Selective laser melting stress deformation calculation method and device based on fluid-solid coupling

Abstract

The invention provides a selective laser melting stress deformation calculation method and device based on fluid-solid coupling, and relates to the technical field of additive manufacturing process simulation, the method can dynamically simulate formation and evolution of a molten pool in the laser scanning process by combining computational fluid dynamics (CFD) and finite element (FEM) technologies, and the deformation of the selective laser melting stress is calculated. And obtaining temperature and volume fraction (VOF) data related to the molten pool. Furthermore, a life-death unit coefficient in the FEM model is determined by using VOF data, a thermal-elastic-plastic model is constructed, and stress calculation is realized through a CFD-FEM data mapping algorithm. According to the method, the limitation of a traditional fixed heat source model is overcome, an efficient and universal calculation framework is provided for the forming process with the flow characteristic, the calculation precision is remarkably improved, and the method has great significance in optimizing the additive manufacturing process and improving the performance of a formed part. The method aims at improving the prediction precision of stress and deformation in a complex forming process.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing process simulation technology, specifically to a method and apparatus for calculating stress and deformation in laser selective melting based on fluid-structure interaction. Background Technology

[0002] With the rapid development of modern manufacturing technologies, additive manufacturing (AM), as an emerging manufacturing process, has gradually become an important component of industrial production. Unlike traditional subtractive manufacturing processes, additive manufacturing constructs objects by adding materials layer by layer, offering significant advantages such as high design freedom, less material waste, and shorter production cycles. This technology is particularly suitable for manufacturing complex structures, enabling high-performance and customized applications in fields such as aerospace, automotive, medical, and construction that are difficult to achieve with traditional processes.

[0003] Among numerous additive manufacturing technologies, Selective Laser Melting (SLM) has attracted considerable attention due to its high precision, high material utilization, and excellent mechanical properties. SLM technology uses a high-energy laser beam to melt metal powder layer by layer, forming complex three-dimensional structures. However, during the SLM forming process, the localized concentration of laser energy leads to significant temperature gradients and complex thermodynamic behaviors. This behavior not only causes stress concentration but may also result in defects such as porosity and cracks due to insufficient powder melting, severely affecting the mechanical properties and service life of the formed parts.

[0004] In the SLM forming process, the dynamic behavior of the molten pool and heat conduction play a crucial role in the stress distribution and deformation of the final formed part. However, traditional simulation methods typically employ simplified, fixed heat source models, assuming a constant laser energy distribution. This model neglects the complex interaction between the laser and the material, as well as the dynamically changing thermodynamic behavior during forming, resulting in low accuracy in predicting residual stress and deformation.

[0005] Furthermore, the influence of melt pool flow and interface changes during SLM forming on stress distribution cannot be ignored. Traditional methods often struggle to accurately capture these flow characteristics and interface changes, thus limiting the precise simulation of the forming process. Therefore, developing a computational method that comprehensively considers flow behavior, thermodynamic properties, and mechanical response is of great significance for improving the performance and reliability of SLM-formed parts.

[0006] In view of the above, this application is hereby submitted. Summary of the Invention

[0007] This invention provides a method and apparatus for calculating stress and deformation in laser selective melting based on fluid-structure interaction, which can at least partially improve the above-mentioned problems.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A method for calculating stress and deformation in laser selective melting based on fluid-structure interaction, comprising: Obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and establish the computational grids of the actual laser selective melting powder particle model, CFD model and FEM model based on the powder particle distribution state and solution scale. Based on the CFD model and the laser energy irradiation and reflection model, computational preprocessing is performed on the computational grid of the CFD model to obtain temperature evolution data and VOF values ​​at all unit nodes during the laser movement process. Based on the VOF values, the Gaussian integration points of all elements in the FEM model are iterated through to find the VOF values ​​of the corresponding computational grid of the CFD model, and the birth and death element coefficients are obtained. The temperature evolution data is optimized and obtained based on the birth and death element coefficients to obtain temperature values, and a thermo-elastoplastic model under the influence of the birth and death element coefficients is constructed based on the temperature values ​​and the birth and death element coefficients. By combining the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained. The laser selective data to be calculated is then transferred to the CFD-FEM stress calculation model to obtain the final stress and deformation calculation results.

[0009] The present invention also provides a laser selective melting stress-deformation calculation device based on fluid-structure interaction, comprising: The first model building unit is used to obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and to establish the actual laser selective melting powder particle model, the computational grid of the CFD model and the computational grid of the FEM model based on the powder particle distribution state and the solution scale. The fluid dynamics solution unit is used to perform computational preprocessing on the computational grid of the CFD model based on the CFD model and the laser energy irradiation and reflection model, to obtain the temperature evolution data and VOF values ​​of all unit nodes during the laser movement process; The birth and death element coefficient calculation unit is used to iterate through all Gaussian integration points of all elements in the FEM model based on the VOF value, find the VOF value of the corresponding computational grid of the CFD model, and obtain the birth and death element coefficients. The second model building unit is used to optimize and obtain the temperature evolution data according to the birth and death unit coefficients, obtain temperature values, and build a thermo-elastic-plastic model under the influence of the birth and death unit coefficients based on the temperature values ​​and the birth and death unit coefficients. The prediction unit combines the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm to obtain the CFD-FEM stress calculation model during the laser selective melting forming process, and transmits the laser selective data to be calculated to the CFD-FEM stress calculation model to obtain the final stress deformation calculation results.

[0010] In summary, the stress-deformation calculation method based on fluid-structure interaction (FSI) in laser selective melting aims to improve the prediction accuracy of stress and deformation in complex forming processes. It establishes computational meshes for both a computational fluid dynamics (CFD) model and a finite element method (FEM) model, depending on the solution scale. Based on the CFD model and the laser energy irradiation and reflection model, the dynamic small-hole molten pool formed during laser scanning is obtained, further enabling the acquisition of temperature data and VOF values ​​at each node in the CFD mesh. Based on the VOF values, the birth and death element coefficients of the Gaussian integral in the FEM model are obtained. Based on the birth and death element coefficients in the FEM model, the temperature data in the FEM model is obtained, and a thermo-elastoplastic model under the influence of the birth and death element coefficients is constructed. Based on the above thermo-elastoplastic model, combined with the aforementioned CFD-FEM interpolation algorithm, the CFD-FEM stress calculation model for the laser selective melting forming process is finally obtained. Compared with traditional methods, this method overcomes the limitations of fixed heat source models, providing a more accurate and universal calculation framework, significantly improving calculation accuracy, and providing strong technical support for optimizing SLM processes and improving the performance of formed parts. This provides a universal computational approach and data mapping framework for calculating thermal stress in forming processes with flow characteristics, improving computational accuracy within a limited computational grid. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating the laser selective melting stress and deformation calculation method based on fluid-structure interaction provided in the first embodiment of the present invention. Figure 2 This is a schematic diagram of the framework of the laser selective melting stress and deformation calculation method based on fluid-structure interaction provided in the first embodiment of the present invention; Figure 3 This is a schematic diagram of the mesh for the laser selective melting stress-deformation calculation method based on fluid-structure interaction provided in this embodiment of the invention. Figure 4 This is a schematic diagram of a module for a laser selective melting stress-deformation calculation device based on fluid-structure interaction provided in the second embodiment of the present invention. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0013] refer to Figure 1 , Figure 2 As shown, the first embodiment of the present invention discloses a method for calculating stress and deformation in laser selective melting based on fluid-structure interaction. This method can be executed by a laser selective melting stress and deformation calculation device based on fluid-structure interaction (hereinafter referred to as the calculation device), specifically by one or more processors within the calculation device, to achieve the following method: S1, obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and establish the calculation grid of the actual laser selective melting powder particle model, the CFD model and the FEM model according to the powder particle distribution state and the solution scale. Specifically, step S1 includes: obtaining the powder particle distribution state in the actual laser selective melting process to be calculated, and using a three-dimensional scanning method, a particle distribution assumption method, or a DEM simulation method to model the powder particle distribution state to obtain the actual laser selective melting powder particle model. Obtain the preset solution scale, and establish the computational grid of the CFD model and the computational grid of the FEM model according to the solution scale. The computational grid of the FEM model is a finite element grid that is uniform in all spaces. In the process of establishing the computational grid for the CFD model, speed optimization is achieved by using an octree grid, a finite difference grid, or a finite volume algorithm grid based on tetrahedrons.

[0014] In this embodiment, it is first necessary to obtain the powder particle distribution state and the preset solution scale in the actual laser selective melting process to be calculated. The powder particle distribution state is obtained using a three-dimensional scanning method, a particle distribution hypothesis method, or a discrete element method (DEM) simulation method. Each of these three methods has its advantages: the three-dimensional scanning method can directly obtain the accurate distribution of actual powder particles and is suitable for known powder bed conditions; the particle distribution hypothesis method can quickly generate the powder distribution based on experience or theoretical models and is suitable for preliminary simulations or theoretical studies; while the DEM simulation method can simulate the dynamic behavior of powder particles, providing results that more closely approximate the actual processing. Through these methods, a powder particle model of the actual laser selective melting process can be obtained, providing a foundation for subsequent calculations.

[0015] Next, based on the preset solution scale, computational meshes for the Computational Fluid Dynamics (CFD) model and the Finite Element Method (FEM) model are established respectively. The FEM model uses a uniform finite element mesh across all spaces. This mesh generation method ensures the uniformity and stability of the calculation, providing a reliable foundation for subsequent mechanical analysis. When establishing the CFD model's computational mesh, octree meshing, finite difference meshing, or a tetrahedral-based finite volume algorithm mesh can be selected for speed optimization. The octree meshing method can adaptively adjust the mesh density according to the distribution and complexity of powder particles, thereby improving computational efficiency while ensuring computational accuracy; the finite difference meshing method is suitable for simulating regular geometries due to its simplicity and efficiency; while the tetrahedral-based finite volume algorithm mesh performs well in handling complex geometries and flow characteristics, and can more accurately capture the details of molten pool flow and temperature distribution. Through these different mesh generation methods, the most suitable scheme can be flexibly selected according to specific process requirements and computational resources, thereby achieving efficient simulation of the laser selective melting process with limited computational resources.

[0016] According to step S1, not only can the powder particle model and computational mesh be accurately established, but the mesh generation method can also be flexibly selected according to different process requirements and computational conditions, thereby improving computational efficiency while ensuring computational accuracy. This flexible and efficient method lays a solid foundation for subsequent fluid-structure interaction calculations and stress-deformation analysis.

[0017] Please see Figure 3 S2, Based on the CFD model and the laser energy irradiation and reflection model, preprocessing is performed on the computational grid of the CFD model to obtain the temperature evolution data and VOF values ​​of all unit nodes during the laser movement process; Specifically, step S2 includes: calculating the dynamic small-aperture molten pool formed during laser scanning by performing calculations on the computational grid of the CFD model based on the CFD model and the laser energy irradiation and reflection model. The mathematical expressions for the heat transfer-convection and molten pool fluid motion model corresponding to the dynamic small-aperture molten pool are as follows: , , ,in, The velocity vector in the three-dimensional model. Density in a 3D model For pressure in a three-dimensional model, The dynamic viscosity in the three-dimensional model. This represents the gravity vector in the 3D model. This refers to the molten pool temperature in the 3D model. The external ambient temperature in the 3D model. The coefficient of thermal expansion in the three-dimensional model. This refers to the specific heat capacity in the three-dimensional model. The thermal conductivity coefficient in the three-dimensional model. The Kozeny–Carman constant in the three-dimensional model. For gradient, For time; Using an algorithm for handling the surface forces of a continuous equilibrium medium, the boundary conditions of the dynamic small-hole molten pool are processed based on the recoil pressure, surface tension, and Marangoni force at the interface, resulting in the final coupled forces. ,in, The surface tension coefficient, For the interface curvature, A unit vector on the metal surface. This represents the rate of change of the surface tension coefficient over time. For recoil pressure, The VOF value of the unit; Based on the heat transfer-convection and molten pool fluid motion model, and the final coupled forces, the CFD model is solved to obtain the temperature evolution data and VOF values ​​of all unit nodes during the laser movement process.

[0018] In this embodiment, the computational mesh of the CFD model is preprocessed based on the computational fluid dynamics (CFD) model and the laser energy irradiation and reflection model. The core objective of this process is to obtain the temperature evolution data and volume fraction (VOF) values ​​at all element nodes during laser movement. This step is crucial for subsequent stress-deformation calculations because it provides accurate temperature and flow information that directly affects the thermodynamic behavior and mechanical response during the forming process.

[0019] First, calculations are performed within the computational grid of the CFD model, based on the CFD model and the laser energy irradiation and reflection model. This calculation process simulates the dynamic keyhole molten pool formed during laser scanning. The formation of the dynamic keyhole molten pool is a key phenomenon in the SLM process, directly affecting the temperature distribution and flow characteristics of the molten pool. To accurately describe this phenomenon, this embodiment employs a heat transfer-convection and molten pool fluid motion model. This model describes the physical behavior within the molten pool through a series of mathematical expressions, including velocity vector, density, pressure, dynamic viscosity, gravity vector, molten pool temperature, ambient temperature, coefficient of thermal expansion, specific heat capacity, thermal conductivity, and the Kozeny-Carman constant (K). These parameters work together to determine the flow and heat transfer behavior within the molten pool.

[0020] In simulating dynamic small-hole molten pools, this method pays particular attention to the treatment of molten pool boundary conditions. The molten pool boundary is a key region where the molten pool interacts with its surrounding environment, and its behavior is influenced by various forces, including recoil pressure, surface tension, and Marangoni forces. To accurately simulate the coupling effects of these forces, an algorithm is employed to handle the surface forces of a balanced continuous medium. This algorithm comprehensively considers recoil pressure, surface tension, and Marangoni forces, transforming them into a final coupled force (F_total). In this way, the behavior of the molten pool boundary can be accurately described, thus providing more accurate boundary conditions for the dynamic simulation of the molten pool.

[0021] Finally, based on the aforementioned heat transfer-convection model and the final coupled forces, the CFD model is solved. This solution process yields temperature evolution data and VOF values ​​at all element nodes during laser movement. The temperature evolution data reflects the heat distribution within the molten pool, while the VOF values ​​describe the distribution of different phases (such as liquid metal and gaseous environment) within the molten pool. These data provide crucial input information for subsequent finite element analysis, enabling stress-deformation calculations to more accurately reflect the thermodynamic behavior during the actual forming process.

[0022] In addition, the Level-Set method can be used to track dynamic small-hole molten pool flow. Tracking dynamic small-hole molten pool flow mainly considers heat transfer-convection during laser scanning, fluid motion in the small-hole molten pool, and molten pool boundary conditions.

[0023] S3, based on the VOF values, perform a cyclic traversal of all Gaussian integration points in the FEM model to find the corresponding VOF values ​​of the computational grid of the CFD model, and obtain the birth and death element coefficients. Specifically, step S3 includes: using a coordinate search method to find the Gaussian integration point in the computational grid of each FEM model, finding the corresponding cell in the computational grid of the CFD model, and obtaining the birth and death cell coefficients of the Gaussian integration point in the FEM model. The computational grid of the FEM model is an unstructured grid containing multiple hexahedral cells in the first-order integral format. The birth and death coefficients of each hexahedral element are calculated from the values ​​of the eight Gaussian integration points within the element, using the following formula: , The number of unit nodes. Let be the VOF value at the i-th cell and j-th node in the FEM model.

[0024] In this embodiment, the Gaussian integration points of all elements in the finite element model are processed based on the volume fraction (VOF) values ​​to obtain the birth and death element coefficients. This step plays a crucial role in fluid-structure interaction calculations, closely linking the flow information of the dynamic molten pool in the CFD model with the mechanical analysis in the FEM model, thereby achieving accurate simulation of the forming process.

[0025] The first step involves iteratively traversing all Gaussian integration points in the FEM model. Since the FEM model's computational mesh is unstructured, containing multiple hexahedral elements in first-order integral schemes, this mesh format better adapts to complex geometries and flow boundaries, while also providing a foundation for accurately capturing the dynamic behavior of the melt pool. During the iterative traversal, a coordinate lookup method is used to locate the corresponding element in the CFD model's computational mesh for each Gaussian integration point. This method allows the acquisition of the VOF value for each Gaussian integration point, thereby determining its lifespan.

[0026] According to the formula for the birth and death element coefficients, for Gaussian integration points in an FEM element, when the VOF values ​​at all integration points are close to 1, it indicates that the element is completely inside the molten pool, i.e., within the free surface, and is a liquid or solid element, with a coefficient close to 1. When the VOF values ​​at all integration points are close to 0, it indicates that the element is completely outside the molten pool, i.e., outside the free surface, and is a gaseous element, with a coefficient close to 0. If the VOF values ​​at the integration points are between 0 and 1, it indicates that the element is located at the interface of the molten pool, and the coefficient is also between 0 and 1. This method of calculating the birth and death element coefficients based on VOF values ​​can accurately reflect the dynamic boundary changes of the molten pool, providing accurate input for subsequent mechanical analysis.

[0027] S4, optimize the temperature evolution data according to the birth and death unit coefficient to obtain the temperature value, and construct a thermo-elastic-plastic model under the influence of the birth and death unit coefficient based on the temperature value and the birth and death unit coefficient. Specifically, step S4 includes: optimizing the temperature evolution data based on the birth and death unit coefficients to obtain temperature values, using the following formula: , Let be the temperature value of the m-th node in the j-th element of the CFD model. For ambient temperature, This is the element interpolation function established based on hexahedral elements in the CFD model.

[0028] The thermo-elastoplastic model is an incremental thermo-elastoplastic model, and its mathematical expression is as follows: ,in, For the element's stress increment vector, For the strain increment vector, This is the vector of the linear expansion coefficients of the material. For plastic constitutive matrix, For the elastic constitutive matrix, It is the elastic modulus.

[0029] In this embodiment, the temperature evolution data is optimized based on the birth and death element coefficients, and a thermo-elastoplastic model influenced by these coefficients is constructed. This step is crucial in the entire computational framework, as it closely links flow characteristics and thermodynamic behavior with mechanical response, providing a foundation for accurately predicting stress and deformation during selective laser melting (SLM) forming.

[0030] The temperature evolution data obtained from the CFD model is optimized based on the birth and death element coefficients. Since the CFD model provides temperature evolution data for all element nodes during laser scanning, this data includes temperature information both inside and outside the molten pool. However, not all node temperature data directly contributes to the mechanical analysis. By filtering through the birth and death element coefficients, nodes inside the molten pool (liquid or solid) and outside the molten pool (gaseous) can be distinguished, thus optimizing the acquisition of temperature data.

[0031] Based on the formula for ambient temperature, by weighting the temperature data using the life and death element coefficients, it is ensured that only the temperature data inside the molten pool (liquid or solid state) is included in subsequent analysis, while the temperature data outside the molten pool (gaseous state) is ignored. Simply put, the temperature data includes both living elements (liquid / solid elements) and interface elements, while the impact of temperature changes on structural stress is not considered for dead elements (gaseous elements). This optimization method not only improves computational efficiency but also reduces errors caused by boundary effects, making the temperature data more consistent with the actual heat distribution during the forming process.

[0032] Based on the optimized temperature values ​​and the birth and death element coefficients, a thermo-elasto-plastic model was further constructed. This model is an incremental thermo-elasto-plastic model that comprehensively considers the thermal expansion, elastic deformation, and plastic deformation behavior of materials. Through this model, the thermal expansion effect caused by temperature changes can be combined with the elasto-plastic deformation behavior of materials, thus comprehensively considering the coupling effects of thermodynamics and mechanics during the forming process. The birth and death element coefficients play a crucial role in constructing the thermo-elasto-plastic model. They not only optimize the acquisition of temperature data but also ensure that the model can dynamically reflect the flow characteristics and heat distribution changes of the molten pool. In this way, the model can more accurately predict stress and deformation during the forming process, especially in critical areas such as the molten pool boundary and the heat-affected zone. This thermo-elasto-plastic model based on birth and death element coefficients not only improves computational accuracy but also significantly enhances adaptability to complex forming processes, providing strong technical support for optimizing SLM processes and improving the performance of formed parts.

[0033] S5. Combining the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained. The laser selective data to be calculated is then transferred to the CFD-FEM stress calculation model to obtain the final stress and deformation calculation results.

[0034] Specifically, step S5 includes: performing loading calculations sequentially according to a preset order within the computational framework of the thermo-elastoplastic model; Within the computational framework of thermo-elastic-plastic sequential coupling, and combined with the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained.

[0035] In this embodiment, the thermo-elastoplastic model is combined with the CFD-FEM interpolation algorithm to form a complete CFD-FEM stress calculation model for the laser selective melting forming process, and the final stress-deformation calculation results are obtained through this model. This step is the core of the entire calculation method, which integrates the coupling effect of fluid mechanics and solid mechanics through numerical calculation, thereby achieving accurate simulation of complex physical phenomena in the additive manufacturing process.

[0036] Specifically, within the computational framework of the thermo-elastic-plastic model, loading calculations are performed sequentially according to a preset order. This process considers the thermal expansion, elastic deformation, and plastic deformation behavior of the material during the forming process, as well as the interaction between these behaviors and temperature changes. By sequentially loading, the melting, solidification, and cooling processes of each layer of material during the forming process can be simulated step by step, thereby ensuring that the calculation results can accurately reflect the physical phenomena in the actual process.

[0037] Within the thermo-elastic-plastic sequential coupling computational framework, and combined with the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is further obtained. The CFD-FEM interpolation algorithm is a key technology connecting fluid dynamics (CFD) and finite element analysis (FEM). By transferring and converting data between the CFD and FEM models, it achieves dynamic capture of the molten pool flow characteristics and thermodynamic behavior, as well as accurate calculation of the stress and deformation of the formed part. This coupling algorithm considers not only the flow and temperature distribution within the molten pool but also the interaction between the molten pool and the surrounding materials, enabling the computational model to more comprehensively reflect the physical phenomena during the forming process.

[0038] By transmitting the laser selective area data to be calculated to the CFD-FEM stress calculation model, the final stress-deformation calculation results can be obtained. This process not only provides detailed information on the internal stress distribution of the formed part but also predicts possible deformation during the forming process. In this way, engineers can optimize the forming process during the design phase, reduce defects, and improve the performance and reliability of the formed part.

[0039] In summary, the proposed method for calculating stress and deformation in laser selective melting based on fluid-structure interaction (FSI) first establishes computational meshes for the powder particle model, CFD model, and FEM model based on the actual powder particle distribution and preset solution scale in the laser selective melting process. This process, through flexible selection of mesh generation methods (such as octree mesh, finite difference mesh, or finite volume algorithm mesh), provides an efficient and accurate foundation for subsequent FSI calculations. Subsequently, based on the CFD model and the laser energy irradiation and reflection model, the dynamic small-hole molten pool formed during laser scanning is calculated, and temperature evolution data and volume fraction of free flow (VOF) values ​​are obtained. By processing the equilibrium continuous medium surface force algorithm, the molten pool boundary conditions are further accurately processed, thus providing high-precision input for the simulation of molten pool flow and heat transfer.

[0040] In the FEM model, the introduction of birth and death element coefficients tightly couples the flow information and mechanical analysis from the CFD model. The calculation of these coefficients is based on VOF values, and precise capture of the dynamic boundary of the molten pool is achieved through coordinate lookup and calculation of Gaussian integral points. This process not only optimizes the acquisition of temperature data but also provides crucial input for the construction of the thermo-elasto-plastic model. The incremental thermo-elasto-plastic model further considers the thermal expansion, elastic deformation, and plastic deformation behavior of the material, enabling the calculation to comprehensively reflect the thermodynamic and mechanical coupling effects during the forming process.

[0041] Finally, the calculation results from fluid mechanics and solid mechanics are integrated using a CFD-FEM interpolation algorithm to form a complete CFD-FEM stress calculation model. This model can dynamically capture the flow characteristics and heat distribution of the molten pool, while accurately calculating the stress and deformation inside the formed part. By inputting the laser selection area data to be calculated into this model, the final stress and deformation calculation results can be obtained, providing strong support for process optimization and defect prevention.

[0042] The beneficial effect of this method lies in its ability to overcome the limitations of traditional fixed heat source models and provide a computational framework that can dynamically reflect the complex thermodynamic behavior and mechanical response in SLM processes. Through fluid-structure interaction technology and data mapping algorithms, this method significantly improves computational accuracy, particularly excelling in handling complex phenomena such as molten pool flow, heat transfer, and stress concentration. Furthermore, this invention provides an efficient and universally applicable solution for optimizing additive manufacturing processes and improving the performance of formed parts, possessing significant theoretical and practical application value.

[0043] Please see Figure 4 The second embodiment of the present invention provides a laser selective melting stress-deformation calculation device based on fluid-structure interaction, which includes: The first model building unit 101 is used to obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and to establish the actual laser selective melting powder particle model, the calculation grid of the CFD model and the calculation grid of the FEM model based on the powder particle distribution state and the solution scale. The fluid dynamics solution unit 102 is used to perform computational preprocessing on the computational grid of the CFD model based on the CFD model and the laser energy irradiation and reflection model, to obtain the temperature evolution data and VOF values ​​on all unit nodes during the laser movement process. The birth and death element coefficient calculation unit 103 is used to iterate through all Gaussian integration points of all elements in the FEM model based on the VOF value, find the VOF value of the corresponding computational grid of the CFD model, and obtain the birth and death element coefficients. The second model building unit 104 is used to optimize and obtain the temperature evolution data according to the birth and death unit coefficients, obtain temperature values, and build a thermo-elastic-plastic model under the influence of the birth and death unit coefficients according to the temperature values ​​and the birth and death unit coefficients. The prediction unit 105 is used to combine the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm to obtain the CFD-FEM stress calculation model in the laser selective melting forming process, and to transmit the laser selective data to be calculated to the CFD-FEM stress calculation model to obtain the final stress deformation calculation result.

[0044] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for calculating stress and deformation in laser selective melting based on fluid-structure interaction, characterized in that, include: Obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and establish the computational grids of the actual laser selective melting powder particle model, CFD model and FEM model based on the powder particle distribution state and solution scale. Based on the CFD model and the laser energy irradiation and reflection model, computational preprocessing is performed on the computational grid of the CFD model to obtain temperature evolution data and VOF values ​​at all unit nodes during the laser movement process. Based on the VOF values, the Gaussian integration points of all elements in the FEM model are iterated through to find the VOF values ​​of the corresponding computational grid of the CFD model, and the birth and death element coefficients are obtained. The temperature evolution data is optimized and obtained based on the birth and death element coefficients to obtain temperature values, and a thermo-elastoplastic model under the influence of the birth and death element coefficients is constructed based on the temperature values ​​and the birth and death element coefficients. By combining the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained. The laser selective data to be calculated is then transferred to the CFD-FEM stress calculation model to obtain the final stress and deformation calculation results.

2. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 1, characterized in that, The powder particle distribution and preset solution scale in the actual laser selective melting process to be calculated are obtained. Based on the powder particle distribution and solution scale, the computational meshes of the actual laser selective melting powder particle model, the CFD model, and the FEM model are established. Specifically: The distribution state of powder particles in the actual laser selective melting process to be calculated is obtained, and the distribution state of powder particles is modeled by three-dimensional scanning method, particle distribution assumption method or DEM simulation method to obtain the actual laser selective melting powder particle model. Obtain the preset solution scale, and establish the computational grid of the CFD model and the computational grid of the FEM model according to the solution scale. The computational grid of the FEM model is a finite element grid that is uniform in all spaces. In the process of establishing the computational grid for the CFD model, speed optimization is achieved by using an octree grid, a finite difference grid, or a finite volume algorithm grid based on tetrahedrons.

3. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 1, characterized in that, Based on the CFD model and the laser energy irradiation and reflection model, computational preprocessing was performed on the computational grid of the CFD model to obtain the temperature evolution data and VOF values ​​at all unit nodes during the laser movement process, specifically: Based on the CFD model and the laser energy irradiation and reflection model, calculations are performed within the CFD model's computational grid to obtain the dynamic small-aperture molten pool formed during laser scanning. The mathematical expressions for the heat transfer-convection and molten pool fluid motion model corresponding to the dynamic small-aperture molten pool are as follows: , , ,in, The velocity vector in the three-dimensional model. Density in a 3D model For pressure in a three-dimensional model, The dynamic viscosity in the three-dimensional model. This represents the gravity vector in the 3D model. This refers to the molten pool temperature in the 3D model. The external ambient temperature in the 3D model. The coefficient of thermal expansion in the three-dimensional model. This refers to the specific heat capacity in the three-dimensional model. The thermal conductivity coefficient in the three-dimensional model. The Kozeny–Carman constant in the three-dimensional model. For gradient, For time; Using an algorithm for handling the surface forces of a continuous equilibrium medium, the boundary conditions of the dynamic small-hole molten pool are processed based on the recoil pressure, surface tension, and Marangoni force at the interface, resulting in the final coupled forces. ,in, The surface tension coefficient, For the interface curvature, A unit vector on the metal surface. This represents the rate of change of the surface tension coefficient over time. For recoil pressure, The VOF value of the unit; Based on the heat transfer-convection and molten pool fluid motion model and the final coupled force, the CFD model is solved to obtain the temperature evolution data and VOF values ​​of all unit nodes during the laser movement process.

4. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 3, characterized in that, Based on the VOF values, the Gaussian integration points of all elements in the FEM model are iteratively traversed to find the corresponding VOF values ​​of the computational grid of the CFD model, thus obtaining the birth and death element coefficients, specifically: The coordinate lookup method is used to find the Gaussian integration points in the computational grid of each FEM model, find the corresponding cells in the computational grid of the CFD model, and obtain the birth and death cell coefficients of the Gaussian integration points in the FEM model. The computational grid of the FEM model is an unstructured grid containing multiple hexahedral cells in the first-order integral format. The birth and death coefficients of each hexahedral element are calculated from the values ​​of the eight Gaussian integration points within the element, using the following formula: , The number of unit nodes. Let be the VOF value at the i-th cell and j-th node in the FEM model.

5. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 1, characterized in that, The temperature evolution data is optimized and obtained based on the birth and death unit coefficients to obtain temperature values, specifically as follows: The temperature evolution data is optimized based on the birth and death unit coefficients to obtain the temperature value, using the following formula: , Let be the temperature value of the m-th node in the j-th element of the CFD model. For ambient temperature, This is the element interpolation function established based on hexahedral elements in the CFD model.

6. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 1, characterized in that, Combining the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained, specifically: Within the computational framework of the thermo-elastic-plastic model, loading calculations are performed sequentially according to a preset order; Within the computational framework of thermo-elastic-plastic sequential coupling, and combined with the CFD-FEM interpolation algorithm, a CFD-FEM stress calculation model for the laser selective melting forming process is obtained.

7. The method for calculating stress and deformation in laser selective melting based on fluid-structure interaction according to claim 4, characterized in that, The thermo-elastoplastic model is an incremental thermo-elastoplastic model, and its mathematical expression is as follows: ,in, For the element's stress increment vector, For the strain increment vector, This is the vector of the linear expansion coefficients of the material. For plastic constitutive matrix, For the elastic constitutive matrix, It is the elastic modulus.

8. A laser selective melting stress-deformation calculation device based on fluid-structure interaction, characterized in that, include: The first model building unit is used to obtain the powder particle distribution state and preset solution scale in the actual laser selective melting process to be calculated, and to establish the actual laser selective melting powder particle model, the computational grid of the CFD model and the computational grid of the FEM model based on the powder particle distribution state and the solution scale. The fluid dynamics solution unit is used to perform computational preprocessing on the computational grid of the CFD model based on the CFD model and the laser energy irradiation and reflection model, to obtain the temperature evolution data and VOF values ​​of all unit nodes during the laser movement process; The birth and death element coefficient calculation unit is used to iterate through all Gaussian integration points of all elements in the FEM model based on the VOF value, find the VOF value of the corresponding computational grid of the CFD model, and obtain the birth and death element coefficients. The second model building unit is used to optimize and obtain the temperature evolution data according to the birth and death unit coefficients, obtain temperature values, and build a thermo-elastic-plastic model under the influence of the birth and death unit coefficients based on the temperature values ​​and the birth and death unit coefficients. The prediction unit combines the thermo-elastic-plastic model and the CFD-FEM interpolation algorithm to obtain the CFD-FEM stress calculation model during the laser selective melting forming process, and transmits the laser selective data to be calculated to the CFD-FEM stress calculation model to obtain the final stress deformation calculation results.

Images