart of a beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in an embodiment of the present invention;
[0031] Figure 2 A visualization of the beam pattern model of the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in an embodiment of the present invention;
[0032] Figure 3 The distribution effect of the sampling sub-beam centers of the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in the embodiments of the present invention is shown in the figure.
[0033] Figure 4 A Gaussian beam model diagram of the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in an embodiment of the present invention;
[0034] Figure 5 This is a diagram showing the Gaussian sub-beam single-point convolution reconstruction effect of the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in an embodiment of the present invention.
[0035] Figure 6 This is a comparison of the reconstruction effects of the Gaussian sub-beam pattern of the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided in the embodiments of the present invention. Detailed Implementation
[0036] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0037] Various embodiments and features of this application are described herein with reference to the accompanying drawings.
[0038] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.
[0039] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application, which have the features described in the claims and are therefore all within the scope of protection defined herein.
[0040] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.
[0041] Specific embodiments of this application are described below with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to ascertain the true intent based on the user's historical operations, and to avoid unnecessary or redundant details that would obscure this application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in various ways with substantially any suitable detailed structure.
[0042] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.
[0043] This invention provides a beam decomposition method for electromagnetic acceleration calculations of electrically large targets, such as... Figures 1 to 6 As shown, it includes:
[0044] Step 1: Obtain the sampling data file and convert the relevant parameter information of the antenna pattern uniform sampling grid representation into a data organization format that can be queried and processed;
[0045] Step 2: Uniform sampling is performed on the sphere using the golden spiral method to determine the sub-beam center, thereby realizing the automatic allocation and load balancing of computing tasks among multiple beams and laying the foundation for efficient parallel computing;
[0046] Step 3: By limiting the amplitude attenuation threshold at the edge of the Gaussian beam and calculating the beam variance as a constraint, the shape and width of the Gaussian beam to be used are controlled, effectively suppressing non-physical scattering generated at the beam cutoff point and ensuring high accuracy of local area calculations.
[0047] Step 4: High-fidelity beam decomposition of the radiation pattern is completed using local convolution, amplitude alignment, and phase cloning operations to ensure that each sub-beam strictly inherits the original field characteristics. That is, a beam reconstruction algorithm based on local convolution, amplitude alignment, and phase cloning is used to ensure that each sub-beam strictly inherits the amplitude and phase characteristics of the original radiation pattern, supporting independent and high-fidelity parallel computation of each sub-region.
[0048] To facilitate understanding of the above technical solutions, a detailed explanation is provided below with reference to the accompanying drawings and specific examples:
[0049] Step 1: Import the uniformly sampled gridded data file with equal spacing Δd for the m-row n-column antenna pattern (e.g., Figure 2 (As shown). Reading the antenna pattern using a uniformly sampled grid at different elevation angles θ and azimuth angles. The following belong to θ polarization, Amplitude A and phase information θ in the polarization direction phase And convert it into a data organization format that facilitates subsequent query processing.
[0050] Step 2: Use the golden spiral method on a unit sphere S 2 A total of T sampling points were obtained by uniform sampling at approximately equal solid angles (e.g., Figure 3 (As shown). The sampling center obtained from the i-th (i = 1, 2, ..., T) sampling is denoted as ω. i The solid angle of the center is denoted as Ω. i solid angle Ω i Corresponding unit sphere S 2 The neighboring region on is denoted as O. i The obtained series of sampling points are used as the center of the preset sub-beam, and the center coordinates ω are recorded. i (x,y,z).
[0051] Step 3: For each sampling center ω i The corresponding initial beam setting is a Gaussian beam G. i (like Figure 4 As shown), and limit the attenuation threshold variable δ of the Gaussian beam edge amplitude, thereby using the constraint condition to calculate and determine the variance parameter σ to control the shape and width of the Gaussian beam to be used;
[0052] Step 4: For the radiation pattern given and imported in Step 1, use a Gaussian beam G with a fixed variance of σ, i.e., a fixed shape and width. i (x i ,y i ,z i ,σ) around the sub-beam center ω uniformly acquired in step one i The neighborhood O of (x,y,z) i Perform convolution operations (such as) Figure 5 (as shown);
[0053] Step 5: For each center ω i The main pointer points to μ i Fixed sub-beam G i , for θ polarization and The amplitude of both polarization directions is adjusted and matched to achieve the desired θ polarization at the center of the sub-beam. Polarization gain amplitude A θ , Align with the previously imported original radiation pattern information (e.g.) Figure 6 (as shown);
[0054] Step 6: Copy and clone the phase at the corresponding position of the original radiation pattern to the phase θ at the corresponding position of the sub-beam reconstruction beam. phase To be incorporated into subsequent electromagnetic calculations.
[0055] As can be seen from the above technical solutions, the beam decomposition method for electromagnetic acceleration calculation of electrically large targets provided by the above embodiments of the present invention effectively solves the problems of traditional full-wave electromagnetic calculation methods, which suffer from huge computational resource consumption and memory requirements when processing electrically large targets, resulting in low computational efficiency and difficulty in completing accurate simulation and efficient solution of ultra-electrically large targets within a limited time; and the problems of existing high-frequency approximation methods (such as physical optics methods) easily generating non-physical edge diffraction effects at the sub-region truncation points, which seriously disrupt the continuity of the field distribution, leading to a significant decrease in computational accuracy and affecting the reliability of simulation results. In addition, it can also solve the problems of strong electromagnetic coupling between sub-domains in traditional domain decomposition techniques, making it difficult to achieve truly independent calculations, lacking a parallelizable and scalable computing architecture, limiting the scalability of parallel scale, making it difficult to cope with the needs of large-scale distributed computing, and failing to effectively utilize high-performance computing resources.
[0056] In addition, it also has the following beneficial effects:
[0057] (1) Significantly improved computational efficiency: By decomposing the original problem into multiple independently computable beam sub-problems and adopting an efficient parallel computing architecture, the computation time for electromagnetic scattering of electrically large targets is reduced by more than an order of magnitude, greatly improving the simulation speed.
[0058] (2) Effective suppression of edge diffraction effect: By introducing a conical beam and strictly controlling the edge attenuation threshold, the non-physical edge scattering caused by truncation in the traditional domain decomposition method is significantly reduced, and the numerical stability and accuracy of the calculation are improved.
[0059] (3) It has good scalability: Based on the beam-region precise mapping mechanism, this method can be naturally extended to parallel processing in ultra-large-scale computing clusters, breaking through the limitations of traditional methods in terms of memory and computing scale, and providing a feasible path for fine electromagnetic modeling of electrically large targets.
[0060] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.
Claims
1. A beam decomposition method for electromagnetic acceleration calculations of electrically large targets, characterized in that, include: Step 1: Obtain the sampling data file and convert the relevant parameter information of the antenna pattern uniform sampling grid representation into a data organization format that can be queried and processed; Step 2: Use the golden spiral method to uniformly sample the sphere and determine the center of the sub-beam; Step 3: By limiting the amplitude attenuation threshold at the edge of the Gaussian beam and calculating the beam variance, the shape and width of the Gaussian beam to be used are controlled as a constraint. Step 4: Use local convolution, amplitude alignment and phase cloning operations to complete high-fidelity beam decomposition of the radiation pattern, ensuring that each sub-beam strictly inherits the original field characteristics.
2. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 1, characterized in that, The process of acquiring the sampled data file and converting the relevant parameter information from the uniformly sampled grid representation of the antenna pattern into a data organization format that can be queried and processed includes: Import the uniformly sampled gridded data file of the m-row n-column antenna pattern with equal spacing Δd; When reading the antenna pattern using a uniformly sampled grid representation, different elevation angles θ and azimuth angles... The following belong to θ polarization, Amplitude A and phase information θ in the polarization direction phase And convert it into a data organization format that facilitates subsequent query processing.
3. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 2, characterized in that, The process of determining the sub-beam center by uniformly sampling on a sphere using the golden spiral method includes: Using the golden spiral method on a unit sphere S 2 A total of T sampling points are obtained by uniformly sampling at approximately equal solid angles. The sampling center obtained from the i-th sampling is denoted as ω. i The solid angle of the center is denoted as Ω. i solid angle Ω i Corresponding unit sphere S 2 The neighboring region on is denoted as O. i Where i = 1, 2, ..., T, and T is a positive integer; The obtained series of sampling points are used as the center of the preset sub-beam, and the center coordinates ω are recorded. i (x,y,z).
4. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 3, characterized in that, The method of controlling the shape and width of the Gaussian beam to be used by limiting the amplitude attenuation threshold at the edge of the Gaussian beam and calculating the beam variance as a constraint includes: Each sampling center ω i The corresponding initial beam setting is a Gaussian beam G. i Furthermore, the attenuation threshold variable δ of the Gaussian beam edge amplitude is defined as a constraint condition, and the variance parameter σ is calculated and determined to control the shape and width of the Gaussian beam to be used.
5. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 4, characterized in that, The high-fidelity beam decomposition of the radiation pattern is achieved by employing local convolution, amplitude alignment, and phase cloning operations, ensuring that each sub-beam strictly inherits the original field characteristics, including: For the radiation pattern already given and imported in step one, use the Gaussian beam G with a defined shape and width. i (x i ,y i ,z i ,σ) around the sub-beam center ω uniformly acquired in step one i The neighborhood O of (x,y,z) i Perform convolution operations.
6. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 5, characterized in that, The method of employing local convolution, amplitude alignment, and phase cloning operations to achieve high-fidelity beam decomposition of the radiation pattern, ensuring that each sub-beam strictly inherits the original field characteristics, also includes: For each center ω i The main pointer points to μ i Fixed sub-beam G i , for θ polarization and The amplitude of both polarization directions is adjusted and matched to achieve the desired θ polarization at the center of the sub-beam. Polarization gain amplitude A θ , Align with the previously imported original orientation pattern information.
7. The beam decomposition method for electromagnetic acceleration calculation of electrically large targets according to claim 6, characterized in that, The method of employing local convolution, amplitude alignment, and phase cloning operations to achieve high-fidelity beam decomposition of the radiation pattern, ensuring that each sub-beam strictly inherits the original field characteristics, also includes: The phase copy of the corresponding position in the original radiation pattern is cloned as the phase θ of the corresponding position in the sub-beam reconstruction beam. phase To be incorporated into subsequent electromagnetic calculations.