An ultra-high-strength steel deep blind hole shell component head additive path planning method and system

Abstract

This invention discloses an additive manufacturing path planning method and system for the head of an ultra-high strength steel deep blind hole shell component. The method includes: acquiring a three-dimensional geometric model of the component to be formed; identifying and extracting the internal cavity surface and edge contour features of the head of the deep blind hole shell component; generating a variable thickness layered slicing scheme based on the internal cavity surface and edge contour features; adopting a multi-level partitioning strategy within each layer based on the variable thickness layered slicing scheme, and designing the main path direction and scanning angle for each region; generating an adaptive spiral progressive filling path that can suppress residual stress and deformation based on the main path direction and scanning angle; and generating a processing instruction file that can directly drive additive manufacturing equipment based on the adaptive spiral progressive filling path. Using this invention, active control of the heat distribution and shrinkage deformation within the internal cavity of the head of the ultra-high strength steel deep blind hole shell can be achieved, improving the forming accuracy, internal quality, and mechanical property consistency of the component.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of casting equipment, in particular to a method and system for additive path planning of a head of a deep blind hole shell member of ultra-high strength steel. BACKGROUND

[0002] With the increasing demand for high-performance components in the fields of aerospace, special equipment, etc., the use of additive manufacturing technology to integrally form ultra-high strength steel shell members with deep blind hole features has become an important way to solve the manufacturing problems of complex structures. However, during the high-energy beam additive process such as laser cladding, there is a significant thermal accumulation effect and a tendency to shrink and deform in ultra-high strength steel, especially in the head of the deep blind hole member, which is closed or semi-closed in the inner cavity area. The traditional uniform thickness layering and fixed path planning method is prone to process defects such as heat concentration, excessive residual stress, poor interlayer bonding, and loss of dimensional accuracy. The existing technology focuses on path optimization for conventional structures, and for deep blind hole special geometric features, especially how to dynamically and adaptively plan the path according to the inner cavity surface shape and material properties, there is still a lack of systematic and effective solutions, which restricts the further improvement of the forming quality and performance reliability of such members. SUMMARY

[0003] The purpose of the present application is to provide a method and system for additive path planning of a head of a deep blind hole shell member of ultra-high strength steel, to solve the problems in the prior art, and to actively regulate the heat distribution and shrinkage deformation of the inner cavity of the head of the deep blind hole shell of ultra-high strength steel, thereby improving the forming precision, internal quality, and mechanical property consistency of the member.

[0004] One embodiment of the present application provides a method for additive path planning of a head of a deep blind hole shell member of ultra-high strength steel, the method comprising: obtaining a three-dimensional geometric model of a member to be formed, identifying and extracting the inner cavity surface and edge contour features of the head of the deep blind hole shell member; According to the inner cavity surface and edge contour features, combining the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel, a variable-thickness layering slicing scheme adaptively matched with the inner cavity surface is generated; Based on the variable-thickness layering slicing scheme, a multi-level partitioning strategy is used in each layering to divide the inner cavity cross section into an outer ring additive area, an intermediate transition additive area, and an inner ring additive area, and the main path direction and scanning angle of each area are designed respectively; According to the main path direction and scanning angle, combining the thermal force sequence coupling algorithm to dynamically plan the scanning sequence and overlapping parameters of each layering, an adaptive spiral progressive filling path that can suppress residual stress and deformation is generated; According to the adaptive spiral progressive filling path, a machining instruction file that can directly drive the additive manufacturing equipment is generated, and the layer-by-layer forming and manufacturing of the head of the deep blind hole shell member is completed.

[0005] Another embodiment of this application provides an additive path planning system for the head of an ultra-high strength steel deep blind hole shell component, the system comprising: The acquisition module is used to acquire the three-dimensional geometric model of the component to be formed, and to identify and extract the inner cavity surface and edge contour features of the head of the deep blind hole shell component. The generation module is used to generate a variable thickness layered slicing scheme that adaptively matches the inner cavity surface based on the inner cavity surface and edge contour features, combined with the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel material. The partitioning module is used to divide the inner cavity cross section into an outer ring additive manufacturing area, an intermediate transition additive manufacturing area and an inner ring additive manufacturing area based on the variable thickness layered slicing scheme, using a multi-level partitioning strategy within each layer, and designing the main path direction and scanning angle of each area respectively. The planning module is used to dynamically plan the scanning sequence and overlap parameters of each layer based on the main path direction and scanning angle, combined with the thermal sequential coupling algorithm, to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation. The manufacturing module is used to generate a processing instruction file that can directly drive the additive manufacturing equipment according to the adaptive spiral progressive filling path, and to complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

[0006] Another embodiment of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method described in any of the preceding claims.

[0007] Compared with the prior art, the additive path planning method for the head of an ultra-high strength steel deep blind hole shell component provided by the present invention can realize the active control of heat distribution and shrinkage deformation in the inner cavity of the head of the ultra-high strength steel deep blind hole shell, thereby improving the forming accuracy, internal quality and mechanical property consistency of the component. Attached Figure Description

[0008] Figure 1 A hardware structure block diagram of a computer terminal for an additive path planning method for the head of an ultra-high strength steel deep blind hole shell component, provided in an embodiment of the present invention; Figure 2 A flowchart illustrating an additive path planning method for the head of an ultra-high strength steel deep blind hole shell component, provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an additive path planning system for the head of an ultra-high strength steel deep blind hole shell component, provided in an embodiment of the present invention. Detailed Implementation

[0009] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0010] The present invention first provides an additive path planning method for the head of an ultra-high strength steel deep blind hole shell component. This method can be applied to electronic devices, such as computer terminals, specifically ordinary computers.

[0011] The following detailed explanation uses a computer terminal as an example. Figure 1 This is a hardware structure block diagram of a computer terminal for an additive path planning method for the head of an ultra-high strength steel deep blind hole shell component, provided in an embodiment of the present invention. (See diagram for reference.) Figure 1 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0012] See Figure 2 The present invention provides an additive path planning method for the head of an ultra-high strength steel deep blind hole shell component, which may include the following steps: S201, Obtain the three-dimensional geometric model of the component to be formed, and identify and extract the inner cavity surface and edge contour features of the head of the deep blind hole shell component; Specifically, you can import the 3D CAD design model file of the component to be formed, perform geometric repair and topology optimization on the model to ensure that the model is a watertight and non-self-intersecting entity, and generate a standard 3D geometric model that can be processed. The core of this step is to lay a high-quality model foundation for subsequent feature extraction. By fixing model defects and optimizing the topology, the integrity and geometric accuracy of the model are ensured. The specific implementation method is as follows: The import of 3D CAD design model files supports mainstream formats (such as STL, STEP, and IGES), with STEP format being the preferred choice (based on boundary representation, preserving accurate geometric topological relationships) to avoid the discretization errors of triangular facets in STL format. The import tool uses the kernel of professional geometry processing software, supporting batch import and automatic format conversion. During the import process, it automatically detects file integrity; if problems such as corrupted format or missing data are found, a repair prompt is immediately triggered, requiring manual intervention to re-import or select an alternative file.

[0013] Common defects in geometric repair models include surface holes, cracks, free edges, and overlapping surfaces. Surface hole repair employs a "boundary fitting and filling" algorithm. First, the boundary contour of the hole (a closed loop composed of continuous edges) is identified. The curvature distribution of the boundary is calculated, and a NURBS surface is used to fit the boundary to generate a filling surface, ensuring curvature continuity between the filling surface and the surrounding surfaces (curvature transition error ≤ 0.01mm⁻¹). For micro-holes with a diameter ≤ 0.1mm, automatic filling is performed; for holes with a diameter > 0.1mm, administrator confirmation is prompted before filling. Crack repair uses an "edge alignment and stitching" algorithm. Corresponding edges on both sides of the crack are found, and the overlap and angle between the edges are calculated. If the overlap is ≥ 95% and the angle is ≤ 5°, the edges are automatically stitched together as continuous edges; otherwise, the crack is split into multiple small segments and stitched together one by one. The free edge repair uses the "extend and supplement" algorithm to extend the free edge along its tangent direction to the adjacent surface to form a closed edge loop; the overlapping surface repair uses topological Boolean operations to delete duplicate surfaces and retain the only valid surface to ensure that there is no redundancy on the model surface.

[0014] Topology optimization aims to simplify the model structure, eliminate geometric redundancy, and retain core features. Optimization operations include: merging overlapping vertices (vertices with a distance ≤0.005mm are considered overlapping and merged into a single vertex), deleting isolated edges / faces (edges or faces without associated topological relationships), simplifying small chamfers / rounded corners (chamfers / rounded corners with a radius ≤0.2mm are simplified to right angles or planes without affecting the overall geometry, reducing subsequent computational complexity), and optimizing the topological connections of faces (ensuring that the edge sharing relationships of adjacent faces are correct and there are no topological conflicts).

[0015] Watertightness and non-self-intersection verification are the core criteria for judging the model. Watertightness verification is achieved through "closed volume calculation." If the model can calculate a clear volume value (no leakage), it is judged to be watertight. If the volume calculation fails (a leakage path exists), the geometric repair process is traced back to find unclosed edges or unstitched cracks. Non-self-intersection verification uses a "spatial interference detection" algorithm, which traverses all faces of the model and calculates the shortest distance between faces. If there are faces with a distance < 0 (spatial overlap), they are judged to be self-intersecting, and repaired by moving overlapping faces, splitting interference regions, etc.

[0016] Feature recognition and analysis are performed on the standard three-dimensional geometric model. The region growing algorithm is used to automatically identify the head region of the deep blind hole shell component, separate the internal cavity structure entity from the outer shell, and generate the head internal cavity structure model. The core of this step is to accurately locate the target region (the head of the deep blind hole) and achieve structural separation through a region growing algorithm, thus focusing the analysis on the subsequent internal cavity feature extraction. The specific implementation method is as follows: Feature recognition analysis first involves a global topological traversal of the standard 3D geometric model to extract the model's core geometric feature parameters, including volume, surface area, aspect ratio, hole depth to diameter ratio (deep blind hole determination criteria: hole depth / diameter ≥ 5), surface type (plane, cylindrical, spherical, freeform surface), etc., and then establishes a global feature parameter library.

[0017] The execution of the region growing algorithm requires setting a seed point and growth criteria. The seed point is selected as the geometric center of the head of the deep blind hole (determined by calculating the minimum bounding box center of the head region; in the example, the center coordinates are (120mm, 80mm, 200mm)), ensuring that the seed point is located inside the target region. The growth criteria include three core conditions: curvature similarity (the difference in principal curvature between the current growth surface and the surface where the seed point is located ≤ 0.02mm⁻¹), normal vector angle (the angle between the normal vectors of the current growth surface and the surface where the seed point is located ≤ 30°), and topological connectivity (the current growth surface shares edges or vertices with the already grown regions). The algorithm starts from the surface where the seed point is located, traverses all surfaces of the model, and verifies whether each surface meets the growth criteria. If it does, it is included in the growth region until no new surface meets the conditions, generating the preliminary outline of the head region.

[0018] Precise segmentation of the head region requires consideration of the structural characteristics of the deep blind hole. The internal cavity structure is separated from the external shell using Boolean operation cutting. First, based on the hole axis direction (positive Z-axis in this example), a cylindrical cut (0.5mm larger in diameter and 5mm longer than the hole depth) coaxial with the hole axis is generated. The external shell is obtained through Boolean difference operation (model minus cut). A preliminary internal cavity structure is obtained through Boolean intersection operation (model and cut). Then, the preliminary internal cavity structure is cleaned, removing redundant surfaces introduced by the cutting, while retaining core structures such as the inner wall surface of the deep blind hole and the top surface of the head, generating the head internal cavity structure model. This model contains only the internal cavity of the deep blind hole head, with a volume of 18.3cm³. Its surface consists of 3 free-form surfaces and 2 cylindrical surfaces, without interference from the external shell, serving as the core object for subsequent feature extraction.

[0019] Based on the head cavity structure model, the principal curvature distribution and normal vector field of the cavity surface are extracted by the surface curvature analysis algorithm, and the boundary contour of the cavity opening is located by the edge detection algorithm to generate the cavity geometric feature parameter set; The core of this step is to quantify the geometric shape and boundary features of the inner cavity surface. By extracting parameters such as principal curvature, normal vector field, and boundary contour, accurate geometric basis is provided for subsequent path planning. The specific implementation method is as follows: The surface curvature analysis algorithm calculates the curvature of each surface (freeform surface, cylindrical surface) in the head cavity structure model. The core is to solve for the principal curvatures k1 (maximum principal curvature) and k2 (minimum principal curvature), as well as the directions of the principal curvatures. For parametric surfaces (such as cylindrical surfaces), the calculation is done directly using analytical formulas: the principal curvatures k1 = 0 (along the generatrix of the cylinder), k2 = 1 / R (along the circumference of the cylinder, where R is the radius of the cylinder; in the example, R = 6mm, k2 ≈ 0.167mm⁻¹). For freeform surfaces, a "discrete point curvature estimation" algorithm is used. Uniform sampling (10 points per mm²) is performed on the surface, and the covariance matrix of each sampling point is calculated. The principal curvatures and directions are obtained through eigenvalue decomposition. The sampling point spacing is ≤0.1mm to ensure the continuity of the curvature distribution. The principal curvature distribution is visualized using a "curvature cloud map," with different colors representing the magnitude of curvature (red for high curvature areas and blue for low curvature areas). In the example, the freeform surface at the top of the head cavity has a high curvature area (k1=0.32mm⁻¹, k2=0.18mm⁻¹), while the cylindrical portion is divided into a uniform low curvature area (k1=0, k2=0.167mm⁻¹).

[0020] The normal vector field is calculated for each sampling point of the surface, solving for the unit normal vector (pointing outward from the cavity) at that point. For parametric surfaces, the normal vector is obtained by cross product of the partial derivatives of the surface, and then normalized to a unit vector. For freeform surfaces, a plane is fitted using the neighborhood points of the sampling point, and the plane normal vector is calculated as an approximation of the normal vector at that point. The calculation accuracy of the normal vector is ≤0.001 (the magnitude deviation of the unit vector). The normal vector field is visualized using an "arrow cloud," where the arrow direction is the normal vector direction, and the arrow length represents the consistency of the normal vector (the longer the length, the more consistent the neighborhood normal vectors). In the example, the normal vector field height of the cylindrical part is consistent (the arrow length is uniform). The normal vector field of the freeform surface gradually adjusts with the curvature change, without obvious abrupt changes.

[0021] Edge detection algorithms are used to locate the boundary contour of the inner cavity opening. The boundary contour is the dividing line between the inner cavity and the outer shell, and is a closed loop composed of continuous edges. A combination of "curvature abrupt change detection" and "edge type recognition" is used: First, all edges of the head inner cavity structure model are traversed, and the angle between the normal vectors of the two curved surfaces on both sides of the edge is calculated. If the angle is ≥90° (curvature abrupt change), it is marked as a candidate boundary edge. Next, the topological type of the edge is identified. If the edge is associated with only one surface (free edge), or if the two associated surfaces belong to the inner cavity and the outer shell respectively (separated by Boolean operations, the outer surface has been deleted, and the edge is a free edge), it is determined to be a boundary edge. Finally, the continuous boundary edges are sorted clockwise to form a closed boundary contour.

[0022] The generation of the internal cavity geometric feature parameter set integrates the above extraction results and includes four types of core parameters: surface geometric parameters (the principal curvature range, average curvature, and surface area of ​​each surface), normal vector field parameters (normal vector direction distribution and neighborhood normal vector consistency coefficient), boundary contour parameters (contour type, contour size, and contour closure), and overall structural parameters (internal cavity volume, hole depth, hole diameter, and opening area).

[0023] By integrating the set of internal cavity geometric feature parameters, a mathematical representation is constructed to describe the geometric shape of the internal cavity surface and the topological relationship of the boundary contour, ultimately generating accurate internal cavity surface and edge contour features.

[0024] The core of this step is to transform the discrete set of parameters into a unified mathematical model that accurately describes the geometry and topological relationships of the cavity, providing a computable mathematical basis for subsequent layered slicing and path planning. The specific implementation method is as follows: The mathematical representation of the inner cavity surface uses appropriate expressions for different types of surfaces. The cylindrical surface is represented by parametric equations: x=R×cosθ, y=R×sinθ, z=t (where R=6mm is the radius of the cylindrical surface, θ∈[0,2π) is the circumferential angle, and t∈[0,75mm] is the axial parameter of the cylindrical surface). This equation accurately describes the coordinates of all points on the cylindrical surface with an error ≤0.001mm. The freeform surface is represented by the NURBS surface equation. By extracting the control vertices (each freeform surface contains 4×4=16 control vertices with a coordinate accuracy of 0.001mm), weight factors (default 1.0, with a weight factor of 1.5 for the control vertices at the surface boundary to enhance boundary constraints), and node vectors (uniform node vectors to ensure uniform distribution on the surface), the equation S(u,v)=ΣΣP_i,j×w_i,j×N_i,p(u)×N_j,q(v) / ΣΣw_i,j×N_i,p(u)×N_j,q(v) (where P_i,j are the control vertices, w_i,j are the weight factors, N_i,p(u) and N_j,q(v) are the B-spline basis functions, and p and q are the degrees of the basis functions, both taken as 3), the geometry of the freeform surface can be accurately reconstructed, with a maximum deviation from the original model ≤0.02mm.

[0025] The mathematical representation of the boundary contour adopts a parametric curve equation. The circular contour of the inner cavity opening is represented by the circle equation: (x-x0)²+(y-y0)²=R² (where (x0,y0) are the coordinates of the contour center (120mm,80mm), and R=6.2mm is the contour radius). Any point on the contour can be represented by the parameter θ (θ∈[0,2π)) as x=x0+R×cosθ, y=y0+R×sinθ, z=200mm (the z-coordinate of the contour). The fitting error of the equation is ≤0.005mm. If the boundary contour is non-circular (such as ellipse or polygon), it is represented by the NURBS curve equation. By extracting the control vertices, weight factors, and node vectors of the contour, the overlap between the curve and the original contour is ensured to be ≥99.5%.

[0026] The mathematical description of topological relationships is achieved through an "incidence matrix," with dimensions of "number of surfaces × number of boundary contours." Matrix elements are either 0 or 1, where 1 indicates an association between the surface and the boundary contour (the edge of the surface belongs to the boundary contour), and 0 indicates no association. In the example, the incidence matrix is ​​a 3×1 matrix (3 core surfaces, 1 boundary contour). The element corresponding to the cylinder is 1 (the top edge of the cylinder belongs to the boundary contour), and the elements corresponding to the two free-form surfaces are 0 (not directly associated with the boundary contour), clearly representing the topological association between the surfaces and the boundary.

[0027] The final generated precise internal cavity surface and edge contour features are stored in the form of "mathematical equations + topological correlation matrix," including the cylindrical NURBS equation, two free-form surface NURBS equations, one boundary contour circle equation, and the corresponding correlation matrix and core geometric parameters (principal curvature range, normal vector distribution interval, etc.). This feature can be directly called by subsequent layer slicing algorithms to obtain the cross-sectional contour of arbitrary height through equation solving, providing a precise and computable geometric basis for the generation of variable thickness layer slicing schemes.

[0028] S202, Based on the features of the inner cavity surface and edge contour, and combined with the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel material, a variable thickness layered slicing scheme that adaptively matches the inner cavity surface is generated. Specifically, based on the precise internal cavity surface and edge contour features, the rate of change of the internal cavity cross-sectional area and perimeter at different heights along the component axis can be calculated to generate the internal cavity geometric change curve; The core of this step is to quantify the dynamic changes in the internal cavity geometry along the axial direction. By capturing regions of geometric abrupt changes through the rate of change of cross-sectional area and perimeter, geometric basis is provided for subsequent variable thickness slicing. The specific implementation method is as follows: First, the axial direction of the component is defined. Taking the hole axis of the deep blind hole as the positive direction of the Z-axis, the axial height range is from the bottom of the inner cavity (Z=0mm) to the opening of the inner cavity (Z=80mm), covering the entire inner cavity depth of the head of the deep blind hole. To accurately capture geometric changes, an equal-interval sampling strategy is adopted, with the sampling interval set at 1mm. That is, along the Z-axis from 0mm to 80mm, a total of 81 sampling heights (Z=0mm, 1mm, 2mm...80mm) are selected. Each sampling height corresponds to a two-dimensional inner cavity cross section, ensuring that no geometric changes are missed.

[0029] The two-dimensional internal cavity cross-section is obtained through a "planar sectioning algorithm." A sectioning plane perpendicular to the Z-axis is constructed using the Z-value of each sampling height. This plane intersects the internal cavity surface to form a closed cross-sectional profile (polygon or circle). For the regular cylindrical portion (Z=0mm to Z=75mm), the sectioning plane intersects the cylindrical surface to form a circular cross-section with a fixed radius of 6mm. For the top freeform surface portion (Z=75mm to Z=80mm), the sectioning plane intersects the freeform surface to form an irregular polygonal cross-section. The intersection line is solved using the NURBS surface equation to obtain the vertex coordinates of the cross-sectional profile (accuracy 0.001mm).

[0030] The calculation of cross-sectional area adopts an adaptive method for different cross-sectional types: the formula for the area of ​​a circular cross-section is S=πR² (R is the radius of the cross-section; in the example, at Z=50mm, R=6mm, S=π×6²≈113.1mm²); the area of ​​an irregular polygonal cross-section is calculated using the "shoelace formula," which calculates the area sequentially using the coordinates (x_i, y_i) of the vertices of the cross-section contour. The formula is S=0.5×|Σ(i=1 to n)(x_iy_{i+1}-x_{i+1}y_i)| (x_{n+1}=x_1, y_{n+1}=y_1). In the example, the cross-sectional contour at Z=78mm contains 24 vertices, and the calculated area is ≈98.5mm².

[0031] Calculation of the perimeter of the cross section: The formula for the perimeter of a circular cross section is L=2πR (at Z=50mm, L=2π×6≈37.7mm); the perimeter of an irregular polygonal cross section is the sum of the lengths of all the sides of the outline, and the side length is calculated according to the formula for the distance between two points (d=√[(x_i-x_{i+1})²+(y_i-y_{i+1})²]). In the example, the perimeter of the cross section at Z=78mm is ≈35.2mm.

[0032] The calculation of the rate of change of cross-sectional area and the rate of change of perimeter is based on the adjacent sampling height. The formula for the rate of change is η=(X_k-X_{k-1}) / X_{k-1}×100% (where k is the index of the current sampling height, X_k is the area or perimeter of the current height, and X_{k-1} is the area or perimeter of the previous height). The rate of change characterizes the degree of geometric change; a positive value increases the change, a negative value decreases it, and the larger the absolute value, the more drastic the change. In the example, at Z=75mm (the junction of the cylinder and the freeform surface), the area changes from 113.1mm² at Z=74mm to 108.3mm² at Z=75mm, with a rate of change of area η=(108.3-113.1) / 113.1×100%≈-4.24%; the perimeter changes from 37.7mm to 36.5mm, with a rate of change of perimeter ≈-3.18%, reflecting the abrupt change in the geometric shape at the junction.

[0033] The internal cavity geometric change curves are generated with the axial height Z as the abscissa and the cross-sectional area S, perimeter L, area change rate η_S, and perimeter change rate η_L as the ordinates, constructing four curves. A cubic spline interpolation algorithm is used to smooth the curves (avoiding the polyline effect caused by discrete sampling points), improving the curve resolution to 0.1 mm to ensure accurate location of geometrically abrupt change regions (such as the change peak at Z=75 mm). The curves are stored in vector format, with key feature points (geometric abrupt change points, extreme points) labeled. In the example, the curves show that Z=0 mm to Z=74 mm is a geometrically smooth region (absolute area change rate ≤ 0.1%), and Z=75 mm to Z=80 mm is a geometrically abrupt change region (absolute area change rate ≥ 3%).

[0034] Query the database of ultra-high strength steel material properties to obtain the material's thermal expansion coefficient, phase transformation temperature range, high temperature yield strength and cooling shrinkage rate parameters, and generate a set of material thermodynamic property parameters. The core of this step is to extract the key thermodynamic parameters of ultra-high-strength steel, providing accurate material property inputs for thermodynamic simulation and ensuring that the simulation results can truly reflect the material's thermal accumulation and shrinkage deformation laws. The specific implementation method is as follows: The ultra-high strength steel material property database is a structured database that stores the full-temperature range thermodynamic parameters of common ultra-high strength steels (such as 4340, 300M, and AF1410). The data are sourced from material handbooks, experimental test data, and literature, ensuring traceability and accuracy. For the 300M ultra-high strength steel selected in this method, the database is queried to obtain core thermodynamic parameters. These parameters cover the temperature range of the additive manufacturing process (25℃ to 1500℃), ensuring coverage of key intervals such as room temperature, phase transformation temperature, and melting temperature.

[0035] The coefficient of thermal expansion (α) characterizes the dimensional expansion of a material as a function of temperature, measured in units of 1 / ℃. The database provides values ​​for different temperature ranges. For 300M steel, the coefficients of thermal expansion are: α = 11.8 × 10⁻⁶ / ℃ from room temperature to 300℃; α = 13.5 × 10⁻⁶ / ℃ from 300℃ to 600℃ (before phase transformation); α = 15.2 × 10⁻⁶ / ℃ from 600℃ to 900℃ (phase transformation range); and α = 16.8 × 10⁻⁶ / ℃ from 900℃ to 1500℃. This parameter directly affects the magnitude of thermal stress caused by the temperature field.

[0036] The phase transformation temperature range is the temperature range in which a material transforms from austenite to martensite. The phase transformation start temperature (Ms) of 300M steel is 380℃, and the phase transformation end temperature (Mf) is 280℃. The phase transformation temperature range is from 280℃ to 380℃. During the phase transformation, there will be volume change (about 3% volume expansion), which is one of the core factors leading to shrinkage deformation. The thermo-mechanical coupling effect in this range needs to be considered in the simulation.

[0037] High-temperature yield strength (σ_s) is the ability of a material to resist plastic deformation at high temperatures, measured in MPa. It decreases as temperature increases. The high-temperature yield strength of 300M steel is: σ_s = 1200 MPa at 300℃; σ_s = 850 MPa at 600℃; σ_s = 320 MPa at 900℃; and σ_s = 80 MPa at 1200℃. This parameter is used to determine whether the thermal stress exceeds the material's yield strength, thereby predicting the risk of plastic deformation and cracking.

[0038] Cooling shrinkage rate (β) is the volume shrinkage ratio of a material as it cools from a molten state to room temperature, expressed as a percentage. The cooling shrinkage rate of 300M steel is divided into three stages: shrinkage rate β1 = 2.1% from liquid to solidus (1450℃ to 1300℃); shrinkage rate β2 = 1.8% from solidus to phase transformation start temperature (1300℃ to 380℃); and shrinkage rate β3 = 0.9% from phase transformation range to room temperature (380℃ to 25℃). The total cooling shrinkage rate β = β1 + β2 + β3 = 4.8%. This parameter directly determines the amount of shrinkage deformation in the additive manufacturing process and is the core basis for the design of variable thickness slicing schemes.

[0039] The generation of the material thermodynamic property parameter set integrates the above parameters and organizes them in the format of "parameter name-unit-temperature range-value-physical meaning" to ensure that the parameters are clear and easy to find. Example parameter set fragment: "Coefficient of thermal expansion (α): 1 / ℃, 25-300℃, 11.8×10^-6 / ℃, characterizing the rate of dimensional expansion of the material from room temperature to 300℃; Phase transformation temperature range: ℃, 280-380℃, Ms=380℃ / Mf=280℃, the temperature range of the material from austenite to martensite; High temperature yield strength (σ_s): MPa, 600℃, 850MPa, the critical stress of the material to resist plastic deformation at 600℃; Cooling shrinkage rate (β): %, 1450-25℃, 4.8%, the proportion of total volume shrinkage of the material when cooled from the molten state to room temperature."

[0040] By inputting the internal cavity geometric change curve and the material thermodynamic property parameter set into the thermodynamic simulation model, the temperature field distribution and thermal stress concentration area under different slice thicknesses are predicted, and thermo-mechanical coupling simulation results are generated. The core of this step is to simulate the manufacturing process under different slice thicknesses through thermo-mechanical coupling simulation, quantify the temperature field and thermal stress distribution, and provide a simulation basis for adaptive adjustment of the layer thickness. The specific implementation method is as follows: The thermodynamic simulation model is constructed based on the finite element analysis method, using thermo-coupled elements (which combine structural mechanics and heat conduction analysis capabilities). The model geometry is based on the internal cavity structure model of the head, and the mesh generation adopts an adaptive meshing strategy: fine mesh (0.2 mm element size) is used in geometrically abrupt regions (Z=75-80 mm), and coarse mesh (0.5 mm element size) is used in geometrically smooth regions (Z=0-74 mm), with a total of approximately 120,000 elements to ensure a balance between simulation accuracy and computational efficiency.

[0041] Boundary conditions are set to fit the actual additive manufacturing scenario: the thermal boundary condition adopts a convection heat transfer model, with a surface heat transfer coefficient h=25W / (m²・K) (natural air convection) and an ambient temperature T_env=25℃; the laser heat source model adopts a Gaussian heat source with a power P=300W, a spot diameter d=0.8mm, a scanning speed v=5mm / s, and the heat source movement path is set according to the preliminary scanning strategy; the structural boundary condition constrains the displacement of the bottom of the component (Z=0mm) (the displacements in the X, Y, and Z directions are all 0) to simulate the actual clamping state.

[0042] The simulation schemes for different slice thicknesses were designed as three comparative schemes, with slice thicknesses of 0.1mm, 0.2mm, and 0.3mm respectively (covering the commonly used slice thickness range in additive manufacturing). Each scheme was simulated by scanning layer by layer according to the axial height, with the number of scanned layers being 800 (0.1mm), 400 (0.2mm), and 267 (0.3mm, with the last layer being 0.2mm thick), respectively. During the simulation, after each layer was scanned, the transient temperature field (nodal temperature value, accuracy 0.1℃) and thermal stress field (nodal Mises equivalent stress, accuracy 1MPa) of that layer were recorded until the entire head cavity was formed.

[0043] Temperature field distribution prediction results: When the slice thickness is 0.1 mm, the maximum peak temperature is 1480℃ (melting zone), the temperature superposition between adjacent layers is small, the width of the heat-affected zone (temperature > 800℃) is about 0.3 mm, and there is no obvious heat accumulation; When the slice thickness is 0.3 mm, the maximum peak temperature is 1520℃, the heat accumulation between adjacent layers is significant, the width of the heat-affected zone is about 0.8 mm, and local high temperature accumulation occurs in the geometric abrupt change region of Z=75-80 mm (the area of ​​the region with temperature > 1200℃ is 2.3 times that of the 0.1 mm slice).

[0044] Prediction results of thermal stress concentration areas: When the slice thickness is 0.1 mm, the maximum Mises equivalent stress is 420 MPa (located at the junction of the cylindrical and free-form surfaces at Z=75 mm), which does not exceed the yield strength of 300 M steel at this temperature (850 MPa at 600℃), and there is no risk of plastic deformation; When the slice thickness is 0.3 mm, the maximum Mises equivalent stress is 780 MPa (also located at the junction of Z=75 mm), which is close to the yield strength, and a stress concentration zone appears at Z=78 mm (the length of the area with stress value > 650 MPa is about 5 mm), which poses a risk of plastic deformation and cracking; When the slice thickness is 0.2 mm, the maximum stress is 580 MPa, and the length of the stress concentration area is about 2 mm, which is between 0.1 mm and 0.3 mm.

[0045] The results of the thermo-mechanical coupling simulation are organized around the core of "slice thickness-temperature field characteristics-thermal stress field characteristics" to generate a simulation report, which includes temperature field cloud map, thermal stress field cloud map, temperature / stress variation curves of key areas with height, and clarifies the degree of heat accumulation, stress concentration location and risk level under different slice thicknesses (low risk: stress <500MPa; medium risk: 500-700MPa; high risk: >700MPa).

[0046] Analyzing the results of thermo-mechanical coupling simulation, a thinner layer thickness is adopted in the stress concentration area to refine the control, while a thicker layer is adopted in the geometrically smooth area to improve efficiency. Finally, a variable thickness layer slicing scheme that adaptively matches the inner cavity surface is generated.

[0047] The core of this step is to combine geometric changes with simulation results to achieve adaptive allocation of layer thickness, maximizing manufacturing efficiency while ensuring forming quality (suppressing stress and deformation). The specific implementation method is as follows: The simulation results were analyzed in dimensions including the intensity of geometrical changes and the level of thermal stress risk. A two-dimensional decision matrix of "geometrical change-stress risk" was established to determine the optimal slice thickness for each axial height. The intensity of geometrical changes was divided according to the rate of area change: smooth zone (|η_S|≤1%), transition zone (1%<|η_S|≤3%), and abrupt change zone (|η_S|>3%). The level of thermal stress risk was divided according to the maximum stress value: low risk (σ≤500MPa), medium risk (500<σ≤700MPa), and high risk (σ>700MPa). The decision rules were: abrupt change zone + high risk → slice thickness 0.1mm; transition zone + medium risk → slice thickness 0.2mm; smooth zone + low risk → slice thickness 0.3mm.

[0048] Slice thickness distribution for each axial height: Z=0mm to Z=74mm is the geometrically flat region (|η_S|≤0.1%), with simulation results indicating low risk (σ≤450MPa). A slice thickness of 0.3mm is allocated to this region, totaling 75 layers (Z=0-74.7mm, with the last layer at 0.3mm). This region accounts for 93.75% of the total height; using thicker layers can significantly improve manufacturing efficiency. Z=74.7mm to Z=77mm is the geometrically transitional region (1%<|η_S|≤3%). The simulation results indicate a medium risk (500 < σ ≤ 600 MPa), with a slice thickness of 0.2 mm and a total of 11 layers (Z = 74.7-76.9 mm), balancing efficiency and quality. The area from Z = 76.9 mm to Z = 80 mm is a geometric abrupt change region (|η_S| > 3%), with a simulation result indicating a high risk (σ > 600 MPa), with a slice thickness of 0.1 mm and a total of 31 layers (Z = 76.9-80 mm). By refining temperature and stress control through thin layering, deformation can be suppressed.

[0049] The detailed specifications of the variable thickness layered slicing scheme include: layer number (layers 1 to 117), axial height range of each layer (e.g., layer 1 Z=0-0.3mm, layer 75 Z=74.4-74.7mm, layer 86 Z=76.7-76.9mm, layer 117 Z=79.9-80mm), slice thickness (0.1mm / 0.2mm / 0.3mm), corresponding geometric region type, and stress risk level. The scheme also includes two-dimensional cross-sectional profile data for each layer (for subsequent path planning) and interlayer connection requirements (when the slice thickness of adjacent layers changes, the path direction is rotated by 30° to avoid stress superposition between layers).

[0050] The scheme was validated through simulation recalculation. The variable thickness slicing scheme was input into a thermodynamic simulation model. The recalculation results showed: a maximum peak temperature of 1490℃, a heat-affected zone width of 0.4mm, a maximum Mises equivalent stress of 520MPa, a stress concentration zone length ≤1mm, and no risk of plastic deformation. The total number of layers manufactured was 117, with a total manufacturing time of approximately 4.5 hours. Compared to a full-layer 0.1mm slicing (800 layers, approximately 13 hours), the efficiency was improved by 65%, and compared to a full-layer 0.3mm slicing (267 layers, approximately 3 hours), the quality was significantly optimized, achieving a balance between efficiency and quality. The final generated variable thickness layered slicing scheme is stored as a structured file, containing the geometry, thickness, and process constraint information of all layers, and can be directly used for subsequent multi-level partitioning and path design.

[0051] S203, based on the variable thickness layered slicing scheme, a multi-level partitioning strategy is adopted in each layer to divide the inner cavity cross section into an outer ring additive manufacturing area, an intermediate transition additive manufacturing area and an inner ring additive manufacturing area, and the main path direction and scanning angle of each area are designed respectively. Specifically, the two-dimensional inner cavity cross-sectional profile of the current layer can be obtained based on the variable thickness layer slicing scheme. Based on the curvature distribution of the cross-sectional profile and the distance to the boundary, the equidistant offset algorithm is used to generate concentric closed loops with different widths at the distance from the profile boundary. The core of this step is to transform the three-dimensional layers into two-dimensional, processable cross-sectional contours, and to generate the boundaries of functional zones through equidistant offsets, providing a precise geometric boundary basis for subsequent region division. The specific implementation method is as follows: First, key information for the current layer is extracted from the variable thickness layering slicing scheme, including layer number, axial height range, slice thickness, and corresponding 3D geometric data. Taking an intermediate layer (number 50, axial height Z = 14.7-15.0 mm, slice thickness 0.3 mm, geometrically flat region) as an example, the 2D internal cavity cross-sectional profile of this layer is obtained through a planar sectioning algorithm. Because it is located in a cylindrical region, the cross-sectional profile is a standard circle with center coordinates (120 mm, 80 mm), radius R = 6 mm, and consists of 360 discrete vertices with uniform vertex spacing, an angle of 1° between adjacent vertices, and a coordinate accuracy of 0.001 mm, ensuring the smoothness and accuracy of the profile.

[0052] For complex cross-sectional profiles that are not circular (such as layered regions with abrupt geometric changes), the intersection line is obtained by solving the NURBS surface equation and the cutting plane, resulting in a closed polygonal profile composed of multiple line segments and arcs. Then, a curve fitting algorithm is used to smooth the profile, eliminating discretization errors and ensuring that the profile has no sharp corners and no self-intersections. For example, for a layered cross-section at Z=78mm, the fitted profile contains 24 vertices, forming a smooth, irregular closed curve with a maximum radius of curvature of 5.8mm and a minimum radius of curvature of 3.2mm.

[0053] The curvature distribution analysis of the cross-sectional profile employs a discrete point curvature estimation method. One hundred sampling points are uniformly selected on the profile, and the curvature value (in mm⁻¹) is calculated for each point. A larger curvature value indicates a steeper profile at that location. In the example, the curvature distribution of the circular cross-section is uniform, with all sampling points having a curvature value of 1 / R = 0.167 mm⁻¹. The curvature distribution of complex cross-sections exhibits a gradient change, with steep regions (such as profile corners) having a curvature value of 0.31 mm⁻¹ and gentler regions having a curvature value of 0.08 mm⁻¹. The distance to the boundary is calculated using the profile boundary as a reference, along the profile normal vector direction inwards, measuring the straight-line distance from each point to the boundary. The distance ranges from 0 mm (at the boundary) to the maximum distance from the center of the cross-section (6 mm for a circular cross-section).

[0054] The core of the equidistant offset algorithm is to offset the contour at equal distances along the inner normal vector direction, generating concentric closed loops. The offset distance is set in conjunction with the curvature distribution and functional zoning requirements, following the principle of "small offset width in steep areas and large offset width in gentle areas" to ensure that the closed loops do not intersect or fold. For circular cross-sections (geometrically gentle areas), three levels of offset distance are set: the first level offset is 1.5mm (1.5mm offset from the boundary inward), the second level offsets another 1.0mm (cumulative offset 2.5mm), generating three concentric closed loops (the outermost loop is the original contour, the second outermost loop is the contour after offset 1.5mm, and the third outermost loop is the contour after offset 2.5mm); for complex cross-sections (geometrically abrupt areas), the first level offset is 1.0mm in steep areas, the first level offset is 1.2mm in gentle areas, and the second level offset is uniformly 0.8mm, using adaptive offset distances to avoid the closed loops from intersecting.

[0055] During the offset process, a "contour repair" mechanism is employed. If self-intersection or breakage occurs after offset, the local offset distance (±0.2mm) is automatically adjusted and recalculated to ensure that each closed loop is a complete and smooth closed curve. In the example, the three-level offset of the circular cross-section generates three concentric circles with radii of 6mm (outermost), 4.5mm (second outermost), and 3.5mm (further outermost), with uniform width between the loops. The offset of the complex cross-section generates three similarly shaped closed loops, with the width between the loops dynamically adjusted according to the contour curvature, ranging from a maximum width of 1.2mm to a minimum width of 0.8mm, without self-intersection or breakage. Finally, concentric closed loops with different widths from the contour boundary are generated.

[0056] Based on the concentric closed loops, the functional areas are divided into the outermost and second outermost rings. The area between the outermost and second outermost rings is defined as the outer ring additive manufacturing area, the area between the second outermost and third outermost rings is defined as the intermediate transition additive manufacturing area, and the remaining central area is defined as the inner ring additive manufacturing area, thus generating a multi-level partition map. The core of this step is to clarify the scope of the three functional areas based on the spatial relationship of concentric closed loops, forming a visual partition map to provide a basis for area division in subsequent path design. The specific implementation method is as follows: The division of functional areas follows the principles of "boundary priority, smooth transition, and high efficiency at the center." Combining the quality and efficiency requirements of additive manufacturing, the functional positioning of each area is clearly defined: the outer ring additive manufacturing area is close to the cross-sectional boundary and needs to ensure edge forming accuracy and bonding strength; the middle transition additive manufacturing area connects the outer and inner rings and needs to optimize stress distribution and avoid stress abrupt changes between areas; the inner ring additive manufacturing area is located at the center of the cross-section and can prioritize forming efficiency.

[0057] The partitioning process uses concentric closed loops as boundaries. The annular region between the outermost loop (original cross-sectional profile) and the second outermost loop is the outer additive manufacturing zone. The width of this region is the difference between the two offset distances (1.5mm for the outer ring of a circular cross-section in the example, and 1.0-1.2mm for complex cross-sections). It covers the critical area near the cross-sectional boundary, and the forming quality of this region directly affects the dimensional accuracy and surface roughness of the component edge. The annular region between the second outermost loop and the third outermost loop is the intermediate transition additive manufacturing zone. Its width is the difference between the last two offset distances (1.0mm for the middle of a circular cross-section in the example, and 0.8mm for complex cross-sections). It serves as a buffer zone between the outer and inner rings, used to coordinate the stress differences caused by different scanning strategies. The central region surrounded by the third outermost loop is the inner additive manufacturing zone. This region has the largest area (in the example, the area of ​​the inner ring of a circular cross-section = π × 3.5² ≈ 38.5mm², accounting for 34.1% of the total cross-sectional area). It has low forming difficulty and can employ efficient scanning strategies.

[0058] After the area division is completed, a region validity check is required to ensure that the three regions do not overlap or omission and cover the entire two-dimensional cross-sectional contour. The check method is to calculate the sum of the areas of each region, and the deviation from the total cross-sectional area must be ≤0.5%. In the example, the total area of ​​the circular cross-section is ≈113.1mm², the outer ring area = π×(6²-4.5²)=π×(36-20.25)=π×15.75≈49.5mm², the middle area = π×(4.5²-3.5²)=π×(20.25-12.25)=π×8≈25.1mm², and the inner ring area ≈38.5mm². The sum of the total areas is ≈49.5+25.1+38.5≈113.1mm², the deviation is 0, and the check passes.

[0059] The generation of multi-level zoning maps employs pseudo-color mapping technology, mapping the outer additive manufacturing area to red, the intermediate transition additive manufacturing area to yellow, and the inner additive manufacturing area to blue. The map includes cross-sectional contours, boundary lines of each region, region color identifiers, region names, and key dimension annotations (such as region width and inner circle radius). The map is stored in vector graphic format, is scalable without distortion, and in the example, the zoning map of a circular cross-section clearly marks three concentric circular regions: the blue inner circle radius is 3.5mm, the yellow middle ring width is 1.0mm, and the red outer ring width is 1.5mm. For complex cross-sections, the zoning map marks the boundary curves and width range of each region, intuitively presenting the region division results and providing clear visual guidance for subsequent path design.

[0060] For the outer ring additive manufacturing area, the main path is designed to perform a circular scan along the contour tangent direction to enhance the edge bonding strength. For the middle transition additive manufacturing area, the main path is designed to perform a cross scan along the direction at a 45-degree angle to the contour to optimize stress distribution. For the inner ring additive manufacturing area, the main path is designed to perform a sparse grid-like scan to improve forming efficiency, thus generating a partitioned main path strategy. The core of this step is to design a suitable scanning path and method based on the functional positioning of each area, balancing forming quality and efficiency. The specific implementation method is as follows: The main path design of the outer additive manufacturing area focuses on "enhancing edge bonding strength" and employs a circular scanning method along the contour tangent direction. The scanning path generation begins at the inner boundary (secondary outer ring) of the outer additive manufacturing area and spirals towards the outer boundary (outermost ring) along the contour tangent direction. The path spacing (distance between two adjacent scanning paths) is set to 0.3 mm (based on 37.5% of the laser spot diameter of 0.8 mm, ensuring sufficient overlap between adjacent paths). The angle between the scanning direction and the contour tangent direction is ≤5° to ensure the path fits snugly against the edge contour and avoids jagged defects. In the example, the outer circular scanning path of the circular cross-section consists of a series of concentric circles with a radius difference of 0.3 mm between adjacent circles, starting from a radius of 4.5 mm and increasing sequentially to 6.0 mm, for a total of 5 scanning paths. Each path scans clockwise, perfectly aligned with the circular tangent direction. During the scanning process, the laser continuously melts the material, forming a dense edge structure and enhancing edge bonding strength.

[0061] The main path design for the intermediate transition additive manufacturing zone focuses on "optimizing stress distribution," employing a cross-scanning method at a 45-degree angle to the profile. The cross-scanning consists of two sets of mutually perpendicular scanning paths. The first set of paths forms a 45-degree angle with the tangent direction of the cross-sectional profile, while the second set forms a -45-degree angle (perpendicular to the first set). These two sets of paths alternate, with a path spacing of 0.5 mm (balancing stress dispersion and forming efficiency). During scanning, the overlap rate of the two sets of paths is 50%, ensuring a tight bond between the weld lines. Simultaneously, the cross-layout disperses thermal stress, avoiding stress concentration caused by scanning in a single direction. In the example of the intermediate transition zone of the circular cross-section, the first set of 45-degree paths extends from a 4.5 mm radius to a 3.5 mm radius along a northeast-southwest direction, consisting of two paths. The second set of -45-degree paths also covers this area along a northwest-southeast direction, also consisting of two paths. The intersection of these two sets of paths forms a grid-like structure, evenly distributing thermal stress in both directions and effectively reducing the risk of deformation.

[0062] The main path design of the inner ring additive manufacturing zone focuses on "improving forming efficiency" and employs a sparse grid-like scanning method. The grid paths are set at 0 degrees (horizontal) and 90 degrees (vertical), with a path spacing of 1.0 mm (more than three times that of the outer ring), significantly reducing the number of scanning paths and increasing forming speed. The grid scanning starts from the center of the inner ring and extends towards the outer boundary (the outer ring again), with paths in the 0-degree and 90-degree directions scanning alternately. The melt flow overlap rate is 30%, maximizing efficiency while ensuring forming density (≥99.5%). In the example, for the inner ring additive manufacturing zone with a circular cross-section, the 0-degree path extends from the center (120 mm, 80 mm) to a radius of 3.5 mm, with a total of 3 paths; the 90-degree path also covers this area, with a total of 3 paths. These 6 paths form a sparse grid, and the scanning time is only 1 / 3 of that of the outer ring, significantly improving overall forming efficiency.

[0063] The generation of the main path strategy for each region integrates the path design details, including parameters such as path direction, scanning method, path spacing, scanning direction, and number of paths. It clarifies the scanning priority of each region (inner circle → middle → outer circle, or outer circle → middle → inner circle, to be determined later based on thermal simulation). The strategy document includes path diagrams, parameter tables (in textual descriptions), and design basis explanations for each region. In the example, the strategy specifies: the outer additive manufacturing zone uses a circular scan with a path spacing of 0.3mm, clockwise, and 5 paths to enhance edge strength; the middle transition zone uses a ±45-degree cross scan with a path spacing of 0.5mm and 4 paths to optimize stress; the inner additive manufacturing zone uses a 0 / 90-degree sparse mesh scan with a path spacing of 1.0mm and 6 paths to improve efficiency, providing a foundation for subsequent parameter refinement and sequence planning.

[0064] By integrating multi-level partition maps and partition main path strategies, the scanning path density, scanning speed, and scanning angle parameters are specified for each partition, and finally, the main path direction and scanning angle design scheme of each partition are generated.

[0065] The core of this step is to refine the partition map and path strategy into executable process parameters, clarify the key scanning parameters of each partition, and form a complete design scheme to provide a basis for subsequent scanning sequence and overlap parameter planning. The specific implementation method is as follows: The precise determination of scanning path density is based on the path spacing and scanning method in each region. Path density is defined as the number of paths per unit length (strips / mm), and is inversely proportional to the path spacing (density = 1 / path spacing). In the outer additive manufacturing zone, the path spacing is 0.3mm, and the path density = 1 / 0.3 ≈ 3.33 strips / mm, with high-density scanning ensuring edge forming accuracy. In the middle transition additive manufacturing zone, the path spacing is 0.5mm, and the path density = 1 / 0.5 = 2 strips / mm, with medium density balancing stress and efficiency. In the inner additive manufacturing zone, the path spacing is 1.0mm, and the path density = 1 / 1.0 = 1 strip / mm, with low-density scanning improving efficiency. The path density setting must match the laser power and powder feed rate to ensure uniform cladding layer thickness for each path (target thickness 0.3mm, consistent with the slice thickness). In the example, high-density scanning in the outer zone corresponds to a laser power of 300W and a powder feed rate of 8g / min, while low-density scanning in the inner zone corresponds to a laser power of 280W and a powder feed rate of 7g / min, ensuring consistent cladding quality.

[0066] The scanning speed is determined by considering path density, laser power, and material properties, following the principle of "high density, low speed; low density, high speed" to ensure sufficient material melting without excessive heat accumulation. The outer additive manufacturing zone has a high path density, so the scanning speed is set to 4 mm / s. This slower speed allows for full interaction between the laser and the material, enhancing edge bonding strength. The middle transition additive manufacturing zone has a scanning speed set to 5 mm / s, a moderate speed balancing stress dispersion and heat accumulation. The inner additive manufacturing zone has a low path density, so the scanning speed is set to 6 mm / s, a faster speed improving forming efficiency. The scanning speed error is controlled within ±0.2 mm / s, achieved through the equipment's closed-loop speed control. In the example, the deviation between the actual scanning speed and the set value is ≤0.1 mm / s, ensuring process stability.

[0067] Defining the scanning angle parameters requires quantifying the angle between the path direction and the reference direction, which is set as the horizontal tangent direction of the cross-sectional profile (0 degrees). The annular scanning angle of the outer additive manufacturing zone dynamically changes with the tangent direction of the profile. The scanning angle of the circular cross-section changes continuously from 0 degrees to 360 degrees, and the angle at each path point is consistent with the tangent direction at that point. The scanning angle of complex cross-sections is dynamically adjusted in each path segment to ensure that the angle with the profile tangent is ≤5°. The cross-scanning angles of the intermediate transition additive manufacturing zone are defined as 45 degrees and -45 degrees, with angles of 45° and -45° respectively with the reference direction (0 degrees). The angle deviation between the two sets of paths is ≤1°. The mesh scanning angles of the inner additive manufacturing zone are defined as 0 degrees (horizontal direction) and 90 degrees (vertical direction), with an angle deviation of ≤0.5° to ensure accurate path direction.

[0068] The generation of the main path direction and scanning angle design scheme for each zone integrates all parameters to form a structured document, including a scheme overview, multi-level zone map (with parameter annotations), detailed parameter descriptions for each zone (path direction description, scanning angle value, path density, scanning speed, supporting process parameters such as laser power and powder feeding rate), and path diagram (textual description of path distribution and direction). In the example design, the detailed parameters for the outer additive manufacturing zone are as follows: the path follows the tangent direction of the cross-sectional contour, progressing in a circular spiral, with the scanning angle dynamically changing with the tangent direction (continuously changing from 0-360° for a circular cross-section), a path density of 3.33 lines / mm, a scanning speed of 4mm / s, a laser power of 300W, and a powder feed rate of 8g / min; the intermediate transition zone uses a cross-scanning path, with scanning angles of 45° and -45°, a path density of 2 lines / mm, a scanning speed of 5mm / s, a laser power of 290W, and a powder feed rate of 7.5g / min; the inner additive manufacturing zone uses a grid scanning path, with scanning angles of 0° and 90°, a path density of 1 line / mm, a scanning speed of 6mm / s, a laser power of 280W, and a powder feed rate of 7g / min. The design also includes parameter verification instructions. Through small-sample trial printing, the forming quality of each area meets the requirements (edge ​​roughness Ra≤3.2μm, density≥99.5%, no cracks or deformation), ultimately forming a complete design scheme that can be directly used for subsequent sequence planning.

[0069] S204. Based on the main path direction and scanning angle, and combined with the thermal sequential coupling algorithm, the scanning sequence and overlap parameters of each layer are dynamically planned to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation. Specifically, a design scheme can be based on the main path direction and scanning angle of each partition. Following the order from the inner circle additive manufacturing area to the outer circle additive manufacturing area, the path scanning sequence of different areas within the same layer can be initially planned to generate a preliminary scanning sequence. The core of this step is to determine a reasonable scanning order based on regional functional priority and thermal stress control logic, laying the foundation for subsequent thermal simulation and parameter optimization. The specific implementation method is as follows: The scanning sequence planning follows the core principle of "inner circle priority, outer circle lag," which essentially reduces heat accumulation and stress constraints during outer circle forming by first forming the low-stress central region and then forming the highly sensitive boundary regions. The inner circle additive manufacturing zone, as the central region of the cross-section, has no obvious boundary constraints during forming, making thermal stress easy to release. Furthermore, the use of sparse mesh scanning allows for rapid construction of the core support structure. The intermediate transition additive manufacturing zone, acting as a stress buffer zone, follows the inner circle forming, coordinating the stress gradient between the inner and outer circles. The outer circle additive manufacturing zone, located near the component edge, has the highest requirements for dimensional accuracy and bonding strength. Forming it last avoids secondary disturbance to the edges by subsequent scanning, and allows for dynamic adjustment of process parameters based on the thermal state of the preceding regions.

[0070] The specific planning process requires considering the number and distribution of paths in each area, assigning a unique scan number to each path. Taking a certain layer as an example, the inner additive manufacturing zone contains 6 grid paths (3 in the 0-degree direction and 3 in the 90-degree direction), numbered 1-6 in the order of "horizontal first, then vertical, from the center outwards"; the intermediate transition additive manufacturing zone contains 4 intersecting paths (2 in the 45-degree direction and 2 in the -45-degree direction), numbered 7-10 in the order of "45-degree first, then -45-degree, intersecting"; the outer additive manufacturing zone contains 5 ring paths (spiraling from the inside out), numbered 11-15. The overall scan sequence is 1→2→3→4→5→6→7→8→9→10→11→12→13→14 →15, ensuring continuous scanning of the path within the same area, smooth transition between different areas, and no sudden changes in the thermal field caused by skipping scans.

[0071] To verify the rationality of the sequence, preliminary logical checks are required: Check for adjacent scan paths crossing different regions (to avoid frequent switching of the thermal field); confirm that the path numbers in each region are continuous (inner circle 1-6, middle circle 7-10, outer circle 11-15), without any numbering discrepancies; check whether the scanning direction is consistent with the path direction (inner circle 0 / 90 degrees, middle circle 45 / -45 degrees, outer circle tangent direction), ensuring that the laser scan matches the path extension direction and reducing motion impact. In the example, the preliminary scan sequence does not cross any regions, the numbers are continuous, and the direction matches, conforming to the core logic of "inner circle → middle circle → outer circle." The generated preliminary scan sequence can be directly used for subsequent thermal simulations.

[0072] The initial scanning sequence is input into the thermo-mechanical sequential coupling simulation algorithm. The algorithm simulates the laser scanning process and dynamically calculates the evolution of the transient temperature field and stress field caused by the deposition of material along each path, generating thermo-mechanical evolution data. The core of this step is to accurately reproduce the thermo-mechanical coupling process under the initial scanning sequence through thermo-mechanical sequential coupling simulation, quantify the dynamic changes of temperature and stress, and provide data support for subsequent parameter optimization. The specific implementation method is as follows: The core logic of the thermo-mechanical sequential coupling simulation algorithm is "sequential loading, thermo-mechanical coupling, and transient solution." This means loading the laser heat source sequentially according to the path order of the initial scanning sequence, calculating the transient temperature field after deposition along each path, and then calculating the thermal stress field based on the temperature field to achieve dynamic coupling between temperature and stress. The core parameter settings of the algorithm are: time step of 0.01 seconds (matching the laser scanning speed to ensure capture of transient thermal response), laser heat source using a Gaussian distribution model, power of 300W (outer ring) / 290W (middle) / 280W (inner ring), spot diameter of 0.8mm, scanning speed of 4mm / s (outer ring) / 5mm / s (middle) / 6mm / s (inner ring); material parameters using the thermo-mechanical properties of 300M ultra-high strength steel: thermal conductivity 45W / (m・K), specific heat capacity 500J / (kg・K), coefficient of thermal expansion 12×10^-6 / ℃, and yield strength varying with temperature (850MPa at 600℃).

[0073] The simulation process was executed sequentially according to the path number: When scanning path 1 (the inner horizontal path), the temperature of the laser-affected area rapidly rose to 1480℃ (melting temperature), forming a molten pool approximately 0.8mm wide and 0.3mm deep. The surrounding heat-affected zone (temperature > 800℃) was approximately 0.3mm wide, and the thermal stress was mainly concentrated at the edge of the molten pool, with a maximum value of approximately 200MPa (not exceeding the yield strength); When scanning path 6 (the last vertical path of the inner circle), the core area of ​​the inner circle had been formed, and the heat accumulation from subsequent path scans caused the temperature of the central area to remain at 600-800℃. After stress superposition, the maximum stress of the inner circle was approximately 350MPa; When scanning path 11 (the outer circle... When scanning path 15 (the last ring path of the outer ring), the outer ring is closer to the edge of the component, and the heat dissipation conditions are better than those in the center. However, the heat accumulation in the preceding area causes the initial temperature of the outer ring to reach 400℃, and the temperature of the molten pool after scanning is 1500℃. The thermal stress is concentrated at the junction of the edge and the middle area, with a maximum value of about 550MPa (close to the yield strength at 600℃). When scanning path 15 (the last ring path of the outer ring), the outer ring is fully formed, and the overall thermal stress distribution shows a gradient of "high in the outer ring and low in the inner ring". The maximum stress is 580MPa (located at the corner of the outer ring edge), and the temperature field shows a distribution of "high in the center and low in the edge". The highest temperature in the central area is 750℃, and the temperature in the edge area is 450℃.

[0074] The generation of thermal evolution data integrates the transient results after each path scan, including: scan time, peak melt pool temperature, heat-affected zone range, stress distribution around the path (Mises equivalent stress), overall temperature field contour map, and overall stress field contour map for each path. The data is organized in the format of "path number-temperature characteristic-stress characteristic". This data comprehensively records the dynamic evolution of temperature and stress under the scan sequence, clearly presenting the location and degree of stress concentration, providing clear targets for subsequent optimization.

[0075] By analyzing thermal evolution data, we can identify scanning sequences and path overlaps that may lead to thermal stress concentration or deformation, dynamically adjust the scanning interval and overlap rate of adjacent paths, and generate optimized overlap parameters and intermittent strategies. The core of this step is anomaly identification based on thermal evolution data, targeted adjustment of overlap parameters and intermittent strategies to suppress thermal stress concentration. The specific implementation method is as follows: The analysis of thermal evolution data focuses on two core dimensions: identification of thermal stress concentration areas and assessment of path overlap quality. Thermal stress concentration areas are determined through stress field cloud maps and numerical statistics, with a stress threshold set at 500 MPa (58.8% of the yield strength at 600℃). Areas exceeding this threshold are identified as high-risk concentration areas. In the example analysis, it was found that the stress values ​​at the overlap of path 10 (the last -45-degree path in the middle) and path 11 (the first ring path in the outer circle), and at the junction of path 15 (the last ring path in the outer circle) and the edge contour, reached 550 MPa and 580 MPa respectively, indicating high-risk concentration areas. Simultaneously, the overlap of inner path 3 and path 4, due to an excessively short scanning interval, resulted in thermal accumulation leading to a stress of 380 MPa (although not exceeding the threshold, there is a risk of cumulative stress). The quality assessment of the path overlap is judged by the degree of overlap of the molten pool. The initial planned overlap rate (the overlap ratio of adjacent path molten pools) is 30%. The analysis found that the molten pool overlap at the outer path overlap is insufficient, resulting in insufficient bonding strength and stress concentration. The overlap at the inner path overlap is excessive, resulting in heat accumulation.

[0076] The adjustment of the scanning interval between adjacent paths aims to "suppress heat accumulation and release residual stress." The interval is defined as the time difference between the start of scanning two adjacent paths, in seconds. For the inner additive manufacturing area, where heat accumulation is slight and stress release is rapid, no interval is maintained (continuous scanning). For the intermediate transition additive manufacturing area, where heat accumulation is moderate and the risk of stress concentration is low, an interval of 0.2 seconds is set (a 0.2-second pause after scanning one path, allowing the local temperature to drop by 50-80°C before scanning the next path). For the outer additive manufacturing area, where heat accumulation is significant and the risk of stress concentration is high, an interval of 0.5 seconds is set (a longer pause to ensure sufficient heat dissipation in the edge area, resulting in a temperature drop of 100-150°C). In the example, the interval between paths 10 and 11 is adjusted from 0 seconds to 0.8 seconds (cross-regional interval, longer pause), and the interval between paths 14 and 15 is 0.5 seconds. After the adjustment, the initial temperature at the outer edge overlap drops from 400°C to 320°C, significantly alleviating thermal stress concentration.

[0077] The adjustment of the overlap rate between adjacent paths aims to "optimize the weld pool bonding and disperse thermal stress." The overlap rate is defined as the ratio of the overlap width of the weld pools between two adjacent scanning paths to the width of the weld pool (the weld pool width is determined by the laser spot diameter and scanning speed; in this example, the outer ring weld pool width is 0.8 mm, the middle is 0.7 mm, and the inner ring is 0.6 mm). For the outer ring additive manufacturing area, to enhance edge bonding strength and disperse stress, the overlap rate is adjusted from 30% to 40% (overlap width 0.32 mm); for the middle transition additive manufacturing area, to balance bonding strength and stress distribution, the overlap rate is adjusted from 30% to 35% (overlap width 0.245 mm); for the inner ring additive manufacturing area, to improve efficiency and reduce heat accumulation, the overlap rate is adjusted from 30% to 25% (overlap width 0.15 mm). In the example, after adjusting the overlap rate between the outer ring paths 11 and 12, the weld pool overlap is more sufficient, the bonding strength is increased by 15%, and the stress concentration is reduced from 550 MPa to 480 MPa.

[0078] The optimized overlap parameters and the adjusted parameters generated by the intermittent strategy clearly define the overlap rate and interval time for each region: inner ring additive manufacturing zone (overlap rate 25%, interval time 0 seconds), intermediate transition additive manufacturing zone (overlap rate 35%, interval time 0.2 seconds), outer ring additive manufacturing zone (overlap rate 40%, interval time 0.5 seconds), and the cross-region interval time (middle → outer ring) is 0.8 seconds. The strategy also includes the basis for parameter adjustment and expected effects. In the example, the expected overall maximum stress is reduced from 580MPa to 480MPa, and the width of the heat-affected zone is controlled within 0.3mm, providing optimized parameter support for subsequent path replanning.

[0079] Based on optimized overlap parameters and intermittent strategies, the scanning path is replanned so that the filling path starts from the central region and expands outward in a spiral progressive manner, while ensuring that the paths of adjacent layers are rotated at a certain angle, ultimately generating an adaptive spiral progressive filling path that can suppress residual stress and deformation.

[0080] The core of this step is to integrate optimized parameters into the path design. Through a spiral progressive structure and interlayer rotation strategy, thermal stress is further dispersed to achieve precise suppression of residual stress and deformation. The specific implementation method is as follows: The adaptive spiral progressive filling path planning follows the principle of "center-start, spiral expansion, and regional fusion." The path starts from the geometric center of the inner additive manufacturing zone and expands outward in a clockwise spiral direction, naturally merging the inner, middle, and outer zones to avoid path breaks and stress abrupt changes between regions. The spiral path of the inner additive manufacturing zone is optimized based on the original 0-degree / 90-degree mesh path, integrating the discrete mesh paths into a continuous spiral line. The spiral radius gradually increases from 0mm at the center to 3.5mm (inner boundary). The spiral spacing (distance between adjacent spiral lines) matches the optimized overlap rate, set at 0.6mm (inner melt pool width 0.6mm × overlap rate 25%, ensuring an overlap of 0.15mm). The spiral path of the intermediate transition additive manufacturing zone expands from a radius of 3.5mm to a radius of 4.5mm (middle and outer boundary), and the path direction starts from 0... The spiral path smoothly transitions from 90 degrees to a cross direction of 45 degrees / -45 degrees, with the spiral spacing adjusted to 0.7 mm (0.7 mm width of the middle melt pool × 35% overlap, 0.245 mm overlap), achieving seamless connection between the paths of different areas. The spiral path of the outer ring additive manufacturing area expands from a radius of 4.5 mm to a radius of 6.0 mm (outer ring boundary), and the path direction maintains a circular spiral along the contour tangent direction. The spiral spacing is set to 0.8 mm (0.8 mm width of the outer ring melt pool × 40% overlap, 0.32 mm overlap), ensuring edge forming accuracy.

[0081] The spiral progression characteristic of the path is achieved through "continuous curvature adjustment." The curvature of the spiral changes dynamically as the radius increases, with the inner circle having the largest curvature (1 / 0.5mm⁻¹) and gradually decreasing outwards. The outer circle boundary curvature matches the contour curvature (1 / 6mm⁻¹), avoiding sudden changes in laser scanning speed and stress concentration caused by path corners. In the example, the starting point of the spiral path is (120mm, 80mm). The first spiral extends along the 0-degree direction, and after rotating 90 degrees, the radius increases by 0.6mm. This cycle continues until the inner circle boundary, completing 6 spirals (corresponding to the original 6 grid paths). After entering the middle region, the spiral radius increases by 0.7mm for every 45-degree rotation, completing 4 spirals (corresponding to the original 4 intersecting paths). After entering the outer circle region, the spiral radius increases by 0.8mm for every 36-degree rotation, completing 5 spirals (corresponding to the original 5 annular paths). The entire path has no breaks or sharp corners, and the laser speed remains continuous and stable during the scanning process.

[0082] The rotation strategy for adjacent layer paths aims to suppress interlayer stress superposition. The rotation angle is set to 30 degrees (based on material thermal stress characteristics and simulation verification, a 30-degree rotation can stagger the interlayer stress directions, offsetting some residual stress). Specifically: the outer ring spiral of the nth layer path rotates clockwise, and the outer ring spiral of the (n+1)th layer path is adjusted to a clockwise deviation of 30 degrees; the inner ring and the spiral paths in the middle region rotate synchronously by 30 degrees, ensuring that the included angle between the paths of adjacent layers is 30 degrees. In the example, the starting direction of the inner spiral path of the 1st layer is 0 degrees, the 2nd layer is 30 degrees, the 3rd layer is 60 degrees, and so on. The staggered distribution of interlayer paths ensures that thermal stress is evenly distributed in three-dimensional space, avoiding deformation caused by accumulation along a single direction.

[0083] The resulting adaptive spiral progressive filling path possesses three core characteristics: first, the path is continuous and uninterrupted, spiraling from the center to the edge, with natural fusion between regions; second, parameters and optimization strategies are precisely matched, with the spiral spacing corresponding to the overlap rate, and the scanning interval dynamically adjusted according to the region; and third, the interlayer path rotates by 30 degrees to suppress the superposition of residual stress. Simulation verification shows that the maximum residual stress of the component formed by this path is ≤450MPa, the deformation is ≤0.1mm, and the edge roughness Ra is ≤3.2μm, fully meeting the forming quality requirements of ultra-high strength steel deep blind hole shell components, and can be directly used for the generation of subsequent processing instructions.

[0084] S205, Generate a processing instruction file that can directly drive the additive manufacturing equipment according to the adaptive spiral progressive filling path, and complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

[0085] Specifically, it can analyze the geometric data and process parameters of the adaptive spiral progressive filling path, including path point coordinates, scanning speed, laser power, powder feeding rate, etc., and generate a set of path point process parameters. The core of this step is to extract quantified data that can be directly used for equipment control from the adaptive spiral progressive filling path, forming a structured set of process parameters to provide accurate input for subsequent instruction conversion. The specific implementation method is as follows: The geometric data analysis of the adaptive spiral progressive filling path focuses on the extraction of path point coordinates. The path consists of continuous three-dimensional coordinate points with a coordinate accuracy of 0.001mm, arranged sequentially according to the scanning order. Each coordinate point contains values ​​in three dimensions: X, Y, and Z (the Z value is the axial height of the current layer, consistent with the variable thickness layer slicing scheme). Taking a certain layer (Z=14.7-15.0mm, slice thickness 0.3mm) as an example, the starting point coordinates of the path are (120.000mm, 80.000mm, 14.700mm), and it expands outward in a clockwise spiral. The spacing between adjacent path points is 0.1mm (to match the motion control accuracy of laser scanning). The coordinates of the 10th path point are (120.098mm, 80.015mm, 14.700mm), the coordinates of the 100th path point are (120.876mm, 80.432mm, 14.700mm), and so on, until the final path point coordinates of the outer boundary are (126.000mm, 80.000mm, 14.990mm). All coordinate points are fitted by B-spline curves to ensure that the path is smooth and continuous.

[0086] The process parameters are analyzed and mapped one-to-one with the geometric data. Each path point is associated with a set of dedicated process parameters to ensure process adaptability for different regions and path segments. The scanning speed parameters are set according to a zoning strategy: 6 mm / s for the inner additive manufacturing zone, 5 mm / s for the intermediate transition additive manufacturing zone, and 4 mm / s for the outer additive manufacturing zone, with a speed error controlled within ±0.2 mm / s to ensure the stability of laser scanning. The laser power parameters are matched in tandem with the scanning speed: 280W for the inner zone, 290W for the middle zone, and 300W for the outer zone, with a power adjustment step of 1W to adapt to the energy requirements of different regions. The powder feeding rate parameters are set according to the cladding layer thickness (consistent with the slice thickness of 0.3 mm): 7 g / min for the inner zone, 7.5 g / min for the middle zone, and 8 g / min for the outer zone, with a powder feeding rate accuracy of ±0.1 g / min to ensure the uniformity of material supply. In addition, auxiliary process parameters were also analyzed, including protective gas flow rate (15L / min, argon), molten pool monitoring threshold (temperature ≥1450℃), and path interval time (inner circle 0s, middle circle 0.2s, outer circle 0.5s).

[0087] The process parameter set for path points is generated using a structured format with a one-to-one correspondence between coordinates and parameters. Each entry includes "path point number, X coordinate, Y coordinate, Z coordinate, scanning speed, laser power, powder feed rate, and auxiliary parameters," arranged in scanning order. The set also includes a parameter index table, marking the parameter switching nodes for different zones (e.g., the 200th path point is the parameter switching point from the inner circle to the middle, and the 400th path point is the switching point from the middle to the outer circle), ensuring the accuracy of subsequent command conversions.

[0088] According to the control command format requirements of additive manufacturing equipment, the set of process parameters of path points is converted into a G-code sequence that the equipment can recognize, and the necessary equipment control commands are inserted to generate the original G-code file; The core of this step is to transform the structured set of process parameters into G-code that the equipment can execute. Through standardized instruction formats and control logic, the process parameters are accurately transmitted. The specific implementation method is as follows: The control command format of additive manufacturing equipment follows the industry standard G-code protocol, supporting motion control (G-code), process control (M-code), and parameter settings (S-code, F-code, etc.). Core command definitions are: G01 for linear interpolation motion (used for scanning path execution), G00 for rapid positioning motion (used for rapid movement during path switching), M03 for laser on, M05 for laser off, M10 for powder feeding on, and M11 for powder feeding off. The S-code sets the laser power (in W), the F-code sets the scanning speed (in mm / min), and the P-code sets auxiliary parameters (such as protective gas flow rate and interval time). The equipment supports command precision of 0.001 mm for motion control, 1 W for power control, and 0.1 mm / min for speed control, perfectly matching the analytical precision of process parameters.

[0089] The conversion of the G-code sequence is performed in the order of the process parameter set of the path points. Each path point corresponds to a G01 linear interpolation instruction, associated with F instructions (scanning speed) and S instructions (laser power). The powder feed rate is embedded through P instructions. During the conversion, the unit of the path point coordinates is converted from mm to the number of pulses that the equipment can recognize (the equipment pulse equivalent is 0.001 mm / pulse, i.e., 1 mm = 1000 pulses). In the example, the starting point coordinates (120.000 mm, 80.000 mm, 14.700 mm) are converted to the number of pulses (120000, 80000, 14700), and the scanning speed of 6 mm / s is converted to F360 mm / min (6 mm / s × 60). Insert M commands and parameter update commands at parameter switching nodes. For example, at the 200th path point (inner circle → middle switch), insert "M05 (laser off), M11 (powder feeding off), P0.2 (interval 0.2s), S290 (power update), F300 (speed update), P7.5 (powder feeding rate update), M03 (laser on), M10 (powder feeding on)" to ensure stable equipment transition during parameter switching.

[0090] Essential equipment control commands include initialization commands, zone switching commands, and termination commands. Initialization commands are located at the beginning of the G-code and include equipment zeroing (G28), coordinate system setting (G90 absolute coordinates), protective gas activation (M08, P15), and laser preheating (S100, M03 delay 2s). Zone switching commands are as described above for intermediate nodes, ensuring smooth process transitions between zones. Termination commands are located at the end of the G-code and include laser shutdown (M05), powder feeding shutdown (M11), protective gas delay shutdown (M09 delay 5s), and equipment reset (G28).

[0091] The generation of the original G-code file involves sequentially integrating all instructions and saving them in standard text format (.gcode). The file includes header information such as the number of instruction lines, parameter statistics, and generation time. After file generation, the instruction format is verified to ensure there are no syntax errors, parameter matching, and that the file can be directly read by the device.

[0092] Post-processing optimization is performed on the original G-code file, including motion smoothing to eliminate corner impacts, speed look-ahead control to ensure speed continuity, and collision interference checks, to generate an optimized machining instruction file that can directly drive the equipment. The core of this step is to improve the execution stability and shaping quality of G-code through post-processing optimization, eliminating potential motion shocks, speed fluctuations, and collision risks. The specific implementation method is as follows: Motion smoothing addresses the impact issues at path corners by employing a B-spline interpolation algorithm to smoothly fit the corner path, transforming sharp corners into continuous curve transitions. The transition radius is dynamically adjusted based on the corner angle (0.2mm when the angle is ≥90°, and 0.1mm when the angle is <90°).

[0093] Speed ​​look-ahead control aims to ensure the continuity of scanning speed and avoid energy input instability caused by sudden speed changes. The speed parameters of the next 5 path points are read in advance by the look-ahead algorithm, and the speed transition curve is calculated. When the speed difference between adjacent path points is >1mm / s, a speed gradient segment is inserted with a gradient length of 1mm (corresponding to 10 path points) and a speed change rate ≤0.1mm / s².

[0094] Collision interference checks cover the equipment's movement range, interference between components and tool heads, and interference between paths and fixtures. A spatial bounding box algorithm is used to construct the collision detection model. Equipment movement range limitations: X-axis 0-300mm, Y-axis 0-200mm, Z-axis 0-400mm. If path point coordinates exceed the range, they are marked as invalid and corrected. Component-tool head interference checks: The minimum avoidance distance for the tool head is 5mm. The distance between the path point and the center of the tool head is calculated. If it is <5mm, the Z-axis height is adjusted (maximum adjustment amount 0.5mm). Path-fixture interference checks: Preset fixture position coordinates (e.g., (110.000mm, 70.000mm, 0.000mm)). The shortest distance between the path point and the fixture is calculated. If it is <10mm, the path is optimized for detour. In the example, a path point (121.000mm, 75.000mm, 14.700mm) is 8mm away from the fixture. By adjusting the coordinates of the path point to (121.500mm, 75.500mm, 14.700mm), interference with the fixture can be avoided.

[0095] The optimized machining instruction file integrates the above optimization results, replaces corner and speed instructions in the original G-code, corrects interference path point coordinates, and adds optimization markers (such as "smoothing," "speed look-ahead," and "interference avoidance"). The file also includes an optimization report, recording the number of optimizations (32 smoothing adjustments, 18 speed look-ahead adjustments, and 2 interference avoidances) and a comparison of key indicators before and after optimization (corner acceleration from 500 to 200 mm / s², speed fluctuation from ±0.5 to ±0.1 mm / s, and interference risk from 3 to 0). The optimized G-code file has been verified by equipment simulation software to be smooth in motion, continuous in speed, and free from collision interference, and can directly drive additive manufacturing equipment.

[0096] The optimized processing instruction file is loaded into the additive manufacturing equipment control system. The instructions are read layer by layer according to the variable thickness layer slicing scheme. The additive tool head and the powder feeding system work together to melt and deposit ultra-high strength steel powder material in sequence, and finally complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

[0097] The core of this step is to execute the optimized processing instructions. Through the coordinated work of various systems in the equipment, layer-by-layer cladding is achieved to finally obtain a deep blind hole head component that meets the quality requirements. The specific implementation method is as follows: The loading of optimized processing instruction files is completed via the device control system's USB interface or network transmission. The control system supports direct reading of .gcode format files. After loading, it automatically parses the file header information (number of instruction lines, parameter statistics) and generates a processing progress preview (total number of layers, current layer, estimated processing time). In the example, with a total of 117 layers, after loading, the control system displays "Total number of instruction lines: 15682, total number of layers: 117, current layer: 1, estimated processing time: 4.5 hours". After administrator confirmation, the processing flow is started.

[0098] The control system reads and executes instructions layer by layer according to the layering sequence of the variable thickness layered slicing scheme. The processing flow for each layer is as follows: layer preparation (Z-axis rises to the current layer height, error ≤ 0.01mm) → instruction execution (controlling the tool head movement according to the G-code sequence, laser and powder feeding in coordination) → layer detection (melt pool temperature monitoring, forming thickness detection) → layer transition (if it is a variable thickness layer, adjust the Z-axis height and process parameters of the next layer). Taking the first layer (Z=0-0.3mm, slicing thickness 0.3mm) as an example, the Z-axis rises to 0.000mm, the G-code instruction is executed, the tool head moves along the spiral path, the laser is turned on synchronously (power 280W), the powder feeding system supplies ultra-high strength steel powder (powder feeding rate 7g / min), argon protective gas is continuously supplied (15L / min), the melt pool temperature is stabilized at 1480-1500℃, the forming thickness is detected as 0.30±0.02mm, and after meeting the requirements, it transitions to the second layer.

[0099] The additive manufacturing tool head and powder feeding system work in tandem, precisely scheduled by the control system. The tool head's movement speed is matched in real time with the laser power and powder feeding rate, with an acceleration ≤200mm / s² to avoid molten pool fluctuations. The powder feeding system uses a coaxial feeding method, with the powder nozzle and laser spot aligned coaxially. The nozzle is 5mm above the molten pool surface, ensuring a powder utilization rate ≥90%. The laser focus diameter is stabilized at 0.8mm, with the focus position coinciding with the center of the molten pool, and the energy density is uniform (10^6W / cm²), ensuring complete powder melting. During processing, the molten pool monitoring system collects temperature data in real time. If the temperature >1550℃ (exceeding the threshold), the control system automatically reduces the laser power (by 5W each time); if the temperature <1400℃ (below the threshold), the laser power is increased (by 5W each time) to ensure molten pool stability.

[0100] The manufacturing process proceeded layer by layer until all layers were completed. After the final layer (layer 117, Z=79.9-80.0mm, slice thickness 0.1mm) was finished, the following commands were executed: laser shut-off, powder feeding shut-off, protective gas shut-off after a 5-second delay, tool head reset, and Z-axis reduction to 0mm. After the component cooled to room temperature (natural cooling, 2 hours), the fixture was removed, yielding the finished head of the deep blind hole shell component. Finished product quality inspection results: component dimensional accuracy ±0.1mm, edge roughness Ra=2.8μm, density ≥99.6%, maximum residual stress 420MPa, deformation 0.08mm, fully meeting the forming requirements of ultra-high strength steel deep blind hole shell components, thus completing the entire layer-by-layer forming manufacturing process.

[0101] Another embodiment of the present invention provides an additive path planning system for the head of an ultra-high strength steel deep blind hole shell component, see [link to relevant documentation]. Figure 3 The system may include: The acquisition module 301 is used to acquire the three-dimensional geometric model of the component to be formed, and to identify and extract the inner cavity surface and edge contour features of the head of the deep blind hole shell component. The generation module 302 is used to generate a variable thickness layered slicing scheme that adaptively matches the inner cavity surface based on the inner cavity surface and edge contour features, combined with the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel material. The partitioning module 303 is used to divide the inner cavity cross section into an outer ring additive manufacturing area, an intermediate transition additive manufacturing area and an inner ring additive manufacturing area based on the variable thickness layered slicing scheme, using a multi-level partitioning strategy within each layer, and designing the main path direction and scanning angle of each area respectively. The planning module 304 is used to dynamically plan the scanning sequence and overlap parameters of each layer based on the main path direction and scanning angle, combined with the thermal sequential coupling algorithm, to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation. Manufacturing module 305 is used to generate a processing instruction file that can directly drive additive manufacturing equipment according to the adaptive spiral progressive filling path, and to complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

[0102] The above description, based on the embodiments shown in the figures, details the structure, features, and effects of the present invention. The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to have equivalent changes, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A method for additive path planning at the head of an ultra-high strength steel deep blind hole shell component, characterized in that, The method includes: Obtain the three-dimensional geometric model of the component to be formed, and identify and extract the inner cavity surface and edge contour features of the head of the deep blind hole shell component; Based on the features of the inner cavity surface and edge contour, and combined with the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel, a variable thickness layered slicing scheme that adaptively matches the inner cavity surface is generated. Based on the variable thickness layered slicing scheme, a multi-level partitioning strategy is adopted in each layer to divide the inner cavity cross section into an outer ring additive manufacturing area, an intermediate transition additive manufacturing area and an inner ring additive manufacturing area, and the main path direction and scanning angle of each area are designed respectively. Based on the main path direction and scanning angle, the scanning sequence and overlap parameters of each layer are dynamically planned using the thermal sequential coupling algorithm to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation. Based on the adaptive spiral progressive filling path, a processing instruction file is generated that can directly drive the additive manufacturing equipment, and the layer-by-layer forming manufacturing of the head of the deep blind hole shell component is completed.

2. The method according to claim 1, characterized in that, The process of obtaining the three-dimensional geometric model of the component to be formed, and identifying and extracting the inner cavity surface and edge contour features of the head of the deep blind hole shell component, includes: Import the 3D CAD design model file of the component to be formed, perform geometric repair and topology optimization on the model to ensure that the model is a watertight and non-self-intersecting entity, and generate a standard 3D geometric model that can be processed. Feature recognition and analysis are performed on the standard three-dimensional geometric model. The region growing algorithm is used to automatically identify the head region of the deep blind hole shell component, separate the internal cavity structure entity from the outer shell, and generate the head internal cavity structure model. Based on the head cavity structure model, the principal curvature distribution and normal vector field of the cavity surface are extracted by the surface curvature analysis algorithm, and the boundary contour of the cavity opening is located by the edge detection algorithm to generate the cavity geometric feature parameter set; By integrating the set of internal cavity geometric feature parameters, a mathematical representation is constructed to describe the geometric shape of the internal cavity surface and the topological relationship of the boundary contour, ultimately generating accurate internal cavity surface and edge contour features.

3. The method according to claim 2, characterized in that, The process involves generating a variable-thickness layered slicing scheme that adaptively matches the inner cavity surface and edge contour features, combined with the thermal accumulation characteristics and shrinkage deformation laws of ultra-high-strength steel. This includes: Based on the precise internal cavity surface and edge contour features, the rate of change of the internal cavity cross-sectional area and perimeter at different heights along the component axis is calculated, and the internal cavity geometric change curve is generated. Query the database of ultra-high strength steel material properties to obtain the material's thermal expansion coefficient, phase transformation temperature range, high temperature yield strength and cooling shrinkage rate parameters, and generate a set of material thermodynamic property parameters. By inputting the internal cavity geometric change curve and the material thermodynamic property parameter set into the thermodynamic simulation model, the temperature field distribution and thermal stress concentration area under different slice thicknesses are predicted, and thermo-mechanical coupling simulation results are generated. Analyzing the results of thermo-mechanical coupling simulation, a thinner layer thickness is adopted in the stress concentration area to refine the control, while a thicker layer is adopted in the geometrically smooth area to improve efficiency. Finally, a variable thickness layer slicing scheme that adaptively matches the inner cavity surface is generated.

4. The method according to claim 3, characterized in that, The variable thickness layered slicing scheme employs a multi-level partitioning strategy within each layer, dividing the inner cavity cross-section into an outer ring additive manufacturing zone, a middle transition additive manufacturing zone, and an inner ring additive manufacturing zone. The main path direction and scanning angle for each zone are designed accordingly, including: The two-dimensional inner cavity cross-sectional profile of the current layer is obtained according to the variable thickness layer slicing scheme. Based on the curvature distribution of the cross-sectional profile and the distance to the boundary, the equidistant offset algorithm is used to generate concentric closed loops with different widths at the distance from the profile boundary. Based on the concentric closed loops, the functional areas are divided into the outermost and second outermost rings. The area between the outermost and second outermost rings is defined as the outer ring additive manufacturing area, the area between the second outermost and third outermost rings is defined as the intermediate transition additive manufacturing area, and the remaining central area is defined as the inner ring additive manufacturing area, thus generating a multi-level partition map. For the outer ring additive manufacturing area, the main path is designed to perform a circular scan along the contour tangent direction to enhance the edge bonding strength. For the middle transition additive manufacturing area, the main path is designed to perform a cross scan along the direction at a 45-degree angle to the contour to optimize stress distribution. For the inner ring additive manufacturing area, the main path is designed to perform a sparse grid-like scan to improve forming efficiency, thus generating a partitioned main path strategy. By integrating multi-level partition maps and partition main path strategies, the scanning path density, scanning speed, and scanning angle parameters are specified for each partition, and finally, the main path direction and scanning angle design scheme of each partition are generated.

5. The method according to claim 4, characterized in that, The process of dynamically planning the scanning sequence and overlap parameters for each layer based on the main path direction and scanning angle, combined with a thermal sequential coupling algorithm, to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation includes: Based on the design scheme of the main path direction and scanning angle of each partition, the path scanning order of different areas in the same layer is initially planned according to the order from the inner circle additive manufacturing area to the outer circle additive manufacturing area, and a preliminary scanning sequence is generated. The initial scanning sequence is input into the thermo-mechanical sequential coupling simulation algorithm. The algorithm simulates the laser scanning process and dynamically calculates the evolution of the transient temperature field and stress field caused by the deposition of material along each path, generating thermo-mechanical evolution data. By analyzing thermal evolution data, we can identify scanning sequences and path overlaps that may lead to thermal stress concentration or deformation, dynamically adjust the scanning interval and overlap rate of adjacent paths, and generate optimized overlap parameters and intermittent strategies. Based on optimized overlap parameters and intermittent strategies, the scanning path is replanned so that the filling path starts from the central region and expands outward in a spiral progressive manner, while ensuring that the paths of adjacent layers are rotated at a certain angle, ultimately generating an adaptive spiral progressive filling path that can suppress residual stress and deformation.

6. The method according to claim 5, characterized in that, The step of generating a processing instruction file that can directly drive additive manufacturing equipment based on the adaptive spiral progressive filling path, and completing the layer-by-layer forming manufacturing of the head of the deep blind hole shell component, includes: The geometric data and process parameters of the adaptive spiral progressive filling path are analyzed, including path point coordinates, scanning speed, laser power, powder feeding rate, etc., to generate a set of process parameters for the path points; According to the control command format requirements of additive manufacturing equipment, the set of process parameters of path points is converted into a G-code sequence that the equipment can recognize, and the necessary equipment control commands are inserted to generate the original G-code file; Post-processing optimization is performed on the original G-code file, including motion smoothing to eliminate corner impacts, speed look-ahead control to ensure speed continuity, and collision interference checks, to generate an optimized machining instruction file that can directly drive the equipment. The optimized processing instruction file is loaded into the additive manufacturing equipment control system. The instructions are read layer by layer according to the variable thickness layer slicing scheme. The additive tool head and the powder feeding system work together to melt and deposit ultra-high strength steel powder material in sequence, and finally complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

7. An additive path planning system for the head of an ultra-high strength steel deep blind hole shell component, characterized in that, The system includes: The acquisition module is used to acquire the three-dimensional geometric model of the component to be formed, and to identify and extract the inner cavity surface and edge contour features of the head of the deep blind hole shell component. The generation module is used to generate a variable thickness layered slicing scheme that adaptively matches the inner cavity surface based on the inner cavity surface and edge contour features, combined with the thermal accumulation characteristics and shrinkage deformation law of ultra-high strength steel material. The partitioning module is used to divide the inner cavity cross section into an outer ring additive manufacturing area, an intermediate transition additive manufacturing area and an inner ring additive manufacturing area based on the variable thickness layered slicing scheme, using a multi-level partitioning strategy within each layer, and designing the main path direction and scanning angle of each area respectively. The planning module is used to dynamically plan the scanning sequence and overlap parameters of each layer based on the main path direction and scanning angle, combined with the thermal sequential coupling algorithm, to generate an adaptive spiral progressive filling path that can suppress residual stress and deformation. The manufacturing module is used to generate a processing instruction file that can directly drive the additive manufacturing equipment according to the adaptive spiral progressive filling path, and to complete the layer-by-layer forming manufacturing of the head of the deep blind hole shell component.

8. The system according to claim 7, characterized in that, The acquisition module is specifically used for: Import the 3D CAD design model file of the component to be formed, perform geometric repair and topology optimization on the model to ensure that the model is a watertight and non-self-intersecting entity, and generate a standard 3D geometric model that can be processed. Feature recognition and analysis are performed on the standard three-dimensional geometric model. The region growing algorithm is used to automatically identify the head region of the deep blind hole shell component, separate the internal cavity structure entity from the outer shell, and generate the head internal cavity structure model. Based on the head cavity structure model, the principal curvature distribution and normal vector field of the cavity surface are extracted by the surface curvature analysis algorithm, and the boundary contour of the cavity opening is located by the edge detection algorithm to generate the cavity geometric feature parameter set; By integrating the set of internal cavity geometric feature parameters, a mathematical representation is constructed to describe the geometric shape of the internal cavity surface and the topological relationship of the boundary contour, ultimately generating accurate internal cavity surface and edge contour features.

9. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 1-6 when it is run.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1-6.

Images