A laser powder bed fusion variable layer thickness forming method, system, apparatus, and medium

Abstract

The application belongs to the field of three-dimensional models, and discloses a laser powder bed fusion variable layer thickness forming method, system, device and medium, which comprises the following processes: extracting a cross section profile of a three-dimensional model of a part on a specified plane to generate a cross section profile image; extracting a profile in the cross section profile image, simplifying the profile to obtain a simplified polygon profile; traversing vertices of the simplified polygon profile, calculating an angle value at each vertex, and screening out vertices satisfying a preset angle threshold as corner points; dividing the simplified polygon profile into different types of geometric feature regions; according to a preset geometric feature and forming layer thickness mapping rule, assigning different layer thicknesses to the identified different geometric feature regions to generate a variable layer thickness slicing scheme; and generating a processing instruction according to the variable layer thickness slicing scheme to drive a laser powder bed fusion device to form the part layer by layer.

Description

Technical Field

[0001] This invention belongs to the field of three-dimensional modeling and relates to a method, system, equipment and medium for laser powder bed melting variable layer thickness forming. Background Technology

[0002] Laser powder bed melting, as one of the core technologies of metal additive manufacturing, plays an irreplaceable role in the manufacturing of complex and precision components in aerospace, biomedical and other fields due to its extremely high degree of design freedom and forming accuracy. In particular, for high-performance materials such as nickel-based superalloys, laser powder bed melting technology can achieve the integrated forming of complex internal flow channels, lightweight lattice structures and other structures that are difficult to complete by traditional processing methods.

[0003] However, the inherent rapid melting and high temperature gradient characteristics of laser powder bed melting processes easily introduce significant residual stress into components, leading to deformation, cracking, and even forming failure. This is particularly prominent for high-temperature alloys with high heat sensitivity. Current industrial practice commonly employs a constant, conservative small layer thickness strategy (e.g., 20-40 μm) to ensure forming quality, but this severely restricts manufacturing efficiency, especially for parts with many thick areas, resulting in high time costs. Furthermore, different geometric features of parts (e.g., thin walls, overhangs, thick blocks) respond differently to heat input and cooling rates. A single fixed layer thickness and process parameters cannot achieve global optimization; often, heat accumulation in thin-walled areas leads to deformation, while in thick areas, inefficiency wastes resources. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a laser powder bed melting variable layer thickness forming method, system, equipment and medium that can identify the geometric features of the part and adaptively match the layer thickness with the process parameters accordingly, thereby improving the quality, accuracy and efficiency of laser powder bed melting forming components.

[0005] To achieve the above objectives, the present invention employs the following technical solution: A method for forming variable layer thickness using laser powder bed melting includes the following steps: Extract the cross-sectional profile of the 3D model of the part on a specified plane and generate a cross-sectional profile image; The contour is extracted from the cross-sectional contour image, and an adaptive polygon approximation algorithm is used to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour. The simplified polygon contour is traversed by an adaptive step size, the angle value at each vertex is calculated, the vertices that meet the preset angle threshold are selected as corner points, and the angle values ​​of the corner points are recorded. Statistical analysis was performed on the side lengths, angle values, and corner density between adjacent corner points. Based on the statistical analysis results, the simplified polygon outline was divided into different types of geometric feature regions. Based on the preset mapping rules between geometric features and forming layer thickness, differentiated layer thicknesses are assigned to the identified regions with different geometric features to generate a variable layer thickness slicing scheme. Based on the variable layer thickness slicing scheme, processing instructions are generated to drive the laser powder bed melting equipment to form parts layer by layer.

[0006] Optionally, the process of extracting the cross-sectional profile of the 3D model of the part on a specified plane and generating a cross-sectional profile image includes: Obtain the vertex coordinates of all triangle faces in the 3D model, and calculate the center coordinates of all vertices; Define a plane that passes through the center coordinates and is parallel to the specified coordinate plane as the cutting plane; Traverse the three sides of each triangular facet and determine whether the two endpoints of each side are located on opposite sides of the cutting plane; Given that the two endpoints of an edge are located on opposite sides of a cutting plane, calculate the coordinates of the intersection point between the edge and the cutting plane using linear interpolation; Collect the coordinates of all intersection points to form a cross-sectional profile point set, convert the cross-sectional profile point set into a binary image, and obtain the cross-sectional profile image.

[0007] Optionally, the process of extracting the contour from the cross-sectional contour image and using an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy based on the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygonal contour, includes: Extract the contour within the cross-sectional contour image and calculate the arc length of each inflection point in the contour; Based on the numerical range of the arc length, a corresponding approximate accuracy scaling factor is assigned to the contour, and the product of the arc length and the approximate accuracy scaling factor is used as the contour threshold. Using a contour threshold as a condition, polygon approximation is performed on the contour to obtain multiple simplified polygon vertex sets; A sequence of closed line segments is constructed based on the simplified polygon vertex set to restore the polygon's geometric structure and generate a simplified polygon outline.

[0008] Optionally, after generating the simplified polygonal outline, the process also includes filtering the simplified polygonal outline: Identify the hierarchical relationships of simplified polygonal outlines, exclude internal outlines that belong to internal holes, and retain external outlines; Calculate the area of ​​the outer contour and filter out outer contours with an area smaller than the area threshold; The filtered outer contours are arranged in descending order of area, and the outer contours with an area greater than a preset ratio of the maximum contour area are retained as the final simplified polygonal contours.

[0009] Optionally, the process of traversing the vertices of the simplified polygonal contour using an adaptive step size, calculating the angle value at each vertex, and selecting vertices that meet a preset angle threshold as corner points includes: Obtain the number of vertices of the simplified polygonal outline, and determine the adaptive step size based on the number of vertices; Centered on the current vertex, select the first and second adjacent vertices forward and backward respectively in the polygon vertex sequence according to the adaptive step size; Based on the coordinates of the current vertex, the first adjacent vertex, and the second adjacent vertex, calculate the vectors from the current vertex to the first adjacent vertex and from the current vertex to the second adjacent vertex, respectively. The cosine value is calculated using the dot product and magnitude of two vectors, and the angle value of the current vertex is obtained through inverse cosine operation; If the angle value falls within the preset angle threshold range, the current vertex will be used as the corner point.

[0010] Optionally, if corner points are not detected using the adaptive step size method, the Harris corner detection method is used for supplementary detection, including: Create a temporary image, scale and center the outline, and perform Harris corner detection; The detection results are expanded and connected region analyzed to obtain the coordinates of the corner center. Transform the center coordinates of the corner points back to the original coordinate system and uniformly mark them as 90° angles.

[0011] Optionally, after marking the point as a corner and recording the angle value, it also includes: Assign a color to the simplified polygon outline, and draw the simplified polygon outline on the image with the corresponding color; Draw marker graphics at the corner points, using different colors of marker graphics according to the range of difference between the angle value and the reference angle; Label the angle values ​​near the corner points and draw the lines connecting the corner points to generate a corner connection diagram with angle labels and connection relationships.

[0012] A laser powder bed melting variable layer thickness forming system includes: The image generation module is used to extract the cross-sectional contour of the 3D model of the part on a specified plane and generate a cross-sectional contour image. The contour simplification module is used to extract the contour from the cross-sectional contour image and adopts an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour. An angle detection module is used to traverse the vertices of a simplified polygonal contour with an adaptive step size, calculate the angle value at each vertex, filter out vertices that meet the preset angle threshold as corner points, and record the angle values ​​of the corner points. The region recognition module is used to perform statistical analysis on the side length, angle value and corner density between adjacent corner points, and divide the simplified polygon outline into different types of geometric feature regions based on the statistical analysis results; The slicing scheme generation module is used to assign differentiated layer thicknesses to the identified different geometric feature regions according to the preset geometric features and forming layer thickness mapping rules, and generate variable layer thickness slicing schemes. The drive processing module is used to generate processing instructions based on the variable layer thickness slicing scheme, and drive the laser powder bed melting equipment to form parts layer by layer.

[0013] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the laser powder bed melting variable layer thickness forming method.

[0014] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the laser powder bed melting variable layer thickness forming method.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention effectively overcomes the high time cost caused by the small layer thickness strategy used in laser powder bed melting process and the inability of a single fixed layer thickness to achieve global optimization by combining polygonal contour geometric feature analysis with a variable layer thickness slicing strategy. This scheme first simplifies the cross-sectional contour image using an adaptive polygon approximation algorithm, and then uses an adaptive step size to traverse the polygon vertices and calculate angle values ​​to filter and mark corner points. Subsequently, by statistically analyzing the side lengths, angle values, and corner point density between adjacent corner points, the simplified polygon contour is accurately divided into different types of geometric feature regions based on the statistical analysis results. Finally, according to the preset mapping rules between geometric features and forming layer thickness, differentiated layer thicknesses are assigned to the identified different geometric feature regions, generating a variable layer thickness slicing scheme and driving the equipment to form layer by layer. This scheme can adapt to the different geometric features of the parts and their varying responses to heat input and cooling rates. By assigning differentiated layer thicknesses, it avoids the problem of heat accumulation leading to deformation in thin-walled areas, and eliminates the waste of resources due to low efficiency in thick areas. Thus, while alleviating residual stress caused by rapid melting and high temperature gradients and preventing cracking or forming failure, it achieves a simultaneous improvement in forming quality and manufacturing efficiency.

[0016] Furthermore, the cutting plane is defined by calculating the center coordinates of all facet vertices, ensuring that the extracted cross-section reflects the overall shape of the 3D model to the greatest extent. When extracting intersection points, the spatial positional relationship of the two endpoints of the triangle sides on both sides of the cutting plane is used for filtering, and the exact intersection point coordinates are calculated by combining linear interpolation. This maps the continuous spatial topology of the 3D mesh to a two-dimensional discrete point set, effectively avoiding spatial geometric distortion during the dimensionality reduction projection process, thus providing a high-precision cross-sectional contour image foundation for subsequent processing.

[0017] Furthermore, by linking the arc length of the cross-sectional profile with the approximation accuracy ratio, the adaptive polygon approximation algorithm can adaptively allocate computing power to profiles of different scales. For large-span profiles, a larger downsampling force is applied to eliminate redundant calculations, while for small-scale profiles, high accuracy is maintained to lock in local geometric details. While filtering out high-frequency boundary noise, the algorithm restores the low-frequency backbone structure of the polygon by reconstructing the closed line segment sequence, achieving an optimal balance between data dimensionality reduction and preservation of key geometric features.

[0018] Furthermore, by combining image topology with geometric measures, the system first analyzes the hierarchical relationships of the cross-sectional contours to remove invalid boundaries belonging to internal holes, ensuring that the analysis focus is concentrated on the external contours. Subsequently, a dual measure of area threshold and preset ratio of maximum contour area is introduced, which can eliminate discrete free points and small area noise caused by poor mesh quality or cutting interpolation, thus purifying the data source and preventing invalid contours from reducing the efficiency and accuracy of subsequent angle calculations.

[0019] Furthermore, by employing adaptive changes in polygon vertex density and using the number of vertices to deduce the adaptive step size, it is possible to overcome minute jagged edges on the boundary caused by discrete sampling or local defects. Using this step size, the first and second adjacent vertices are selected and a spatial vector is constructed. Then, the angle value is calculated using the vector dot product and inverse cosine operation to obtain the true turning angle of the vertex over the macroscopic geometric span, rather than the microscopic local tangent changes. This macroscopic smoothing mechanism based on vector algebra can shield against interference from uneven fluctuations and accurately pinpoint the true corner point.

[0020] Furthermore, when the vector traversal method fails, the analysis domain is smoothly switched from vector space to pixel space. By invoking the Harris corner detection algorithm, the autocorrelation matrix changes of pixel gradients in various directions within the local window are calculated, enabling responses to regions with significant orthogonal edge strength. Supplemented by scaling and centering, dilation processing, and connected component analysis, the positioning drift caused by pixel discretization is effectively overcome. Ultimately, the high-gradient response center is mapped back to the original coordinate system, filling in the feature detection blind spots caused by local topological anomalies or extreme smoothness. Attached Figure Description

[0021] Figure 1 This is a flowchart of the laser powder bed melting variable layer thickness forming method of Embodiment 1 of the present invention; Figure 2 These are the different geometric features of the three typical geometric feature structures of the present invention; Figure 3 This is a schematic diagram of the model of three typical geometric feature structures of the present invention after feature recognition; Figure 4 The deformation cloud diagrams are for three typical geometric feature structures of the present invention under different layer thickness strategies (before and after optimization). (a) and (b) are thin-walled conical cylinder structures, (c) and (d) are sandwich structures, and (e) and (f) are boss structures. Figure 5 This is a schematic diagram illustrating the adaptive matching of process parameters for three typical geometric feature structures according to the present invention. Figure 6 This is a schematic diagram comparing the forming effects of the conventional fixed layer thickness (a) and the variable layer thickness method (b) of the present invention for three typical geometric feature structural components. Detailed Implementation

[0022] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0023] Example 1: like Figure 1 As shown, the laser powder bed melting variable layer thickness forming method described in this embodiment includes the following steps: Step 1: Extract the cross-sectional profile of the 3D model of the part on the specified plane and generate a cross-sectional profile image.

[0024] Step 2: Extract the contour from the cross-sectional contour image and use an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour.

[0025] Step 3: Use an adaptive step size to traverse the vertices of the simplified polygonal contour, calculate the angle value at each vertex, select vertices that meet the preset angle threshold as corner points, and record the angle values ​​of the corner points.

[0026] Step 4: Perform statistical analysis on the side lengths, angle values, and corner density between adjacent corner points, and divide the simplified polygon outline into different types of geometric feature regions based on the statistical analysis results; Step 5: Based on the preset geometric features and forming layer thickness mapping rules, assign differentiated layer thicknesses to the identified different geometric feature regions to generate a variable layer thickness slicing scheme; Step 6: Generate processing instructions based on the variable layer thickness slicing scheme, and drive the laser powder bed melting equipment to form the part layer by layer.

[0027] The above process will be explained in detail below.

[0028] Step 1: Extract the STL section profile.

[0029] First, read the 3D model file in Standard Template Library (STL) format. The 3D model consists of multiple triangular facets, each containing the 3D coordinates of three vertices. Then, reshape the vertices of all triangular facets into an N×3 array, where N is the total number of vertices.

[0030] Calculate the center coordinates of all vertices: center = (center_x, center_y, center_z), where x, y, and z are 3D coordinate values. Define the cutting plane as a plane parallel to the XZ plane, which passes through the geometric center of the 3D model to obtain a representative cross-sectional profile.

[0031] Iterate through (visit each element in the set) the three edges of each triangular facet. For each edge, determine if the Y-coordinates of its two endpoints lie on opposite sides of the cutting plane, i.e., whether each edge intersects the cutting plane. If (p1_y - plane_y) × (p2_y - plane_y) < 0, where p1_y represents the Y-coordinate of the first endpoint of the current edge and p2_y represents the Y-coordinate of the second endpoint, and plane_y represents the position of the cutting plane on the Y-axis, this condition indicates that the Y-coordinates of the two endpoints are greater than and less than plane_y, i.e., one is above the plane and the other is below the plane. Therefore, the line segment must intersect the plane, and the edge intersects the cutting plane. Calculate the intersection point coordinates t using linear interpolation. t=(plane_y- p1_y) / (p2_y- plane_y) After collecting the coordinates t of all intersection points, a cross-sectional profile point set is formed, and the cross-sectional profile point set is converted into a binary image, i.e., a cross-sectional profile image, for subsequent processing.

[0032] Step 2: Adaptive polygon approximation.

[0033] The extracted cross-sectional contour image is preprocessed, including grayscale conversion and adaptive thresholding, to extract the contour and calculate the arc length (arc_length) of each inflection point in the contour.

[0034] The contour threshold ε of the polygon is dynamically determined based on the arc length: When arc_length < 500mm, ε = 0.01 × arc_length (higher precision is used for small contours to preserve details).

[0035] When 500mm≤arc_length<2000mm, ε=0.007×arc_length (medium precision is used for medium profiles).

[0036] When arc_length ≥ 2000mm, ε = 0.003 × arc_length (lower precision is used for large contours to simplify calculations).

[0037] Among them, 0.01, 0.007 and 0.003 are approximate accuracy scaling factors.

[0038] The Douglas-Peucker algorithm (a curve simplification algorithm that reduces the number of points and thus compresses data while preserving the approximate shape of the curve) is employed to approximate the contour into polygons using a contour threshold ε, resulting in multiple simplified polygon vertex sets. Based on these simplified polygon vertex sets, a sequence of closed line segments is constructed by sequentially connecting adjacent vertices, thereby recovering the geometric contour of the polygon and obtaining multiple simplified polygon contours. This algorithm recursively removes points whose distance from the connecting lines is less than the threshold, preserving the key shape features of the contour.

[0039] Simultaneously, the contours are filtered: external contours are identified through hierarchical relationships, and internal holes are excluded; the contour area is calculated, and small noise contours with areas smaller than the area threshold are filtered out; the contours are sorted in descending order of area, and effective contours with areas greater than the maximum contour area preset ratio (preferably 5% in this embodiment) are retained.

[0040] Step 3: Adaptive step size angle detection.

[0041] For simplified polygonal contours, obtain the number of vertices n, and calculate the adaptive step size step_size=max(1, n / / 20). The adaptive step size ensures that a large step size is used for large contours (with many points) to avoid false detections caused by small local fluctuations; and a small step size is used for small contours (with few points) to ensure sufficient detection accuracy.

[0042] Iterate through each vertex, taking the current vertex b as the center, move forward by step_size vertices a and backward by step_size vertices c, and calculate the angle value θ. The specific calculation method is as follows: ba = ab, bc = c – b cosθ=(ba·bc) / (|ba|×|bc|) θ = arccos(cosθ) × (180 / π) If the angle value is within the range of 90°±10°, then the vertex is taken as the corner point and the angle value is recorded.

[0043] When the adaptive step-size traversal method fails to detect significant angles, the Harris corner detection method is used as an alternative for supplementary detection. Specifically: a temporary image is created, the contour is scaled and centered; Harris corner detection is performed with block_size=3, aperture_size=3, k=0.04, and quality_level=0.01, where block_size is the pixel neighborhood side length considered when calculating the matrix; aperture_size is the size of the Sobel kernel (a discrete differential operator used for edge detection and gradient calculation in image processing) used to calculate the image gradient; k is an empirical coefficient controlling the corner response sensitivity; and quality_level is a threshold coefficient used to filter effective corners (a ratio relative to the maximum response value, set to 0.03 in this embodiment); subsequently, dilation and connected component analysis are performed on the detection results to obtain the corner center coordinates; the coordinates are then converted back to the original coordinate system and uniformly marked as 90° angles.

[0044] Corner distance visualization.

[0045] like Figure 3 As shown, different colors are assigned to the simplified polygonal outlines, and the outlines are drawn on the image using the corresponding colors. Marker circles are drawn at the corner points: solid red circles are used for corners of 90°±3°, and solid green circles are used for other corners. Angle values ​​are labeled near the corner points; corners of 90°±3° are labeled "Angle (90°)". Simultaneously, lines are drawn connecting the corner points to form a corner point connection diagram. This diagram is a visual image including angle labels, thus intuitively showing the connection relationships between the corner points and facilitating observation of the outline's geometric structure.

[0046] The final output includes a corner connection diagram and an analysis report containing detailed statistical information, including the perimeter, area, number of angles, angle details, and classification statistics for each contour.

[0047] Step 4: Geometric feature region identification.

[0048] In simplified polygonal contours, statistical analysis is performed on the side lengths, angle values, and corner density between adjacent corner points: when the angle values ​​at multiple consecutive vertices are close to 90° and the local curvature of the contour changes drastically, it can be identified as a suspended structure, and this area requires additional support in additive manufacturing; when two nearly parallel contour segments with a very small gap are detected, and the length of the line connecting their corner points is short and the angle is less than 90°, it indicates that the wall thickness of this area is relatively thin and can be marked as a thin-walled area; if corner points are sparse in a certain local area, the contour tends to be a straight line or a large-radius arc, and the contour area is large, it corresponds to a thick area, which has a stable structure and good heat conduction; and the part located between the thin-walled and thick areas, if the corner density, angle value, and contour width show a continuous gradual change trend, is identified as a transition zone.

[0049] Step 5: Generate a variable layer thickness slicing scheme.

[0050] The mapping rule between geometric features and forming layer thickness is as follows: smaller forming layer thickness (preferably 20-40 μm) is assigned to overhanging structures and thin-walled areas that are sensitive to thermal deformation; larger forming layer thickness (preferably 50-100 μm) is assigned to thick areas that are mainly composed of deposits; and a transition region with gradual layer thickness change is set at the point where the layer thickness changes.

[0051] Based on the variable layer thickness slicing scheme, the laser process parameters and scanning strategies are adaptively adjusted to match the layer thickness of each region: for regions with small layer thickness, a lower volumetric energy density and a zoned skip scanning strategy are used to control heat accumulation; for regions with large layer thickness, a higher volumetric energy density and a continuous long-range scanning strategy are used to improve melt channel stability and deposition efficiency.

[0052] Step 6: Drive the machining process.

[0053] Based on the variable layer thickness slicing scheme, laser process parameters, and scanning strategy, executable adaptive processing instructions are generated to drive the laser powder bed melting equipment to form parts layer by layer; the parts are simulated to verify the effectiveness of the variable layer thickness strategy.

[0054] The method described in this embodiment is particularly suitable for laser powder bed melting forming of easily deformable metal materials such as nickel-based superalloys, titanium alloys, and high-strength aluminum alloys.

[0055] This embodiment employs a two-level adaptive strategy. The first level adjusts the polygon approximation accuracy based on the contour arc length, while the second level adjusts the angle calculation step size based on the simplified vertex count. This enables angle detection of contours of different scales and complexities, effectively suppressing local fluctuation interference and improving the accuracy of corner point recognition.

[0056] This embodiment replaces the traditional fixed layer thickness strategy, realizing an intelligent closed loop from geometric model to manufacturing strategy, significantly improving the intelligence level of laser powder bed melting process. By assigning optimal layer thickness to different geometric features and matching customized processes, synergistic optimization of quality and efficiency is achieved: in easily deformable areas such as thin walls and overhangs, small layer thickness ensures forming accuracy and surface quality; in thick areas, large layer thickness significantly increases the material deposition rate, improving overall manufacturing efficiency by 30%-50%, while effectively suppressing overall deformation and residual stress of parts.

[0057] Example 2: A digital 3D model of a thin-walled conical cylinder structure is established, key geometric features in the 3D model are identified, and the 3D model is divided into regions according to feature type.

[0058] By calling the preset geometric features and forming layer thickness mapping rules, an adaptive variable layer thickness slicing scheme (layer thickness transitions from 60μm to 40μm) is generated, and the specific layer thickness change locations are clearly defined.

[0059] Based on the layer thickness information of each partition, the corresponding laser process parameters (such as power and speed) are matched and optimized to form a complete process package: layer thickness of 60μm, laser power of 260W, scanning speed of 900mm / s, and scanning spacing of 0.1mm; layer thickness of 40μm, laser power of 240W, scanning speed of 1000mm / s, and scanning spacing of 0.09mm.

[0060] Comparative Example 1: A fixed layer thickness strategy (layer thickness of 60 μm) and a matching laser powder bed melting process window were adopted: laser power of 260 W, scanning speed of 900 mm / s, layer thickness of 60 μm, and scanning interval of 0.1 mm. The slicing model file designed according to the process parameters was imported to verify the laser powder bed melting forming of stainless steel samples and thin-walled conical cylinder components. Figure 4 (a) and (b) in the figure are deformation cloud diagrams under different layer thickness strategies for thin-walled structures, compared with the deformation amount obtained in Example 2. Figure 4 The deformation amount obtained in (b) and Comparative Example 1 ( Figure 4 As can be seen from (a) in Example 2, the variable layer thickness strategy can improve the deformation of thin-walled conical structures during additive manufacturing.

[0061] Example 3: A digital 3D model of the sandwich structure is established, key geometric features in the 3D model are identified, and the 3D model is divided into regions according to feature type.

[0062] By calling the preset geometric features and forming layer thickness mapping rules, an adaptive variable layer thickness slicing scheme (layer thickness transitions from 40μm to 80μm) is generated, and the specific layer thickness change locations are clearly defined.

[0063] Based on the layer thickness information of each partition, the corresponding laser process parameters (such as power and speed) are matched and optimized to form a complete process package: layer thickness of 40μm, laser power of 240W, scanning speed of 1000mm / s, and scanning spacing of 0.09mm; layer thickness of 80μm, laser power of 320W, scanning speed of 900mm / s, and scanning spacing of 0.12mm.

[0064] Comparative Example 2: A fixed layer thickness strategy (layer thickness of 40 μm) and a matching laser powder bed melting process window were adopted: laser power of 240 W, scanning speed of 1000 mm / s, layer thickness of 40 μm, and scanning interval of 0.09 mm. The slicing model file designed according to the process parameters was imported to verify the laser powder bed melting forming of stainless steel samples and thin-walled ring components.

[0065] Figure 4 (c) and (d) in the figure are deformation cloud diagrams under different layer thickness strategies for the sandwich structure, compared with the deformation amount obtained in Example 3. Figure 4 The deformation amount obtained in (d) and Comparative Example 2 ( Figure 4 As can be seen from (c) in Example 3, the variable layer thickness strategy can improve the deformation of the sandwich structure during the additive manufacturing process.

[0066] Example 4: A digital 3D model of the boss structure is established, key geometric features in the 3D model are identified, and the 3D model is divided into regions according to feature type.

[0067] By calling the preset geometric features and forming layer thickness mapping rules, an adaptive variable layer thickness slicing scheme (layer thickness transitions from 40μm to 60μm) is generated, and the specific layer thickness change locations are clearly defined.

[0068] Based on the layer thickness information of each partition, the corresponding laser process parameters (such as power and speed) are matched and optimized to form a complete process package: layer thickness of 40μm, laser power of 240W, scanning speed of 1000mm / s, and scanning spacing of 0.09mm; layer thickness of 80μm, laser power of 260W, scanning speed of 900mm / s, and scanning spacing of 0.12mm.

[0069] Comparative Example 3: A fixed layer thickness strategy (60 μm) and a matching laser powder bed melting process window were adopted: laser power of 260 W, scanning speed of 900 mm / s, layer thickness of 60 μm, and scanning interval of 0.1 mm. The slicing model file designed according to the process parameters was imported to verify the laser powder bed melting forming of stainless steel samples and boss components. Microstructure analysis and mechanical property testing were performed on the samples, and roughness and residual stress were measured on the boss components.

[0070] Figure 4 (e) and (f) in the figure are deformation cloud diagrams under different layer thickness strategies for the process boss structure, compared with the deformation amount obtained in Example 4. Figure 4 The deformation amount obtained in (f) and Comparative Example 3 (f) Figure 4 As can be seen from (e) in Example 4, the variable layer thickness strategy can improve the deformation of the boss structure during the additive manufacturing process.

[0071] Example 5: Establish a digital 3D model containing three typical feature structures, identify key geometric features in the 3D model, and divide the 3D model into regions according to feature type, such as... Figure 2 As shown, Figure 2 The left side is a thin-walled conical cylinder structure, the middle is a sandwich structure, and the right side is a boss structure.

[0072] By invoking preset geometric features and forming layer thickness mapping rules, an adaptive variable layer thickness slicing scheme is generated, clearly defining the specific areas of layer thickness variation, such as... Figure 5 As shown.

[0073] Based on the layer thickness information of each zone, the corresponding laser process parameters are matched and optimized to form a complete process package.

[0074] Comparative Example 4: A fixed layer thickness strategy (40 μm) and a matching laser powder bed melting process window were adopted: laser power of 240 W, scanning speed of 900 mm / s, layer thickness of 40 μm, and scanning interval of 0.09 mm. The slicing model file designed according to the process parameters was imported to verify the laser powder bed melting forming of stainless steel samples and boss components. Microstructure analysis and mechanical property testing were performed on the samples, and roughness and residual stress were measured on the boss components.

[0075] Figure 6 The deformation cloud maps of three typical structural features under two layer thickness strategies are compared with the deformation obtained in Example 5. Figure 6 The deformation obtained in (b) and Comparative Example 4 ( Figure 6 As can be seen from (a) in Example 5, the variable layer thickness strategy can improve the deformation of the three typical feature structures during the additive manufacturing process.

[0076] Example 6: In this embodiment, a laser powder bed melting variable layer thickness forming system is provided. This laser powder bed melting variable layer thickness forming system can be used to implement the above-mentioned laser powder bed melting variable layer thickness forming method. Specifically, the laser powder bed melting variable layer thickness forming system includes an image generation module, a contour simplification module, an angle detection module, a region recognition module, a slicing scheme generation module, and a driving processing module.

[0077] The image generation module is used to extract the cross-sectional contour of the 3D model of the part on a specified plane and generate a cross-sectional contour image.

[0078] The contour simplification module is used to extract the contour from the cross-sectional contour image and adopts an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour.

[0079] The angle detection module is used to traverse the vertices of the simplified polygonal contour with an adaptive step size, calculate the angle value at each vertex, select vertices that meet the preset angle threshold as corner points, and record the angle values ​​of the corner points.

[0080] The region identification module is used to perform statistical analysis on the side length, angle value and corner density between adjacent corner points, and divide the simplified polygon outline into different types of geometric feature regions based on the statistical analysis results.

[0081] The slicing scheme generation module is used to assign differentiated layer thicknesses to the identified different geometric feature regions according to the preset geometric features and forming layer thickness mapping rules, and generate variable layer thickness slicing schemes.

[0082] The drive processing module is used to generate processing instructions based on the variable layer thickness slicing scheme, and drive the laser powder bed melting equipment to form parts layer by layer.

[0083] Example 7: This embodiment provides a computer device including a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the computer device, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or function. The processor described in this embodiment can be used in the operation of a laser powder bed melting variable layer thickness forming method, including: extracting the cross-sectional contour of a three-dimensional model of a part on a specified plane, generating a cross-sectional contour image; extracting the contour from the cross-sectional contour image, and using an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour. The contour is then simplified to obtain a simplified polygonal contour. An adaptive step size is used to traverse the vertices of the simplified polygonal contour, calculating the angle value at each vertex. Vertices that meet a preset angle threshold are selected as corner points, and their angle values ​​are recorded. Statistical analysis is performed on the side lengths, angle values, and corner point density between adjacent corner points. Based on the statistical analysis results, the simplified polygonal contour is divided into different types of geometric feature regions. According to a preset mapping rule between geometric features and forming layer thickness, differentiated layer thicknesses are assigned to the identified different geometric feature regions, generating a variable layer thickness slicing scheme. Processing instructions are generated based on the variable layer thickness slicing scheme to drive the laser powder bed melting equipment to form parts layer by layer.

[0084] Example 8: This embodiment provides a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the operating system of the computer device. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be a high-speed memory or non-volatile memory, such as at least one disk storage device.

[0085] One or more instructions stored in a computer-readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the laser powder bed melting variable layer thickness forming method in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: reading the STL format three-dimensional model file, extracting the cross-sectional contour of the three-dimensional model of the part on a specified plane, and generating a cross-sectional contour image; extracting the contour in the cross-sectional contour image, and using an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour to obtain a simplified polygon wheel. The simplified polygonal contour is traversed using an adaptive step size, and the angle value at each vertex is calculated. Vertices that meet the preset angle threshold are selected as corner points, and their angle values ​​are recorded. Statistical analysis is performed on the side lengths, angle values, and corner point density between adjacent corner points. Based on the statistical analysis results, the simplified polygonal contour is divided into different types of geometric feature regions. According to the preset geometric feature and forming layer thickness mapping rules, differentiated layer thicknesses are assigned to the identified different geometric feature regions to generate a variable layer thickness slicing scheme. Processing instructions are generated based on the variable layer thickness slicing scheme to drive the laser powder bed melting equipment to form the part layer by layer.

[0086] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0087] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0091] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0092] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

[0093] It should be understood that the above description is for illustrative purposes and not for limitation. Many embodiments and applications beyond the provided examples will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of this application should not be determined by reference to the above description, but rather by reference to the foregoing claims and the full scope of their equivalents. For purposes of completeness, all articles and references, including patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the foregoing claims is not intended as a waiver of that subject matter, nor should it be construed as an indication that the applicant has not considered that subject matter as part of the disclosed inventive subject matter.

Claims

1. A method for forming variable-thickness laser powder bed fusion layers, characterized in that, The process includes the following: Extract the cross-sectional profile of the 3D model of the part on a specified plane and generate a cross-sectional profile image, including: Obtain the vertex coordinates of all triangle faces in the 3D model, and calculate the center coordinates of all vertices; Define a plane that passes through the center coordinates and is parallel to the specified coordinate plane as the cutting plane; Traverse the three sides of each triangular facet and determine whether the two endpoints of each side are located on opposite sides of the cutting plane; Given that the two endpoints of an edge are located on opposite sides of a cutting plane, calculate the coordinates of the intersection point between the edge and the cutting plane using linear interpolation; Collect the coordinates of all intersection points to form a cross-sectional profile point set, convert the cross-sectional profile point set into a binary image, and obtain the cross-sectional profile image; The contour is extracted from the cross-sectional contour image, and an adaptive polygon approximation algorithm is used to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour. The simplified polygon contour is traversed by an adaptive step size, the angle value at each vertex is calculated, the vertices that meet the preset angle threshold are selected as corner points, and the angle values ​​of the corner points are recorded. Statistical analysis is performed on the side lengths, angle values, and corner density between adjacent corner points. Based on the statistical analysis results, the simplified polygon outline is divided into different types of geometric feature regions. Based on the preset mapping rules between geometric features and forming layer thickness, differentiated layer thicknesses are assigned to the identified regions with different geometric features to generate a variable layer thickness slicing scheme. Based on the variable layer thickness slicing scheme, processing instructions are generated to drive the laser powder bed melting equipment to form parts layer by layer.

2. The laser powder bed melting variable layer thickness forming method according to claim 1, characterized in that, The process of extracting the contour from the cross-sectional contour image and using an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy based on the arc length of each inflection point in the contour to simplify the contour and obtain a simplified polygonal contour includes: Extract the contour within the cross-sectional contour image and calculate the arc length of each inflection point in the contour; Based on the numerical range of the arc length, a corresponding approximate accuracy scaling factor is assigned to the contour, and the product of the arc length and the approximate accuracy scaling factor is used as the contour threshold. Using a contour threshold as a condition, polygon approximation is performed on the contour to obtain multiple simplified polygon vertex sets; A sequence of closed line segments is constructed based on the simplified polygon vertex set to restore the polygon's geometric structure and generate a simplified polygon outline.

3. The laser powder bed melting variable layer thickness forming method according to claim 2, characterized in that, After generating the simplified polygon outline, the process also includes filtering the simplified polygon outline: Identify the hierarchical relationships of simplified polygonal outlines, exclude internal outlines that belong to internal holes, and retain external outlines; Calculate the area of ​​the outer contour and filter out outer contours with an area smaller than the area threshold; The filtered outer contours are arranged in descending order of area, and the outer contours with an area greater than a preset ratio of the maximum contour area are retained as the final simplified polygonal contours.

4. The laser powder bed melting variable layer thickness forming method according to claim 1, characterized in that, The process of traversing the vertices of a simplified polygonal contour using an adaptive step size, calculating the angle value at each vertex, and selecting vertices that meet a preset angle threshold as corner points includes: Obtain the number of vertices of the simplified polygonal outline, and determine the adaptive step size based on the number of vertices; Centered on the current vertex, select the first and second adjacent vertices forward and backward respectively in the polygon vertex sequence according to the adaptive step size; Based on the coordinates of the current vertex, the first adjacent vertex, and the second adjacent vertex, calculate the vectors from the current vertex to the first adjacent vertex and from the current vertex to the second adjacent vertex, respectively. The cosine value is calculated using the dot product and magnitude of two vectors, and the angle value of the current vertex is obtained through inverse cosine operation; If the angle value falls within the preset angle threshold range, the current vertex will be used as the corner point.

5. The laser powder bed melting variable layer thickness forming method according to claim 1, characterized in that, When adaptive step size fails to identify corner points, the Harris corner detection method is used for supplementary detection, including: Create a temporary image, scale and center the outline, and perform Harris corner detection; The detection results are expanded and connected region analyzed to obtain the coordinates of the corner center. Transform the center coordinates of the corner points back to the original coordinate system and uniformly mark them as 90° angles.

6. The laser powder bed melting variable layer thickness forming method according to claim 1, characterized in that, The mapping rule between geometric features and forming layer thickness is as follows: Small forming layer thicknesses are allocated to overhanging structures and thin-walled areas, while large forming layer thicknesses are allocated to thick areas. A transition zone with gradual layer thickness changes is set between adjacent areas where the layer thickness difference exceeds a threshold.

7. A laser powder bed melting variable layer thickness forming system, characterized in that, include: The image generation module is used to extract the cross-sectional contour of the 3D model of the part on a specified plane and generate a cross-sectional contour image, including: Obtain the vertex coordinates of all triangle faces in the 3D model, and calculate the center coordinates of all vertices; Define a plane that passes through the center coordinates and is parallel to the specified coordinate plane as the cutting plane; Traverse the three sides of each triangular facet and determine whether the two endpoints of each side are located on opposite sides of the cutting plane; Given that the two endpoints of an edge are located on opposite sides of a cutting plane, calculate the coordinates of the intersection point between the edge and the cutting plane using linear interpolation; Collect the coordinates of all intersection points to form a cross-sectional profile point set, convert the cross-sectional profile point set into a binary image, and obtain the cross-sectional profile image; The contour simplification module is used to extract the contour from the cross-sectional contour image and adopts an adaptive polygon approximation algorithm to dynamically adjust the approximation accuracy according to the arc length of each inflection point in the contour, thereby simplifying the contour and obtaining a simplified polygon contour. An angle detection module is used to traverse the vertices of a simplified polygonal contour with an adaptive step size, calculate the angle value at each vertex, filter out vertices that meet the preset angle threshold as corner points, and record the angle values ​​of the corner points. The region recognition module is used to perform statistical analysis on the side length, angle value and corner density between adjacent corner points, and divide the simplified polygon outline into different types of geometric feature regions based on the statistical analysis results; The slicing scheme generation module is used to assign differentiated layer thicknesses to the identified different geometric feature regions according to the preset geometric features and forming layer thickness mapping rules, and generate variable layer thickness slicing schemes. The drive processing module is used to generate processing instructions based on the variable layer thickness slicing scheme, and drive the laser powder bed melting equipment to form parts layer by layer.

8. A computer 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 computer program, it implements the steps of the laser powder bed melting variable layer thickness forming method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the laser powder bed melting variable layer thickness forming method as described in any one of claims 1 to 6.

Images