Method for generating a surface geometry model of a formed layer for lpbf powder spreading simulation

Abstract

This invention discloses a method for generating a geometric model of a solidified layer surface for LPBF powder laying simulation, comprising: obtaining a configuration parameter set, wherein the configuration parameter set includes at least melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters; generating a multi-melt channel basic morphology height field by stacking the melt channel geometry parameters and the scanning spacing parameters; generating a random roughness height field with the target surface roughness parameters, and superimposing the random roughness height field with the multi-melt channel basic morphology height field to obtain a composite height field; performing boundary normalization and solidification processing based on the composite height field to obtain a geometric model of the solidified layer surface; extending the substrate model for LPBF powder laying simulation from an ideal flat plate to a solidified layer surface with multi-melt channel morphology and random roughness, making the substrate conditions for DEM powder laying simulation closer to the real multilayer manufacturing conditions, and providing a key basic input model for multilayer powder laying process optimization.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing numerical simulation technology, and in particular to a method for generating a geometric model of the surface of a pre-formed layer for LPBF powder spreading simulation. Background Technology

[0002] Laser Powder Bed Fusion (LPBF) is one of the most widely used processes in the field of metal additive manufacturing. Its basic principle is to selectively melt metal powder in a powder bed using a high-energy laser beam along a preset scanning path, layer by layer, to form a three-dimensional solid part. The powder spreading process is a crucial step, involving the uniform spreading of powder onto a forming platform using a scraper or roller. The uniformity and density of the powder layer directly affect the stability of the subsequent laser melting process, as well as the density, surface quality, and mechanical properties of the final formed part. Currently, scholars both domestically and internationally widely employ the Discrete Element Method (DEM) to conduct numerical simulations of the LPBF powder spreading process, aiming to reveal the flow, accumulation, and force chain evolution mechanisms of powder particles under the action of a scraper / roller.

[0003] Existing powder-laying simulation studies have significant shortcomings: current schemes typically assume the powder-laying substrate to be an ideal smooth flat plate. However, actual LPBF manufacturing is a layer-by-layer process. Except for the first layer, all subsequent layers of powder are spread on the surface of the solidified shaped layer above. This shaped layer surface is not a smooth plane but has complex geometric features directly related to its process history and melt pool behavior. For example, there are periodic ripple morphologies formed by multiple overlapping melt channels; random micro-roughness caused by melt pool solidification shrinkage and partial molten powder adhesion; and random protrusions caused by spatter deposition or local spheroidization defects. The ideal flat plate substrate completely ignores the influence of these features on powder-laying behavior.

[0004] Therefore, existing technologies use an idealized flat substrate model, ignoring the influence of the complex morphology of the actual formed layer surface on the subsequent powder spreading behavior, resulting in deviations between simulation results and actual working conditions. Summary of the Invention

[0005] This invention provides a method for generating a geometric model of the surface of a formed layer for LPBF powder spreading simulation, in order to solve the defect in the prior art that "the powder spreading substrate model in the existing LPBF powder spreading DEM simulation is too idealized".

[0006] This invention provides a method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation, comprising: Obtain a set of configuration parameters, which includes at least melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters; Based on the geometric parameters of the melt channel and the scanning interval parameters, a multi-melt channel basic topography height field is generated by stacking each channel sequentially. A random roughness height field with the target surface roughness parameters is generated, and the random roughness height field is superimposed with the multi-melt channel basic morphology height field to obtain a composite height field; Based on the composite height field, boundary normalization and solidification processes are performed to obtain the geometric model of the formed layer surface.

[0007] According to the present invention, a method for generating a surface geometric model of a formed layer for LPBF powder spreading simulation is provided. The step of generating a multi-melt-channel basic topography height field by stacking the melt channel geometric parameters and the scanning spacing parameters, specifically includes: Based on the geometric parameters of the melt channel and the scanning spacing parameters, and according to the preset mathematical model of the melt channel cross section, the basic topography height field of the multi-melt channel is generated by superimposing the melt channels one by one along the direction perpendicular to the scanning direction. Each melt channel generates a cross-sectional shape according to a preset mathematical model of the melt channel cross-section, and adds random perturbations to its height and width.

[0008] According to the present invention, a method for generating a geometric model of a formed layer surface for LPBF powder spreading simulation is provided. In the process of generating a multi-melting channel base morphology height field, for any two adjacent melting channels that overlap in the horizontal direction, the larger of the cross-sectional height values ​​of the two melting channels at the overlapping position is taken as its final height value in the multi-melting channel base morphology height field.

[0009] According to the present invention, a method for generating a geometric model of the surface of a formed layer for LPBF powder spreading simulation is provided, wherein the preset mathematical model of the melt channel section is a cosine section model or a Gaussian section model.

[0010] The present invention provides a method for generating a geometric model of a formed layer surface for LPBF powder spreading simulation. Based on Gaussian random field theory, the method includes the step of generating a random roughness height field with the target surface roughness parameters, specifically comprising: Generate a white noise field of the same size as the height field of the multi-melt channel base morphology; The standard deviation parameter of the Gaussian filter is determined based on the roughness autocorrelation length in the target surface roughness parameters, so as to perform Gaussian spatial filtering on the white noise field to obtain the original roughness field with spatial correlation. The original roughness field is subjected to zero-mean processing, and the current arithmetic mean roughness of the original roughness field after zero-mean processing is calculated. The random roughness height field is obtained by scaling the target arithmetic mean roughness to the current arithmetic mean roughness in the target surface roughness parameters.

[0011] The present invention provides a method for generating a geometric model of a formed layer surface for LPBF powder spreading simulation, which, based on the power spectral density method, includes the step of generating a random roughness height field with the target surface roughness parameters, specifically comprising: Construct a frequency grid to generate a power spectral density distribution that satisfies the frequency decay law in Fourier space; A random phase field is generated, and the power spectral density distribution is combined with the random phase field to form a complex spectrum; Perform an inverse Fourier transform on the complex spectrum to obtain the roughness height field in real space; The roughness height field is scaled to the target arithmetic mean roughness in the target surface roughness parameters to obtain the random roughness height field.

[0012] According to the present invention, a method for generating a geometric model of a formed layer surface for LPBF powder spreading simulation is provided. The step of performing boundary normalization and solidification processing based on the composite height field to obtain the geometric model of the formed layer surface specifically includes: The composite height field is subjected to minimum value zeroing processing so that the lowest point of the composite height field is located on the zero reference plane; A flat plate of equal thickness is added to the bottom of the composite height field after it has been reduced to zero to form a closed solid base. The composite height field after the additional solid base is triangularly meshed to generate a closed three-dimensional surface mesh model enclosed by multiple triangular facets, and output in a general three-dimensional geometry file format to obtain the surface geometry model of the formed layer.

[0013] According to the present invention, a method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation is provided, the method further comprising: Multiple preset scanning spacing parameters are available; Iterate through each of the scanning spacing parameters and generate the surface geometry model of the formed layer corresponding to each scanning spacing parameter, so as to output the three-dimensional visualization image, cross-sectional profile and roughness parameter statistics table of all the surface geometry models of the formed layer.

[0014] This invention also provides a system for generating geometric models of pre-formed layer surfaces for LPBF powder spreading simulation, comprising: The parameter configuration module is used to obtain a set of configuration parameters, which includes at least melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters. The initial morphology generation module is used to generate a multi-melt channel basic morphology height field by stacking each channel based on the melt channel geometric parameters and the scanning interval parameters. The surface generation core module is used to generate a random roughness height field with the target surface roughness parameters, and to superimpose the random roughness height field with the multi-melt channel basic morphology height field to obtain a composite height field. The solidification output module is used to perform boundary normalization and solidification processing based on the composite height field to obtain the geometric model of the formed layer surface.

[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for generating a surface geometry model of a formed layer for LPBF powder spreading simulation as described above.

[0016] This invention provides a method for generating a geometric model of the formed layer surface for LPBF powder spreading simulation. Based on melt channel geometry parameters and scanning spacing parameters, a multi-melt channel basic morphology height field is generated by superimposing each melt channel to form a macroscopic melt channel overlap wave morphology. A random roughness height field with target surface roughness parameters is generated and superimposed with the multi-melt channel basic morphology height field to obtain a composite height field, thus superimposing the macroscopic melt channel overlap morphology with the microscopic random roughness. Finally, boundary regularization and solidification processing are performed based on the composite height field to obtain the geometric model of the formed layer surface, forming a closed and usable three-dimensional solid model for DEM simulation import. Therefore, the substrate model for LPBF powder spreading simulation is extended from an ideal flat plate to a solidified layer surface with multi-melt channel morphology and random roughness, making the substrate conditions for DEM powder spreading simulation closer to real multilayer manufacturing conditions, providing a key basic input model for multilayer powder spreading process optimization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts illustrating the method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation provided by the present invention. Figure 2 This is the second flowchart of the method for generating the surface geometric model of the formed layer for LPBF powder spreading simulation provided by the present invention. Figure 3This is a schematic diagram illustrating the superposition principle of the roughness field and the base height field in the method for generating the geometric model of the formed layer surface for LPBF powder spreading simulation provided by this invention. Figure 4 These are comparison images of the cross-sectional contours of the surface of the formed layer under different parameter combinations of the method for generating geometric models of the surface of the formed layer for LPBF powder spreading simulation provided by this invention. Figure 5 These are three-dimensional morphology images of the surface of a formed layer under different scanning intervals, based on the method for generating geometric models of the surface of a formed layer for LPBF powder spreading simulation provided by this invention. Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0019] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0020] In the description of this invention, it should be noted that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0021] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] See Figure 1 This application provides a method for generating a geometric model of the surface of a pre-formed layer for LPBF powder spreading simulation, including the following steps: S110: Obtain the configuration parameter set, which includes at least the melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters; S120: Based on the geometric parameters of the melt channel and the scanning spacing parameters, the basic morphology height field of the multi-melt channel is generated by stacking the channels one by one; S130: Generate a random roughness height field with target surface roughness parameters, and superimpose the random roughness height field with the multi-melt channel basic morphology height field to obtain a composite height field; S140: Based on the composite height field, boundary regularization and solidification are performed to obtain the geometric model of the formed layer surface.

[0023] This invention provides a method for generating a geometric model of the formed layer surface for LPBF powder spreading simulation. Based on melt channel geometry parameters and scanning spacing parameters, a multi-melt channel basic morphology height field is generated by superimposing each melt channel to form a macroscopic melt channel overlap wave morphology. A random roughness height field with target surface roughness parameters is generated and superimposed with the multi-melt channel basic morphology height field to obtain a composite height field, thus superimposing the macroscopic melt channel overlap morphology with the microscopic random roughness. Finally, boundary regularization and solidification processing are performed based on the composite height field to obtain the geometric model of the formed layer surface, forming a closed and usable three-dimensional solid model for DEM simulation import. Therefore, the substrate model for LPBF powder spreading simulation is extended from an ideal flat plate to a solidified layer surface with multi-melt channel morphology and random roughness, making the substrate conditions for DEM powder spreading simulation closer to real multilayer manufacturing conditions, providing a key basic input model for multilayer powder spreading process optimization.

[0024] Figure 2 This is a comprehensive technical flowchart illustrating the method for generating the surface geometry model of the formed layer for LPBF powder spreading simulation, showcasing the complete technical path of this application. The following sections provide a detailed explanation of the technical solution of this application, focusing on the specific steps at each stage.

[0025] In S110, user-defined modeling parameters are received to configure surface generation parameters that define the macroscopic morphology and microscopic texture features of the surface geometry of the formed layer.

[0026] In this embodiment of the application, the configuration parameter set includes: Geometric parameters of the runner: runner height track_height, runner cross-section model type track_model (cosine or gaussian), runner width track_width; Scan spacing parameters: Multiple scan spacing values ​​can be set to form a scan spacing variable group; Surface roughness parameters: target arithmetic mean roughness_Ra, roughness autocorrelation length, and random seed roughness_seed; More specifically, the configuration parameter set may also include: Domain size parameters: X-direction length domain_x (along the melt path), Y-direction length domain_y (perpendicular to the melt path), grid resolution grid_resolution; Output control parameters: output directory (output_dir), base plate additional switch (add_base_plate), base plate thickness (base_thickness).

[0027] In S120, the macroscopic periodic ripples on the surface of the formed layer, determined by the laser scanning path, are simulated. Traditional simulations often use ideal flat plates, completely ignoring the "peaks" and "canyons" formed by multiple overlapping weld channels. This step reproduces this macroscopic feature by generating and superimposing the morphology of individual weld channels one by one. By introducing weld channel geometry parameters and scanning spacing parameters, the generated macroscopic morphology is directly correlated with the specific LPBF process parameters.

[0028] Specifically, the step of generating a multi-melt-channel basic topography height field by stacking each channel based on the melt channel geometric parameters and scanning spacing parameters includes: according to the melt channel geometric parameters and scanning spacing parameters, and according to the preset melt channel cross-section mathematical model, generating a multi-melt-channel basic topography height field by stacking each channel along the direction perpendicular to the scanning direction; wherein, each melt channel generates a cross-sectional topography according to the preset melt channel cross-section mathematical model, and adds random perturbations to the height and width.

[0029] This application describes the cross-sectional shape of each melt channel using mathematical curves such as cosine or Gaussian cross-section models, ensuring that the center of the melt channel is the highest, the sides transition smoothly, and the height returns to zero or approaches zero at the edges. Random perturbations in height and width are then added to simulate the inhomogeneity of the melt channel caused by factors such as laser power fluctuations, scanning speed deviations, and uneven powder distribution in actual LPBF processes.

[0030] The cosine section model expression is as follows: ; In the formula, The local x-coordinate is defined with the center of the melt channel as the zero point. Let y be the height of the melt channel at a horizontal distance y from the center of the melt channel. W is the height of the melt channel and W is the width of the melt channel.

[0031] The Gaussian section model expression is: ; In the formula, σ is the standard deviation of the Gaussian distribution, which controls the full width at half maximum (FWHM) of the melt channel.

[0032] In this process, random perturbations are added to the height and width of each melt channel to simulate the non-uniformity of the melt channel in actual processes. The random perturbations for the height and width of each melt channel are determined according to the following formula: ; ; In the formula, Here, represents the fluctuation amplitude coefficient (ranging from 0.05 to 0.20), and N(0,1) represents a standard normally distributed random number. The height of the molten channel with disturbance. The width of the runner with disturbance.

[0033] In the embodiments of this application, during the process of generating the multi-melting channel base morphology height field, for any two adjacent melting channels that overlap in the horizontal direction, the larger of the two melting channel cross-sectional height values ​​at the overlapping position is taken as its final height value in the multi-melting channel base morphology height field.

[0034] This step simulates the remelting and superposition effect when adjacent melt channels overlap in the LPBF process. In physical reality, the later melt channel partially remelts the edge of the previous melt channel, resulting in material accumulation rather than simple linear superposition or averaging. By adopting a strategy of taking the larger value, this material accumulation effect can be effectively simulated, avoiding physically unrealistic depressions in the overlapping area caused by improper superposition algorithms.

[0035] See S130. Figure 3 A random roughness field is generated and superimposed with the basic height field obtained in the above steps to obtain a composite height field.

[0036] This step is used to simulate the microscopic and random surface undulations on the surface of a real formed layer caused by complex physical processes such as molten pool solidification and powder adhesion. By generating an independent random roughness field and superimposing it on the macroscopic morphology, the shortcomings of existing models that cannot simultaneously take into account periodic ripples and random roughness are solved, thus improving the fidelity of the model.

[0037] In this embodiment of the application, the step of generating a random roughness height field with target surface roughness parameters based on Gaussian random field theory specifically includes: generating a white noise field of the same size as the height field of the multi-channel base morphology; determining the standard deviation parameter of the Gaussian filter according to the roughness autocorrelation length in the target surface roughness parameters, so as to perform Gaussian spatial filtering on the white noise field to obtain a roughness original field with spatial correlation; performing zero-mean processing on the roughness original field, calculating the current arithmetic mean roughness of the zero-mean roughness original field; and scaling it according to the ratio of the target arithmetic mean roughness in the target surface roughness parameters to the current arithmetic mean roughness to obtain the random roughness height field.

[0038] Specifically, firstly, a white noise field following a standard normal distribution is generated, with the same size as the height field mesh of the multi-channel base topography. Then, the standard deviation parameter of the Gaussian filter is calculated. ( =roughness_correlation_length / grid_resolution), and the standard deviation of the white noise field is calculated as follows: Gaussian spatial filtering is applied to remove high-frequency components to eliminate spatial correlation. The mean of the original roughness field is removed to make it a zero-mean field, thus eliminating the overall offset. The current arithmetic mean roughness of the zero-mean roughness field is measured, and the roughness field is scaled to the target Ra value by a scaling factor.

[0039] In another embodiment, the step of generating a random roughness height field with target surface roughness parameters based on the power spectral density method specifically includes: constructing a frequency grid to generate a power spectral density distribution that satisfies the frequency decay law in Fourier space; generating a random phase field and combining the power spectral density distribution and the random phase field into a complex spectrum; performing an inverse Fourier transform on the complex spectrum to obtain a roughness height field in real space; and scaling the roughness height field to the target arithmetic mean roughness in the target surface roughness parameters to obtain a random roughness height field.

[0040] Specifically, a frequency grid is first constructed to generate a power spectral density distribution in Fourier space that satisfies the frequency decay law PSD∝f^{-(2H+2)}, where PSD is the power spectral density, f is the spatial frequency, and H is the Hurst exponent, which ranges from 0.5 to 1.0. It is used to control the fractal characteristics of the surface. The larger the value of H, the wider and smoother the surface undulations; the smaller the value of H, the finer and more chaotic the surface texture.

[0041] Simultaneously, a random phase field uniformly distributed in the interval [0, 2π) is generated. The power spectral density amplitude is combined point-by-point with the random phase field to form a complex spectrum, and then the frequency domain information is converted to real space through inverse Fourier transform to obtain the roughness height field. Finally, the roughness height field is scaled as a whole according to the ratio of the target arithmetic mean roughness to the current Ra value, so that the arithmetic mean roughness of the roughness field is equal to the preset target Ra value, thus obtaining the final random roughness height field.

[0042] Furthermore, the random roughness height field is superimposed with the multi-melting channel basic morphology height field to obtain a composite height field, namely the surface height layer of the formed layer, so as to jointly constitute a complete characterization of the multi-scale morphology features of the solidified surface of the real LPBF.

[0043] Therefore, this application provides a two-layer superposition modeling framework for multi-melt channel morphology and random roughness. By hierarchically modeling and superimposing the macroscopic melt channel overlap morphology (determined by the scanning strategy) and the microscopic random roughness (determined by the solidification process), a complete characterization of the multi-scale features of the solidified layer surface is achieved.

[0044] In S140, the steps of performing boundary normalization and solidification based on the composite height field to obtain the surface geometry model of the formed layer specifically include: performing minimum value zeroing processing on the composite height field so that the lowest point of the composite height field is located on the zero reference plane; adding a plate of equal thickness to the bottom of the composite height field after minimum value zeroing processing to form a closed solid base; performing triangular meshing on the composite height field after adding the solid base to generate a closed three-dimensional surface mesh model enclosed by multiple triangular facets, and outputting it in a general three-dimensional geometry file format to obtain the surface geometry model of the formed layer.

[0045] Specifically, the composite height field is translated as a whole, reducing its minimum height value to zero and eliminating negative value regions to ensure that all height values ​​are above the reference plane. Simultaneously, according to preset base plate addition options, a flat plate of uniform thickness is added to the bottom of the zeroed height field, forming a solid base with a closed bottom surface. Through these processes, the final generated 3D surface geometry model is ensured to be geometrically closed and physically plausible.

[0046] Specifically, the composite height field after the above processing is triangulated to generate a closed STL triangular mesh model containing the upper surface triangular face, the bottom triangular face, and the four side triangular faces, and output as a general STL format file.

[0047] For example, generating an STL format file includes: cell conversion, converting the units of the processed composite height field data from micrometers (μm) to millimeters (mm); traversing the height field mesh, dividing each rectangular mesh cell into two triangular patches to generate a set of upper surface triangular faces; generating a set of lower surface triangular faces corresponding to the upper surface in the Z=0 plane, ensuring that the vertex order and normal vector direction are consistent with the outward normal of the closed surface; traversing the mesh boundary to generate the front side (j=0) and back side (j=n) respectively. y -1), left side (i=0) and right side (i=n) x -1) is a set of triangular faces, where i is the X-direction index of the height field grid, j is the Y-direction index of the height field grid, and n x This represents the total number of grid points in the X direction and the total number of grid points in the Y direction; it assembles all triangular faces into a complete triangular mesh object and writes it to an STL file in binary or ASCII format.

[0048] Therefore, the STL format file generated by this application can be directly imported into mainstream DEM simulation platforms such as LIGGGHTS, EDEM, and YADE, which greatly reduces the preprocessing threshold for multi-layer powder spreading simulation research.

[0049] In this embodiment of the application, the method further includes: presetting multiple different scanning spacing parameters; traversing each scanning spacing parameter and generating a geometric model of the surface of the formed layer corresponding to each scanning spacing parameter, so as to output a three-dimensional visualization image, cross-sectional contour diagram and roughness parameter statistical table of all the geometric models of the surface of the formed layer.

[0050] In one example, based on the batch generation of the single variable scan spacing, a target arithmetic mean roughness variable was further added. Specifically, a list of scan spacings and a list of target arithmetic mean roughness were preset, and by traversing the Cartesian product of the two, all parameter combinations were exhausted, automatically generating a library of STL models of the formed layer surface corresponding to all combinations, providing a more comprehensive geometric input dataset for systematic parameter studies.

[0051] Therefore, this application supports batch parameter matrix generation, which is suitable for systematic parameter sensitivity studies. By traversing the Cartesian product of multiple parameter dimensions such as scanning spacing and roughness, a complete parameter matrix STL model library can be automatically generated, providing batch data support for establishing the full-chain mapping relationship of "process parameters → surface morphology → powder spreading quality".

[0052] In summary, this application achieves the automatic generation of a powder-laying substrate model that approximates the real manufacturing conditions of multilayer LPBFs. It extends the powder-laying substrate from the traditional ideal flat plate to a solidified layer surface model with multiple melt channel overlap morphology and random roughness, providing more realistic basic input conditions for DEM simulation of multilayer LPBF powder-laying process, and helping to reveal the interlayer evolution law of powder-laying quality in actual process.

[0053] This application achieves parameterized, repeatable, and adjustable modeling of surface morphology features. Through a random seed fixing mechanism and parameterized interface, it ensures that the results generated by the same set of parameters are completely reproducible. At the same time, researchers and engineers can systematically explore the powder spreading behavior under different process conditions by adjusting parameters such as scanning spacing and roughness Ra value, providing theoretical support for process parameter optimization.

[0054] In a specific example, this embodiment demonstrates the batch generation process of an STL model of a pre-formed layer surface under the condition that the scanning spacing is the single scanning variable and the roughness Ra=10μm is fixed.

[0055] Configuration parameters: Set the domain size to 3000μm × 2000μm (X direction is the melt path direction, Y direction is perpendicular to the melt path direction), and the mesh resolution to 5μm / point. Set the melt path height to 15μm, and use a cosine section model. Set the target arithmetic mean roughness Ra = 10μm, the roughness autocorrelation length to 80μm, and the random seed to 42 to ensure repeatable results. Set the scan spacing variable group to [60, 80, 100, 120]μm. Enable the base plate attachment function, and set the base plate thickness to 20μm.

[0056] Mathematical model selection for the weld channel cross-section: A cosine cross-section model is selected. This model reaches its maximum height H at the center of the weld channel and smoothly transitions to zero height at the edge, exhibiting good physical rationality. For each weld channel, a random perturbation of ±15% (fluctuation coefficient α=0.15) is added to the height, and a random perturbation of ±7.5% is added to the width to simulate the non-uniformity of the weld channel caused by factors such as laser power fluctuations and scanning speed deviations in actual processes.

[0057] Roughness height field generation: Random roughness is generated using Gaussian filtering. First, a white noise field (standard normal distribution) with the same size as the height field grid is generated. Then, Gaussian spatial filtering is performed with a standard deviation σ = 80μm / 5μm = 16 pixels to remove high-frequency white noise components and introduce spatial correlation. The filtered field is then zero-mean processed, the current arithmetic mean roughness Ra is calculated, and then scaled to the target Ra = 10μm by a coefficient (10 / current Ra).

[0058] Superposition and boundary treatment: The height field of the multi-track base topography and the random roughness height field are superimposed point by point: Z_total = Z_tracks + Z_roughness. The superimposed height field is then minimized to zero, and a 20μm thick base plate is added at the bottom to make the entire model a closed solid.

[0059] STL format conversion: Triangulates the stacked height field. Unit conversion: Multiplies the micrometer values ​​by 10. - ³ is converted to millimeters. The height field mesh (approximately 601 × 401 mesh points) is traversed, and each rectangular mesh cell is divided into two triangular faces, generating approximately 480,000 triangular faces on the upper surface. An equal number of bottom triangular faces are generated in the Z=0 plane. Side triangular faces are generated at the four boundaries. The final STL file contains approximately 960,000 triangular faces and can be directly imported into DEM software such as EDEM and LIGGGHTS.

[0060] Visualization and Statistics: Views are generated for each scan interval, including a 3D surface topography map, a top-view heatmap, and a cross-sectional profile map. The center position of each melt channel is marked with a red dashed line in the cross-sectional profile map. The statistics table lists parameters such as scan interval, melt channel width, overlap rate, Ra, Rq (root mean square deviation of the profile), Rz (maximum profile height), and Rsk (skewness).

[0061] In another specific example, a matrix-based batch generation of two parameters, scan spacing and Ra roughness, is demonstrated.

[0062] Parameter matrix definitions: Scan spacing list: [60, 80, 100, 120] μm (4 levels); Ra roughness list: [5, 10, 15, 20] μm (4 levels); A total of 16 STL models are generated. The random seed is fixed at 42 to ensure comparability between different parameter combinations.

[0063] Iterative generation: Iterates through all 16 combinations of scan spacing and Ra value. For each parameter combination, an independent parameter configuration object is instantiated to generate the corresponding height field, and then an STL file is output through the STL conversion module. Each STL file is saved with the naming format "asbuilt_h{scan spacing}μm_Ra{Ra value}μm.stl".

[0064] Output Results: A model library consisting of 16 STL files will be generated in the output directory. This model library can be used for systematic parameter sensitivity studies in subsequent DEM powder spreading simulations, such as analyzing "how the scanning spacing affects the surface morphology of the solidified layer, and thus affects the powder filling density and uniformity of the next powder layer".

[0065] Figure 4 This displays a comparison of the cross-sectional profiles of the formed layer under different parameter combinations. Overlap represents the overlap rate, h s This refers to the scanning spacing; Figure 5 The three-dimensional topography of the formed layer surface is shown under different scanning intervals (60μm, 80μm, 100μm, 120μm).

[0066] The following describes the system for generating geometric models of the surface of a formed layer for LPBF powder spreading simulation provided by the present invention. The system for generating geometric models of the surface of a formed layer for LPBF powder spreading simulation described below can be referred to in correspondence with the method for generating geometric models of the surface of a formed layer for LPBF powder spreading simulation described above.

[0067] This application provides a system for generating geometric models of pre-formed layer surfaces for LPBF powder spreading simulation, including: The parameter configuration module is used to obtain a set of configuration parameters, which includes at least the melt channel geometry parameters, the scanning spacing parameters, and the target surface roughness parameters. The initial morphology generation module is used to generate a multi-channel basic morphology height field by stacking channels one by one based on the channel geometry parameters and scanning spacing parameters. The surface generation core module is used to generate a random roughness height field with target surface roughness parameters, and superimpose the random roughness height field with the multi-melt channel basic morphology height field to obtain a composite height field; The solidification output module is used to perform boundary normalization and solidification processing based on the composite height field to obtain the geometric model of the surface of the formed layer.

[0068] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include a processor 610, a communication interface 620, a memory 630, and a communication bus 640. The processor 610, communication interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call logic instructions from the memory 630 to execute a method for generating the surface geometry model of the formed layer for LPBF powder spreading simulation.

[0069] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation, characterized in that, include: Obtain a set of configuration parameters, which includes at least melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters; Based on the geometric parameters of the melt channel and the scanning interval parameters, a multi-melt channel basic topography height field is generated by stacking each channel sequentially. A random roughness height field with the target surface roughness parameters is generated, and the random roughness height field is superimposed with the multi-melt channel basic morphology height field to obtain a composite height field; Based on the composite height field, boundary normalization and solidification processes are performed to obtain the geometric model of the formed layer surface.

2. The method for generating the surface geometry model of a pre-formed layer for LPBF powder spreading simulation according to claim 1, characterized in that, The step of generating a multi-channel basic topography height field by stacking channels one by one based on the melt channel geometric parameters and the scanning spacing parameters specifically includes: Based on the geometric parameters of the melt channel and the scanning spacing parameters, and according to the preset mathematical model of the melt channel cross section, the basic topography height field of the multi-melt channel is generated by superimposing the melt channels one by one along the direction perpendicular to the scanning direction. Each melt channel generates a cross-sectional shape according to a preset mathematical model of the melt channel cross-section, and adds random perturbations to its height and width.

3. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 2, characterized in that, In the process of generating the height field of the multi-melting channel base morphology, for any two adjacent melting channels that overlap in the horizontal direction, the larger of the two melting channel cross-sectional height values ​​at the overlapping position is taken as its final height value in the height field of the multi-melting channel base morphology.

4. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 2, characterized in that, The preset mathematical model for the melt channel cross section is a cosine cross section model or a Gaussian cross section model.

5. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 1, characterized in that, The steps for generating a random roughness height field with the target surface roughness parameters based on Gaussian random field theory specifically include: Generate a white noise field of the same size as the height field of the multi-melt channel base morphology; The standard deviation parameter of the Gaussian filter is determined based on the roughness autocorrelation length in the target surface roughness parameters, so as to perform Gaussian spatial filtering on the white noise field to obtain the original roughness field with spatial correlation. The original roughness field is subjected to zero-mean processing, and the current arithmetic mean roughness of the original roughness field after zero-mean processing is calculated. The random roughness height field is obtained by scaling the target arithmetic mean roughness to the current arithmetic mean roughness in the target surface roughness parameters.

6. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 1, characterized in that, The step of generating a random roughness height field with the target surface roughness parameters based on the power spectral density method specifically includes: Construct a frequency grid to generate a power spectral density distribution that satisfies the frequency decay law in Fourier space; A random phase field is generated, and the power spectral density distribution is combined with the random phase field to form a complex spectrum; Perform an inverse Fourier transform on the complex spectrum to obtain the roughness height field in real space; The roughness height field is scaled to the target arithmetic mean roughness in the target surface roughness parameters to obtain the random roughness height field.

7. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 1, characterized in that, The step of performing boundary normalization and solidification based on the composite height field to obtain the geometric model of the formed layer surface specifically includes: The composite height field is subjected to minimum value zeroing processing so that the lowest point of the composite height field is located on the zero reference plane; A flat plate of equal thickness is added to the bottom of the composite height field after it has been reduced to zero to form a closed solid base. The composite height field after the additional solid base is triangularly meshed to generate a closed three-dimensional surface mesh model enclosed by multiple triangular facets, and output in a general three-dimensional geometry file format to obtain the surface geometry model of the formed layer.

8. The method for generating a geometric model of a pre-formed layer surface for LPBF powder spreading simulation according to claim 1, characterized in that, The method further includes: Multiple preset scanning spacing parameters are available; Iterate through each of the scanning spacing parameters and generate the surface geometry model of the formed layer corresponding to each scanning spacing parameter, so as to output the three-dimensional visualization image, cross-sectional profile and roughness parameter statistics table of all the surface geometry models of the formed layer.

9. A system for generating geometric models of pre-formed layer surfaces for LPBF powder spreading simulation, characterized in that, include: The parameter configuration module is used to obtain a set of configuration parameters, which includes at least melt channel geometry parameters, scanning spacing parameters, and target surface roughness parameters. The initial morphology generation module is used to generate a multi-melt channel basic morphology height field by stacking each channel based on the melt channel geometric parameters and the scanning interval parameters. The surface generation core module is used to generate a random roughness height field with the target surface roughness parameters, and to superimpose the random roughness height field with the multi-melt channel basic morphology height field to obtain a composite height field. The solidification output module is used to perform boundary normalization and solidification processing based on the composite height field to obtain the geometric model of the formed layer surface.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for generating the surface geometry model of the formed layer for LPBF powder spreading simulation as described in any one of claims 1 to 8.

Images