Thin-wall structure lightweight design method based on additive manufacturing

Abstract

The invention discloses a thin-wall structure lightweight design method based on additive manufacturing, and belongs to the technical field of additive manufacturing, and the thin-wall structure lightweight design method comprises the following steps: S100, designing an initial structure based on function and performance mapping; s200, an additive manufacturing process knowledge base is constructed, and the adaptability of the additive manufacturing process is evaluated; s300, carrying out process and structure bidirectional coupling finite element analysis; s400, collaborative optimization of a process and a structure driven by a multi-objective optimization algorithm; s500, feature part manufacturing verification and design and process scheme iterative optimization are carried out; and S600, generating a design and process integrated scheme, wherein the integrated scheme comprises design parameters and manufacturing parameters. The technical problem that in the prior art, the structure optimization target and the manufacturing adaptive capacity are lack of collaboration can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of additive manufacturing, and particularly to a lightweight design method for thin-walled structures based on additive manufacturing. Background Art

[0002] With the higher requirements for the performance of structural components in high-end equipment fields such as aerospace and automotive manufacturing, achieving the lightweighting of structures, functional integration, and improvement of manufacturing freedom have become the key development directions. Due to its advantages in significant weight reduction effect and high structural efficiency, thin-walled structures are widely used in complex load-bearing components such as engine casings, support skeletons, and heat exchange channels. Such structures not only need to meet strict mechanical performance indicators but also face practical engineering challenges such as high manufacturing difficulty and high dimensional stability requirements.

[0003] As an important method for realizing lightweight structural design, topology optimization can generate an optimal structural configuration that meets the performance requirements through the automatic regulation of material distribution based on given boundary conditions and performance constraints. However, most of the existing topology optimization methods only take mechanical performance as the optimization goal and ignore the actual constraints in the additive manufacturing process, such as minimum forming size, overhang angle limit, support removability, and powder cleaning ability inside the component. These unconsidered process limitations often lead to the optimization results being difficult to be directly used for actual printing or even unable to complete manufacturing, resulting in a serious disconnection between structural design and process implementation.

[0004] In addition, traditional design methods lack effective linkage between the structural generation stage and the manufacturing process. The improvement of structural performance often comes at the cost of manufacturing feasibility; moreover, the influence of manufacturing parameters on factors such as structural deformation, density, and microdefects has not been incorporated into the analysis during the design stage. Especially for complex thin-walled structures, they are extremely vulnerable to problems such as thermal stress accumulation, deformation warping, and microcrack formation during the additive manufacturing process, and the existing optimization paths are difficult to provide the ability to predict and control these manufacturing problems. Summary of the Invention

[0005] The present invention provides a lightweight design method for thin-walled structures based on additive manufacturing to solve the technical problem of the lack of coordination between the structural optimization goal and the manufacturing adaptability in the prior art.

[0006] According to one aspect of the present invention, there is provided a lightweight design method for thin-walled structures based on additive manufacturing, including the following steps: S100, design an initial structure based on the mapping of functions and performance, obtain mechanical performance indicators according to the functional requirements of the thin-walled structure, and based on the mechanical performance indicators, use the topology optimization method to design the initial structure to obtain an initial lightweight structure model that meets the performance constraints; S200, Construction of the additive manufacturing process knowledge base and evaluation of the adaptability of the additive manufacturing process. Build the additive manufacturing process knowledge base, and evaluate the process adaptability of the initial lightweight structure model according to the additive manufacturing process knowledge base to identify the areas and features that do not meet the requirements of the additive manufacturing process; S300, Two-way coupled finite element analysis of process and structure. Establish a two-way coupled finite element simulation model between the structure model and manufacturing process parameters to analyze the influence of process parameters on the structural performance and manufacturing quality; S400, Collaborative optimization of process and structure driven by multi-objective optimization algorithm. Based on the multi-objective optimization algorithm, with the structural quality, performance stability and manufacturing efficiency as the goals, co-optimize the structural parameters and process parameters; S500, Manufacturing verification of characteristic parts and iterative optimization of design and process plans. Select representative characteristic structures for physical manufacturing verification, evaluate whether the manufacturing defects, deformation degree and performance meet the design goals, and iteratively optimize the design parameters and process parameters according to the verification results until the structural performance and manufacturing process requirements are met; S600, Generate an integrated design and process plan. The integrated plan includes design parameters and manufacturing parameters.

[0007] Optionally, the topology optimization method in step S100 includes the following steps: S110, Digital characterization of additive manufacturing process constraints; S120, Force / thermal weakly coupled multi-objective topology optimization; S130, Optimized design of structural stiffening and grille layout.

[0008] Optionally, step S110 includes the following steps: S111, Analyze and identify the process constraints that affect the forming power and structural quality during the additive manufacturing process, including self-supporting forming constraints and preparation process constraints; The self-supporting forming constraints include overhang angle constraints and printing direction constraints; the preparation process constraints include manufacturability constraints, machinability constraints and connectivity constraints; S112, Digitalize all constraint conditions.

[0009] Optionally, step S120 includes the following steps: S121; Determine two basic constraints: volume constraint and external interface constraint; S122; Construct the objective function and normalize it. Convert multiple design goals into a unified optimization objective function through dimensionless normalization; S123; Select and apply an optimization algorithm to obtain the optimal topology structure and transmission path. The optimization algorithms include at least one of the return matrix method, the adjoint method and the MMA algorithm.

[0010] Optionally, in step S200, the additive manufacturing process knowledge base includes: a process classification module, a typical feature structure module, a material matching information module, an additive manufacturing process parameter module, and a post-processing and inspection and testing module.

[0011] Optionally, step S400 includes the following steps: S410, perform geometric reduction processing on the topology optimization result to obtain a structural entity model with engineering recognizability; S420, based on the structural entity model, construct an analysis surrogate model for multi-physical field performance prediction; S430, based on the constructed analysis model, further jointly optimize the outer shape boundary and geometric dimension parameters of the structure.

[0012] Optionally, step S410 includes the following steps: S411, use non-uniform B-spline basis functions to perform continuous surface reconstruction on the topology structure boundary to achieve high-smoothness modeling of complex topology contours; S412, based on the point cloud density distribution of the topology result, execute an automatic cropping and boundary extraction algorithm to identify and separate the effective load path and redundant area of the structure; S413, construct the processed boundary data into a high-precision three-dimensional entity model to provide basic input for subsequent parametric expression and analysis modeling.

[0013] Optionally, step S420 includes the following steps: Step S421, use sampling technology to effectively cover the design variable space; Step S422, introduce an optimization algorithm and an adaptive surrogate modeling method to establish an approximate expression of the variable and response relationship; Step S423, perform parametric modeling on the key features in the structure, express them in variable form and import them into the optimization framework; Step S424, based on the parametric model, integrate the innovative structure feature template of typical stator components to improve the functional adaptability and manufacturing rationality of the structure in specific application scenarios.

[0014] Optionally, step S430 includes the following steps, S431, use the compromise programming method to perform collaborative trade-offs on multiple optimization objectives of the structure; S432, apply the Pointer intelligent optimization method to perform multiple rounds of rapid iteration on the design variables to improve the optimization efficiency and convergence stability; S433, input multiple optimization constraint conditions, including: multi-point loading criterion, shape boundary control rule, and dimension parameter limit.

[0015] Optionally, in step S500, the process plan iterative optimization includes the following steps: S510, clarify the requirements of the printed workpiece, including mechanical property requirements, geometric accuracy requirements, and functional integration. These requirements serve as the starting point of the process, driving the formulation of subsequent design and manufacturing strategies; S520, perform topology optimization to minimize the material usage while meeting the structural performance goals, and preliminarily determine the material distribution and component layout form; S530, consider the placement method of parts on the printing platform, which affects residual stress, cooling path, and powder usage rate; S540, design the support structure, including type, layout, and removable analysis; S550, set the core parameters of the manufacturing process, including controlling the energy input, printing path, printing spacing, and layer thickness.

[0016] In summary, the present application includes at least one of the following beneficial technical effects: Starting from S100, the present invention introduces the concept of "function-performance mapping", converts the functional requirements of the structure into quantifiable mechanical property indicators, and combines the additive manufacturing process knowledge base constructed in S200 for process adaptability evaluation during the topology optimization process, excluding non-manufacturable features from the initial design stage to ensure the printability of the structure. Subsequently, in S300, process-structure bidirectional coupled finite element analysis is introduced to clarify the influence of manufacturing process parameters on structural deformation, residual stress, and density, realizing the linkage prediction of manufacturing path and structural performance. Through the multi-objective collaborative optimization algorithm set in S400, the structural parameters and manufacturing parameters co-evolve in the same optimization framework, comprehensively balancing the trade-off relationship among structural quality, performance stability, and manufacturing efficiency, breaking the traditional single-objective optimization path. Further, the representative feature part manufacturing verification mechanism in S500 forms feedback signals on manufacturing defects, deformation laws, and material responses through physical verification, realizing a closed-loop iterative adjustment mechanism of optimization, verification, and re-optimization. Finally, in S600, an integrated solution including structural design and manufacturing process parameters is generated, improving the implementation ability from virtual structure to physical component. The entire process not only makes the structural design results have actual manufacturability, but also continuously improves the optimization accuracy and engineering reliability through data-driven iteration, substantially solving the problem of insufficient coordination and path fragmentation between existing structural design and additive manufacturing processes, and realizing the transformation from performance-oriented optimization to manufacturing constraint-driven collaborative optimization.

[0017] In addition to the purposes, features, and advantages described above, the present invention has other purposes, features, and advantages. The present invention will be further described in detail below with reference to the drawings. Description of the Drawings

[0018] The accompanying drawings, which form a part of this application, are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings: Figure 1 is a flowchart of the lightweight design method for thin-walled structures based on additive manufacturing according to the present invention; Figure 2 is a schematic diagram of the initial structure designed by mapping function and performance according to the present invention; Figure 3 is a schematic diagram of the additive manufacturing process knowledge base according to the present invention; Figure 4 is a schematic diagram of the process and structure collaborative optimization driven by the multi-objective optimization algorithm according to the present invention; Figure 5 is a schematic diagram of the iterative optimization of the process plan according to the present invention. Detailed Embodiment

[0019] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways defined and covered by the following.

[0020] The following combines the attached Figures 1-5 to further describe the present application in detail.

[0021] The embodiments of the present application disclose a lightweight design method for thin-walled structures based on additive manufacturing.

[0022] Referring to Figure 1 , the lightweight design method for thin-walled structures based on additive manufacturing provided by the present invention is carried out around the collaborative integration of the structural optimization design goal and the manufacturing feasibility control path, and constructs a complete closed-loop process from structural design input, process adaptability analysis, process-structure coupling simulation, collaborative optimization, verification iteration to solution output. Specifically, it includes the following steps: S100, Initial design input and function-performance mapping modeling. Based on the mapping of function and performance, an initial structure is designed. According to the functional requirements of the thin-walled structure, mechanical property indexes are obtained, and based on the mechanical property indexes, a topology optimization method is used to design the initial structure, and an initial lightweight structure model that meets the performance constraints is obtained.

[0023] S200, Construction of the additive manufacturing process knowledge base and evaluation of the adaptability of the additive manufacturing process. The additive manufacturing process knowledge base is constructed, and the initial lightweight structure model is evaluated for process adaptability according to the additive manufacturing process knowledge base, and the regions and features that do not meet the requirements of the additive manufacturing process are identified.

[0024] S300, Two-way Coupled Finite Element Analysis of Process and Structure. In two-way coupled finite element analysis of process and structure, a two-way coupled finite element simulation model is established between the structural model and manufacturing process parameters to analyze the influence of process parameters on structural performance and manufacturing quality.

[0025] S400, Collaborative Optimization of Process and Structure Driven by Multi-objective Optimization Algorithm. Based on the multi-objective optimization algorithm, the structural parameters and process parameters are collaboratively optimized with the goals of structural quality, performance stability, and manufacturing efficiency.

[0026] S500, Manufacturing Verification of Feature Parts and Iterative Optimization of Design and Process Plans. Select representative feature structures for physical manufacturing verification, evaluate whether manufacturing defects, deformation degrees, and performance meet the design goals, and iteratively optimize the design parameters and process parameters based on the verification results until the structural performance and manufacturing process requirements are met.

[0027] S600, Generate an Integrated Design and Process Plan. The integrated plan includes design parameters and manufacturing parameters.

[0028] By introducing a collaborative control mechanism for structural design and manufacturing processes, a full-process system covering design input, process adaptability evaluation, structure-process coupling simulation, multi-objective collaborative optimization, manufacturing verification, and closed-loop output of the plan is established, which has significant technical advantages. On the one hand, this method introduces an additive manufacturing process knowledge base and manufacturability evaluation at the initial stage of structural optimization, effectively avoiding manufacturing obstacles such as non-printable structures and non-removable supports in traditional topology optimization schemes, and improving the engineering feasibility of design schemes; on the other hand, by establishing a two-way coupled finite element model between process parameters and structural performance, feedforward prediction of the influence of process variables such as laser power, scanning path, and layer thickness on structural stress, deformation, defects, etc. is realized, significantly enhancing the adaptive control ability of structural design for the manufacturing process; in addition, this method uses a multi-objective optimization algorithm to realize the joint optimization of structural parameters and process parameters, and through physical verification and feedback adjustment of representative feature parts, a closed-loop iterative mechanism of performance-manufacturing-verification is constructed, and finally an integrated manufacturing plan with unified design parameters and manufacturing parameters is output. This method not only achieves the collaborative optimization of thin-walled structures under performance indicators such as lightweight, stiffness, and density, but also significantly improves the printing success rate, product consistency, and manufacturing efficiency in the additive manufacturing process.

[0029] The topology optimization method in step S100 includes the following steps: S110, Digital Characterization of Additive Manufacturing Process Constraints; S120, Force / Thermal Weak Coupling Multi-objective Topology Optimization; S130, Optimized Design of Structural Reinforcement and Grid Arrangement.

[0030] In step S100, after establishing the initial design model based on the function-performance mapping, structural topology optimization is further carried out in combination with the characteristics of additive manufacturing. To ensure the practical manufacturability of the optimization results, the digital representation of additive manufacturing process constraints is first introduced into the optimization modeling. This step extracts typical manufacturing limitation conditions, including self-supporting forming constraints and preparation process constraints, and parameterizes and makes them computable. Among them, the self-supporting constraint limits the angle between the surface and the printing direction in the structure not to exceed a given limit value, so as to reduce the dependence on the support structure and improve the printing forming quality. The preparation constraints include minimum wall thickness, minimum hole diameter, structural connectivity, and support removability requirements. Geometric modeling is carried out through methods such as surface normal analysis, volume constraint mask, and path accessibility analysis, and is integrated into the topology optimization constraint terms.

[0031] Subsequently, in step S120, multi-objective topology optimization of force / thermal weak coupling is carried out. First, the basic constraints of the optimization problem are determined, including the volume constraint for controlling the overall material usage and the setting of non-changeable regions for key installation interfaces. Then, a multi-objective function system is constructed. Objectives such as maximizing structural stiffness and minimizing thermal deformation are subjected to dimensionless normalization and unified into the optimization objective function. The sensitivity information of the design variables to the objective function is obtained through the adjoint method or the return matrix method, and the moving asymptote method is used as the iterative solution strategy to globally optimize and update the design variables, and finally the optimal distribution path of the material in the design space is obtained. This optimization process can effectively identify the main load-bearing path and the key heat conduction path, form a preliminary structural topology model with a coordinated distribution, and have reasonable structural continuity and efficient force / thermal performance distribution.

[0032] After obtaining the preliminary topology results, step S130 is entered to carry out the detailed design of stiffening and grid layout for key structures. For the high stress concentration areas and large-area thin-walled areas generated in the topology optimization, by introducing directional stiffening structures and lattice filling units, the local stability is improved and the overall stiffness is enhanced. The stiffening design determines the stiffening direction and position based on the stress distribution map of the topology results. The grid layout adopts a periodic or hierarchical density control strategy to keep the structure lightweight while maintaining good structural integrity and manufacturing continuity. During the above structural refinement process, the minimum forming size, smooth connection transition, and connectivity of the powder discharge channel are considered synchronously to ensure that the optimization results can meet the process requirements of the additive manufacturing process and provide a feasible geometric basis for subsequent process adaptability evaluation and collaborative optimization.

[0033] Step S110 includes the following steps: S111, analyze and identify the process constraints that affect the forming power and structural quality in the additive manufacturing process, including self-supporting forming constraints and preparation process constraints; the self-supporting forming constraints include overhang angle constraints and printing direction constraints; the preparation process constraints include manufacturability constraints, machinability constraints, and connectivity constraints; S112, digitize all the constraint conditions.

[0034] In this embodiment, step S110 is the digital characterization of the additive manufacturing process constraints, which includes analyzing and identifying the main process limiting factors that affect the forming power and structural quality of the additive manufacturing process, and integrating these limitations into the topology optimization model in a quantifiable and modelable form, so as to establish an effective linkage with manufacturing feasibility at the initial stage of structural optimization. Specifically, in step S111, combined with the process characteristics of typical metal laser powder bed fusion (SLM) additive manufacturing processes, the process constraints are systematically identified and divided into two dimensions: self-supporting forming constraints and preparation process constraints.

[0035] The self-supporting forming constraints mainly include two aspects: overhang angle limitation and printing direction limitation. The overhang angle constraint is used to prevent the structure surface from collapsing or sintering failure during printing due to an excessive overhang angle. Usually, the maximum unsupported forming angle is set to 45°. By analyzing the angle between the normal direction of each surface of the structure and the printing direction and establishing a vector field, all structural areas that do not meet the angle requirements can be identified and removed. The printing direction constraint is directly related to the scanning path, heat conduction direction, and support setting, and needs to be jointly determined in combination with the functional partition and thermal flow path of the designed structure, and is embedded as an input parameter into the topology optimization solution framework.

[0036] The preparation process constraints include manufacturability constraints, machinability constraints, and connectivity constraints. The manufacturability constraint is used to prevent the appearance of detailed structures smaller than the minimum forming ability of the additive manufacturing process, and ensure that the structure boundaries, channels, and thin-walled areas are within the controllable accuracy range of the equipment. The machinability constraint is for post-processing requirements, such as heat treatment deformation repair, support removal, and finish machining of functional surfaces, and requires the structure to have necessary clamping surfaces, accessible areas, and machining paths. The connectivity constraint is mainly used to ensure the flow and cleaning of powder during the forming process. Especially for internal complex cavity structures, the generation of closed dead cavities or closed spaces without through holes should be avoided to improve the printing success rate and product consistency.

[0037] In step S112, the aforementioned identified process constraints are modeled and expressed by digital means so that they can be directly embedded into the topology optimization calculation process. The specific methods include establishing a spatial geometric constraint mask, defining a minimum entity unit filtering function, constructing an angle constraint field based on the printing direction, and constructing a structure accessibility map in combination with the feasibility of support removal. These parameters are incorporated into the objective function and constraint function of topology optimization through means such as Boolean operations, shape penalty functions, and sensitivity weighting factors, to achieve dynamic screening of non-manufacturable features and guiding control of the structural optimization path, thereby maintaining the consistency between the structural design and the additive manufacturing process capabilities in real time during the topology evolution process.

[0038] Step S120 includes the following steps: S121, determining two basic constraints: volume constraint and external interface constraint; S122, constructing an objective function and normalizing it, converting multiple design objectives into a unified optimization objective function through dimensionless normalization processing; S123, selecting and applying an optimization algorithm to obtain the optimal topological structure and transfer path, and the optimization algorithm includes at least one of the return matrix method, the adjoint method, and the MMA algorithm.

[0039] In this embodiment, step S120 involves multi-objective topology optimization of weak force / heat coupling, aiming to obtain an initial structural topology model with the advantages of structural lightweight, stiffness, and thermal performance through multi-physical field performance coupling optimization means on the premise of considering the feasibility of the additive manufacturing process. This step first sets the basic constraint conditions of the optimization problem in S121, including two types: volume constraint and external interface constraint. Among them, the volume constraint is used to limit the maximum usage ratio of the structural material. For example, the material usage rate is controlled within 30% - 50% of the total volume of the design space to ensure the achievement of the lightweight goal; the external interface constraint is set for the unchangeable functional areas, such as mounting holes, flange surfaces, interface channels, etc., as non-design domains, and the fixed boundary attributes are maintained during the topology optimization iteration process to ensure the assembly compatibility and functional continuity between the structure and other components of the system.

[0040] Subsequently, in S122, an optimization objective function is constructed and normalized to unify the dimensions of different performance indicators, facilitating collaborative processing in the multi-objective optimization framework. Aiming at the typical requirements of thin-walled structures during actual service, the present invention simultaneously considers multiple design objectives such as maximizing stiffness, minimizing thermal stress, and controlling thermal deformation. The dimensionless normalization means is used to standardize each objective function, construct a unified form of the objective expression, and set the weighting coefficients between each performance indicator according to actual needs to achieve trade-off optimization between multiple objectives. Dynamic weighting or piecewise penalty strategies can be introduced into the normalization function to strengthen the optimization guiding ability for key performance objectives.

[0041] In S123, a suitable topology optimization algorithm is selected and applied to solve the above model to obtain the optimal material distribution path and structural configuration. One or more combinations of mainstream topology optimization methods such as return matrix method, adjoint method and MMA can be selected during the optimization process. The return matrix method is suitable for small-size problems with fast convergence, the adjoint method can be used for sensitivity analysis and gradient-guided optimization, and the MMA algorithm is preferred in the present invention due to its high convergence efficiency for nonlinear objective functions and the advantages of solving large-scale variables. The SIMP material interpolation method is combined in the optimization algorithm, and the unit density is used as a continuous variable to control the distribution state of the material in the design space. The penalty coefficient is adjusted to strengthen the differentiation ability of the "0-1" structure during the iteration process, and finally a three-dimensional structural initial mold that takes into account mechanical stiffness, thermal conductivity and manufacturing processability is obtained. The topology optimization result generated in this step has a clear force / heat load transfer path, efficient material utilization efficiency and good structural continuity, which can effectively support subsequent geometric refinement design, manufacturing adaptability evaluation and process parameter collaborative optimization.

[0042] In this embodiment, step S200 involves the construction of an additive manufacturing process knowledge base and process adaptability assessment, aiming to provide systematic and callable knowledge support for the subsequent manufacturing feasibility analysis and optimization of the topological structure. The process knowledge base described in this step includes multiple functional modules, which are respectively used to store and manage core information resources related to the additive manufacturing process, and constitute a basic database for structure-process linkage design.

[0043] The process knowledge base first includes a process classification module, which is used to systematically divide the available manufacturing processes according to the additive manufacturing technology path, including but not limited to laser selective melting, electron beam melting, directed energy deposition, binder jetting, etc. Each process path corresponds to different forming accuracy, material applicability, heat input characteristics and typical defect modes. This module classifies and organizes the process windows, forming mechanisms, and equipment constraints of various processes, and is used to preliminarily determine the forming technology route suitable for the target structure.

[0044] The typical feature structure module is used to record the morphological features and manufacturing difficulties related to common structural units in additive manufacturing, including overhanging structures, ribs, thin-walled channels, bridging structures, internal closed cavities, etc. Each type of structural feature is equipped with its printability evaluation index, support strategy, printing direction sensitivity, post-processing method and typical failure mode, which can be used for local process adaptability evaluation after structural modeling or topology optimization.

[0045] The material matching information module is used to store information on commonly used materials under different process routes, including material grades, powder characteristics, thermal physical parameters, printability grades, typical defect sensitivities, etc. By matching the division of the target structural functional area with the material applicability, the collaborative selection of process and materials is realized, avoiding forming failures, element segregation or internal defects caused by improper material selection.

[0046] The additive manufacturing process parameter module includes recommended parameter ranges and empirical parameter models for various process paths, involving key variables such as laser power, scanning speed, layer thickness, scanning spacing, preheating temperature, platform temperature, atmosphere control, etc. This module can provide a default parameter template, or output a recommended process window by combining process simulation and actual verification data for subsequent process-structure coupling modeling and collaborative optimization calculations. The post-processing and inspection module is used to manage the process chain information after the structure is printed, including the applicability judgment and parameter recommendation of post-processing processes such as hot isostatic pressing, stress relief annealing, support removal, surface polishing, and finishing, as well as the applicable standards, equipment requirements, and inspection node layout logic of quality inspection methods. This module provides a full-process technical basis for structure verification, process closed-loop control, and quality assurance.

[0047] By constructing the above process knowledge base module and embedding it into the structural design process, a systematic process adaptability analysis of the model can be carried out after the structure optimization is completed, identifying potential manufacturing difficulty areas, warning local structural features that do not meet the process window, and providing data support for subsequent coupled simulation, process optimization, and manufacturing plan generation, ensuring the manufacturing feasibility, cost rationality, and quality stability of the structural plan, and providing a knowledge basis for the design of high-performance thin-walled structures oriented by additive manufacturing.

[0048] In this embodiment, step S300 aims to establish a two-way coupling analysis mechanism between process parameters and structural performance, quantify the influence of the manufacturing process on the structural response by means of finite element simulation, and provide coupled data support for subsequent optimization. Different from traditional structural optimization methods that only take the geometric model as the input, the present invention introduces key process variables such as heat input, melt pool evolution, scanning path, and cooling rate in the additive manufacturing process into the analysis model to realize physical coupling modeling between the structure and the process.

[0049] First, a finite element structural grid is constructed based on a high-precision three-dimensional solid model, and thermal-mechanical physical parameters consistent with the material are assigned, including elastic modulus, coefficient of thermal expansion, thermal conductivity, specific heat capacity, etc. The process parameter modeling part is based on the parameters of typical additive manufacturing equipment, and input variables such as laser power, scanning speed, layer thickness, scanning strategy, and platform preheating temperature are set. In the simulation environment, the layer-by-layer scanning sequence and laser trajectory are defined based on the process path generation module, and the laser energy input behavior is simulated through a heat source modeling method.

[0050] In the thermal analysis stage, the transient non-steady-state heat conduction equation is used to solve the point-by-point temperature evolution process during the printing process, and the thermal history information at key moments, such as peak temperature, temperature rise rate and cooling gradient, is recorded. Then, in the mechanical analysis module, the thermal stress distribution, residual deformation, displacement accumulation and overall warping trend of the component caused by the temperature field are solved based on the thermal-elastic coupling or thermal-plastic-elastic coupling constitutive relationship. This simulation can identify the local stress concentration, high deformation area and stress release path caused by the manufacturing process, which helps to avoid the performance degradation risk caused by the thermal manufacturing process in advance during the structural design stage.

[0051] At the same time, in order to reflect the two-way feedback mechanism of process and structure, the present invention supports the reaction of structural response to process path design and scanning strategy adjustment. For example, when the simulation finds that the thermal stress of a specific area exceeds the limit due to uneven cooling, the scanning sequence, laser power or support structure setting of the area can be locally optimized and adjusted; if there is an overall warping trend, feedforward compensation can be performed through the platform preheating strategy or scanning direction distribution.

[0052] In addition, the bidirectional coupling model also supports defect prediction and process robustness analysis. By combining the typical defect formation thresholds in the material database, the defect-prone areas can be predicted based on the simulated temperature field, and the process parameters can be adjusted in a linked manner to avoid high-risk thermal history areas. Finally, the coupled analysis model not only provides structural response prediction capabilities, but also establishes a quantitative mapping relationship between "manufacturing-performance", providing a core data foundation for objective function construction, constraint setting, and parameter optimization in multi-objective collaborative optimization.

[0053] Step S400 includes the following steps: S410, geometrically restoring the topology optimization results to obtain a structural entity model with engineering identifiability; S420, constructing an analytical proxy model for multi-physical field performance prediction based on the structural entity model; S430, further jointly optimizing the shape boundary and geometric dimension parameters of the structure based on the constructed analytical model.

[0054] In this embodiment, step S400 involves the collaborative optimization of process and structure driven by a multi-objective optimization algorithm, aiming to achieve deep collaboration between structural performance and manufacturing processes based on the initial topology optimization model through geometric reconstruction, surrogate modeling, and parametric optimization means. First, in step S410, geometric restoration processing is performed on the topology optimization result to obtain a three-dimensional solid structure with engineering recognizability and modeling continuity. Since the output of topology optimization is usually a density distribution map or an irregular contour, it is necessary to convert it into a solid model that meets the modeling accuracy requirements through boundary extraction, smoothing processing, and surface reconstruction techniques. The present invention uses non-uniform rational B-spline (NURBS) functions to perform continuous surface fitting on the boundary region of the topology result, uses the point cloud density distribution to cut the unstructured region, and combines an automatic closing algorithm to generate a high-precision geometric model with a smooth contour and complete volume definition. This model has clear structural boundaries, connection surfaces, and functional unit contours, and is suitable for subsequent parametric analysis and engineering calls.

[0055] After completing the high-precision geometric modeling, step S420 is entered, and an analytical surrogate model for predicting the performance of multiple physical fields is constructed on the basis of the solid model. To improve the efficiency of structural optimization and avoid the computational burden of direct finite element simulation, the present invention introduces adaptive surrogate modeling methods such as response surface models and Kriging models, generates training samples in the design variable space through advanced sampling techniques such as Latin hypercube sampling and Sobol sequences, and fits the mapping relationship between design variables and structural responses based on the sample data. On this basis, an approximate model with fast prediction ability is constructed to evaluate the influence of different structural forms and process parameter combinations on structural performance. To enhance the engineering applicability of the model, template features of typical stator components are also introduced as basic units of the structural expression form during the modeling process to realize the integrated modeling of structural features and manufacturing rules.

[0056] After the analytical surrogate model is constructed, step S430 is executed to jointly optimize the structural outer shape boundary and geometric dimension parameters to achieve a collaborative improvement in performance, quality, and manufacturing efficiency. The present invention is based on a multi-objective optimization framework and introduces the compromise programming method to balance and adjust multiple optimization objectives. For example, taking the minimum structural mass, maximum stiffness, and minimum manufacturing time as objective functions, the optimal solution is selected by setting a multi-objective weighting function or Pareto front screening method. At the same time, the Pointer intelligent optimization method or a similar efficient algorithm based on the combination of gradient guidance and population evolution is used to perform iterative search in a large-scale variable space to improve the optimization speed and convergence efficiency. During the optimization process, structural variables such as wall thickness distribution, hole diameter size, and rib beam spacing, and process variables such as scanning speed, laser power, and powder spreading thickness are all input into the optimization solver as adjustable parameters. The optimization constraints include multi-point load conditions, boundary smoothness control, structural continuity maintenance, manufacturability of the manufacturing process window, and constraints on thermally stressed risk areas.

[0057] The compromise programming method is a multi-objective optimization strategy aimed at establishing a balanced relationship among multiple conflicting design goals. By setting weights, priorities, or satisfaction functions for each optimization goal, the compromise programming method can transform the original multi-objective problem into an equivalent single-objective problem or solve the Pareto optimal solution set, thereby achieving a reasonable trade-off among different dimensions such as performance, cost, and manufacturability, and is widely applied to the scheme selection and prioritization in complex engineering design. The Pointer intelligent optimization method is a multi-objective topology and parameter collaborative optimization method based on the swarm intelligence evolution mechanism. This method integrates intelligent optimization strategies such as particle swarm, genetic algorithm, and surrogate modeling, and realizes the global search and local convergence of design variables through multiple rounds of rapid iteration. It has strong adaptability and high search efficiency, and is particularly suitable for solving non-linear, multi-modal, and multi-constrained problems. The Pointer method can effectively improve the optimization efficiency and solution stability in the structure-process coupling design process, and has significant engineering practical value.

[0058] Step S410 includes the following steps: S411, using non-uniform B-spline basis functions to perform continuous surface reconstruction on the topology structure boundary to achieve high-light smoothness modeling of complex topology contours; S412, based on the point cloud density distribution of the topology result, executing an automatic clipping and boundary extraction algorithm to identify and separate the effective load path and redundant regions of the structure; S413, constructing the processed boundary data into a high-precision three-dimensional solid model to provide the basic input for subsequent parametric expression and analysis modeling.

[0059] In this embodiment, step S410 is used to perform geometric restoration processing on the topology optimization result to solve the problem that the initial model of topology optimization cannot be directly used for CAD modeling, simulation analysis, and engineering manufacturing. Since the topology optimization algorithm usually outputs a continuous density distribution or an approximate voxel grid, its boundary morphology often has problems such as irregularity, serrated edges, or topological discontinuity. Therefore, it is necessary to convert it into a high-precision three-dimensional structure model that can be recognized, analyzed, and manufactured through geometric reconstruction means.

[0060] In sub-step S411, non-uniform rational B-spline basis functions are used to perform continuous surface reconstruction on the topology structure boundary. This method can achieve high-order expression of complex boundary curvature through parameter space mapping, and has strong local controllability and fitting accuracy, and is suitable for modeling and smoothing the edge transition region of high-degree-of-freedom topology forms. In actual operation, first sample the boundary point set of the topology optimization result, extract the main path contour line and key cross-section points, then construct a control polygon mesh in the three-dimensional parameter domain, and use the NURBS surface interpolation method to generate a smooth and closed structure shape, ensuring that the model surface has continuous first-order derivatives and local adjustability in the main path direction, providing a geometric basis for subsequent structure identification and analysis modeling.

[0061] In sub-step S412, an automatic cropping and boundary extraction algorithm is executed based on the point cloud density distribution of the topology result to further clean and optimize the main structure contour. By setting a density threshold, low-density regions are identified and removed, separating redundant regions in the material distribution that have no bearing significance from the model. A local sub-region division method based on voxel grids or octrees is used to divide the topology region into blocks, and the structure boundary lines are extracted through gradient recognition technology. Combining with the main structure path recognition algorithm, the payload transfer regions in the structure can be further distinguished and automatically separated from the non-structural domains, achieving simultaneous improvement in geometric modeling accuracy and structural function recognition.

[0062] In sub-step S413, the processed boundary data is converted into a high-precision three-dimensional solid model. Voxel-recognizable entity geometry data is generated through 3D modeling software or an integrated CAD kernel, realizing the conversion process from point sets and surfaces to closed bodies. During the modeling process, the closure of the structure surface and the continuity of the boundary are ensured, eliminating geometric errors such as surface stitching gaps, duplicate surfaces, or normal anomalies. The generated solid model not only has the main structural form and force flow characteristics of the topology optimization result but also meets the geometric input requirements for finite element mesh generation, parametric feature extraction, and process adaptability evaluation, providing a complete and high-fidelity modeling basis for subsequent process-structure coupling analysis, multi-objective collaborative optimization, and manufacturing verification phases.

[0063] Step S420 includes the following steps: Step S421, using sampling techniques to effectively cover the design variable space; Step S422, introducing an optimization algorithm and an adaptive surrogate modeling method to establish an approximate expression of the variable and response relationship; Step S423, parametrically modeling the key features in the structure, expressing them in variable form and importing them into the optimization framework; Step S424, based on the parametric model, integrating the innovative structural feature templates of typical stator components to improve the functional adaptability and manufacturing rationality of the structure in specific application scenarios.

[0064] In this embodiment, step S420 is used to construct an analytical surrogate model for multi-physics field performance prediction based on a high-precision structural model, aiming to significantly reduce the computational cost of full-order finite element analysis while maintaining the accuracy of performance evaluation and providing a fast response ability for subsequent collaborative optimization. In step S421, to ensure that the design variable space is fully covered and representative sample data is obtained, advanced sampling techniques are used to construct a training sample set. Specific methods include Latin hypercube sampling, Sobol sequence sampling, or orthogonal experimental design, etc., which are used to uniformly select sample points in the multi-dimensional design variable space to avoid problems such as uneven sample distribution or aggregation imbalance. The sample points include structural geometric variables and process parameter variables, and the corresponding structural response results of each sample point, such as maximum stress, overall deformation, thermal gradient, displacement peak, etc., need to be obtained through simulation calculations.

[0065] In step S422, an optimization algorithm and an adaptive surrogate modeling method are introduced based on the above sample set to construct an approximate expression between the design variables and the performance responses. The surrogate models adopted include but are not limited to response surface models, Kriging models, radial basis function networks, or support vector regression, etc. During the model establishment process, the cross-validation method is used to evaluate the accuracy of the surrogate model, dynamically adjust the distribution of sample points and the complexity of the fitting function, and improve the generalization ability and prediction stability of the model. At the same time, global optimization means such as genetic algorithms and multi-objective particle swarm optimization algorithms are combined to perform rapid optimization and solution on the surrogate model, significantly reducing the computational resources and time overhead required for traditional finite element iterations.

[0066] In step S423, to make the optimization model have structural adjustability and manufacturing feasibility, parametric modeling is performed on the key features in the topological geometry model. This process first decomposes the structure model into physical forms, extracts typical design units, such as round holes, rib plates, stiffening ribs, shell segments, etc., and defines their variable parameters in forms such as dimensions, positions, and boundary contours. Subsequently, each feature variable is embedded into the optimization framework, enabling the surrogate model to achieve structural deformation, configuration adjustment, and function adaptation by directly adjusting the parameter values. This parametric modeling process ensures that the structural design has high controllability and modification flexibility, and is an important basis for realizing multiple rounds of optimization iterations and manufacturing adjustment feedback.

[0067] In step S424, further based on the parametric model, innovative structural feature templates of typical stator components are integrated to improve the functional adaptability and manufacturing rationality of the structure in specific application scenarios. The templates include rib-shell integrated structures for bearing radial / axial combined loads, multi-channel variable cross-section configurations with local strengthening functions, honeycomb-lattice hybrid filling structures that take into account thermal protection and weight reduction, etc. These structural templates are refined from historical design experience and process verification data, and have good mechanical properties and forming stability. They are integrated with the current optimization model by being embedded in the parametric features. The integration methods include technical routes such as geometric Boolean combination, structural area substitution, and local feature migration, ensuring the formability within the manufacturing window while realizing structural innovation.

[0068] Step S430 includes the following steps: S431, using the compromise programming method to make a coordinated balance of multiple optimization objectives of the structure; S432, applying the Pointer intelligent optimization method to perform multiple rounds of rapid iteration on the design variables to improve the optimization efficiency and convergence stability; S433, inputting multiple optimization constraint conditions, including: multi-point loading criterion, shape boundary control rule, and dimension parameter limit.

[0069] In this embodiment, step S430 is used to further jointly optimize the outer shape boundary and geometric dimension parameters of the structure based on the analysis agent model, so as to achieve an overall balance among structural performance, manufacturing efficiency, and process adaptability. In sub-step S431, the compromise programming method is first used to jointly weigh multiple optimization objectives. The optimization objectives in the present invention mainly include the structural lightweight objective, the stiffness and strength stability objective, the thermal stress control objective, the manufacturing time minimization objective, and the energy consumption or powder utilization rate maximization objective. By constructing a multi-objective function group and introducing a weighting coefficient or establishing a Pareto front set, in the forms of the weighted sum method, the ε-constraint method, or the fuzzy satisfaction function, etc., the multi-objective problem is transformed into an equivalent single-objective programming problem, so as to establish an optimization priority and adjustability among different performance indicators, and obtain the solution with the optimal overall performance.

[0070] In sub-step S432, to improve the solution efficiency and convergence stability of the structure-process multi-variable problem, the Pointer intelligent optimization method is introduced to perform multiple rounds of rapid iteration on the design variables. This optimization strategy integrates the swarm intelligence algorithm, local gradient-guided search, and dynamic adaptive learning mechanism, and has strong global search ability and anti-trap convergence performance. By performing population initialization on the initial distribution of the design variable space, the particle swarm, genetic algorithm, or differential evolution algorithm is used for global exploration, and the high-potential area is encrypted and iterated in combination with the local gradient path. At the same time, the response surface function output by the agent model is introduced as a fast evaluation sub-module, which is used to evaluate the change trend of the objective function and the solution space distribution state in each round of iteration, accelerate the convergence of the search path, and reduce the consumption of computing resources. During the optimization process, the search step size and mutation strategy are dynamically adjusted to achieve adaptive iteration depth control until the convergence tolerance or the maximum number of iterations is satisfied.

[0071] In sub-step S433, multiple optimization constraints are input to standardize the feasible region of the design solution, ensuring that the final result not only has excellent performance but also meets the requirements of engineering applications. The constraint conditions include but are not limited to the following three categories: First, the multi-point loading criterion requires that the optimized structure maintain the stiffness and strength requirements under multiple working conditions, and set response upper and lower limit intervals for key positions; second, the shape boundary control rule limits that the structure shape needs to maintain a certain geometric continuity, boundary smoothness, and no severe contour mutation to ensure processability and aesthetics; third, the dimension parameter limit clarifies the dimension range of each key structural unit, including wall thickness, rib height, hole diameter, minimum connection dimension, etc., to prevent the appearance of detailed structures smaller than the additive manufacturing resolution ability and ensure manufacturing feasibility. The above constraint conditions are embedded in the optimization solver in a mathematical form, and the legality of the design variables is screened and response penalties are imposed in each round of iteration to ensure the dual compliance of the final optimized solution in terms of structural performance and manufacturing implementation path.

[0072] In step S500, the iterative optimization of the process plan includes the following steps: S510, defining the requirements of the printed workpiece, including mechanical property requirements, geometric accuracy requirements, and functional integration. These requirements serve as the starting point of the process, driving the formulation of subsequent design and manufacturing strategies; S520, topology optimization, minimizing the material usage while meeting the structural performance goals, and preliminarily determining the material distribution and the layout form of components; S530, component layout, considering the placement method of parts on the printing platform, which affects residual stress, cooling path, and powder usage rate; S540, support structure design, including type, layout, and removability analysis; S550, setting the core parameters of the manufacturing process, including controlling the energy input, printing path, printing spacing, and layer thickness.

[0073] In this embodiment, step S500 realizes the multi-round iterative optimization of the structure and process plan through the physical manufacturing and response performance detection of representative feature parts, constructs a closed-loop control mechanism of structure design - manufacturing verification - feedback adjustment, and ensures that the final structure not only has the theoretically optimized performance but also has practical manufacturability and service reliability. In sub-step S510, first, the key application requirements of the workpiece to be printed are defined. This requirement mainly includes three categories of content: First, the mechanical property requirements, such as the safety factor under the maximum working load, the equivalent stress limit value, the fatigue life cycle, etc.; Second, the geometric accuracy requirements, such as dimensional tolerance, form and position tolerance, surface roughness grade, etc.; Third, the functional integration, such as whether it includes internal cooling channels, guide holes, connection thread structures, vibration damping areas, or socket joint structures, etc. The above requirements serve as the starting point of the entire design - manufacturing process, driving the selection of subsequent structural modeling paths and manufacturing strategies.

[0074] In sub-step S520, based on the above performance constraints and functional requirements, the topology optimization process is performed again to further minimize the material usage without reducing the mechanical and functional performance. This optimization process can combine the results of the previous analysis proxy model and the manufacturing adaptability knowledge base, dynamically adjust the volume constraint weight, load path setting, and manufacturing direction guiding function, so as to output a new round of more optimal material distribution results. After the optimization is completed, the material retention area, cavity distribution form, and rib orientation are preliminarily determined to form a preliminary layout form of components that takes into account both structural performance and printing rationality.

[0075] In sub-step S530, an optimized layout design is carried out around the spatial positioning method of components on the printing platform. This layout method has a direct impact on the residual stress of components, the distribution of the cooling path, and the powder stacking efficiency. Specifically, it includes factors such as component orientation, tilt angle, distance from the platform, and partition printing strategy. By simulating and analyzing the thermal field distribution, gravity deformation trend, and support dependence relationship in different postures, the optimal printing posture is selected to control the warping deformation caused by thermal gradient or support influence. At the same time, optimize the arrangement density of components on the platform to improve the printing batch efficiency and powder utilization rate.

[0076] In sub-step S540, process design is carried out around the type and layout of the support structure. The support design not only determines the forming quality of the component but also affects the post-processing operability and process safety. The types of support structures include various typical structures such as conical column type, tree type, multi-point type, and bridge type, and the selection needs to be adapted according to the local characteristics of the component. The support layout needs to take into account the force direction, heat dissipation path, minimum contact area, and removable path. By establishing an analysis model for support removability, the geometric accessibility, contact strength, and machining residue amount in the support area are evaluated to ensure that all support structures can be safely and efficiently removed after manufacturing, avoiding damage to the main structure or residual stress concentration.

[0077] In sub-step S550, the core parameters in the manufacturing process are set, including but not limited to laser power, scanning speed, scanning spacing, layer thickness, scanning strategy, powder laying thickness, and platform preheating temperature, etc. These parameters determine the energy input distribution, melt pool stability, interlayer fusion degree, and forming density. Based on the data in the previous process knowledge base and the results of structural response analysis in this embodiment, the above parameter combinations are dynamically adjusted to establish the mapping relationship between structure, process, and performance, and the physical test of printing characteristic parts is used to verify whether the manufacturing accuracy, surface quality, and internal integrity meet the design goals. If the verification result shows out-of-tolerance or defects, it will be automatically transmitted back to S520 topological model, S530 component layout, or S540 support setting for correction, forming a complete closed-loop optimization system.

[0078] In the specific implementation manner, a lightweight design method for a thin-walled structure based on additive manufacturing is used to design a certain type of aero-engine bearing housing.

[0079] 1) Initial structure design based on function-performance mapping: The bearing housing is a main load-bearing component of the aero-engine, and is simultaneously subjected to the actions of loads such as aerodynamic force, temperature, inertial force, and external rotor vibration. It has the characteristics of complex structural stress and important functions. The bearing housing is integrated with a lubricating oil supply oil path, a return oil path, a cooling air path, and an emergency oil path inside. A reliable bearing housing plays an important role in reducing the eccentric vibration of the shaft and ensuring the normal operation of the aero-engine.

[0080] According to the principles of gas dynamics and structural mechanics, the functional requirements are quantified into specific mechanical performance indicators: the maximum stress of the bearing housing under the ultimate working load shall not exceed 400 MPa, and the overall deformation shall not be greater than 0.3 mm.

[0081] With the goal of lightweighting, a topology optimization algorithm based on the constraints of additive manufacturing processes is adopted to perform topology optimization on the initial structure of the bearing housing while meeting the above performance constraints. Through multiple rounds of complex calculations, the materials in the non-critical stress-bearing areas inside the casing are removed, and a bearing housing model with an optimized topology structure is initially generated. Compared with the traditional design, the weight is reduced by 5%.

[0082] 2) Construction of manufacturing process knowledge base and process pre-selection: The structure of the bearing housing after topology optimization is complex and has extremely high requirements for dimensional accuracy, internal quality, and weight index. The main structure is a thin-walled stiffened structure: diamond-shaped weight-reducing grooves with different depths of 0.5 mm - 1 mm are arranged on the outer wall. The material of the bearing housing is selected as TC4, and the SLM process, which is widely used, has high forming accuracy, and stable TC4 printing performance, is selected as the main process.

[0083] 3) Process-structure two-way coupled finite element simulation analysis: For the additive manufacturing process, using the simufact finite element simulation software, a process-structure two-way coupled analysis model is established. Simulate the influence of process parameters such as laser scanning path, laser power, and scanning speed on the temperature field and stress field during the additive manufacturing process. Through two-way coupled simulation, it is found that during the manufacturing of the bearing housing by the additive manufacturing process, due to thermal stress concentration, small deformations may occur in some corner and thin-walled areas of the casing. 4) Process-structure collaborative optimization driven by multi-objective optimization algorithm: With the multi-objectives of lightweighting, optimal mechanical properties, and lowest manufacturing cost of the bearing housing, using advanced sampling techniques, efficient optimization algorithms, and adaptive surrogate models, the bearing housing collaborative optimization is completed based on the ANSYS software. For example, on the premise of ensuring the strength and stiffness of the bearing housing, by optimizing the wall thickness and rib beam width, the weight of the bearing housing is further reduced; at the same time, adjusting the laser power and scanning speed to reduce thermal stress concentration and the risk of bearing housing deformation, and improving the manufacturing efficiency and reducing the manufacturing cost by optimizing the layer thickness and scanning strategy. After multiple rounds of iterative calculations, a set of optimized design-process parameters is determined. The weight of the optimized bearing housing is reduced by 3% compared with the initial design, and at the same time, all mechanical property requirements are fully met. 5) Manufacturing verification of characteristic parts and iterative optimization of design-process solutions: Design typical characteristic parts, conduct physical manufacturing to verify the parameters optimized in the previous step. According to the verification results, fine-tune the design-process parameters and conduct virtual manufacturing verification again. After three rounds of iterative optimization, the virtual manufacturing verification results show that the bearing housing can be manufactured with high quality and high efficiency under the preset additive manufacturing process conditions, and all performance indicators far exceed the design requirements. 6) Generation of integrated design and process plan and quality control plan: After the manufacturing verification of characteristic parts and iterative optimization, a detailed integrated design and process plan for the aeroengine bearing housing is generated. The plan includes: (1)The optimized 3D model of the bearing housing, with detailed dimension markings for each part, and design parameters such as the layout, dimensions, and shape of the ribs.

[0084] (2)Precise additive manufacturing process parameters, for the solid area (laser power: 200 - 400W, scanning speed: 800 - 1200mm / s, spot diameter: 90 - 120μm, powder layer thickness: 60 - 90μm, platform temperature: 120 - 200°C); for the support area (laser power: 200 - 350W, scanning speed: 900 - 1350mm / s, spot diameter: 90 - 120μm, powder layer thickness: 60 - 90μm, platform temperature: 120 - 200°C).

[0085] (3)Product quality inspection methods and standards: such as in - furnace mechanical property inspection (GB / T228.1) and fluorescent inspection (GJB2367A).

[0086] Through actual manufacturing verification, the aero - engine bearing housing manufactured using the design method of the present invention has stable and reliable quality. Compared with traditional design and manufacturing methods, its structural weight is reduced by 8%, and the manufacturing efficiency is increased by 15%. It fully meets the high - performance requirements of aero - engines, providing strong support for the performance improvement and reliability enhancement of aero - engines. The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A lightweight design method for thin-walled structures based on additive manufacturing, characterized in that: It includes the following steps: S100. Design the initial structure based on the mapping of function and performance. Obtain the mechanical property indexes according to the functional requirements of the thin-walled structure, and based on the mechanical property indexes, adopt the topology optimization method to design the initial structure and obtain the initial lightweight structure model that meets the performance constraints; S200. Construct the additive manufacturing process knowledge base and evaluate the adaptability of the additive manufacturing process. Construct the additive manufacturing process knowledge base, and evaluate the process adaptability of the initial lightweight structure model according to the additive manufacturing process knowledge base, and identify the regions and features that do not meet the requirements of the additive manufacturing process; S300. Two-way coupled finite element analysis of process and structure. Establish a two-way coupled finite element simulation model between the structure model and the manufacturing process parameters, and analyze the influence of the process parameters on the structure performance and manufacturing quality; S400. Process and structure collaborative optimization driven by multi-objective optimization algorithm. Based on the multi-objective optimization algorithm, with the structure quality, performance stability and manufacturing efficiency as the goals, jointly optimize the structure parameters and process parameters; S500. Manufacture and verify the characteristic parts and iteratively optimize the design and process plan. Select the representative characteristic structure for physical manufacturing verification, evaluate whether the manufacturing defects, deformation degree and performance meet the design goals, and iteratively optimize the design parameters and process parameters according to the verification results until the requirements of the structure performance and manufacturing process are met; S600. Generate an integrated design and process plan, and the integrated plan includes design parameters and manufacturing parameters.

2. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 1, characterized in that: The topology optimization method in step S100 includes the following steps: S110. Digital characterization of additive manufacturing process constraints; S120. Force / thermal weak coupling multi-objective topology optimization; S130. Optimized design of structure stiffening and grille layout.

3. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 2, characterized in that: Step S110 includes the following steps: S111. Analyze and identify the process constraints that affect the forming power and structure quality during the additive manufacturing process, including self-supporting forming constraints and preparation process constraints; The self-supporting forming constraints include overhang angle constraints and printing direction constraints; the preparation process constraints include manufacturability constraints, machinability constraints and connectivity constraints; S112. Digitalize all constraint conditions.

4. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 3, characterized in that: Step S120 includes the following steps: S121. Determine two basic constraints: volume constraint and external interface constraint; S122. Construct the objective function and normalize it, and convert multiple design objectives into a unified optimization objective function through dimensionless normalization processing; S123. Select and apply the optimization algorithm to obtain the optimal topology structure and transmission path, and the optimization algorithm includes at least one of the return matrix method, the adjoint method and the MMA algorithm.

5. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 4, characterized in that: In step S200, the additive manufacturing process knowledge base includes: a process classification module, a typical feature structure module, a material matching information module, an additive manufacturing process parameter module, a post-processing and inspection and testing module.

6. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 5, wherein: Step S400 includes the following steps: S410, perform geometric restoration processing on the topology optimization result to obtain a structural entity model with engineering recognizability; S420, based on the structural entity model, construct an analysis surrogate model for multi-physical field performance prediction; S430, based on the constructed analysis surrogate model, further jointly optimize the outer shape boundary and geometric dimension parameters of the structure.

7. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 6, wherein: Step S410 includes the following steps: S411, use non-uniform B-spline basis functions to perform continuous surface reconstruction on the topology structure boundary to achieve high-smoothness modeling of complex topology contours; S412, based on the point cloud density distribution of the topology result, execute an automatic cropping and boundary extraction algorithm to identify and separate the effective load path and redundant area of the structure; S413, construct the processed boundary data into a high-precision three-dimensional entity model to provide basic input for subsequent parametric expression and analysis modeling.

8. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 7, wherein: Step S420 includes the following steps: Step S421, use sampling technology to effectively cover the design variable space; Step S422, introduce an optimization algorithm and an adaptive surrogate modeling method to establish an approximate expression of the variable and response relationship; Step S423, perform parametric modeling on the key features in the structure, express them in variable form and import them into the optimization framework; Step S424, based on the parametric model, integrate the innovative structure feature template of typical stator components to improve the functional adaptability and manufacturing rationality of the structure in specific application scenarios.

9. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 8, wherein: Step S430 includes the following steps, S431, use the compromise programming method to make a coordinated trade-off of multiple optimization objectives of the structure; S432, apply the Pointer intelligent optimization method to perform multiple rounds of rapid iteration on the design variables to improve the optimization efficiency and convergence stability; S433, input multiple optimization constraint conditions, including: multi-point loading criterion, shape boundary control rule, and dimension parameter limit.

10. The lightweight design method for thin-walled structures based on additive manufacturing according to claim 9, wherein: In step S500, iteratively optimize the design parameters and process parameters according to the verification result, specifically including the following steps: S510, clarify the requirements of the printed workpiece, including mechanical property requirements, geometric accuracy requirements, and functional integration, which serve as the starting point of the process and drive the formulation of subsequent design and manufacturing strategies; S520, Topological optimization, to minimize the material usage while meeting the structural performance goals, and preliminarily determine the material distribution and the layout form of components; S530, Component layout, considering the placement of parts on the printing platform, which affects residual stress, cooling path and powder utilization rate; S540, Support structure design, including type, layout and removability analysis; S550, Set the core parameters of the manufacturing process, including controlling the energy input, printing path, printing spacing and layer thickness.

Images