Method for additive manufacturing of light-weight metal mold blank of automobile by using composite material

Abstract

The invention discloses a method for additive manufacturing of an automobile lightweight metal mold blank through a composite material, and relates to the technical field of additive manufacturing and automobile mold manufacturing, and the method comprises the following steps that a mold three-dimensional discretization partition model is constructed, and a mold is divided into a wear-resistant shell domain and a tough matrix domain; mechanical field characteristics of a tough matrix domain are extracted, a principal stress driven non-uniform lattice structure is generated, lattice rod pieces grow in the dominant principal stress direction, and the distribution density is mapped according to stress numerical values; and establishing a process heat accumulation model of the lattice nodes, and calculating a heat accumulation index. In the manufacturing process, heterogeneous material partition deposition and interface topology interlocking scanning are executed, laser power is dynamically modulated according to a heat accumulation index, and cooling compensation of an extreme heat node is implemented. According to the method, cooperation of high wear resistance of the surface of the die and an internal light-weight tough base body is achieved, the force transfer path is optimized, and the thermal defect in the complex lattice additive manufacturing process is effectively restrained.

Description

Technical Field

[0001] This application relates to the fields of additive manufacturing and automotive mold manufacturing technology, and in particular to a method for additive manufacturing of lightweight metal mold blanks for automobiles using composite materials. Background Technology

[0002] As a core process equipment in car body manufacturing, automotive stamping dies typically have enormous size and weight. Traditional manufacturing processes mainly employ integral casting or forging of blanks combined with computer numerical control (CNC) subtractive processing. This method not only has a high material removal rate and long processing cycle, but also results in high energy consumption and operational difficulties during the transportation, installation, and debugging of the manufactured solid dies due to their excessive weight. With the automotive industry's increasing demands for production efficiency and energy conservation and emission reduction, lightweight die design has become a key focus of the industry.

[0003] While metal additive manufacturing technology offers a new path for mold structure optimization and lightweight manufacturing, it still faces dual technical bottlenecks in practical applications: matching material properties and controlling forming processes. In terms of materials, existing mold additive manufacturing often uses a single homogeneous material, making it difficult to simultaneously achieve the high hardness and wear resistance required for the mold's working surface and the high strength, toughness, and impact resistance required for the matrix structure. If the entire mold is made of high-hardness maraging steel or tool steel, although the wear resistance requirement is met, it leads to insufficient impact toughness in the matrix, making it prone to brittle fracture under alternating stamping loads. Furthermore, the high material cost limits its large-scale application. Conversely, if high-toughness structural steel is used, the wear life requirements of the formed surface cannot be met.

[0004] In terms of structure and process, lattice or crystal structures are often introduced inside the mold to achieve lightweighting. However, conventional filling structures often employ uniform distribution or simple gradient distribution based solely on geometry, lacking precise matching with the complex stress state under actual stamping service conditions. This can easily lead to yielding or fatigue failure in local high-stress areas due to insufficient load-bearing capacity. More critically, when laser powder bed melting is performed on such complex non-uniform lattice structures, the drastic changes in scanning cross-sectional dimensions and local heat dissipation conditions can easily cause heat accumulation effects at lattice nodes and small rods. Existing commercial slicing and control software typically employs constant laser power or simple contour filling strategies, which are difficult to adapt to rapid changes in microscopic geometry. This often results in excess energy at nodes causing remelting collapse, severe powder adhesion, or warping deformation due to excessive thermal stress. This limits the manufacturing precision and service reliability of high-performance composite lightweight molds. Summary of the Invention

[0005] The purpose of this invention is to provide a method for additive manufacturing of lightweight metal mold blanks for automobiles using composite materials, which at least solves one of the technical problems in the additive manufacturing process of metal molds: the difficulty for a single homogeneous material to simultaneously meet the performance requirements of high wear resistance on the mold surface and high strength and toughness of the internal structure; the lack of matching between conventional lightweight lattice structures and the actual service stress field of the mold, resulting in low structural load-bearing efficiency; and the forming defects and precision loss caused by local heat accumulation effects when manufacturing complex non-uniform lattice structures.

[0006] This invention provides a method for additive manufacturing of lightweight metal mold blanks for automobiles using composite materials. The method is executed by a manufacturing system that includes a digital processing terminal and a multi-material additive manufacturing execution unit. It aims to solve the contradiction between wear resistance and lightweighting in mold manufacturing, as well as the problem of thermal accumulation deformation in additive manufacturing of complex lattice structures.

[0007] The method of the present invention mainly includes the following:

[0008] In the data processing stage, a three-dimensional discretized partitioned model of the mold is constructed. Using the functional surfaces of the mold as a reference, the geometric space of the mold is divided into an outer wear-resistant shell domain and an inner tough matrix domain, and the heterogeneous material transition interface between the two is defined. Specifically, the initial geometric data of the mold is acquired and manifold preprocessing is performed to identify the functional surfaces on the outer boundary surface. Using the functional surfaces as the zero-level set surface, a signed distance field algorithm or a voxelized bias algorithm is used to calculate isosurfaces within the model, generating a set of points that satisfy the distance constraint conditions as the wear-resistant shell domain. Subsequently, through Boolean difference operations, the remaining portion of the total mold volume domain after deducting the wear-resistant shell domain is defined as the tough matrix domain.

[0009] To meet the process requirements of subsequent subtractive manufacturing, a machining compensation layer is generated for the functional surface. The discretized mesh of the functional surface is reconstructed using normal vectors, and the vertex normal vectors are calculated using an area-weighted average algorithm. Based on preset machining allowance parameters, an offset coordinate transformation along the normal is performed on the vertices of the functional surface to generate an initial offset mesh surface. Through voxel-based self-intersection culling and contour stitching, a closed solid compensation layer is formed, which is then merged with the wear-resistant shell domain as the deposition boundary of the wear-resistant material.

[0010] To optimize the internal structure of the mold, the eigenvectors of the mechanical field in the tough matrix domain are extracted, and a non-uniform lattice structure driven by principal stress is generated. After mesh discretization and interface topology sharing processing of the wear-resistant shell domain and the tough matrix domain, the maximum reaction force of stamping is mapped to the functional surface to construct a static boundary value problem. The Cauchy stress tensor field in the tough matrix domain is solved and eigenvalue decomposition is performed. The principal stress with the largest absolute value is selected as the dominant principal stress, and its corresponding spatial characteristic vector direction is extracted. When constructing the lattice, the growth axis of the lattice rods is controlled to be parallel to the spatial characteristic vector direction, and a quantitative mapping relationship between the dominant principal stress value and the cross-sectional radius of the lattice rods is established to match the lattice distribution density with the local stress level. In addition, a curved orthogonal mesh is generated by calculating stress streamlines, and continuous torsion of the lattice attitude is achieved by using node coordinate transformation to ensure that the principal stiffness direction of the lattice element always follows the stress transmission path.

[0011] To ensure the integrity of the structure, it is also necessary to construct a transition connection between the lattice and the shell: identify the suspended end nodes in the skeleton of the lattice solid model that are close to the transition interface of the heterogeneous materials, and use these nodes as the starting point to execute a ray projection algorithm along the direction of the spatial feature vector, calculate the intersection of the ray and the transition interface and generate connecting rods, and finally perform a Boolean union operation to merge the lattice solid model and the wear-resistant shell domain into a single closed solid model.

[0012] During the process planning phase, a process thermal accumulation model for lattice nodes is established. Node locations within the solid lattice model are identified, and a local bounding sphere for thermal analysis is constructed. Taking into account the number of connecting members passing through the bounding sphere, their cross-sectional area, and the angle between the member axes and the construction direction, a local geometric heat dissipation factor characterizing the node's heat dissipation capacity is calculated. Simultaneously, based on preset laser scanning process parameters, the equivalent volumetric energy input of the lattice nodes during the forming process is calculated. Based on the principle of energy conservation, a thermal accumulation index is constructed. This index is positively correlated with the equivalent volumetric energy input and negatively correlated with the local geometric heat dissipation factor, thereby quantifying the risk of local thermal retention at each node.

[0013] During the additive manufacturing execution phase, the deposition of heterogeneous materials is performed in zones with dynamic control based on the thermal accumulation index. High-hardness metallic materials are deposited in the wear-resistant shell domain and the solid compensation layer, while high-toughness structural steel materials are deposited in the strong and tough matrix domain. A metallurgical bonding transition zone is constructed at the interface between the heterogeneous materials. By controlling the powder spreading device, a mixed powder band with gradient composition is formed, and an interface interlacing scanning strategy is adopted to control the laser beam to extend and scan along the topological interlocking path, thereby enhancing the interface bonding strength.

[0014] Simultaneously, variable energy power modulation and cooling compensation for extreme hot nodes are implemented. Discrete scan vectors are resampled into a sequence of control points with equal time steps to construct a spatiotemporal mapping between the scan path and the thermal field. When depositing a tough substrate domain, hardware control commands are generated based on the thermal accumulation index, and a piecewise smooth modulation strategy is used to dynamically adjust the heat source input power: when the thermal accumulation index is higher than a preset threshold, the laser power is reduced through negative feedback, and time axis translation compensation is performed for the physical hysteresis of the laser generator and the servo hysteresis of the galvanometer system. For lattice nodes where the thermal accumulation index reaches a critical value, the critical thermal relaxation time required for natural cooling to a safe state is calculated based on the transient heat conduction equation. A farthest-point skip scan strategy is adopted, interrupting the current region scan and jumping to the far-end partition when the cooling time does not meet the requirements, using spatial jump to achieve temporal cooling.

[0015] The above solution achieves the following beneficial technical effects:

[0016] This application achieves functional decoupling between the high wear resistance of the mold surface and the high strength and toughness of the internal structure by constructing a heterogeneous material partition between a wear-resistant shell and a tough matrix. While ensuring that the working surface of the mold meets the hardness requirements of stamping conditions, the overall weight of the mold is effectively reduced by utilizing the internal high-toughness structural steel to absorb impact energy. In addition, by constructing a metallurgical bonding transition zone with a composition gradient at the heterogeneous material interface and implementing topological interlocking scanning, the risks of interface bonding fragility and delamination common in traditional insert or composite structures are eliminated, thereby improving the fatigue life of the mold under alternating loads.

[0017] This application employs a principal stress-driven non-uniform lattice generation technique to achieve optimal material distribution along the mechanical transmission path. By extracting the stress field characteristics within the strong and tough matrix domain, the growth axis of the lattice rods is aligned with the direction of the local principal stress vector, and the cross-sectional dimensions of the rods are dynamically adjusted based on the stress values. This removes redundant material in low-stress areas while enhancing the stiffness of the main load-bearing path. This design significantly reduces the consumption of mold blank material and printing time while ensuring sufficient deformation resistance and structural stability of the mold under extreme stamping loads.

[0018] This application establishes a dynamic manufacturing control strategy based on a lattice node process thermal accumulation model, effectively solving the problem of overheating deformation in additive manufacturing of complex lattice structures. By comprehensively calculating local geometric heat dissipation conditions and energy input, the real-time thermal retention risk of nodes is quantified, and the laser power is negatively fed back for smooth modulation and cooling compensation is implemented for extreme hot nodes accordingly. This method avoids energy accumulation and remelting collapse at small rods and nodes, ensuring the forming accuracy and density of the lattice structure, thereby improving the overall manufacturing quality and yield of composite material molds. Attached Figure Description

[0019] Figure 1This is a flowchart of a method for additive manufacturing of lightweight metal mold blanks for automobiles using composite materials, provided in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the construction process of the three-dimensional discretization partition model of the mold provided in the embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the process for generating the functional surface machining compensation layer according to an embodiment of the present invention.

[0022] Figure 4 This is a schematic diagram of the process for extracting the feature vector of the mechanical field in a strong and tough matrix domain provided in an embodiment of the present invention.

[0023] Figure 5 This is a schematic diagram of the principal stress-driven non-uniform lattice structure generation process provided in an embodiment of the present invention.

[0024] Figure 6 This is a schematic diagram of the process for establishing a thermal accumulation model for lattice node technology provided in an embodiment of the present invention.

[0025] Figure 7 This is a schematic diagram of the partitioned switching deposition process of heterogeneous materials provided in an embodiment of the present invention.

[0026] Figure 8 This is a schematic diagram of the variable energy power modulation process based on the thermal accumulation index provided in an embodiment of the present invention.

[0027] Figure 9 This is a schematic diagram of the extreme hot node cooling compensation process provided in an embodiment of the present invention. Detailed Implementation

[0028] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] See attached document Figure 1 The method in this embodiment is executed by a manufacturing system comprising a digital processing terminal and a multi-material additive manufacturing execution unit. The digital processing terminal is responsible for data processing and path planning, while the multi-material additive manufacturing execution unit is responsible for performing physical deposition.

[0030] This invention provides a method for additive manufacturing of lightweight metal mold blanks for automobiles using composite materials, such as... Figure 1 As shown, the method includes the following steps:

[0031] Step S100: Construct a three-dimensional discretized partitioned model of the mold. The initial three-dimensional geometric model of the mold is imported into the digital processing terminal, and the functional surfaces of the mold are identified. Based on the preset normal distance extending inward from the functional surfaces, the geometric solid space of the mold is divided into a wear-resistant shell domain and a tough matrix domain. The wear-resistant shell domain envelops the outer layer of the mold, and the tough matrix domain fills the interior of the mold; the two domains do not overlap spatially and are seamlessly connected.

[0032] Step S200: Generate a machining compensation layer for the functional surface. The digital processing terminal establishes a solid compensation layer of preset thickness along the outward direction of the normal to the functional surface of the mold. This compensation layer merges with the wear-resistant shell domain, serving as the boundary for subsequent wear-resistant material deposition in additive manufacturing, to meet the allowance requirements of subsequent subtractive finishing.

[0033] Step S300: Extract the mechanical field characteristic vector of the toughened matrix domain. The digital processing terminal applies the mechanical boundary conditions of the stamping process to the toughened matrix domain and performs finite element numerical calculations. Extract the dominant principal stress values ​​and their corresponding spatial characteristic vector directions of each discrete element within the toughened matrix domain.

[0034] Step S400: Generate a principal stress-driven non-uniform lattice structure. The digital processing terminal constructs a truss-like lattice structure within a strong and tough matrix domain. The growth axes of the lattice members are controlled to be parallel to the first principal stress characteristic vector extracted in step S300, and the lattice distribution density is mapped according to the magnitude of the local principal stress modulus to generate a non-uniform anisotropic support structure.

[0035] Step S500: Establish a process heat accumulation model for the lattice nodes. The digital processing terminal identifies the node locations where the bars converge in the lattice structure generated in step S400. The number of bars converging at each node, the cross-sectional area of ​​the bars, and the spatial angle parameters of the bars are obtained. The heat accumulation index of each node is calculated; this index characterizes the heat dissipation resistance at that location during manufacturing.

[0036] Step S600: Perform partitioned switching deposition of heterogeneous materials. The multi-material additive manufacturing execution unit receives layered slicing data. When the deposition path is located within the wear-resistant shell domain and the machining compensation layer, the feeding system delivers high-hardness metal material; when the deposition path is located within the tough matrix domain, the feeding system switches to delivering high-toughness structural steel material.

[0037] Step S700: Implement variable energy power modulation based on the thermal accumulation index. During the deposition process in the tough matrix region, the control system reads the thermal accumulation index corresponding to the current deposition path. When the deposition head reaches the lattice node region and the thermal accumulation index is higher than a preset threshold, the system reduces the heat source input power; when the deposition head reaches the lattice rod region, the system restores the reference heat source input power.

[0038] Step S800: Perform cooling compensation for extreme hot nodes. For lattice nodes where the thermal accumulation index reaches a critical value, the control system pauses the heat source input for a preset duration and maintains the protective gas supply after completing the material deposition at that node, and then performs the deposition of subsequent paths after the node region has cooled down.

[0039] The following section will elaborate on the specific implementation principles and technical details of each step of the present invention in conjunction with the above steps S100 to S800.

[0040] See attached document Figure 2 In step S100, the digital processing terminal constructs a three-dimensional discretized partitioned model of the mold, transforming the geometric model of the mold made of a single material into a volume domain model that includes the definition of heterogeneous materials. This step specifically includes the following sub-steps:

[0041] Step S101: Acquire the initial geometric data of the mold and perform manifold preprocessing. The digital processing terminal reads the initial three-dimensional solid model of the mold. The initial 3D solid model's data format includes, but is not limited to, STEP, IGES, Parasolid, or high-precision STL formats. After importing the model, the processing terminal performs a geometric topology check, detects and closes non-manifold edges in the model using a mesh repair algorithm, removes isolated vertices and overlapping faces, and unifies the direction of all face normals to point outwards from the solid, ensuring the initial 3D solid model... It is a closed manifold entity with a unique internal and external definition.

[0042] Step S102: Extract the set of functional surfaces of the mold. The digital processing terminal identifies the initial 3D solid model. outer boundary surface And select the functional surfaces of the mold from them. Mold functional surfaces This refers to the working surface of the mold that directly contacts the stamped sheet metal during actual service. The specific selection method is as follows: calculate the average curvature and normal change rate of each grid vertex on the outer boundary surface, and identify regions with continuous curvature changes and no abrupt changes in the normal vector as candidate surfaces; or read the preset color attributes (Color Map) or layer index from the CAD model as the identification basis. Mathematically, the functional surface of the mold... Defined as outer boundary surface A proper subset, namely .

[0043] Step S103: Construct the volume envelope of the wear-resistant shell domain. Based on the functional surface extracted in step S102. The digital processing terminal defines the wear-resistant zone on the mold surface. This zone is constructed as a three-dimensional volume domain with a specific physical thickness. The preset process thickness of the wear-resistant layer is set to... The digital processing terminal employs either the Signed Distance Field (SDF) algorithm or the Voxel-based Offsetting algorithm for computation. Specifically, it uses functional surfaces... For the zero-level set surface, calculate the isosurfaces towards the interior of the model. Wear-resistant shell domain. The set of points is defined as satisfying the distance constraint, and the calculation formula is as follows:

[0044] ;

[0045] In the formula, Represents a vector of coordinate points in model space; This represents the total volume domain of the mold. Point To the set of surfaces The minimum Euclidean distance; This is the preset wear-resistant layer thickness constant.

[0046] To address sharp features in the mold geometry, chamfered transitions or polyhedral trimming are employed when generating the volume envelope to avoid self-intersecting interference. This calculation generates a solid shell structure that closely adheres to the working surface of the mold; this region is designated as the first material property region.

[0047] Step S104: Construct the solid filling of the tough matrix domain. After determining the wear-resistant shell domain... Subsequently, the digital processing terminal defines the internal support structure region of the mold, namely the strong and tough matrix domain, through Boolean difference operations. Strong matrix domain Defined as the total volume domain of the mold Excluding wear-resistant housing area The remaining volume is calculated using the following formula:

[0048] ;

[0049] Or it can be expressed as:

[0050] ;

[0051] The above Boolean operations ensured the wear-resistant housing domain. With tough matrix domain They do not overlap in space, and their union completely covers the original mold model. This strong matrix domain It is marked as the second material property region.

[0052] Step S105: Define the heterogeneous material transition interface. The digital processing terminal extracts the wear-resistant housing domain. With tough matrix domain Public boundaries, generating transition interfaces The geometry of this interface is defined as follows:

[0053] ;

[0054] In the formula, Indicates the boundary of the wear-resistant shell region; This represents the boundary of the tough matrix domain.

[0055] The processing terminal will use this transition interface Extract the data into independent triangular mesh surfaces or parametric surfaces, and establish a local coordinate system. This interface data serves as the geometric reference surface for applying interface connection mechanical conditions and planning heterogeneous material switching paths in subsequent steps.

[0056] See attached document Figure 3 In step S200, the digital processing terminal generates a solid compensation layer for subsequent subtractive finishing based on the functional surface data of the mold. This step specifically includes the following sub-steps:

[0057] Step S201: Reconstruct the normal vectors of the discretized mesh for the functional surface. The digital processing terminal reads the functional surface of the mold. The processing terminal will have a functional surface. Perform triangular mesh subdivision processing to generate a set of vertices. Dough sheet collection A triangular mesh surface. For any vertex in the vertex set... The system calculates its unit normal vector. To ensure the accuracy of the normal vector direction, the system uses an area-weighted average algorithm to calculate the vertex normal vector. Specifically, this involves identifying all shared vertices. The system calculates the normal vectors of each adjacent triangular face, and then performs a weighted summation using the area of ​​each face as a weighting factor. The summation vector is then normalized. Simultaneously, the system performs a normal vector consistency check, traversing all faces using topological adjacency relationships to ensure that the normal vectors of all vertices are consistent. A unified direction pointing outwards from the geometric exterior of the mold entity features.

[0058] Step S202: Generate equidistant offset point clouds and mesh deformation. The digital processing terminal processes the data according to preset machining allowance parameters. Perform an offset coordinate transformation along the normal direction on all vertices of the functional surface. Machining allowance parameters. The value of is determined based on the surface roughness and predicted thermal deformation of the additive manufacturing process. For any vertex on the functional surface... Its corresponding bias point The formula for calculating coordinates is:

[0059] ;

[0060] In the formula, The vertex coordinate vectors on the original functional surface; The machining allowance thickness is expressed in scalar form. Let be the unit outward normal vector at that vertex.

[0061] The system retains the original mesh's topological connectivity indexes and sets the original vertices unchanged. The coordinates are updated to the offset point. The coordinates are used to generate the initial offset mesh surface. .

[0062] Step S203: Reconstruct the closed envelope of the processing compensation layer entity. The digital processing terminal processes the initial bias mesh surface. Perform geometric corrections and closure. This is for cases where the radius of curvature is smaller than the machining allowance. The system performs self-intersecting interference of the mesh generated by the concave feature regions, and performs self-intersecting culling based on voxel remeshing or mesh Boolean trimming. Specifically, the system calculates the initial bias mesh surface. The bounding box is discretized into a 3D voxel mesh. The signed distance field (SDF) of the voxels is calculated based on the normal information of the mesh patches. Zero isosurfaces are extracted to generate smooth offset surfaces without self-intersection. Subsequently, the system extracts the original functional surfaces. Outer edge contour ring and offset surfaces Corresponding outer edge contour ring By constructing ruled surfaces or minimum area triangular partitions, and A suture connection is made to form a lateral sealing surface. This is achieved by connecting the original functional surfaces. Offset surface In addition, the lateral sealing surfaces are geometrically combined to construct a closed, three-dimensional manifold entity with physical volume, namely the processing compensation layer entity. .

[0063] Step S204: Merge the target domains of wear-resistant material deposition. The digital processing terminal performs a Boolean union operation to merge the processing compensation layer entities generated in step S203. With the wear-resistant housing region generated in step S103 The volumes are merged. The merged volumetric domain is defined as the final target domain for wear-resistant material deposition. The calculation formula is as follows:

[0064] ;

[0065] Through this step, the digital model of the mold is updated to a production-level model that includes machining allowances. In subsequent additive manufacturing processes, the feeding system will... It continuously delivers high-hardness, wear-resistant material within the covered area. This technical feature ensures that subsequent CNC finishing of the mold blank can remove [materials / materials]. After the corresponding physical material layer, the exposed mold surface is still located within the wear-resistant shell region. Within the material range, this ensures that the hardness index of the final mold surface meets the design requirements.

[0066] See attached document Figure 4 In step S300, the digital processing terminal performs mechanical simulation analysis on the defined strong and tough matrix domain, analyzes the internal stress transmission path of the mold under service conditions, and extracts the principal stress eigenvectors to guide subsequent lattice growth. This step specifically includes the following sub-steps:

[0067] Step S301: Establish the preprocessing model for finite element analysis. The digital processing terminal retrieves the three-dimensional discretized partitioned model of the mold constructed in step S100. The system analyzes the wear-resistant shell domain. and tough matrix domain Mesh discretization is performed. To ensure the accuracy of mechanical transfer at heterogeneous material interfaces, the system performs a geometric topology sharing operation before mesh generation to ensure the interface... The mesh nodes on both sides coincide or correspond in spatial position. Second-order tetrahedral elements are used for volume mesh generation to accommodate the complex free-form surface features inside the mold. In the material property assignment stage, corresponding isotropic linear elastic material parameters, including Young's modulus, are assigned to the mesh elements of the wear-resistant shell domain and the tough matrix domain, respectively. Compared to Poisson The system at the interface Define continuous medium connection conditions based on node sharing, or set multi-point constraint (MPC) binding contacts to simulate the continuous structure formed by metallurgical bonding during additive manufacturing.

[0068] Step S302: Apply boundary conditions for the stamping forming process. The digital processing terminal maps the stamping process parameters of the die to the load boundary conditions of the finite element model. In specific implementation, the system reads the functional surfaces of the die. The mesh data will determine the maximum reaction force during the stamping process. A non-uniform pressure field is formed by weighting the distribution of the pressure field to each grid node or integration point on the functional surface based on the area of ​​the facet; simultaneously, a blank holder force load is applied to the grid nodes in the contact area of ​​the die blank holder. Full-degree-of-freedom displacement constraints are applied to the mesh nodes on the bottom plane of the mold or the mounting reference plane to restrict the rigid body motion of the model. Through the above loading, the static boundary value problem of the mold under extreme stamping conditions is constructed.

[0069] Step S303: Solve for the Cauchy stress tensor field within the strong and tough matrix domain. The digital processing terminal, based on the equilibrium equations of continuum mechanics, calls the finite element solver to perform numerical calculations. After the calculation converges, the system extracts the strong and tough matrix domain. Stress state data at all element integration points within the domain. Its stress state is determined by the second-order Cauchy stress tensor. The tensor in matrix form in the global Cartesian coordinate system is described as follows:

[0070] ;

[0071] In the formula, the diagonal elements ( , , ) represents normal stress; off-diagonal elements ( , , , , , ) represents shear stress, which is the shear stress parallel to the cross-section direction.

[0072] Step S304: Solve for the dominant principal stresses and their characteristic direction vectors. The digital processing terminal processes the stress tensor obtained in step S303. Perform eigenvalue decomposition. Solve the characteristic equation. The three principal stress eigenvalues ​​were obtained. , , (in ) and their corresponding normalized eigenvectors , , Given that the die structure may simultaneously bear tensile or compressive loads under stamping conditions, the system executes the dominant stress screening logic, calculates the absolute value of each principal stress, and selects the principal stress with the largest absolute value as the dominant principal stress at that point. Its calculation logic is as follows:

[0073] ;

[0074] At the same time, extract and The corresponding eigenvector is used as the principal stress characteristic direction vector at that point. This characteristic direction vector It represents the direction in which the material bears the maximum load intensity at a given point, i.e., the direction of the main force transmission path.

[0075] Step S305: Construct the principal stress vector field data structure. The digital processing terminal traverses the strong and tough matrix domain. The dominant principal stress modulus will be calculated from all sampling points within the range. With principal stress characteristic direction vector Structured storage is performed. The generated dataset is defined as a set of vector field maps. Its form is:

[0076] ;

[0077] In the formula, This represents the total number of sampling points within the tough matrix domain; For the first The spatial coordinates of the points; Indicates the first Each point corresponds to the principal eigenvector of the dominant principal stress; Indicates principal stress The modulus.

[0078] This vector field mapping set This constitutes the driving source data for the subsequent generation of non-uniform lattice structures, ensuring that the growth direction of the subsequently generated lattice rods is always consistent with the direction of maximum force (whether tensile or compressive) inside the mold, thereby maximizing the utilization of structural stiffness.

[0079] See attached document Figure 5 In step S400, the digital processing terminal uses the principal stress vector field data extracted in step S300. In the tough matrix domain A lightweight lattice structure with an internally generated geometry that matches the mechanical transmission path is generated. This step optimizes the material distribution within the mold by establishing a mapping relationship between the stress field, the microstructure topology, and the dimensions of the rods. Specifically, it includes the following sub-steps:

[0080] Step S401: Define the basic lattice unit and the parameterized mapping space. The digital processing terminal operates in the tough substrate domain. The inner partition is used to fill the background mesh of the lattice. The system selects a topological configuration as the basic unit cell of the lattice, preferably a truss-like structure with anisotropic mechanical properties, such as a body-centered cubic (BCC), octahedral, or Kelvin structure. To facilitate subsequent rod size mapping, the basic unit cell is parameterized as a set of nodes. and rod assembly The geometric diagram, where the initial cross-sectional radius of the member is defined as a variable. The element in parameter space The side length in the equation is defined as a normalization constant.

[0081] Step S402: Construct the mapping function between stress modulus and member cross-section. The digital processing terminal establishes the dominant principal stress modulus. With the cross-sectional radius of the lattice bar The quantitative mapping relationship between them. The system traverses the vector field mapping set generated in step S305. Obtain the maximum dominant stress value across the entire field. and minimum dominant stress value The minimum allowable radius of the rod in the additive manufacturing process is set as follows: and the maximum member radius is For any position in space The lattice rod at the location, its cross-sectional radius The calculation formula is as follows:

[0082] ;

[0083] In the formula, This is the interpolation of the dominant principal stress modulus at that location; The system uses a maximum value function to determine the preset stress cutoff threshold. The mapping term is set to zero when the local stress is below this threshold, thus maintaining the minimum radius. ; This is a distribution control index used to adjust the gradient sensitivity of material distribution. Through this mapping function, large-section load-bearing members are generated in stress concentration regions, while small-section support members are generated in low-stress regions.

[0084] Step S403: Perform conformal lattice mapping based on principal stress directions. This is done to ensure that the principal stiffness directions of the lattice units align with the characteristic vectors of the local principal stresses. To maintain consistency and ensure node connectivity between adjacent lattice units, the digital processing terminal does not directly rotate individual lattice units, but instead constructs a curve coordinate system that conforms to the stress flow lines. Specifically, the system calculates the principal stress vector field. The streamlines generate a set of curvilinear grids aligned with the principal stress trajectories. The calculations are performed from the parameter space. To physical space Mapping transformation matrix This makes the tangent direction of the mapped mesh align with... Parallel. The system fills each voxel of the curved orthogonal grid with standard lattice basic units, and achieves continuous torsion of the overall lattice orientation through nodal coordinate transformation.

[0085] ;

[0086] In the formula, The coordinates of the lattice nodes in standard parameter space; These are the transformed physical space coordinates. This step ensures that the lattice structure conforms to stress flow lines macroscopically, while the nodal coordinates of adjacent elements at the boundary are precisely coincident.

[0087] Step S404 generates a globally continuous lattice solid model. The digital processing terminal generates solid geometry based on the continuously deformable wireframe skeleton model generated in step S403. The system employs implicit geometric modeling technology to construct a skeleton-based signed distance field (SDF) function. :

[0088] ;

[0089] In the formula, For a complete set of poles; For the first The center line segment of the root member; The local radius of the member is calculated based on step S402; For spatial points to line segment The Euclidean distance. The system extracts... The isosurfaces are used to generate a lattice solid model with smooth transition nodes and continuously varying cross-sections. For geometric overlap caused by the intersection of multiple members at a node, the SDF algorithm automatically performs Boolean union fusion to form a smooth rounded corner transition feature, eliminating geometric stress concentration points.

[0090] Step S405: Construct a transition connection between the lattice and the shell. This is to prevent wear-resistant shell domains... With internal lattice solid model There is a gap between them, with the digital processing terminal at the interface. A transitional enhancement structure is generated at the location. The system identifies interfaces within the lattice framework. For suspended end nodes with a distance less than a preset value, a raycasting algorithm is executed, starting from the end node and proceeding along the principal stress characteristic direction. Emit rays and calculate the interaction between the rays and the interface. The intersection coordinates are used to generate connecting rods. Then, a Boolean union operation is performed to transform the lattice solid model. With wear-resistant housing area It is merged into a single closed entity model.

[0091] See attached document Figure 6 In step S500, the digital processing terminal processes the lattice solid model generated in step S400. A mathematical model is established to predict local thermal behavior during additive manufacturing. This step generates a thermal accumulation exponential field by calculating the geometric thermal accumulation effect data at lattice nodes, and specifically includes the following sub-steps:

[0092] Step S501: Extract the topological and geometric features of the lattice nodes. The digital processing terminal processes the lattice solid model. The skeleton data is traversed to identify the locations of all intersecting nodes. For any given lattice node... The system establishes a coordinate system based on this node. Centered on the ball, with the largest member radius of times (recommended value) Thermal analysis of a locally enclosed sphere with radius ) The system identifies all connecting members whose axes pass through the bounding sphere. The digital processing terminal extracts the topological connectivity of the node and the local cross-sectional area of ​​each connecting member at that node. and the angle between the member axis and the additive manufacturing construction direction (usually the positive Z-axis). .

[0093] Step S502: Calculate the local geometric heat dissipation factor. The digital processing terminal calculates the node heat dissipation factor according to the discrete form of Fourier's law of thermal conductivity. Effective heat dissipation capacity at a given location. Define the local geometric heat dissipation factor. This characterizes the geometric resistance to heat diffusion from the node to the surrounding members. To accurately reflect the solid-state heat conduction characteristics, the system treats the connecting members as one-dimensional heat conduction channels, and the calculation formula is as follows:

[0094] ;

[0095] In the formula, The number of members connected to this node; The inherent thermal conductivity of the printing material; For the first Cross-sectional area of ​​the root member at the node; The effective thermal conductivity characteristic length is taken as the smaller of the radius of the locally enclosing sphere and the actual length of the rod. For the first The angle between the root member and the dominant direction of heat dissipation (i.e., the direction of substrate construction, negative Z-axis); As an anisotropy correction factor, this term assigns a higher heat dissipation weight to the rods pointing vertically towards the substrate.

[0096] Step S503: Quantize the volumetric energy density distribution of the input. The digital processing terminal calculates the energy input acting on the node region according to the preset laser scanning process parameters. The system uses the voxel-based Monte Carlo integration method or the Boolean volume calculation method to calculate the lattice solidity. With local enclosing sphere The intersection volume is used to obtain the local solid volume of the node. Subsequently, combined with laser power Scanning speed Hatch spacing and the thickness of the powder layer Calculate the equivalent volumetric energy input received by the node during the forming process. :

[0097] ;

[0098] In the formula, The absorptivity coefficient of the material for the laser wavelength; This is the standard volumetric energy density (VED). This formula reflects the total energy absorbed by the node during the printing process.

[0099] Step S504: Construct the node thermal accumulation index model. Based on the principle of energy conservation, the digital processing terminal constructs the node thermal accumulation index (HAI) by combining the heat dissipation conditions of step S502 and the energy input of step S503. This index is used to approximately characterize the degree of heat retention at the node after laser scanning. For any node in the entire field... Its heat accumulation index The calculation model is as follows:

[0100] ;

[0101] Substitute and simplify the formulas in steps S502 and S503 to eliminate local solid volumes. After the item, the expanded form is:

[0102] ;

[0103] In the formula, The specific heat capacity of the material; The density of the material; This is the interlayer thermal history correction factor. The value of the node The shaft height is related, and the calculation formula is as follows: ,in, The thermal accumulation coefficient was determined experimentally. This represents the absolute height of the node; This is the height of the upper surface of the formed substrate.

[0104] Step S505: Generate the thermal accumulation risk mapping point cloud. The digital processing terminal traverses all lattice nodes and calculates the corresponding... The system sets a critical overheating threshold and assigns this scalar value as an attribute to the lattice framework model. This threshold was determined through process experiments using standard samples. The system will... The nodes are marked as high-risk overheating nodes, and Below the preset densification threshold Nodes with low thermal risk are marked as nodes with under-fusion risk. The final result is a sparse point cloud dataset containing thermal risk attributes. This dataset serves as a direct input for subsequent adjustments to the scanning strategy.

[0105] See attached document Figure 7 In step S600, the control system of the additive manufacturing equipment executes a bimetallic or multi-material partitioned powder spreading and differentiated laser scanning strategy based on the lattice solid model generated in step S400 and the thermal accumulation risk data generated in step S500. This step achieves solid manufacturing of the mold with gradient mechanical properties through physical-level material partitioning and process-level parameter switching, specifically including the following sub-steps:

[0106] Step S601: Generate multi-material discrete slice mask data. The digital processing terminal processes the wear-resistant shell domain. and lattice solid model Perform parallel slicing. The system maintains a fixed layer thickness along the build direction (Z-axis). Discretize the model to generate the first... Two-dimensional cross-sectional data of each layer. The system identifies the material property boundaries of each layer and generates the corresponding first material mask image. Second material mask image At the interface of two phase materials The system build width is [value missing]. The metallurgical bonding transition region. When generating the mask, pixels within the transition region are set to simultaneously belong to both the first and second material masks, forming a geometrically overlapping region. Overlap width. The calculation is based on the difference in the thermal expansion coefficient of the materials. Determination of process temperature difference:

[0107] ;

[0108] In the formula, For safety factor; and These are the linear thermal expansion coefficients of the shell material and the base material, respectively; The equivalent temperature difference during additive manufacturing (i.e., the difference between the melting point or preheating temperature and room temperature). This represents the maximum allowable residual strain at the interface; This represents the length of the feature connection at the interface.

[0109] Step S602: Perform mask-based selective powder spreading. The additive manufacturing equipment is equipped with a multi-nozzle array powder feeding system with piezoelectric ceramic driven microvalve. The control system reads the mask data for the current layer. and The powder feeding module controls the powder spreading device to perform zoned deposition on the worktable. Specifically, the powder feeding module controls the opening and closing frequency of the corresponding nozzles based on the discretized pixel distribution of the mask image. Martensitic aging steel powder is placed in the corresponding area. Low-alloy high-strength steel powder is applied to the affected area. A powder-spreading doctor blade then smooths the deposited heterogeneous powder, forming a mixed powder bed with clear material boundaries. For the transition zone... By controlling the timing of the simultaneous opening of two powder nozzles, a mixed powder band with a gradient of composition is formed in this region.

[0110] Step S603: Configure differentiated laser scanning process parameters. The control system loads independent process parameter packages for different material regions. For the wear-resistant housing region, the high-density parameter group is invoked. High laser power combined with low scanning speed is used to ensure the shell reaches a dense state. For the lattice matrix domain, a low thermal input parameter set is used. This is to suppress heat accumulation deformation in small rods. Specifically, the system uses the heat accumulation index point cloud generated in step S505. The laser power at high-risk nodes is modulated in real time. The corrected real-time power... for:

[0111] ;

[0112] In the formula, Set the power for the reference; Current laser spot position The corresponding thermal accumulation index; This is the critical overheating threshold. The safe thermal threshold; This is the power attenuation coefficient. When... At the same time, maintain the reference power output.

[0113] Step S604: Implement the interface interlacing scanning strategy. To eliminate the bonding weaknesses at the interface of heterogeneous materials, the control system in the transition region... Perform interleaved scanning. The system defines a topologically interlocking path for the scan vector as it crosses the interface. Specifically, when the laser beam scans the housing side, its scan path extends into the substrate side depth. When scanning the substrate side, extend its path into the depth of the shell side. Within the transition region, the laser energy density... The distance function is linearly or nonlinearly transitioned to match the melting point changes of the mixed powders:

[0114] ;

[0115] In the formula, This is the distance of the current scan point relative to the side boundary of the shell; and These are the optimal volumetric energy densities of the two-phase materials, respectively. For the transition distribution function, the Sigmoid function or the cubic smooth polynomial function is preferred to ensure the smooth and continuous energy input at the interface.

[0116] Step S605, interlayer remelting and stress release. After completing the dual-material scan of a single layer, if the current layer is located at a Z-axis height where material properties change drastically, the control system performs a low-energy overall contour remelting, or controls the substrate heating system to perform in-situ tempering treatment on the layer within a specific temperature range. The remelting scan line deviates from the original filling path by a specific distance to eliminate porosity and release interlayer tensile residual stress. For heterogeneous material bonding surfaces, those skilled in the art can verify the effectiveness of the process parameters by observing the dilution rate and the thickness of the element diffusion layer using a metallographic microscope. This verification process is a routine testing method in the field of materials processing and will not be elaborated here.

[0117] See attached document Figure 8In step S700, the control system of the additive manufacturing equipment dynamically modulates the energy output during the laser scanning process in real time and point-by-point, based on the lattice node thermal accumulation index field generated in step S500. This step achieves precise control of heat input at the microscopic level by establishing a real-time mapping relationship between spatial position, thermal field, and laser power, and specifically includes the following sub-steps:

[0118] Step S701: Construct the spatiotemporal mapping relationship between the scanning path and the thermal field. The digital processing terminal reads the laser scanning path data of the current processing layer, which consists of a series of discrete scanning vectors. Composition. For each scan vector, the system operates according to the command clock frequency of the laser control card. (e.g., 100kHz) The geometric path is resampled into a sequence of discrete control points with equal time steps. This ensures that each control cycle corresponds to a unique spatial coordinate. The system utilizes a spatial indexing algorithm (such as KD-Tree) to map the heat accumulation risk point cloud generated in step S505. Query distance from current coordinates Recent The local heat accumulation index at the path point is calculated using the inverse distance weighted interpolation (IDW) method, considering the number of neighboring nodes. :

[0119] ;

[0120] In the formula, For feature nodes within the neighborhood, The pre-calculated heat accumulation index for this feature node. This represents the Euclidean distance between the current path point and the feature node. Through this mapping, the system transforms static spatial thermal field data into a dynamic thermal signal stream that is strictly aligned with the scanning time axis.

[0121] Step S702: Calculate the variable power modulation curve. The control system uses the local heat accumulation index... The target laser power at each moment is calculated using a preset negative feedback adjustment function. To prevent drastic power fluctuations from causing molten pool instability, a segmented smoothing modulation strategy is adopted:

[0122] ;

[0123] ;

[0124] ;

[0125] ;

[0126] In the formula, This is the reference process power for a standard rod cross-section; To enhance power for suspended or small features, and to compensate for the problem of excessive heat dissipation caused by insufficient geometric heat dissipation; To maintain the minimum power required for laser ignition; This is the baseline heat accumulation value; This is the normalized scaling factor; Power suppression coefficient; It is a hyperbolic tangent function used to achieve smooth saturation characteristics of power adjustment, avoiding abrupt changes in power output under extreme heat accumulation values.

[0127] Step S703: Perform laser response hysteresis compensation. Due to the pump response time of the laser generator and the optical path transmission delay, the actual output power lags behind the control command physically. Meanwhile, the mechanical motion of the galvanometer system also exhibits servo lag relative to the position command. To ensure that the modulated laser power is accurately applied to the corresponding geometric position, the control system executes a time-domain compensation algorithm. The system calculates the total relative time delay. The power signal sequence generated in step S702 Perform timeline translation to generate final control commands. :

[0128] ;

[0129] when At that time, the control system sends a power modulation command in advance; when At the same time, the power command is sent with a delay. This compensation ensures that the peak and valley variations of the laser energy density are strictly synchronized with the geometric positions of the lattice nodes on a microsecond timescale.

[0130] Step S704: Generate FPGA-executable hardware control instructions. The control system will then use the compensated power curve. This is converted into corresponding analog voltage values ​​or pulse width modulation (PWM) duty cycle values. The system constructs a composite control frame containing synchronization timestamps, packaging the XY axis galvanometer position commands of the scanning path with the corresponding synchronous laser power control commands. During processing, the underlying field-programmable gate array (FPGA) controller of the additive manufacturing equipment parses this file, simultaneously sending position signals to the galvanometer driver card and analog voltage modulation signals (0-10V) to the laser control interface in each clock cycle, enabling continuous variation of laser power with microsecond-level time steps during the scanning vector motion.

[0131] See attached document Figure 9In step S800, the control system executes a time-domain-based thermal relaxation compensation strategy for local areas where the heat accumulation index still exceeds the safety limit after power modulation in step S700. This step targets extreme hot nodes with extremely dense geometric features or limited heat dissipation channels. When further heat suppression by reducing laser power is not possible, it actively introduces a cooling waiting time or changes the scanning sequence, utilizing the natural thermal diffusion characteristics of the material to reduce the local temperature. Specifically, it includes the following sub-steps:

[0132] Step S801: Identify the set of extreme heat accumulation nodes. The digital processing terminal reads the power command stream corrected in step S703 and, combined with the heat accumulation index field from step S505, performs a secondary thermal risk assessment. The system sets an absolute thermal threshold. This threshold corresponds to the physical limit at which the material experiences severe remelting collapse or porosity defects. The system screens out even the lowest laser power... Below, its predicted heat accumulation index Still satisfied Spatial nodes constitute a set of extreme hot nodes. For each node in the set The system extracts its spatial coordinates and the corresponding scan vector segment.

[0133] Step S802: Calculate the critical thermal relaxation time. The digital processing terminal, based on the lumped parameter approximation of the transient heat conduction equation, calculates the time interval required for an extreme hot node to naturally cool from its current thermal state to a safe state. The system defines the thermal relaxation time constant. This characterizes the rate at which heat diffuses from the node center to the surrounding medium.

[0134] ;

[0135] In the formula, The characteristic geometric scale of the node (usually taken as the radius of the local bounding sphere); The thermal diffusivity of the material, ; It is a geometric correction factor whose value is positively correlated with the solid angle at the node or the number of connecting rods, and is used to correct the deviation of the one-dimensional heat conduction model at the three-dimensional node.

[0136] Based on this time constant, the system computes nodes Minimum required forced cooling time :

[0137] ;

[0138] In the formula, This represents the current predicted heat accumulation value for the node. The target safe thermal threshold; This represents the ambient background calorific value. This formula quantifies the physical time required for excess heat to dissipate through heat conduction without the input of new energy.

[0139] Step S803: Perform scan path reordering based on a skip mechanism. This is done to meet the cooldown time requirement. To maintain processing efficiency, the control system does not directly insert a stop-light-and-wait command, but instead adopts a farthest-point skip-scan strategy. The control system divides the current processing layer into several non-overlapping scanning islands. When the laser beam is about to scan to an extreme hot spot... When the vector segment is in question, the system determines the time interval since the last scan of the neighborhood of that region. Is it less than If so, the system interrupts continuous scanning of the current area and controls the galvanometer to jump to the partition furthest from the current node that has not yet been scanned to perform the operation. Jump distance Insulation requirements must be met:

[0140] ;

[0141] The system dynamically adjusts the execution sequence of the scanning vector and uses the time spent scanning other areas as a natural cooling window for extreme hot nodes, ensuring that when the laser beam returns to the node, its local temperature field has decayed to a safe range.

[0142] Step S804: Insert a mandatory minimum interlayer cooling cycle. In certain high-density lattice layers, if the entire field is a high-thermal-risk region and cooling requirements cannot be met through spatial jumps, the control system activates an interlayer cooling command. The system calculates the maximum residual cooling time required for all extreme hot nodes in the current layer. ,in This represents the theoretical total scan time for the current layer. If... The system inserts a duration of [duration] before the current layer scan ends and the powder spreading action begins. The system receives a mandatory wait command. During this period, the laser shuts down, while the protection gas circulation system remains operational to accelerate convective heat transfer.

[0143] Step S805: Generate machining code with timing constraints. The control system encapsulates the reordered scan path and inserted cooling instructions into the final machine execution code. The code displays the galvanometer jump instruction, jump delay parameters, and laser switching timing. For scan vectors involving extreme hot nodes, to avoid end-point overburn caused by frequent laser switching and galvanometer acceleration / deceleration, the system applies a galvanometer overshoot compensation strategy (i.e., Skywriting mode) at the vector endpoints. Specifically, the control system generates a non-emitting virtual extension path before the start point and after the end point of the scan vector, ensuring the galvanometer maintains a constant speed as it passes through the actual scan segment. This, combined with precise laser on-delay and laser-off-delay, eliminates energy buildup.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of additive manufacturing of a composite material automotive light-weight metal die blank, characterized in that, The method is executed by a manufacturing system comprising a digital processing terminal and a multi-material additive manufacturing execution unit, and comprises the following steps: S100, constructing a three-dimensional discrete partition model of the mold: based on the functional surface of the mold, the mold geometric entity space is divided into an outer wear-resistant shell domain and an internal strong and tough matrix domain, and a heterogeneous material transition interface located between the two is defined; S200, generating a machining compensation layer of the functional surface: establishing an entity compensation layer and merging it with the wear-resistant shell domain as a wear-resistant material deposition boundary; S300, extracting the mechanical field characteristic vector of the strong and tough matrix domain: extracting the dominant principal stress value inside the strong and tough matrix domain and the corresponding spatial characteristic vector direction; S400, generating a non-uniform lattice structure driven by the principal stress: constructing a lattice in the strong and tough matrix domain, controlling the growth axis of the lattice rod to be parallel to the spatial characteristic vector direction and mapping the distribution density according to the dominant principal stress value, and generating a lattice entity model; S500, establishing a process heat accumulation model of the lattice nodes: calculating the heat accumulation index of the lattice nodes in the lattice entity model; S600, executing partition switching deposition of heterogeneous materials: depositing high-hardness metal materials in the wear-resistant shell domain and the entity compensation layer, and depositing high-toughness structural steel materials in the strong and tough matrix domain; S700, implementing variable-energy power modulation based on the heat accumulation index: generating hardware control instructions according to the heat accumulation index for dynamically adjusting the heat source input power when depositing the strong and tough matrix domain; S800, executing cooling compensation of extreme heat nodes: suspending heat source input for the lattice nodes whose heat accumulation index reaches a critical value.

2. The method of claim 1, wherein the method is characterized by: The S100 step specifically comprises: Obtaining mold initial geometric data and performing manifold preprocessing to identify the functional surface in the mold outer boundary surface; Using the functional surface as a zero level set surface, using a signed distance field algorithm or voxelization bias algorithm to calculate an isosurface in the model interior direction, and generating a point set satisfying a distance constraint condition as the wear-resistant shell domain; Defining the remaining volume part after deducting the wear-resistant shell domain from the total volume domain of the mold as the strong and tough matrix domain, and extracting the common boundary of the wear-resistant shell domain and the strong and tough matrix domain as the heterogeneous material transition interface.

3. The method of claim 1, wherein the method is characterized by: The S300 step specifically comprises: Performing grid discretization on the wear-resistant shell domain and the strong and tough matrix domain, and performing geometric topology sharing operation at the heterogeneous material transition interface; Distributing the maximum reaction force in the stamping forming process to the grid nodes of the functional surface according to the area weighting of the surface patch to construct a statics boundary value problem; Solving the Cauchy stress tensor field at all element integration points in the strong and tough matrix domain, and performing eigenvalue decomposition on the Cauchy stress tensor field; Selecting the principal stress with the largest absolute value as the dominant principal stress, and extracting the eigenvector corresponding to the dominant principal stress as the spatial characteristic vector direction.

4. The method of claim 1, wherein the method is characterized by: The S400 step specifically comprises: Defining a lattice basic unit and establishing a quantitative mapping relationship between the dominant principal stress value and the cross-sectional radius of the lattice rod, and generating large-section load-bearing rods in the stress concentration area; A streamline of the spatial feature vector direction is calculated to generate a curved orthogonal grid aligned with the principal stress trajectory; The lattice basic unit is filled into the voxel of the curved orthogonal grid, and the continuous torsion of the lattice posture is realized through node coordinate transformation, so that the principal stiffness direction of the lattice unit is consistent with the spatial feature vector direction; A skeleton-based signed distance field function is constructed to generate the lattice entity model with smooth transition nodes and continuously changing variable cross-section.

5. The method of claim 1, wherein the method is characterized by: The S500 step specifically includes: A local heat analysis enclosing sphere with the lattice node as the sphere center is established, and a set of connecting rods passing through the local heat analysis enclosing sphere range is identified; According to the number, cross-sectional area and angle of the connecting rod axis relative to the additive manufacturing building direction, the local geometric heat dissipation factor is calculated; According to the preset laser scanning process parameters, the equivalent volume energy input received by the lattice node in the forming process is calculated; Based on the principle of energy conservation, the local geometric heat dissipation factor and the equivalent volume energy input are integrated to construct the heat accumulation index, which is positively correlated with the equivalent volume energy input and negatively correlated with the local geometric heat dissipation factor.

6. The method of claim 1, wherein the method is characterized by: The S600 step specifically includes: Multi-material discrete slice mask data is generated, and a metallurgical bonding transition zone is constructed at the heterogeneous material transition interface between the wear-resistant shell domain and the tough matrix domain; A powder laying device is controlled to perform partition deposition on the workbench according to the multi-material discrete slice mask data, and a mixed powder belt with gradient change in composition is formed in the metallurgical bonding transition zone; An interface interweaving scanning strategy is performed in the metallurgical bonding transition zone, and a laser beam is controlled to adopt a topologically interlocking path to extend into the preset depth of the other side material when scanning one side material.

7. The method of claim 1, wherein the method is characterized by: The S700 step specifically includes: A space-time mapping relationship between the scanning path and the thermal field is constructed, discrete scanning vectors are resampled into discrete control point sequences with equal time steps, and the heat accumulation index of the neighborhood nodes of each control point is queried; A segmented smooth modulation strategy is used to calculate the target laser power, and when the heat accumulation index is higher than a preset threshold, the laser power is reduced through a negative feedback adjustment function; The difference between the physical lag of the laser generator and the servo lag of the galvanometer system is calculated, the target laser power is time-axis shifted to compensate, and the hardware control instruction is generated.

8. The method of claim 1, wherein the method is a method of additive manufacturing of a composite material automotive lightweighting metal mold blank. The step S800 specifically includes: Based on the lumped parameter method approximation of the transient heat conduction equation, the critical heat relaxation time required for the lattice node to naturally cool from the current thermal energy state to a safe state is calculated; A farthest point jump scanning strategy is used, when the laser beam is about to scan the vector segment where the lattice node is located and the cooling time does not meet the critical heat relaxation time, the scanning of the current region is interrupted and the farthest unscanned partition from the current node is jumped to for work.

9. The method of claim 1, wherein the method is characterized by: The step S200 specifically includes: The normal vector of the discretized grid of the functional surface is reconstructed, and the area weighted average algorithm is used to calculate the vertex normal vector; According to the preset machining allowance parameter, a normal offset coordinate transformation is performed on all vertices on the functional surface to generate an initial offset mesh surface; The initial offset mesh surface is subjected to self-intersection removal based on voxel reconstruction, and is stitched with the outer edge contour line ring of the functional surface to form a closed entity compensation layer.

10. The method of claim 1, wherein the method is a method of additive manufacturing of a composite material automotive lightweighting metal mold blank. The S400 step further includes constructing a transition connection between the lattice and the shell: An overhanging end node in a skeleton of the lattice entity model is identified, which is less than a preset value from the heterogeneous material transition interface; A ray projection algorithm is performed along the spatial feature vector direction from the overhanging end node as a starting point, the intersection coordinates of the ray and the heterogeneous material transition interface are calculated, and a connecting rod is generated; A Boolean set operation is performed to fuse the lattice entity model and the wear-resistant shell domain to form a single closed entity model.

Images