A bin pre-processing reverse ray blanking method

Abstract

This invention relates to the field of high-frequency electromagnetic calculation technology using physical optics, specifically to a method for reverse ray tracing in preprocessing of surface elements. When reading the triangular mesh information of the model, this invention calculates the normal vector, center coordinates, and axial bounding box (AABB) of each triangular surface element. Triangular surface elements whose dot product of the normal vector and the plane wave incident direction is less than 0 are designated as potential illuminated surfaces and included in the potential illuminated area set. Then, a secondary illumination judgment is performed on the triangular surface elements in each potential illuminated area to obtain the final set of effective illuminated surfaces. Finally, the scattering contributions of each effective illuminated surface element are summed to obtain the total scattering field of the target, and the radar cross section (RCS) is then calculated. This invention systematically solves the core contradictions of existing technologies—insufficient accuracy of the normal vector dot product, two-dimensional approximation of projection clipping, and low efficiency of traditional reverse ray tracing—through an integrated design; it provides a novel technical solution for the electromagnetic scattering characteristics analysis of electrically large targets.

Description

Technical Field

[0001] This invention relates to the field of high-frequency electromagnetic calculation technology using physical optics (PO) methods, specifically to a reverse ray blanking method for surface preprocessing, adapted to plane wave incidence in physical optics methods. It is mainly applied to radar cross section (RCS) estimation, electromagnetic scattering characteristic analysis, and related engineering design scenarios for electrically large targets such as aircraft, satellites, and ships. Background Technology

[0002] The Proof-of-Sight (PO) method is one of the core methods for solving the electromagnetic scattering characteristics of electrically large targets in high-frequency electromagnetic computation. Its core principle is to calculate the induced current on the effectively illuminated surface of the target surface, and then solve for the scattered field and radar cross-section (RCS) through integration. The blanking determination, as a crucial preliminary step in the PO method, aims to accurately distinguish between the effectively illuminated surface (directly illuminated by incident electromagnetic waves without obstruction) and the shadowed area (facing away from the incident wave or obstructed by other surface elements). The accuracy of the determination directly determines the accuracy of the RCS calculation, while computational efficiency affects the method's engineering feasibility.

[0003] In plane wave incident scenarios, existing hidden surface removal detection techniques have the following shortcomings:

[0004] Traditional hidden surface removal methods rely on the dot product of the surface element's normal and the plane wave's incident direction, determining only whether the surface element faces the incident wave. This method is computationally simple, with a complexity of only [missing information - likely a time complexity]. However, it completely ignores scenarios where surface elements face the incident wave but are occluded by other surface elements, and has no ability to recognize multiple occlusions on complex curved surface targets, such as... Figure 1 The normal shown is The triangular facets identified by this method may be misclassified as valid illuminated surfaces, resulting in RCS calculation accuracy failing to meet engineering requirements.

[0005] To address the problem of multiple occlusions, a projection clipping method was introduced. This method projects 3D facets onto a 2D plane perpendicular to the incident direction and determines occlusion by analyzing projection overlap and sorting projection depths. However, this method is essentially a 2D simplification of 3D occlusion relationships, inherently suffering from insufficient clipping due to 2D approximation: if two facets occlude in 3D space but their projection overlap is not significant, they may be missed or misjudged as effective illuminated surfaces. Ultimately, this method still cannot accurately reflect the 3D occlusion relationships of complex targets, resulting in limited accuracy improvement.

[0006] To address the problem of misjudgment, existing technologies have proposed reverse ray tracing methods, such as... Figure 1 The normal shown is The emission direction vector of the triangular facet element Determine its normal direction as The triangular facets are occluded. Accurate judgment is achieved by detecting the intersection of pairwise inverse rays emitted from all facets, but this method has a computational complexity of O(n log n). (N is the total number of face cells). When the target is electrically large (with tens of thousands or even millions of face cells), the computational load increases exponentially. For example, a target with 100,000 face cells requires 10... 10 Intersection detection, without additional acceleration methods, takes too long on ordinary computers and cannot meet the real-time or rapid iterative design requirements of engineering.

[0007] In summary, existing blanking methods suffer from problems such as missed detection, misidentification, and low efficiency when identifying the effective illuminated surface in physical optics. Summary of the Invention

[0008] To address the aforementioned problems, this invention proposes a reverse ray blanking method for surface element preprocessing, which is adapted to plane wavefront incidence in physical optics. The core concept of this invention is to calculate the center coordinates and axial bounding box (AABB) of each triangular facet when reading the model triangular mesh information (vertex coordinates and normal vectors) and store them all in a data structure; to iterate through each triangular facet and obtain the dot product of the normal vector of each triangular facet and the incident direction of the plane wave, the triangular facets with a result less than 0 are marked as potential illumination surfaces and included in the potential illumination area set; then, a secondary illumination judgment is performed on each triangular facet marked as a potential illumination surface: a ray is constructed with the center point of the potential illumination surface as the starting point and the opposite direction of the incident wave as the emission direction, and the intersection calculation of the ray and the triangular facets is performed quickly with other potential illumination surfaces (using pre-processed AABBs). Once an intersection is found, the triangular facet that constructed the ray is excluded from the potential illumination area set, resulting in the final set of effective illumination triangular facets; finally, the induced current of each effective illumination triangular facet is integrated to solve for the scattering contribution, and the scattering contributions of all effective illumination triangular facets are accumulated to obtain the total scattering field of the target, and then the radar cross section (RCS) is solved, achieving a balance between accuracy and efficiency.

[0009] A method for reverse ray tracing in surface element preprocessing, the specific steps of which are as follows:

[0010] Step 1: Target triangular element mesh generation and STL format mesh file reading.

[0011] Step 1.1: Divide the target into triangular elements that meet the requirements of the PO method using CAD. Set the size to 1 / 8 to 1 / 10 of the incident wavelength (the size can be relaxed where the curvature is gentle) to meet the current uniformity assumption of the PO method, balance geometric accuracy and computational load, and provide a reliable geometric basis for subsequent processing.

[0012] Step 1.2: Directly read the STL format mesh file generated by the CAD software, extract the vertex coordinates of each triangular element and the triangular element normal vectors provided in the file (ASCII format is read using the "normal" keyword, binary format is parsed according to the standard structure), and construct the total triangular element set. Each triangular element F records the three-dimensional coordinates of its three vertices and its normal vector.

[0013] Step 2: Analyze the total set of triangular facets constructed in Step 1.2. For each triangular facet F in the array, perform the following preprocessing operations:

[0014] Step 2.1: Calculate the coordinates of the center point by taking the average of the coordinates of the three vertices of the triangular element F. :

[0015] ;

[0016] in , , These are the three vertices of a triangular facet.

[0017] Step 2.2: Construct AABB based on vertex coordinate extrema: 3D range has , , ,in:

[0018] ;

[0019] Step 3: Utilize the normal vector of the triangular element Unit vector of plane wave incident direction The dot product results are used for initial screening by multiplying the normal vectors.

[0020] Based on the propagation characteristics of plane waves, calculate The result, whether positive or negative, directly reflects the relative orientation of the triangular element's normal to the incident direction: When the angle between the normal of a triangular element and the incident direction is greater than or equal to 90°, the triangular element is completely opposite to the incident wave and cannot be directly illuminated by the plane wave. Therefore, it is determined to be an absolutely shadowed region and is included in the set of absolutely shadowed regions. This will prevent further processing. At that time, the normal of the triangular facet faces the incident direction, and there is a possibility that it will be illuminated, so it is included in the set of potential illumination areas. Then proceed to the next step of accurate judgment.

[0021] Step 4: Constructing the reverse ray and determining secondary occlusion.

[0022] Step 4.1: Based on the physical characteristics of plane waves, analyze the potential illumination region set obtained in Step 3. The triangular facet design uses reverse rays to achieve accurate occlusion detection: center point As the launch point, with Construct a reverse ray for the unit ray direction: ;in The unit direction vector representing the reverse ray. It is a scalar parameter of the ray, with the dimension of length, representing the distance from the initial point. along The distance of directional propagation.

[0023] Step 4.2, for Except Other triangular facets Perform secondary occlusion detection:

[0024] AABB Fast Repulsion: Identifying Reverse Rays and If there is no spatial intersection between AABBs, then they are directly excluded. No further testing is required.

[0025] Furthermore, the specific method for determining whether spatial intersection exists is to solve for the reverse ray. Minimum parameters for entering AABB and ;

[0026] First, solve for the reverse ray. Intersection parameters of the boundary plane aligned with each axis of AABB , , , , , :

[0027]

[0028] in, for of Axis coordinates for of Axis coordinates for of Axis coordinates for of Axial components, for of Axial components, for of Axial components.

[0029] In each coordinate axis direction The range is:

[0030]

[0031] Therefore, we get:

[0032]

[0033] If the above expression satisfies and Then determine the reverse ray It has overlap with AABB.

[0034] The Moller-Trumbore intersection method accurately detects intersections with AABBs by identifying reverse rays. The starting point, direction vector and Using the coordinates of the three vertices as input, solve for the correlation coefficient. , , .

[0035] If the correlation coefficient satisfies and and and Then determine Obstructed, terminate the operation. Subsequent testing;

[0036] Furthermore, the Moller-Trumbore intersection method correlation coefficient , , The solution formula is:

[0037] .

[0038] Step 5: Traverse the entire collection For all triangular elements, perform step 4 to complete the secondary occlusion judgment and obtain the final set of effective illuminated surfaces. ;

[0039] Step 6: Calculation of scattered field and RCS.

[0040] Based on the core principles of physical optics, the cumulative set The total scattered field of the target is obtained by taking the scattering contribution of all triangular elements, and then the radar cross section (RCS) can be solved.

[0041] Furthermore, specifically:

[0042] gather In the above, the formula for calculating the induced current density of a single triangular element is:

[0043] ,in The outward normal unit vector of the triangular facet element. The incident magnetic field.

[0044] The formula for calculating the scattered field using physical optics is:

[0045]

[0046] in, For the target surface point, For the target surface element, The position vector of the observation point. The straight-line distance from the observation point to the scattering source on the target surface is given. Let be the unit vector of the scattering direction. The incident wave vector, Let be the unit vector of the incident direction. For wave number, The wavelength of electromagnetic waves.

[0047] The formula for calculating RCS is:

[0048] .

[0049] in, For the incident electric field, This is the scattered electric field.

[0050] In summary, this invention rapidly filters each triangular element by applying its normal vector, and then performs a fast AABB intersection of the filtered triangular elements using reverse rays to obtain the final set of effectively illuminated triangular elements. Compared to traditional hidden surface removal methods, this invention has the following advantages:

[0051] (1) It solves the problem of misjudgment of the dot product of normal vectors in existing hidden vector removal methods;

[0052] (2) It solves the problem of two-dimensional approximation defects in the projection clipping method;

[0053] (3) The efficiency of the reverse ray tracing method is improved by preprocessing the surface elements;

[0054] (4) Furthermore, when solving for surface current, the reverse ray blanking method of the surface element preprocessing of the present invention is applied to the PO method. It is suitable for the analysis of electromagnetic scattering characteristics and radar scattering cross-section prediction of electrically large targets such as aircraft, satellites and ships under plane wave incident scenarios. It can be widely used in engineering design and performance evaluation in aerospace, defense and military industries. Attached Figure Description

[0055] Figure 1 Schematic diagram of illumination and occlusion for a triangular facet;

[0056] Figure 2 The flowchart for effectively illuminating triangular facets in this invention is shown below;

[0057] Figure 3 This is a schematic diagram of a simple composite cube model and its mesh numbering in Example 1;

[0058] Figure 4 This is a large ship model as described in Example 2;

[0059] Figure 5 This is a comparison chart of the reverse ray culling method for surface preprocessing in Example 2 and the CST simulation results; Detailed Implementation

[0060] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings.

[0061] Implementation environment:

[0062] Hardware environment: CPU is Intel Xeon(R) Silver 4210R;

[0063] Software environment: Operating system is Windows 10, compiler is Visual Studio 2015.

[0064] Example 1

[0065] See Figure 3 This embodiment uses a simple composite cube model as an example. The triangular facets of this model occlude with each other. The structural dimensions of a single cube model are as follows: The model material is an ideal electrical conductor (PEC), and the excitation is a plane wave incident with the following incident directions: pitch angle theta = 0° and azimuth angle phi = 0°. Under the above simulation parameters, the final number and index of effectively illuminated triangles are output, and compared with the number and index of effectively illuminated triangles output by the traditional dot product method, proving the accuracy of this method in judging multiple occlusions; the flowchart of the effective illuminated triangle judgment process is as follows. Figure 2 As shown. The specific implementation method is as follows:

[0066] Step 1: Target triangular element subdivision and STL file reading.

[0067] The target to be calculated was divided into a small number of triangular elements using CAD software. The 12 surfaces of the two combined cube models were divided into 56 triangular elements and stored in an STL mesh file for visual verification of its accuracy.

[0068] The algorithm directly reads the STL format mesh file generated by CAD software and extracts the vertex coordinates and normal vectors of the triangular facets. This yields the total set of triangular facets. The vertices and normals of the 56 triangular facets.

[0069] Step 2: Triangular facet preprocessing.

[0070] The coordinates of the center point are calculated by averaging the coordinates of the three vertices. .

[0071] AABB is determined based on the extreme values ​​of vertex coordinates.

[0072] Step 3: Initial screening of normal vectors.

[0073] Using the normal vector of the triangular element Unit vector of plane wave incident direction The dot product results were initially screened. When the normal of a triangular facet faces the incident direction, it has the potential to be illuminated and is therefore identified as a potential illumination area, constituting the set in this embodiment. There are 8 potential illuminated triangles with indices 0-3 and 24-27.

[0074] Step 4: Constructing the reverse ray and determining secondary occlusion.

[0075] Regarding the physical properties of plane waves, the set in step 3... The design of the eight triangular facets in the reverse ray is as follows , To achieve accurate occlusion detection.

[0076] Only for In addition to Other triangular facets Execution detection: Determine the reverse ray and If there is no spatial intersection between AABBs, then they are directly excluded. No further testing is required;

[0077] If intersections exist, the Moller-Trumbore intersection method is used for precise detection: the reverse rays are... The starting point, direction vector and Using the coordinates of the three vertices as input, solve for the correlation coefficient. , , Occlusion determination: If the correlation coefficient satisfies and and Then determine If obstructed, immediately terminate the operation. Subsequent testing.

[0078] Step 5: Traverse After the eight triangular facets in the dataset complete the secondary occlusion judgment in step 4, the set of this embodiment is obtained. There are 4 valid triangular facets, with triangular facet indices 24-27.

[0079] Table 1 compares the effective number of illuminated triangular facets and their illumination indices using the traditional dot product, reverse ray tracing, and facet preprocessing reverse ray tracing methods:

[0080] Table 1:

[0081] method Number of effective illuminated surfaces Illuminated face index (number) Traditional dot product 8 0-3，24-27 Reverse ray tracing method 4 24-27 Reverse Ray Hiding Method for Surface Element Preprocessing 4 24-27

[0082] The illumination results show that the preprocessing reverse ray tracing method further filters the effective illuminated triangular facets identified by the traditional dot product method. The results are consistent with the effective illuminated triangular facets obtained by the reverse ray tracing method, verifying the accuracy of the preprocessing reverse ray tracing method.

[0083] Example 2

[0084] See Figure 4 This embodiment is a large ship model, a typical electrically large structure. The triangular facets of the model also obstruct each other. The model's structural dimensions are: 14 meters wide, 115 meters long, and 22 meters high. The model material is an ideal electrical conductor (PEC), the excitation is a plane wave at a frequency of 5 GHz, the polarization is vertical, and the incident direction is: a fixed elevation angle (theta) of 90° and an azimuth angle (phi) ranging from 0° to 180° (with a step size of 1°).

[0085] Under the above simulation parameters, the single-station RCS scattering value was calculated and compared with commercial software. Simultaneously, the simulation time of the proposed reverse ray tracing method based on surface preprocessing was compared with that of the reverse ray tracing method after dot product filtering. The specific implementation method is as follows:

[0086] Repeating the five steps of Example 1 yields the effective illuminated triangular element set of Example 2. And further implement step 6:

[0087] Step 6: Calculation of scattered field and RCS.

[0088] Based on the core principles of physical optics, the set is first calculated. The induced current density of each triangular element in the set, and then the set The induced current density of each triangular facet element is physically optically integrated to solve for its scattering contribution, and finally the results are summed. The total scattered field of the target is obtained by calculating the scattering contribution of all triangular facet elements, and then the RCS is solved. To verify its accuracy, the same simulation parameters were set in the commercial software CST, and the simulation comparison results are as follows. Figure 5 As shown, from Figure 5 As can be seen, the results of the method proposed in this invention are in high agreement with the CST simulation results, with an average agreement of 99.7%.

[0089] To verify the computational efficiency of the proposed element preprocessing reverse ray tracing method, Table 2 shows the total RCS calculation time and speedup ratio of the reverse ray tracing method after dot product filtering and the element preprocessing reverse ray tracing method proposed in this invention under the same mesh parameters:

[0090] Table 2:

[0091] method Calculation time acceleration ratio Reverse ray tracing method (dot product filtering) 1478.61s / Reverse Ray Hiding Method for Surface Element Preprocessing 383.43s 3.86

[0092] The simulation calculation time and speedup ratio show that the reverse ray culling method based on plane wave incident surface preprocessing has high solution efficiency.

[0093] As can be seen from the above embodiments, this invention systematically solves the core contradictions of existing technologies, such as insufficient accuracy of normal vector dot product, two-dimensional approximation of projection clipping, and low efficiency of traditional reverse ray generation, through an integrated design of surface element preprocessing, normal vector initial screening, reverse ray generation, and rapid intersection calculation. It provides a brand-new technical solution for the electromagnetic scattering characteristics analysis of electrically large targets, and has important engineering application value and national defense significance. It can be applied to radar cross section (RCS) prediction, electromagnetic scattering characteristics analysis, and related engineering design scenarios of electrically large targets such as aircraft, satellites, and ships.

Claims

1. A method for reverse ray tracing in surface element preprocessing, characterized in that, Includes the following steps: Step 1: Target surface element generation and STL format mesh file reading; Step 1.1: Use CAD software to divide the target to be calculated into triangular facets that meet the requirements of the PO method. The facet size is between 1 / 8 and 1 / 10 of the incident wavelength. Step 1.2: Directly read the STL format mesh file generated by the CAD software, extract the vertex coordinates of each triangular element and the triangular element normal vectors provided in the file, and construct the total triangular element set. Each triangular element F records the three-dimensional coordinates of its three vertices and its normal vector; Step 2: Analyze the total set of triangular facets constructed in Step 1.

2. For each triangular facet F in the array, perform the following preprocessing operations: Step 2.1: Calculate the coordinates of the center point by taking the average of the coordinates of the three vertices of the triangular element F. : ； in , , These are the three vertices of a triangular element; Step 2.2: Construct AABB based on vertex coordinate extrema: 3D range has , , ,in: ； Step 3: Utilize the normal vectors of the triangular facets Unit vector of plane wave incident direction The dot product results are used for initial screening by the dot product of the normal vectors: Based on the propagation characteristics of plane waves, calculate The result, whether positive or negative, directly reflects the relative orientation of the triangular element's normal to the incident direction; Triangular facets are classified into the absolute shadow region set. , The triangular facets are included in the potential illumination area set. ; Step 4: Constructing the reverse ray and determining secondary occlusion: Step 4.1: Based on the physical characteristics of plane waves, analyze the potential illumination region set obtained in Step 3. In the design of triangular facets, reverse rays are used: center point As the launch point, with Construct a reverse ray for the unit ray direction , , ;in The unit direction vector representing the reverse ray. It is a scalar parameter of the ray, with the dimension of length, representing the distance from the initial point. along The distance of directional propagation; Step 4.2, for Except Other triangular facets Perform secondary occlusion detection: AABB Fast Repulsion: Identifying Reverse Rays and If there is no spatial intersection between AABBs, then they are directly excluded. No further testing is required; Moller-Trumbore intersection algorithm for accurate detection: This involves identifying reverse rays that intersect with AABB. The starting point, direction vector and Using the coordinates of the three vertices as input, solve for the correlation coefficient. , , ; Occlusion determination: If the correlation coefficient satisfies and and and Then determine Obstructed, terminate the operation. Subsequent testing; Step 5: Traverse the entire collection All triangular elements in the set undergo step 4 to complete the secondary occlusion judgment, resulting in the final set of effective illuminated surfaces. ; Step 6, Scattering Field and RCS Calculation: Based on the core principles of physical optics, the cumulative set is calculated. The total scattered field of the target is obtained by taking the scattering contribution of all triangular elements, and then the radar cross section (RCS) can be solved.

2. The reverse ray blanking method for surface element preprocessing as described in claim 1, characterized in that: In step 4.2, the reverse ray is determined. and The specific method to determine whether AABBs have spatial intersection is as follows: Solving for the reverse ray Minimum parameters to enter AABB and ; First, solve for the reverse ray. Intersection parameters of the boundary plane aligned with each axis of AABB , , , , , : ； in, for of Axis coordinates for of Axis coordinates for of Axis coordinates for of Axial components, for of Axial components, for of Axial components; In each coordinate axis direction The range is: ； Therefore, we get: ； If the above expression satisfies and Then determine the reverse ray. It has overlap with AABB.

3. The reverse ray blanking method for surface element preprocessing as described in claim 1, characterized in that, The correlation coefficient obtained in step 4.2 using the Moller-Trumbore intersection method , , The solution formula is: 。 4. The reverse ray blanking method for surface element preprocessing as described in claim 1, characterized in that, Step 6 specifically involves: gather In the above, the formula for calculating the induced current density of a single triangular element is: ,in The outward normal unit vector of the triangular facet element. For the incident magnetic field; The formula for calculating the scattered field using physical optics is: ； in, For the target surface point, For the target surface element, The position vector of the observation point. The straight-line distance from the observation point to the scattering source on the target surface is given. Let be the unit vector of the scattering direction. The incident wave vector, Let be the unit vector of the incident direction. For wave number, The wavelength of electromagnetic waves; The formula for calculating RCS is: ； in, For the incident electric field, This is the scattered electric field.

Images