Vehicle lamp rear housing structure optimization method and vehicle lamp

Abstract

The application is suitable for the technical field of vehicle lamps, and particularly relates to a vehicle lamp rear shell structure optimization method and a vehicle lamp. The method comprises the following steps: performing topological optimization iteration on a design space of a vehicle lamp rear shell to be optimized to obtain a physical density field; obtaining a macroscopic rib topological configuration of the vehicle lamp rear shell to be optimized based on the physical density field; performing skeleton identification on the macroscopic rib topological configuration to construct a double-layer design variable space; obtaining global optimal design parameters of the vehicle lamp rear shell to be optimized based on the double-layer design variable space; and reconstructing and outputting a final three-dimensional digital model of the vehicle lamp rear shell to be optimized based on the global optimal design parameters. The vehicle lamp rear shell structure optimization method and the vehicle lamp provided in the application embodiment can improve the problem that the structure optimization of an existing lamp rear shell is difficult to integrate strict demolding directional geometric constraints in a differentiable form inside a gradient optimizer in a topological optimization stage.

Description

Technical Field

[0001] This application belongs to the field of automotive lighting technology, and particularly relates to a method for optimizing the rear housing structure of an automotive lighting headlight and an automotive lighting headlight. Background Technology

[0002] The rear housing of a car headlight is the main structural component in an automotive lighting system that supports the lamp cover, optical module, electronic components, and mounting bracket. It is typically manufactured using injection molding and is a typical thin-walled, complex cavity structure. Its inner surface is densely covered with a network of crisscrossing reinforcing ribs to meet mechanical performance requirements such as stiffness and vibration modes while controlling weight.

[0003] In existing technologies, structural optimization of the rear housing of automotive lights is usually performed in two independent steps: the first step is to generate a macroscopic layout of reinforcing ribs by using topology optimization techniques such as variable density method with unit pseudo density as design variable; the second step is for structural engineers to add draft angles, homogenize wall thickness and adjust cross-sectional dimensions of the reinforcing ribs output by topology optimization in CAD environment based on experience, so as to meet the demolding requirements of injection mold and molding process specifications.

[0004] However, the topology optimization stage may be difficult to integrate strict demolding directional geometric constraints in a differentiable form within the gradient optimizer, resulting in the output free configuration containing a large number of indentations, sudden changes in wall thickness, and other structures that cannot be injection molded. This requires manual reconstruction in the later stages. Furthermore, the layout (topology) of the stiffeners and the cross-sectional dimensions (parameters) are strongly coupled. Changes in the cross-sectional dimensions will alter the efficiency of the force transmission path and may even make some originally necessary stiffeners redundant. Summary of the Invention

[0005] This application provides a method for optimizing the structure of the rear housing of a vehicle headlight and a vehicle headlight, which can improve the problem that existing headlight rear housing structure optimization is difficult to integrate strict demolding directional geometric constraints in a differentiable form within the gradient optimizer during the topology optimization stage.

[0006] In a first aspect, embodiments of this application provide a method for optimizing the rear housing structure of a vehicle headlight, including: A topology optimization iteration with manufacturability constraints is performed on the design space of the rear housing of the headlight to be optimized. In each iteration step, a differentiable density field deformation mapping is applied to the test density field to obtain a physical density field that meets the requirements of undercut-free injection molding. The density field deformation mapping is used to force the topology optimization process to always search within the manufacturable feasible region, which is the range of the design space that meets the requirements of undercut-free injection molding. Based on the physical density field, the mechanical response is solved and the design variables are updated. After iterative convergence, the macroscopic reinforcing rib topology of the rear housing of the headlight to be optimized is obtained. The macroscopic stiffener topology is identified by skeletonization, and a two-layer design variable space containing topology layout variables and stiffener cross-sectional size variables is constructed. Based on the dual-layer design variable space, a combined proxy model is constructed to simultaneously fit the coupled influence of the two types of variables. The global optimal design parameters of the rear housing of the headlight to be optimized are obtained through a global optimization algorithm. The combined proxy model is used to characterize the influence of the coordinated changes in topology and cross-sectional dimensions on structural performance. Based on the globally optimal design parameters, the final three-dimensional digital model of the rear housing of the headlight to be optimized is reconstructed and output. The technical solutions described in this application embodiment have at least the following technical effects: The taillight rear housing structure optimization method provided in this application applies a differentiable density field deformation mapping to the experimental density field in each iteration of topology optimization. This integrates the injection molding non-undercut constraint into the iterative loop of the gradient optimizer in a differentiable form, forcing the optimization process to always search within the manufacturable feasible domain. This improves upon the problem that traditional topology optimization results may have unmanufacturable undercut structures that require subsequent manual repair. By constructing a two-layer design variable space and coupled combined proxy model that synchronously covers the topology layout and cross-sectional dimensions, the global collaborative optimization of the macroscopic force transmission path of the stiffener and the microscopic load-bearing cross-section is achieved. This improves upon the performance limitations of traditional serial optimization, which struggles to consider the coupling effects of layout and parameters. Ultimately, it can output a three-dimensional digital model of the taillight rear housing that simultaneously meets the requirements of stiffness, lightweight, and injection molding mass production.

[0007] Secondly, embodiments of this application provide a taillight rear housing structure optimization system, comprising: An optimization iteration unit is used to perform topology optimization iteration with manufacturability constraints on the design space of the rear housing of the headlight to be optimized. In each iteration step, a differentiable density field deformation mapping is applied to the test density field to obtain a physical density field that meets the requirements of undercut-free injection molding. The density field deformation mapping is used to force the topology optimization process to always search within the manufacturable feasible region, which is the range of the design space that meets the requirements of undercut-free injection molding. The topological configuration unit is used to complete the mechanical response solution and design variable update based on the physical density field, and obtain the macroscopic reinforcing rib topological configuration of the rear shell of the headlight to be optimized after iterative convergence. A spatial construction unit is used to perform skeletal identification of the macroscopic stiffener topology and construct a two-layer design variable space containing topology layout variables and stiffener cross-sectional size variables. The parameter design unit is used to construct a combined surrogate model that synchronously fits the coupled influence of two types of variables based on the two-layer design variable space, and to obtain the global optimal design parameters of the rear shell of the headlight to be optimized through a global optimization algorithm. The combined surrogate model is used to characterize the influence of the coordinated changes in topology and cross-sectional dimensions on structural performance. The output unit is used to reconstruct and output the final three-dimensional digital model of the rear housing of the headlight to be optimized based on the global optimal design parameters.

[0008] Thirdly, embodiments of this application provide a device for optimizing the structure of a vehicle headlight rear housing, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any one of the first aspects above.

[0009] Fourthly, embodiments of this application provide an automotive lamp, the automotive lamp including a rear housing, the structure of which is optimized by the method described in the first aspect.

[0010] It is understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a method for optimizing the rear housing structure of a vehicle headlight according to an embodiment of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0015] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0016] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0017] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0019] In related technologies, the rear housing of a vehicle headlight is the main structural component in an automotive lighting system used to support the lamp cover, optical module, electronic components, and mounting bracket. It is usually manufactured using injection molding and is a typical thin-walled complex cavity structure. Its inner surface is densely covered with a network of crisscrossing reinforcing ribs to meet mechanical performance requirements such as stiffness and vibration modes while controlling weight.

[0020] In existing technologies, structural optimization of the rear housing of automotive lights is usually performed in two independent steps: the first step is to generate a macroscopic layout of reinforcing ribs by using topology optimization techniques such as variable density method with unit pseudo density as design variable; the second step is for structural engineers to add draft angles, homogenize wall thickness and adjust cross-sectional dimensions of the reinforcing ribs output by topology optimization in CAD environment based on experience, so as to meet the demolding requirements of injection mold and molding process specifications.

[0021] However, the topology optimization stage may be difficult to integrate strict demolding directional geometric constraints in a differentiable form within the gradient optimizer, resulting in the output free configuration containing a large number of indentations, sudden changes in wall thickness, and other structures that cannot be injection molded. This requires manual reconstruction in the later stages. Furthermore, the layout (topology) of the stiffeners and the cross-sectional dimensions (parameters) are strongly coupled. Changes in the cross-sectional dimensions will alter the efficiency of the force transmission path and may even make some originally necessary stiffeners redundant.

[0022] To address the aforementioned issues, this application provides a method for optimizing the rear housing structure of a vehicle headlight and a vehicle headlight. In this method, a differentiable density field deformation mapping is applied to the experimental density field in each iteration of topology optimization. The injection molding non-undercut constraint is integrated into the iterative loop of the gradient optimizer in a differentiable form, forcing the optimization process to always search within the manufacturable feasible domain. This improves upon the problem that traditional topology optimization results may contain unmanufacturable undercut structures requiring subsequent manual repair. By constructing a two-layer design variable space and coupled surrogate model that synchronously covers the topological layout and cross-sectional dimensions, global collaborative optimization of the macroscopic force transmission path of the stiffeners and the microscopic load-bearing cross-section is achieved. This improves upon the performance limitations of traditional serial optimization, which struggles to simultaneously consider the effects of layout and parameter coupling. Ultimately, a three-dimensional digital model of the rear housing of the vehicle headlight that simultaneously meets the requirements of stiffness, lightweighting, and injection molding mass production can be output.

[0023] To better understand the method for optimizing the rear housing structure of vehicle lights provided in this application, the specific implementation process of the method for optimizing the rear housing structure of vehicle lights provided in this application will be described below by way of example.

[0024] Figure 1 This illustration shows a schematic flowchart of a method for optimizing the rear housing structure of a vehicle headlight according to an embodiment of this application. The method includes: S100, perform topology optimization iteration with manufacturability constraints on the design space of the rear housing of the headlight to be optimized. In each iteration step, apply a differentiable density field deformation mapping to the test density field to obtain a physical density field that meets the requirements of undercut-free injection molding. The density field deformation mapping is used to force the topology optimization process to always search within the manufacturable feasible region, which is the range of the design space that meets the requirements of undercut-free injection molding.

[0025] It is understandable that the rear housing of a vehicle headlight is an injection-molded structural component used in automotive lighting systems to support optical modules, mounting structures, and related electronic components. For example, the rear housing of a vehicle headlight can be the rear housing of an LED taillight or the rear housing of a front combination headlight for passenger vehicles.

[0026] The design space is a three-dimensional geometric region that allows for adjustments to material distribution and modifications to structural form during topology optimization. It can be divided into a design domain that allows for optimization and a non-design domain that requires the original structure to be retained. For example, the design space may include the main thin-walled area of ​​the rear housing of a vehicle headlight, the mounting boss area, etc.

[0027] Topology optimization iteration with manufacturability constraints is a numerical optimization process that aims to optimize the structural mechanical performance and uses injection molding manufacturability requirements as a constraint. It updates the distribution of structural materials through multiple rounds of iterative calculations. Each iteration simultaneously completes the calculation of structural performance, constraint verification, and update of design parameters.

[0028] The experimental density field is field data generated by the optimizer during the topology optimization iteration process. It is characterized by the pseudo density of the elements to represent the material distribution state in the design space. The pseudo density can range from 0 to 1 and is used for structural performance trial calculation and constraint condition verification during the iteration process.

[0029] Differentiable density field deformation mapping is a numerical operator that can convert an experimental density field into a physical density field that meets the demolding directionality requirements, and can also complete sensitivity transfer through the chain rule, thus adapting to the iterative calculation needs of gradient-based optimizers.

[0030] The requirement of undercut-free injection molding is a fundamental constraint on the structure of plastic parts in the injection molding process. It means that there are no concave or suspended structures in the plastic part along the mold opening direction that would interfere with the mold cavity. It is an important compliance indicator for the mold design and molding process of the rear housing of the headlight. For example, the draft structure of the reinforcing rib of the rear housing of the headlight can be designed with reference to the requirements of undercut-free injection molding, and the snap-fit ​​mounting structure of the housing can also meet the requirements of undercut-free injection molding, etc.

[0031] The physical density field is material distribution field data that meets the requirements of injection molding without undercuts after density field deformation mapping and can be used to calculate the mechanical properties of structures. It is the basis for outputting the structural configuration during the topology optimization iteration process.

[0032] Density field deformation mapping is a numerical processing method used to adjust the material distribution pattern of the experimental density field so that the topology optimization process is carried out within the range that meets the injection molding requirements. It can simultaneously realize the verification and adjustment of manufacturability constraints during the iteration process.

[0033] The manufacturable feasible region is the set of all structural design schemes that meet the requirements of injection molding without undercuts and can be used for mold design and injection molding during the topology optimization process. It is the limited range within which the optimizer searches for structural schemes.

[0034] The design space is the boundary of the three-dimensional geometric region that can be adjusted during the topology optimization process. It can be defined according to the structural function and installation requirements of the rear housing of the headlight. For example, the main internal cavity area of ​​the rear housing of the headlight can be included in the design space, while the outer contour edge area of ​​the housing can be excluded from the design space.

[0035] To obtain a physical density field that meets the requirements of undercut-free injection molding by applying a differentiable density field deformation mapping to the test density field, one method is to first divide the design space of the taillight rear housing into multiple continuous layers perpendicular to the demolding direction along the preset main demolding direction. Then, perform maximum value projection calculation layer by layer in the order from the parting surface to the mold opening direction, so that the physical density value of the current layer is the larger value between the test density value of the current layer and the physical density value of the adjacent layer directly below. This geometrically avoids the suspended concave structure along the demolding direction. For example, for the LED taillight rear housing with the main demolding direction being the positive Z-axis, its design space can be divided into 120 continuous slices along the Z-axis. Starting from the bottom layer where the parting surface is located, perform layer-by-layer projection calculation from bottom to top. The final physical density field generated meets the requirements of undercut-free injection molding along the Z-axis and is compatible with conventional two-platen molding. The molding requirements of injection molds can be addressed by first dividing the complex rear housing of a headlight with a side core-pulling structure into corresponding design sub-regions according to different demolding directions, then performing differentiable density field deformation mapping on each sub-region along the corresponding demolding direction, and finally superimposing and fusing the mapping results of each sub-region to obtain a physical density field that meets the requirements of no undercuts in multiple directions for the entire area. For example, for the rear housing of a headlight with a side wiring harness mounting port, its mold includes two demolding directions: Z-direction main demolding and X-direction side core-pulling. The main design sub-region corresponding to the Z-direction main demolding and the side design sub-region corresponding to the X-direction side core-pulling can be divided first, and mapping operations can be performed on each sub-region along the corresponding direction before fusing the results. The resulting physical density field meets the requirements of no undercut injection in both demolding directions, adapting to the molding requirements of complex injection molds with side core-pulling mechanisms.

[0036] By applying a differentiable density field deformation mapping to the experimental density field at each step of the topology optimization iteration with manufacturability constraints, a physical density field that meets the requirements of undercut-free injection molding is generated. The search range of topology optimization is limited by the density field deformation mapping, so that the optimization process can complete the iterative update of the structural scheme within the manufacturable domain, providing a constraint implementation path that adapts to the injection molding process requirements for the topology optimization process of the taillight rear shell.

[0037] In one possible implementation, S100, a differentiable density field deformation mapping is applied to the test density field to obtain a physical density field that satisfies the requirements of undercut-free injection molding, including: S110, along the preset main demolding direction, divides the design space of the rear housing of the headlight to be optimized into multiple continuous layers perpendicular to the main demolding direction.

[0038] It can be understood that the main demolding direction is the reference movement direction in which the moving mold and the fixed mold complete the separation of the main body during the mold opening process of the injection mold, and it is also the guiding direction for the smooth demolding of the main structure of the plastic part.

[0039] The rear housing of the headlight to be optimized refers to the initial design model of the rear housing of the headlight that has not yet completed the optimization of structural performance and molding process and needs to be adjusted through numerical optimization methods to adjust the internal structural layout.

[0040] Continuous layering refers to a series of parallel thin-layer structures that are obtained by dividing the three-dimensional design space along a specified direction without gaps or overlaps between each other. It is the basic unit for realizing layer-by-layer density field mapping calculation. For example, continuous layering can refer to 120 parallel thin layers obtained by dividing the design space of the rear housing of the headlight along the main demolding direction, or 30 parallel thin layers obtained by dividing the design space of the lateral structure along the side core pulling direction, etc.

[0041] By orderly dividing the design space of the rear housing of the headlight to be optimized along the preset main demolding direction, and generating multiple continuous layers perpendicular to the main demolding direction, a layered processing unit that conforms to the demolding direction is provided for the subsequent density field mapping calculation that adapts to the requirements of injection molding without undercuts. This provides a basic geometric calculation framework for embedding injection molding process constraints during topology optimization.

[0042] S120, Perform maximum value projection calculation layer by layer in the order from the parting surface to the main demolding direction; wherein, the physical density value of the current layer is taken as the larger value between the current layer test density value and the physical density value of the adjacent layer directly below.

[0043] It is understandable that the parting surface is the contact surface between the moving mold and the fixed mold in the injection mold, where they fit together and separate when the mold opens. It is also the interface between the plastic part and the molded part. For example, the parting surface can be set at the maximum projection surface of the outer contour of the shell, or at the mating surface between the lamp cover and the rear shell, etc.

[0044] Maximum projection calculation is a numerical calculation method used in topology optimization to adjust the material density distribution within layers and make the structure conform to the molding requirements along the demolding direction. It can impose directional constraints on the material distribution by comparing and assigning density values.

[0045] The current layer is the continuous layer that is performing density value calculation and assignment operations during the layer-by-layer operation. It is the processing object of each projection operation. For example, if there are 120 continuous layers divided along the main demolding direction, and the operation of the 30th layer is being performed, the 30th layer is the current layer.

[0046] The adjacent layer directly below is the previous continuous layer in the layer sequence along the main demolding direction, located on the side of the current layer closest to the parting surface and adjacent to the current layer without gaps. It is the reference object for assigning the density value of the current layer. For example, if the current layer is the 30th layer divided along the main demolding direction, the 29th layer is the adjacent layer directly below.

[0047] The method of performing maximum value projection calculations layer by layer, following the order from the parting surface towards the main demolding direction, can be as follows: First, divide the design space of the taillight rear housing to be optimized into continuous layers with uniform step lengths along the main demolding direction. Then, strictly follow the order from the parting surface towards the main demolding direction, and perform maximum value projection calculations for each layer sequentially. Each step of the calculation is based on the experimental density value of the current layer and the final physical density value of the adjacent layer directly below. For example, for the taillight rear housing of a passenger vehicle with a regular structure, the design space can be divided into 100 continuous layers with uniform step lengths along the main demolding direction. Starting from the first layer where the parting surface is located, perform maximum value projection calculations for the second to the 100th layers sequentially. The maximum value projection calculation with equal step length can be performed. Alternatively, based on the preset minimum draft angle, the design space of the headlight rear housing to be optimized can be divided into continuous layers with step length adjusted according to the draft slope along the main demolding direction. Then, the maximum value projection calculation can be performed layer by layer in the order from the parting surface to the main demolding direction. During the calculation, the physical density field of the adjacent layer directly below can be laterally expanded to adapt to the draft angle. For example, for the headlight rear housing with a complex curved surface, the design space can be divided into 85 continuous layers with variable step length along the main demolding direction according to the minimum draft angle requirement of 1°. The maximum value projection calculation with lateral expansion compensation can be performed layer by layer starting from the parting surface.

[0048] By performing an ordered traversal of consecutive layers from the parting line towards the main demolding direction, and by performing maximum value projection calculations layer by layer and taking the physical density value of the current layer as the larger of the experimental density value of the current layer and the physical density value of the adjacent layer directly below, a directional constraint can be applied to the material distribution in the topology optimization process. This allows the generated physical density field to adapt to the demolding requirements of injection molding, providing a feasible computational path for embedding manufacturability constraints into the topology optimization iteration process. In one possible implementation, S120 involves performing the maximum value projection operation layer by layer, including: S121, based on the preset minimum draft angle, determines the lateral outward offset corresponding to each layer.

[0049] It is understandable that the minimum draft angle is the minimum tilt angle preset on the molding surface where the part contacts the mold to ensure that the injection molded part can be smoothly ejected from the mold cavity. For example, for the rear housing of the taillight of passenger car made of glass fiber reinforced PP material, the preset minimum draft angle can be set to 1°, and for the rear housing of the headlight of commercial vehicle made of ABS material, the preset minimum draft angle can be set to 0.5°, etc.

[0050] The lateral outward offset is the distance that the solid material area of ​​each layer needs to expand outward relative to the adjacent layer directly below in order to meet the preset minimum draft angle requirement in a plane perpendicular to the main demolding direction. For example, for a layer with a layer height of 0.5mm along the main demolding direction, the lateral outward offset corresponding to a preset minimum draft angle of 1° can be 0.0087mm, and for a layer with a layer height of 1mm along the main demolding direction, the lateral outward offset corresponding to a preset minimum draft angle of 1.5° can be 0.0262mm, etc.

[0051] Based on a preset minimum draft angle, the method for determining the lateral expansion offset of each layer can be as follows: First, determine the uniform fixed layer height of all continuous layers along the main demolding direction. Then, using trigonometric functions, calculate the lateral expansion offset of each layer relative to the adjacent layer directly below using the formula: "Single-layer lateral expansion offset = Fixed layer height × tan(preset minimum draft angle)". Simultaneously, the total lateral expansion offset of each layer relative to the parting surface reference layer can be accumulated. For example, for a structurally regular passenger vehicle LED taillight rear housing, with a preset minimum draft angle of 1° and a fixed layer height of 0.5mm along the main demolding direction, the lateral expansion offset of each layer can be calculated to be approximately 0.0087mm. Simultaneously, the total lateral expansion offset of the 100th layer relative to the parting surface reference layer can be accumulated to approximately... 0.87mm; Alternatively, based on the curvature variation of the rear housing of the headlight to be optimized, continuous layers with inconsistent layer heights are divided along the main demolding direction. Then, the basic offset is calculated by combining the preset minimum draft angle with the actual layer height of the corresponding layer. At the same time, the basic offset is compensated and corrected according to the molding shrinkage rate of the injection molding material used in the housing. Finally, the lateral expansion offset corresponding to each layer is obtained. For example, for the rear housing of the headlight of a passenger vehicle with a complex free-form surface, the preset minimum draft angle is 1°, and the molding shrinkage rate of the PP-TD40 material used is 1.2%. For a layer with a layer height of 1mm in a gently curved area, the compensated lateral expansion offset can be calculated to be approximately 0.0176mm. For a layer with a layer height of 0.2mm in a steeply curved area, the compensated lateral expansion offset can be calculated to be approximately 0.0035mm, etc.

[0052] By using the preset minimum draft angle as the core calculation basis and combining the layer height parameters of each layer with the geometric relationship of the main demolding direction, the lateral outward offset of each layer can be determined. This provides quantitative reference parameters for the subsequent morphological adjustment of the density field and provides a numerical basis for the process requirement of simultaneously adapting the minimum draft angle during topology optimization.

[0053] S122, Perform continuously differentiable morphological dilation processing on the physical density field of the adjacent layer directly below according to the lateral outward offset, and use the processed physical density value for the maximum value projection calculation of the current layer.

[0054] It is understandable that continuously differentiable morphological expansion processing is a numerical computation method adapted to topology optimization of continuous density fields. It can achieve directional expansion of the material distribution area along a specified direction without destroying the continuity of the density field gradient. At the same time, it can complete sensitivity backpropagation through the chain rule, adapting to the iterative calculation requirements of gradient-type optimizers. For example, for the physical density field of a certain layer of the rear shell of a car headlight, this processing can be performed according to the preset lateral expansion offset to achieve smooth directional expansion of the material area. For a layered physical density field with multiple reinforcing ribs, this processing can complete the synchronous expansion adjustment of all reinforcing rib areas while preserving the differentiability of the density field.

[0055] The processed physical density value refers to the updated density value corresponding to each computational unit within the layer after the physical density field has undergone continuous differentiable morphological dilation processing. It is the reference data for performing the maximum value projection operation in the current layer.

[0056] By performing a continuously differentiable morphological expansion process on the physical density field of the adjacent layer directly below according to the lateral outward offset, and by using the processed physical density value for the maximum value projection calculation of the current layer, reference density data that adapts to the minimum draft angle requirement can be provided for the maximum value projection calculation. While ensuring the continuity of the density field gradient to adapt to the iterative requirements of the optimizer, the material distribution generated during the topology optimization process can conform to the preset draft process requirements, providing a feasible computational link for embedding injection molding manufacturability constraints into the topology optimization iteration process.

[0057] S130 generates a physical density field without suspended structures or inverted loops, based on the calculation results of all layers.

[0058] It can be understood that the calculation result refers to the physical density assignment and material distribution data of each unit in each layer after performing operations such as maximum value projection calculation and morphological expansion processing on each continuous layer of the design space of the taillight rear housing to be optimized. For example, the calculation result can be the density distribution data of each layer after performing the maximum value projection calculation on each layer of the 100 continuous layers of the design space of the LED taillight rear housing; or it can be the density assignment data of each layer after performing the projection calculation with draft angle compensation on the 85 continuous layers of the design space of the headlight rear housing.

[0059] A non-suspended structure refers to a solid material area of ​​the plastic part that, when viewed along the main demolding direction, does not have a suspended solid structure that is not continuously supported below and is prone to connecting with the mold cavity.

[0060] "No undercut" refers to a plastic part structure that does not have an inward or concave shape along the main demolding direction, and will not interfere with the mold cavity during the mold opening process, allowing for smooth demolding.

[0061] The physical density field refers to the continuous material distribution data field that covers the entire design space of the rear housing of the headlight to be optimized, obtained after integrating all the results of the layered calculations. It is the basis for calculating mechanical properties during the topology optimization process. For example, the physical density field can be the density distribution data of the entire design space obtained by integrating the rear housing of the taillight after layer-by-layer projection calculations; or it can be the density distribution data of the entire design space obtained by integrating the rear housing of the headlight after multi-directional constraint calculations, etc.

[0062] Based on the calculation results of all layers, the method to generate a physical density field without suspended structures or undercuts can be as follows: First, according to the parting surface towards the main demolding direction, the calculation results of all layers are sequentially spliced ​​together according to their spatial positions. During the splicing process, the density continuity and structural compliance of adjacent layers are simultaneously verified. Finally, a physical density field without suspended structures or undercuts is generated throughout the entire design space. For example, for the rear housing of a passenger vehicle LED taillight with a regular structure and only a single main demolding direction, the calculation results of its 100 consecutive layers can be spliced ​​together from bottom to top, simultaneously verifying whether there is a risk of suspension or undercuts in the structure of each layer, ultimately generating a physical density field that meets the injection molding requirements. Alternatively, it can be done by first... The design sub-regions are divided according to different demolding directions. The calculation results of all layers in each sub-region are integrated to generate a sub-region density field. Then, the boundary positions of the density fields of each sub-region are smoothed and verified for compliance. Finally, the physical density field of the whole region is merged to generate a physical density field with no suspended structure and no undercut. For example, for the rear housing of the front headlight of a passenger vehicle with a side wiring harness installation port and including two demolding directions, namely Z-direction main demolding and X-direction side core pulling, the layered calculation results of the sub-regions corresponding to the two demolding directions can be integrated to generate a sub-region density field. Then, the boundary positions of the two sub-regions are smoothed and verified for undercut. Finally, a physical density field with no suspended structure and no undercut is generated to adapt to complex side core pulling molds.

[0063] By continuously stitching together the density data of each layer and verifying structural compliance, a physical density field without suspended structures or inverted fasteners is generated. This provides a material distribution basis that meets the requirements of injection molding process for solving the mechanical response during the topology optimization iteration process, and also provides a compliant reference for the subsequent update and iteration of design variables.

[0064] S200, based on the physical density field, completes the mechanical response solution and design variable update, and after iterative convergence, obtains the macroscopic reinforcing rib topology of the rear shell of the headlight to be optimized.

[0065] It is understandable that solving the mechanical response is a process based on the finite element numerical simulation method, which involves numerically calculating the three-dimensional structure corresponding to the physical density field to obtain the relevant parameters of the structure's mechanical properties.

[0066] Design variable update is the process of numerically adjusting the design variables that characterize material distribution through optimization algorithms based on the mechanical response solution results and preset constraints during topology optimization iteration.

[0067] Iterative convergence is a state in topology optimization where the objective function change rate and constraint violation rate are both below a preset threshold for multiple consecutive iterations, indicating that the optimization process meets the stopping condition. For example, in the topology optimization of the taillight rear housing, iterative convergence is determined when the structural flexibility change rate is below 0.1% for 5 consecutive iterations and all preset constraints are met. Similarly, in the topology optimization of the headlight rear housing, iterative convergence is determined when the volume target deviation is below 0.05% for 8 consecutive iterations and modal constraints are met.

[0068] The macroscopic reinforcing rib topology configuration is extracted from the physical density field that meets the requirements after topology optimization iteration convergence. It refers to the overall layout and force transmission path structure of the reinforcing ribs in the rear housing of the headlight used to improve the structural mechanical performance. For example, after the iteration convergence of the LED taillight rear housing, the main and auxiliary rib combination layout structure connecting each mounting point can be extracted as the macroscopic reinforcing rib topology configuration; after the iteration convergence of the headlight rear housing, the grid-like force transmission path structure adapted to the cavity shape can be extracted as the macroscopic reinforcing rib topology configuration, etc.

[0069] The method of obtaining the macroscopic stiffener topology of the taillight rear housing after iterative convergence by solving the mechanical response and updating design variables based on the physical density field can be as follows: Minimize structural flexibility as the core optimization objective. In each iteration, the static mechanical response is solved based on the physical density field. The sensitivity of the objective function and constraints to the design variables is calculated using the adjoint method. Then, the design variables are updated using a gradient-based optimization algorithm. Iterative convergence is determined when the change in the objective function is below a preset threshold and all constraints are satisfied after multiple consecutive iterations. Finally, the macroscopic stiffener topology of the taillight rear housing is extracted from the converged physical density field. For example, for the LED taillight rear housing, minimizing structural flexibility is the objective. In each iteration, the static response under Z-axis mounting load is solved based on the physical density field. The design variables are updated using the moving asymptote method. Convergence is determined when the flexibility change rate is below 0.1% for five consecutive iterations and the volume and modal constraints are satisfied. Finally, the topology connecting the four mounting points is extracted. The topological configuration of macro-reinforcing ribs can be a combination of main and auxiliary ribs. Alternatively, it can be based on multiple optimization objectives, such as minimizing structural flexibility and volume. In each iteration, static and modal mechanical response solutions are solved simultaneously based on the physical density field. The priority of multiple objectives is adjusted by adaptive weight coefficients, and design variables are updated by combining sensitivity analysis results. When the change in the comprehensive evaluation index of multiple objectives in consecutive iterations is lower than a preset threshold, the iteration is considered to have converged. Finally, the macro-reinforcing rib topological configuration of the rear housing of the headlight to be optimized is extracted from the converged physical density field. For example, for the rear housing of the headlight, with the dual objectives of minimizing flexibility and volume, static response and the first three natural modal frequencies are solved simultaneously based on the physical density field in each iteration. The priority of objectives is adjusted by adaptive weights to update design variables. When the change rate of the comprehensive evaluation index of multiple objectives in eight consecutive iterations is lower than 0.15% and all constraints are satisfied, convergence is determined. Finally, a grid-like macro-reinforcing rib topological configuration adapted to large-span cavities is extracted.

[0070] By solving the mechanical response based on the physical density field that meets the manufacturability requirements of injection molding, the relevant parameters of the mechanical properties of the structure are obtained, providing a numerical basis for adjusting the design variables. By combining the mechanical response solution results with preset constraints, the design variables are updated, driving multiple rounds of iterative topology optimization. After the iteration converges, the macroscopic stiffener topology of the rear housing of the headlight to be optimized is obtained. This provides a basic framework for the structural optimization of the rear housing of the headlight that meets both mechanical performance requirements and injection molding process requirements. It also provides a prerequisite for the subsequent fine optimization of the stiffener cross-sectional parameters.

[0071] In one possible implementation, before S200, which involves solving for the mechanical response and updating the design variables based on the physical density field, the following steps are also included: S200a, based on the volume difference ratio between the physical density field and the density field after corrosion, completes the quantitative evaluation of the minimum wall thickness constraint. The volume difference ratio is used to characterize the proportion of structures smaller than the minimum wall thickness in the physical density field.

[0072] It can be understood that the post-corrosion density field is the density field data obtained after performing a continuously differentiable morphological corrosion operation on the physical density field. It can be used to identify structural regions in the physical density field that are smaller than the preset minimum wall thickness. For example, for the physical density field of the taillight rear housing, the density field obtained after performing a morphological corrosion operation with a corrosion radius of 1 mm can be used as the post-corrosion density field; for the physical density field of the headlight rear housing, the density field obtained after performing a morphological corrosion operation with a corrosion radius of 2 mm can also be used as the post-corrosion density field, and so on.

[0073] The volume difference ratio is the difference between the total volume of the physical density field and the total volume of the density field after corrosion. Dividing it by the total volume of the physical density field gives the proportion of structures in the physical density field that do not meet the minimum wall thickness requirement. For example, if the volume difference ratio between the physical density field and the density field after corrosion of the taillight rear shell is 3%, it means that the corresponding proportion of the structure may be smaller than the preset minimum wall thickness.

[0074] Minimum wall thickness constraint is a limitation on the minimum thickness of the plastic part structure in the injection molding process to ensure the filling quality and structural strength of the plastic part. For example, for the rear housing of a car headlight made of PP-TD40 material, a minimum wall thickness constraint of 2.0 mm can be set; for the rear housing of a car headlight made of ABS material, a minimum wall thickness constraint of 1.5 mm can be set, etc.

[0075] Quantitative assessment is a quantitative evaluation process that uses numerical indicators to assess whether a structure meets preset constraints. It can transform the compliance of process constraints into calculable and iterative numerical indicators. For example, quantitative assessment can be a quantitative evaluation of the minimum wall thickness compliance of the taillight rear housing by the volume difference ratio; or it can be a quantitative evaluation of the achievement of the weight reduction target of the taillight rear housing by the volume ratio.

[0076] The quantitative assessment of minimum wall thickness constraints based on the volume difference ratio between the physical density field and the post-corrosion density field can be achieved by first setting a uniform minimum wall thickness value and a corresponding allowable threshold for the volume difference ratio for the entire design space of the rear housing of the headlight to be optimized. Then, the volume difference ratio between the physical density field and the corresponding post-corrosion density field in the entire design space is calculated. The calculated result is compared with the preset allowable threshold to complete the quantitative assessment of the minimum wall thickness constraint for the entire structure. For example, for the rear housing of a passenger vehicle LED taillight with a regular structure, a globally uniform minimum wall thickness of 2.0 mm is set, and the allowable threshold for the volume difference ratio is 1%. The calculated volume difference ratio for the entire design space is 0.6%, which can be used to determine that the minimum wall thickness of the current structure meets the preset requirements through quantitative assessment. Alternatively, one approach can be to first determine the minimum wall thickness constraint based on the volume difference ratio between the physical density field and the post-corrosion density field in the entire design space. The structural function and mold filling path of the rear housing of the headlight are optimized. The design space is divided into multiple sub-regions. For each sub-region, a corresponding minimum wall thickness value and allowable threshold for volume difference ratio are set. Then, the volume difference ratio of the physical density field and the corresponding post-corrosion density field in each sub-region is calculated. The minimum wall thickness constraint is quantitatively evaluated by region. For example, for the rear housing of the headlight of a passenger vehicle with a complex installation structure, the minimum wall thickness of the area around the mounting point is set to 2.5mm and the allowable threshold for volume difference ratio is 0.5%. The minimum wall thickness of the thin-walled area of ​​the main body of the housing is set to 1.8mm and the allowable threshold for volume difference ratio is 1.2%. The volume difference ratio of the two sub-regions is calculated and compared with the corresponding thresholds. The minimum wall thickness constraint is quantitatively evaluated by region.

[0077] By calculating the proportion of the volume difference between the physical density field and the density field after corrosion, and using this proportion of volume difference as the core indicator to complete the quantitative evaluation of the minimum wall thickness constraint, the minimum wall thickness process requirement of injection molding can be transformed into a calculable and verifiable numerical indicator in topology optimization iteration, providing a quantitative basis for the sensitivity calculation of constraint conditions and the updating of design variables in the topology optimization process.

[0078] S200b determines the expansion radius based on the preset maximum wall thickness value, performs a continuously differentiable morphological expansion operation on the physical density field, and obtains the expanded density field.

[0079] It is understandable that the preset maximum wall thickness value is a limit value set on the maximum thickness of the plastic part in the injection molding process in order to reduce the probability of molding defects such as shrinkage marks and uneven cooling. For example, the preset maximum wall thickness value can be 4.5mm, 3.5mm, etc.

[0080] The expansion radius is a calculation parameter used in morphological expansion operations to determine the outward expansion range of a solid material region. For example, when the preset maximum wall thickness is 4.0 mm, the corresponding expansion radius can be set to 2.0 mm, and when the preset maximum wall thickness is 3.0 mm, the corresponding expansion radius can be set to 1.5 mm, etc.

[0081] Continuously differentiable morphological expansion is a numerical computation method adapted to topology optimization of continuous density fields. It can achieve directional expansion of solid material regions without destroying the continuity of density field gradients. At the same time, it can complete sensitivity backpropagation through the chain rule to adapt to the iterative calculation requirements of gradient-type optimizers. For example, for the physical density field of a certain layer of the rear shell of a car headlight, this operation can be performed according to the corresponding expansion radius to identify local structures that exceed the maximum wall thickness. For the physical density field with thick reinforcing ribs, this operation can be used to complete the wall thickness boundary identification of the entire region while preserving the differentiability of the density field.

[0082] The expanded density field is the updated density field data obtained by performing a continuously differentiable morphological expansion operation on the physical density field according to a preset expansion radius. It can be used for the subsequent quantitative evaluation of the maximum wall thickness constraint.

[0083] The expansion radius is determined based on a preset maximum wall thickness value. A continuously differentiable morphological expansion operation is performed on the physical density field to obtain the expanded density field. One approach is to determine a fixed expansion radius based on a uniform preset maximum wall thickness value across the entire design space of the taillight rear housing to be optimized. An isotropic smooth Gaussian kernel is used as the structuring element, and a continuously differentiable morphological expansion operation is performed on the entire physical density field, ultimately obtaining a uniformly processed expanded density field across the entire region. For example, for the rear housing of a passenger vehicle LED taillight with a regular structure and uniform wall thickness requirements, the preset global maximum wall thickness value is 4.5mm, corresponding to a determined expansion radius of 2.25mm. An isotropic Gaussian kernel is used to perform a continuously differentiable expansion operation on the physical density field of the entire design space, ultimately obtaining an expanded density field adapted to the global maximum wall thickness constraint. Alternatively, one approach is to first determine the expansion radius based on the preset maximum wall thickness value across the entire design space of the taillight rear housing to optimize. Different preset maximum wall thickness values ​​are set for different functional zones of the rear housing of the vehicle headlight. The expansion radius corresponding to each sub-region is determined. Then, an adapted anisotropic structural element is used for different sub-regions to perform continuously differentiable morphological expansion operations on the corresponding regions of the physical density field. Finally, the expanded density field of the entire design space is obtained by fusion. For example, for the rear housing of the headlight of a passenger vehicle with a thickened mounting boss and a thin-walled main body, a preset maximum wall thickness value of 5.0 mm and a corresponding expansion radius of 2.5 mm are set for the mounting boss sub-region, and a preset maximum wall thickness value of 3.5 mm and a corresponding expansion radius of 1.75 mm are set for the thin-walled sub-region of the housing body. After performing continuously differentiable expansion operations on the two sub-regions using adapted anisotropic structural elements, the results are fused to obtain the expanded density field adapted to the differentiated maximum wall thickness requirements of the sub-regions.

[0084] S200c, based on the proportion of the volume difference between the expanded density field and the physical density field, completes the quantitative evaluation of the maximum wall thickness constraint. The proportion of the volume difference is used to characterize the proportion of structures in the physical density field that are greater than the maximum wall thickness.

[0085] It can be understood that the volume difference ratio refers to the ratio obtained by dividing the difference between the total volume of the expanded density field and the total volume of the physical density field by the total volume of the physical density field. This ratio is used to characterize the proportion of structures in the physical density field that exceed the preset maximum wall thickness.

[0086] The quantitative assessment of maximum wall thickness constraints based on the volume difference ratio between the expanded density field and the physical density field can be achieved by first setting a uniform maximum wall thickness value and a corresponding allowable threshold for the volume difference ratio for the entire design space of the rear housing of the headlight to be optimized. Then, the volume difference ratio between the expanded density field and the physical density field within the entire design space is calculated. The calculated result is compared with the preset allowable threshold to complete the quantitative assessment of the maximum wall thickness constraint for the entire structure. For example, for the rear housing of an LED taillight in a passenger vehicle with a regular structure and uniform wall thickness requirements, a globally uniform maximum wall thickness value of 4.5mm is set, and the allowable threshold for the volume difference ratio is 1%. The calculated volume difference ratio for the entire design space is 0.7%, which allows the quantitative assessment to determine that the maximum wall thickness of the current structure complies with the preset requirements. Alternatively, the assessment can be based on the volume difference ratio between the expanded density field and the physical density field within the entire design space. Structural function, mold melt filling path, and appearance requirements divide the design space into multiple sub-regions. For each sub-region, a maximum wall thickness value and an allowable threshold for volume difference percentage are set. Then, the volume difference percentage between the expanded density field and the physical density field within each sub-region is calculated. This allows for a quantitative assessment of the maximum wall thickness constraint by region. For example, for the rear housing of a passenger vehicle's front combination headlight with an exterior surface, mounting structure, and non-exterior main body area, a maximum wall thickness of 3.5mm and an allowable threshold for volume difference percentage are set for the housing area corresponding to the exterior surface. A maximum wall thickness of 5.0mm and an allowable threshold for volume difference percentage are set for the non-exterior reinforcing ribs and mounting points. The volume difference percentage for each sub-region is calculated and compared with the corresponding thresholds to achieve a quantitative assessment of the maximum wall thickness constraint by region.

[0087] By calculating the proportion of the volume difference between the expanded density field and the physical density field, and using this proportion as an indicator to complete the quantitative evaluation of the maximum wall thickness constraint, the maximum wall thickness process requirement of injection molding can be transformed into a calculable and verifiable numerical indicator in topology optimization iteration. This provides a quantitative basis for the sensitivity calculation of constraint conditions and the updating of design variables during topology optimization, and can also be used to evaluate the degree of adaptation of the structure in the current iteration step to the maximum wall thickness process requirement.

[0088] S200d, based on the volume difference between the density field after corrosion and the density field after expansion, determines the wall thickness fluctuation range of the physical density field.

[0089] It can be understood that the volume difference refers to the numerical difference between the total volume of the etched density field and the total volume of the expanded density field after performing morphological etch and expansion operations on the same physical density field. For example, the total volume of the etched density field of the rear housing of an LED taillight is 85 cm³, and the total volume of the expanded density field is 105 cm³, with a volume difference of 20 cm³.

[0090] The wall thickness fluctuation range refers to the range of differences between the maximum and minimum values ​​of the wall thickness in each local area of ​​the plastic part structure corresponding to the physical density field. It is used to characterize the uniformity of the overall wall thickness of the structure. For example, the wall thickness fluctuation range of the taillight rear housing can be 1.8mm~4.2mm, and the wall thickness fluctuation range of the headlight rear housing can be 2.0mm~3.8mm, etc.

[0091] Based on the volume difference between the density field after corrosion and the density field after expansion, the method to determine the wall thickness fluctuation range of the physical density field can be as follows: First, calculate the total volume difference between the density fields after corrosion and expansion within the entire design space. Then, combine the solid projected area of ​​the design space and the number of layers to fit the maximum and minimum values ​​of the overall structural wall thickness, thus determining the global wall thickness fluctuation range of the physical density field. For example, for the rear shell of a passenger vehicle LED taillight with a regular structure, first calculate the total volume difference between the density fields after corrosion and expansion in the entire design space as 18 cm³. Then, combine the solid projected area of ​​the shell and the number of 100 layers to fit the minimum and maximum values ​​of the overall structural wall thickness as 1.9 mm and 4.3 mm, thus determining the overall wall thickness fluctuation range of the physical density field. The wall thickness fluctuation range of the physical density field is 1.9mm~4.3mm; alternatively, the design space of the rear housing of the headlight to be optimized can be divided into multiple continuous sub-regions along the main demolding direction. The volume difference between the density field after corrosion and the density field after expansion in each sub-region can be calculated. Then, the extreme values ​​of the wall thickness in each sub-region can be statistically analyzed segment by segment. Finally, the full range of wall thickness fluctuation of the physical density field can be obtained. For example, for the rear housing of the headlight of a passenger vehicle with complex curved surfaces and multiple installation structures, the design space can be divided into three sub-regions along the main demolding direction: the installation area, the main thin-walled area, and the lateral snap-fit ​​area. The volume difference in each sub-region can be calculated and the extreme values ​​of the wall thickness in the corresponding sub-region can be statistically analyzed. Finally, the wall thickness fluctuation range of the physical density field can be determined.

[0092] S200e achieves quantitative control of wall thickness uniformity by constraining the upper limit of the wall thickness fluctuation range.

[0093] It is understandable that the upper limit of the wall thickness fluctuation range is to limit the maximum dispersion of the wall thickness of the plastic part structure, and to set a numerical boundary for the maximum allowable difference between the maximum and minimum values ​​of the local wall thickness of the structure; for example, for the rear housing of LED taillights for passenger vehicles made of glass fiber reinforced PP material, the upper limit of the wall thickness fluctuation range can be set to 2.0mm, 2.5mm, etc.

[0094] Wall thickness uniformity is an indicator that characterizes the degree of consistency in wall thickness values ​​in different local areas of a plastic part. The quality of wall thickness uniformity directly affects melt filling, cooling shrinkage, and the degree of warpage deformation of the plastic part during the injection molding process.

[0095] Quantitative control is a process of limiting, adjusting, and verifying the state of a target object based on clear numerical indicators. It can transform fuzzy process requirements into calculable, executable, and controllable operations in topology optimization iteration. For example, quantitative control can be used to control the uniformity of the wall thickness of the rear housing of a vehicle headlight by using the upper limit of the wall thickness fluctuation range as an indicator, or to control the compliance of the maximum wall thickness of the rear housing of a vehicle headlight by using the proportion of volume difference as an indicator.

[0096] The quantitative control of wall thickness uniformity can be achieved by constraining the upper limit of the wall thickness fluctuation range. One approach is to set a uniform upper limit for the wall thickness fluctuation range across the entire design space of the rear housing of the headlight to be optimized. This upper limit is then used as a mandatory constraint in the topology optimization iteration process. In each iteration, the wall thickness fluctuation range corresponding to the current physical density field is checked to ensure it meets the constraint requirements. Sensitivity analysis is used to guide the update and adjustment of design variables, ultimately achieving quantitative control of the wall thickness uniformity across the entire structure. Alternatively, the design space can be divided into multiple independent sub-regions based on the appearance grade, melt filling path, and structural functional zoning of the rear housing of the headlight to be optimized. Differentiated upper limits for the wall thickness fluctuation range are set for each sub-region. During the topology optimization iteration process, the wall thickness fluctuation in each sub-region is checked, and the adjustment and optimization of design variables are guided by region, ultimately achieving hierarchical quantitative control of the wall thickness uniformity across the entire structure.

[0097] By setting a clear numerical upper limit constraint for the wall thickness fluctuation range, and by embedding this constraint into the iterative process of topology optimization to guide the updating and adjustment of design variables, the quantitative control of wall thickness uniformity can be achieved. This can transform the process requirements of injection molding for wall thickness uniformity into a feasible numerical constraint in topology optimization, which helps to reduce the probability of molding defects such as uneven cooling and warping deformation in plastic parts, and can also improve the stability of the mechanical properties of the rear shell structure of the headlight.

[0098] In one possible implementation, S200, based on the physical density field, completes the mechanical response solution and design variable update, including: S210, based on the physical density field, constructs a finite element model and solves for the mechanical response quantities of structural flexibility, volume, and modal frequencies in the current iteration step.

[0099] It can be understood that a finite element model is a digital model that discretizes a continuous three-dimensional solid structure into a finite number of interconnected smallest computational units, and is used for numerical simulation calculation of structural mechanical performance. For example, a finite element model can be a modified quadratic tetrahedral element finite element model based on the physical density field discretization of the LED taillight rear housing, or a hexahedral element finite element model based on the physical density field discretization of the headlight rear housing, etc.

[0100] The current iteration step is a single-cycle node in the multi-round loop operation of topology optimization, which is performing mechanical performance solution, constraint condition verification and design variable adjustment.

[0101] Structural flexibility is a mechanical performance index that characterizes the degree of overall deformation of a structure under a specified load, and its value is negatively correlated with the overall stiffness of the structure.

[0102] Volume refers to the total volume of solid material in the rear shell structure of the headlight, corresponding to the physical density field.

[0103] Modal frequency is a core parameter characterizing the inherent vibration characteristics of a structure, referring to the natural frequency of a structure under constrained conditions during free vibration.

[0104] Mechanical response quantities are obtained through finite element numerical simulation. They are various numerical indicators that characterize the mechanical performance of a structure under specified working conditions and are the core basis for constructing optimization objectives and verifying constraints in topology optimization.

[0105] The method of constructing a finite element model based on the physical density field to obtain the mechanical response quantities of structural flexibility, volume, and modal frequencies in the current iteration step can be as follows: First, apply fixed boundaries and static loads based on the actual installation constraints of the taillight rear housing. Then, build a static finite element model based on the material distribution characteristics of the physical density field. Simultaneously calculate and obtain the mechanical response quantities such as structural flexibility, solid volume, and low-order modal frequencies within a single iteration cycle. For example, for the taillight rear housing of a passenger vehicle, set constraint boundaries according to the vehicle assembly limit and apply static compressive loads to construct a finite element model and complete the solution, stably outputting the various mechanical response data corresponding to the current iteration step. Alternatively, combine various actual service conditions such as vehicle driving vibration and assembly stress. Based on the physical density field, build finite element calculation models for each corresponding condition. Through parallel calculation of multiple conditions, comprehensively solve the structural flexibility, volume, and modal frequency indices under different conditions, integrating them to form a complete set of mechanical response quantities for the current iteration step. For example, for the complex-shaped automotive headlight rear housing, simultaneously superimpose static assembly loads and dynamic vibration conditions, couple and construct a multi-condition finite element model for joint solution, comprehensively obtaining mechanical response parameters under multiple scenarios.

[0106] By constructing a finite element model adapted for numerical calculation based on physical density field discretization, and by performing numerical simulation on the finite element model under specified working conditions, the mechanical response quantities of structural flexibility, volume, and modal frequency in the current iteration step can be obtained. This can provide quantitative mechanical performance basis for the evaluation of optimization objectives and the verification of constraint compliance in the topology optimization iteration process, and can also provide basic data support for subsequent sensitivity analysis and design variable updates.

[0107] S220 calculates the sensitivity of each mechanical response quantity and process constraint quantity to the physical density field using the adjoint method, and then uses the chain rule to transfer the sensitivity back to the design variables corresponding to the experimental density field.

[0108] It is understandable that the adjoint method is a numerical method in the field of topology optimization used to efficiently calculate the derivatives of the objective function and constraints with respect to the design variables. It can complete the sensitivity calculation of the entire design domain in a single solution, reducing the computational cost of large-scale topology optimization problems.

[0109] Process constraints are numerical values ​​of various quantitative constraint indicators used to characterize the requirements of injection molding process during topology optimization. For example, process constraints can be the proportion of minimum wall thickness volume difference, the proportion of maximum wall thickness volume difference, and the constraint values ​​corresponding to the wall thickness fluctuation range in the topology optimization of the rear housing of a car headlight.

[0110] Sensitivity refers to the rate of change of the objective function or constraint value with the change of element density in the physical density field in topology optimization. It is used to characterize the degree of influence of design variable adjustments on the optimization objective and constraints. For example, sensitivity can be the rate of change of structural flexibility with element density or the rate of change of minimum wall thickness constraint value with element density in the topology optimization of the rear shell of a car headlight.

[0111] The chain rule is a core rule in calculus used to calculate the derivative of a composite function. In topology optimization, it can be used to transfer the sensitivity corresponding to the physical density field to the design variable corresponding to the front-end experimental density field, ensuring the continuity of gradient calculation.

[0112] The method of calculating the sensitivity of each mechanical response quantity and process constraint quantity to the physical density field using the adjoint method, and then transferring the sensitivity back to the design variables corresponding to the experimental density field using the chain rule, can be as follows: minimizing the total strain energy of the structure as the single optimization objective, with volume constraint as the core mechanical constraint and minimum wall thickness constraint as the core process constraint. The sensitivity of the mechanical response quantity and process constraint quantity to the physical density field is calculated simultaneously in a single finite element solution using the adjoint method, and then the sensitivity is directly transferred back to the element pseudo-density design variables corresponding to the experimental density field using the chain rule. Alternatively, the sensitivity of each mechanical response quantity and each process constraint quantity to the physical density field can be calculated separately in stages using the adjoint method, and then the comprehensive sensitivity can be obtained by weighted summation. After that, the sensitivity is transferred back to the design variables corresponding to the experimental density field after completing the multi-step composite mapping operation step by step using the chain rule. However, this method is not limited to these approaches.

[0113] The adjoint method is used to efficiently calculate the sensitivity of each mechanical response quantity and process constraint quantity to the physical density field, and to obtain data on the degree of influence of design variable adjustment on optimization objectives and constraints. The calculated sensitivity is fed back to the design variables corresponding to the experimental density field through the chain rule, providing gradient basis for the design variable update of the gradient-type optimizer, ensuring the convergence and directional rationality of the topology optimization iteration process, and realizing the synchronous adaptation of injection molding process constraints and mechanical performance objectives in the gradient optimization process.

[0114] S230 inputs the returned sensitivity into the gradient optimizer to update the design variables.

[0115] It can be understood that the sensitivity after feedback refers to the gradient data of the optimization objective and constraint conditions with respect to the corresponding design variables after the mechanical response and process constraint quantities are calculated with respect to the physical density field by the adjoint method and then fed back to the experimental density field by the chain rule.

[0116] Gradient-type optimizers are numerical computation modules that, in the process of topology optimization, use the sensitivity gradient of design variables as the core basis, adjust the design variables according to preset numerical optimization algorithm rules, and drive the optimization objective to converge in the preset direction. For example, gradient-type optimizers can be moving asymptote optimizers suitable for multi-constraint topology optimization scenarios, or sequential quadratic programming optimizers that can adapt to continuous density field iterations.

[0117] The update of design variables refers to the operation by which gradient-based optimizers adjust the design variables corresponding to the experimental density field based on the input sensitivity data, within the preset constraint boundaries and iteration step size.

[0118] By inputting the returned sensitivity into the gradient-type optimizer, quantitative directional guidance and numerical basis are provided for the iterative adjustment of design variables. The gradient-type optimizer updates the design variables according to the preset algorithm rules, promoting the orderly advancement of multiple iterations of topology optimization. This allows the optimization process to gradually converge towards a design scheme that meets the mechanical performance requirements and injection molding process constraints, providing an updated design basis for the density field mapping and mechanical response solution in the next iteration.

[0119] S300 performs skeletal identification of the macroscopic stiffener topology and constructs a two-layer design variable space that includes topological layout variables and stiffener cross-sectional size variables.

[0120] It can be understood that frame recognition is a geometric processing method based on three-dimensional distance field and topology refinement algorithm, which automatically extracts the material distribution of the macroscopic stiffener topology into a linear and parameterizable stiffener central axis network.

[0121] The dual-layer design variable space is a set of full-dimensional design variables composed of topological layout variables in the macro-layout dimension and stiffener cross-sectional size variables in the micro-section dimension. It can simultaneously cover two types of design requirements: stiffener force transmission path adjustment and load-bearing capacity optimization.

[0122] The method for skeletonizing the macroscopic stiffener topology and constructing a two-layer design variable space containing topological layout variables and stiffener cross-sectional size variables can be as follows: First, extract the single-pixel central axis of the macroscopic stiffener topology using a topology refinement algorithm to complete skeletonization and obtain a continuous stiffener axis network without branch redundancy. Then, set the endpoint coordinates, branch direction, and connection position of the stiffener axis as topological layout variables, and set the cross-sectional height and width of a single stiffener as stiffener cross-sectional size variables. Finally, integrate and construct the two-layer design variable space. For example, for the rear housing of a passenger vehicle LED taillight with a regular structure and few stiffener branches, first use a topology refinement algorithm to skeletonize the macroscopic stiffener topology and obtain the eight continuous stiffeners connecting four mounting points. The design involves first establishing a central axis, then setting the endpoint coordinates and branch angles of each axis as 12 topological layout variables, and setting the cross-sectional height and width of each reinforcing rib as 16 reinforcing rib cross-sectional size variables, ultimately constructing a corresponding two-layer design variable space. Alternatively, based on the primary and secondary relationships of the force transmission paths in the macroscopic reinforcing rib topology, the reinforcing ribs can be divided into two levels: primary force transmission ribs and auxiliary reinforcing ribs. Hierarchical skeletonization identification of reinforcing ribs at different levels is then performed in different regions to obtain a reinforcing rib axis network with primary and secondary levels. The axis direction and connection position of the primary force transmission ribs are then set as core topological layout variables, and the number and distribution position of the auxiliary reinforcing ribs are set as secondary topological layout variables. Corresponding cross-sectional size variables are also set for the primary and auxiliary reinforcing ribs, ultimately constructing a two-layer design variable space in a hierarchical manner.

[0123] In one possible implementation, S300 constructs a two-layer design variable space containing topology layout variables and stiffener cross-sectional dimension variables, including: S310 defines the cross-sectional height and cross-sectional width as cross-sectional dimension variables for each stiffener in the stiffener network obtained by skeletonization identification.

[0124] It is understandable that the reinforcing rib network is obtained after skeletonization identification, and is a network structure composed of multiple interconnected reinforcing ribs forming force transmission paths along their central axes.

[0125] The cross-sectional height refers to the extended dimension of the reinforcing rib cross-section in the direction perpendicular to the thin-walled plane of the taillight housing substrate. For example, the cross-sectional height of the reinforcing rib of the taillight housing for passenger vehicles can be set to an adjustable range of 3mm to 8mm, and the cross-sectional height of the main force transmission rib of the headlight housing can be set to an adjustable range of 5mm to 10mm, etc.

[0126] The cross-sectional width refers to the lateral dimension of the reinforcing rib cross-section in the direction parallel to the thin-walled plane of the taillight rear housing substrate; for example, the cross-sectional width of the taillight rear housing reinforcing rib can be set to an adjustable range of 1.5mm to 4mm, and the cross-sectional width of the auxiliary reinforcing rib of the headlight rear housing can be set to an adjustable range of 2mm to 3.5mm, etc.

[0127] Cross-sectional dimension variables refer to the independent variables of cross-sectional height and cross-sectional width defined for each reinforcing rib and adjustable within a preset range; for example, cross-sectional dimension variables can be the adjustable parameters of cross-sectional height and cross-sectional width corresponding to a single reinforcing rib of the taillight rear housing, the adjustable parameters of cross-sectional height and cross-sectional width corresponding to the main force transmission rib of the headlight rear housing, etc.

[0128] For the reinforcing rib network obtained from skeletonization, defining the cross-sectional height and width as cross-sectional dimension variables for each reinforcing rib can be achieved by independently parameterizing each reinforcing rib in the network. This involves setting independent adjustable ranges for cross-sectional height and width for each rib, and defining each set of independent cross-sectional height and width as the cross-sectional dimension variables for the corresponding reinforcing rib. For example, for the rear housing of a passenger vehicle LED taillight with a regular structure and a small number of reinforcing ribs, the eight reinforcing ribs obtained from skeletonization can be independently defined, with each rib... The reinforcing ribs are each set with an adjustable range of cross-sectional height (3mm~8mm) and cross-sectional width (1.5mm~4mm) that are not related to each other. The cross-sectional height and width of each reinforcing rib are defined as independent cross-sectional dimension variables. Alternatively, the reinforcing ribs can be divided into two levels, main force transmission ribs and auxiliary reinforcing ribs, according to the primary and secondary force transmission paths of the reinforcing rib network. The corresponding adjustable range of the foundation is set for the reinforcing ribs of different levels. Then, the cross-sectional height and cross-sectional width based on the foundation range are defined as cross-sectional dimension variables for each reinforcing rib in the same level. At the same time, the association constraints of variables in the same level can be set.

[0129] By defining each stiffener independently in the stiffener network obtained by skeletonization identification, and by setting the cross-sectional height and width of each stiffener as adjustable cross-sectional size variables, an independent and controllable design object can be provided for the subsequent fine parameter optimization of the stiffener. This enables precise control over the load-bearing capacity and forming processability of a single stiffener, and also provides basic data support for the micro-section dimension for constructing a fully parameterized double-layer design variable space.

[0130] S320 defines fine-tunable topology layout variables for the topology layout of stiffener connections and critical load transfer areas.

[0131] It is understandable that the connection point of the stiffener is the node area where the axes of two or more stiffeners in the stiffener network intersect, realizing the transfer of load across the stiffeners.

[0132] The critical area for load transfer is the structural area of ​​the rear housing of the headlight that bears the main load input and concentrates the force transmission under actual service conditions. For example, the critical area for load transfer may be the area around the four mounting points of the LED taillight rear housing that are connected to the body, or the middle section of the main force transmission rib in the rear housing of the headlight that bears the vibration load of the whole vehicle.

[0133] The fine-tunable topology layout variables are design variables that can be adjusted within a preset small range based on the macroscopic stiffener topology configuration, and are used to optimize the local force transmission path and load distribution. For example, the fine-tunable topology layout variables can be the node coordinate offset at the connection of the stiffener at the rear housing of the taillight, the stiffener branch angle adjustment parameter, the stiffener axis orientation fine-tuning parameter in the key area of ​​load transmission of the rear housing of the headlight, the stiffener branch number adjustment parameter, etc.

[0134] By defining finely adjustable topology layout variables for the connection points of stiffeners and key areas of load transfer, a small range of controllable design adjustment parameters are provided for the fine optimization of the local core load-bearing area of ​​the rear housing of the headlight. By incorporating the finely adjustable topology layout variables into a two-layer design variable space, the local precise optimization of the macroscopic force transmission path of the stiffeners and the coordinated adaptation of global mechanical properties are achieved. This helps to improve the efficiency of local load transfer, reduce the probability of stress concentration in the structure, and also provides complete macroscopic layout dimension design support for the subsequent coordinated optimization of stiffener layout and cross-sectional parameters.

[0135] S330 combines cross-sectional dimension variables with topological layout variables to form a two-layer design variable space.

[0136] It is understandable that combining cross-sectional dimension variables and topology layout variables to form a two-layer design variable space can be done in several ways. One approach is to set all independently defined cross-sectional dimension variables and topology layout variables as equally weighted optimization design variables, without setting hierarchical priorities or correlation constraints between variables, and directly combining both types of variables to form a flattened two-layer design variable space. Another approach is to first determine the degree of influence of the variables on structural performance and manufacturability, setting the topology layout variables as upper-level core variables and the cross-sectional dimension variables as lower-level subordinate variables, clarifying the hierarchical priorities of the two types of variables, setting correlation constraint rules between variables, and then combining the two types of variables in an orderly manner according to the hierarchical relationship to form a two-layer design variable space with priorities and correlation constraints.

[0137] By systematically integrating independently defined cross-sectional dimension variables with topological layout variables, a variable set covering different design dimensions is constructed. By combining two types of design variables of different dimensions to form a two-layer design variable space, the full-dimensional design requirements of macroscopic force transmission path fine-tuning and microscopic cross-sectional parameter optimization of stiffeners can be simultaneously covered. This provides a complete variable framework for subsequent multi-objective collaborative optimization of stiffener structures and also provides a feasible design basis for achieving collaborative optimization of the mechanical properties and injection molding processability of the rear shell of the headlight.

[0138] S400 constructs a combined surrogate model based on a two-layer design variable space to simultaneously fit the coupled influence of two types of variables, and obtains the global optimal design parameters of the rear housing of the headlight to be optimized through a global optimization algorithm; among them, the combined surrogate model is used to characterize the influence of the coordinated changes in topology and cross-sectional dimensions on structural performance.

[0139] It is understandable that the combined surrogate model is based on the construction of a two-layer design variable space. By combining the advantages of multiple single surrogate models, it simultaneously fits the coupling effect of topological layout variables and cross-sectional size variables. It is a mathematical approximation model used to characterize the law of influence of the coordinated changes of the two types of variables on structural performance. For example, the combined surrogate model can be a Kriging-response surface combined surrogate model or a radial basis function-support vector machine combined surrogate model, which are built for the rear shell of LED taillights and adapted to the coupling relationship of multiple variables.

[0140] The two-type variable coupling effect refers to the superimposed effect of the mutual linkage and synergistic action between the adjustment of topology layout variables and the change of cross-sectional size variables on the structural mechanical properties and injection molding processability. For example, the positional offset of the intersection node of the reinforcing ribs of the rear housing of the headlight will synchronously change the degree of influence of the corresponding reinforcing rib cross-sectional parameters on the structural stiffness, which belongs to the two-type variable coupling effect. The fine adjustment of the direction of the reinforcing ribs around the mounting point of the rear housing of the headlight will change the sensitivity of the corresponding reinforcing rib cross-sectional size to the structural modal frequency, which also belongs to the two-type variable coupling effect.

[0141] Global optimization algorithms are numerical optimization algorithms that search for design parameters that can achieve a better state of the optimization objective within the entire feasible domain of the two-layer design variable space, based on the output results of the combined surrogate model and under preset constraints. For example, global optimization algorithms can be multi-objective particle swarm optimization algorithms suitable for multi-variable optimization scenarios of automotive headlight rear housings, or second-generation non-dominated sorting genetic global optimization algorithms adapted to automotive headlight housing optimization with injection molding process constraints.

[0142] The globally optimal design parameters are obtained by searching within the feasible region of the two-layer design variable space using a global optimization algorithm. They are combinations of topological layout variables and cross-sectional dimension variables that simultaneously satisfy preset mechanical performance requirements and injection molding process constraints, and achieve a relatively optimal state for the optimization objective. For example, the globally optimal design parameters can be a combination of design parameters for the taillight rear housing obtained through global optimization that simultaneously satisfies stiffness and wall thickness constraints, or a combination of design parameters for the headlight rear housing obtained through global optimization that simultaneously satisfies modal, volume, and molding process constraints.

[0143] By constructing a combined surrogate model based on a two-layer design variable space to simultaneously fit the coupled influence of two types of variables, a mathematical representation of the impact of coordinated changes in topology and cross-sectional dimensions on structural performance is achieved, replacing time-consuming finite element simulation calculations to improve the efficiency of optimization iteration. By calling a global optimization algorithm based on the constructed combined surrogate model, the globally optimal design parameters of the taillight rear shell to be optimized are obtained by searching the entire feasible domain of the two-layer design variable space. Under the premise of taking into account mechanical performance requirements and injection molding process constraints, the coordinated optimization of the macroscopic layout and microscopic cross-sectional parameters of the stiffeners can be achieved, providing a quantitative scheme reference for the final refined structural design of the taillight rear shell.

[0144] In one possible implementation, S400 constructs a combined surrogate model that simultaneously fits the coupled effects of two types of variables based on a two-level design variable space, including: S410 collects a preset number of sample points in a two-layer design variable space.

[0145] It is understandable that the preset quantity is a fixed number of samples determined in advance based on the accuracy requirements of structural optimization and the computing power conditions; for example, the preset quantity for optimizing a simple car headlight housing can be one hundred sets, and the preset quantity for a complex car headlight structure with multiple constraints can be two hundred sets, etc.

[0146] A sample point is an independent design sample within a two-layer design variable space, consisting of a complete set of variable values, which can correspond to a specific stiffener layout and cross-sectional dimension scheme. For example, a sample point can be a combination of variables including stiffener layout offset parameters and cross-sectional height parameters, or a combination scheme including branch direction parameters and cross-sectional width parameters, etc.

[0147] By defining a reasonable sampling range based on a complete two-layer design variable space and collecting a predetermined number of diverse sample points in a standardized manner, it is possible to cover design states under different variable combinations, providing sufficient and representative basic data support for subsequent fitting of variable coupling rules and building of combined proxy models.

[0148] S420 automatically generates a corresponding three-dimensional geometric model for each sample point based on the design parameters, performs mechanical performance and manufacturability verification, and obtains structural response data for each sample point.

[0149] It is understandable that the three-dimensional geometric model is a three-dimensional digital model that is automatically generated by parametric modeling tools based on the design parameters (topology layout variables and cross-sectional size variables) corresponding to the sample points, and is consistent with the actual rear shell structure of the vehicle headlight.

[0150] Mechanical performance verification is the process of performing finite element simulation calculations on automatically generated three-dimensional geometric models to verify whether they meet the preset mechanical performance requirements such as structural flexibility and modal frequencies. For example, mechanical performance verification can be performed by static simulation on the three-dimensional geometric model of sample points to verify the structural stiffness, or by modal simulation on the three-dimensional geometric model of sample points to verify the inherent vibration characteristics.

[0151] Manufacturability verification is a process that checks parameters such as wall thickness, rib layout, and demolding angle of a three-dimensional geometric model in accordance with injection molding process requirements to determine whether it is feasible for actual production. For example, manufacturability verification may check whether the minimum wall thickness of the sample point three-dimensional geometric model meets the injection molding process requirements, or whether the rib layout of the sample point three-dimensional geometric model will lead to injection shrinkage marks.

[0152] Structural response data consists of various quantitative data obtained from the three-dimensional geometric model corresponding to each sample point after mechanical performance and manufacturability verification.

[0153] For each sample point, the corresponding 3D geometric model is automatically generated based on the design parameters, and mechanical performance and manufacturability verification are completed. The method for obtaining the structural response data corresponding to each sample point can be to process them sequentially according to the order of the sample points. First, the 3D geometric model is automatically generated based on the design parameters of a single sample point. Then, the mechanical performance simulation verification and manufacturability parameter verification of the model are completed sequentially. Finally, the structural response data corresponding to the sample point is recorded. This process is repeated until all sample points have been processed. Alternatively, multiple sample points can be processed in parallel using multi-threading. At the same time, the corresponding 3D geometric models are automatically generated based on the design parameters of multiple sample points, and the mechanical performance and manufacturability verification of multiple models are carried out simultaneously. The structural response data corresponding to all sample points is obtained in batches.

[0154] By automatically generating a three-dimensional geometric model that fits the actual structure based on the design parameters corresponding to each sample point, a standardized digital carrier is provided for subsequent performance and manufacturability verification. By performing mechanical performance and manufacturability verification on each of the generated three-dimensional geometric models, the system obtains the structural response data corresponding to each sample point, providing real and reliable quantitative support for subsequent construction of combined proxy models based on sample data and fitting the coupling influence law of topological layout variables and cross-sectional size variables.

[0155] S430, based on the design parameters and corresponding structural response data of all sample points, constructs a Kriging combined surrogate model to characterize the mapping relationship between design parameters and structural response.

[0156] It can be understood that the Kriging combined surrogate model is a mathematical model built on the Kriging interpolation algorithm, used to characterize the mapping relationship between design parameters and structural response.

[0157] The mapping relationship between design parameters and structural response refers to the corresponding relationship between changes in design parameters (such as topology and cross-sectional dimensions) and structural responses (such as stiffness and modal frequencies).

[0158] The design parameters of the sample points refer to the adjustable parameters such as the topology layout and cross-sectional dimensions of all previously collected sample points.

[0159] Based on the design parameters and corresponding structural response data of all sample points, the Kriging combined surrogate model used to characterize the mapping relationship between design parameters and structural response can be constructed in two ways: a single Kriging model construction method, which integrates the design parameters (topology layout, cross-sectional dimensions) and corresponding structural response data of all sample points to construct a single Kriging surrogate model and uniformly fits the mapping relationship of all samples; or a regional Kriging model construction method, which constructs independent Kriging surrogate models for different regions according to the structural partitions (main load-bearing area, auxiliary support area) of the rear housing of the headlight, and then integrates them into a complete combined surrogate model, etc.

[0160] By integrating the design parameters and corresponding structural response data of all sample points, and constructing a Kriging combined surrogate model, the mapping relationship between design parameters and structural response is characterized, providing an efficient computational platform for subsequent global optimization, avoiding repeated complex finite element simulations, and reducing the computational cost of the optimization process.

[0161] In one possible implementation, in S400, the globally optimal design parameters of the rear housing of the headlight to be optimized are obtained through a global optimization algorithm, including: S440 uses minimizing structural flexibility as the optimization objective and lightweighting, modal frequency requirements, allowable stress requirements, and injection molding manufacturability requirements as constraints to construct a global collaborative optimization model.

[0162] It is understandable that the optimization objective is the design direction that needs to be prioritized in the topology optimization process, which is used to clarify the effect that the optimization needs to achieve; for example, the optimization objective can be to minimize the structural flexibility, minimize the structural volume, etc.

[0163] The goal of lightweighting is to constrain the amount of material used in the rear housing of the headlights and control the volume of the structural body. For example, the lightweighting target for the rear housing of LED taillights is set at a structural body volume of no more than 95 cm³, and the lightweighting target for the rear housing of headlights is set at a material usage of no more than 1.3 kg.

[0164] Modal frequency requirements are standards that limit the inherent vibration characteristics of the rear housing of vehicle headlights. They are used to prevent the structure from resonating with the vehicle body vibration during the service of the whole vehicle and to ensure structural stability. For example, the modal frequency requirement for the rear housing of the headlights of commercial vehicles can be set to a first-order constrained modal frequency of not less than 50Hz, and the modal frequency requirement for the rear housing of the taillights of passenger vehicles can be set to a second-order constrained modal frequency of not less than 60Hz, etc.

[0165] Allowable stress requirements are limits on the maximum stress that the rear housing structure of a vehicle headlight can withstand under actual service conditions. They are used to prevent the structure from cracking, deforming, or being damaged due to excessive stress. For example, the allowable stress requirement for a glass fiber reinforced PP material headlight rear housing can be set to no more than 5 MPa, and the allowable stress requirement for an ABS+PC material headlight rear housing can be set to no more than 50 MPa.

[0166] Injection molding manufacturability requirements are specific constraints on the structural parameters of the rear housing of automotive lights, taking into account the characteristics of the injection molding process. These requirements are used to ensure that the design can be actually produced and to avoid injection molding defects. For example, injection molding manufacturability requirements may include a minimum wall thickness of not less than 1.5 mm and a demolding angle of not less than 3°.

[0167] The global collaborative optimization model is a mathematical optimization model that integrates core optimization objectives and various constraints to achieve collaborative adaptation of multiple requirements and indicators. It is used to coordinate and balance structural performance, lightweighting and manufacturability.

[0168] The optimization objective is to minimize structural flexibility, with lightweighting, modal frequency requirements, allowable stress requirements, and injection molding manufacturability requirements as constraints. A global collaborative optimization model can be constructed in several ways. One approach is to prioritize minimizing structural flexibility as the core objective, without distinguishing the priority of various constraints. Lightweighting, modal frequency requirements, allowable stress requirements, and injection molding manufacturability requirements are set as equal constraints, and the objective and constraints are directly integrated to construct the global collaborative optimization model. Another approach is to prioritize minimizing structural flexibility as the core objective, with lightweighting as a secondary auxiliary objective. Modal frequency requirements and allowable stress requirements are classified as core constraints (prioritized), while injection molding manufacturability requirements are classified as secondary constraints (appropriately adapted). The objective and constraints are then integrated hierarchically to construct the global collaborative optimization model.

[0169] By clearly defining minimizing structural flexibility as the core optimization objective, the core direction of optimizing the rear shell structure of the headlight is defined, guiding the optimization process towards improving structural stiffness and reducing deformation. By using lightweight objectives, modal frequency requirements, allowable stress requirements, and injection molding manufacturability requirements as constraints, the feasible scope of optimization is limited, avoiding problems such as substandard performance or inability to produce optimized solutions, thereby improving the practicality and engineering adaptability of the optimized solutions.

[0170] In one possible implementation, in S440, the verification of injection molding manufacturability requirements includes: for each set of design parameters to be solved, firstly, verifying the absence of undercut constraints through density field deformation mapping, and then verifying the wall thickness requirements through continuously differentiable morphological constraints, ensuring that all solution processes are performed within the manufacturable feasible domain.

[0171] Understandably, undercut constraint verification is a structural compliance test conducted for injection molding demolding processes. It is used to identify geometric features in the structure that are not conducive to mold demolding, such as reverse snaps and reverse protrusions. For example, in a car, undercut constraint verification can be carried out by adjusting the spatial posture of local ribs through density field mapping and by detecting the spatial position of lateral protrusions based on density field deformation.

[0172] Continuously differentiable morphological constraints are constraint rules centered on continuous numerical filtering and geometric shape control. They can smooth local geometric abrupt changes in the structure and stably control the size boundaries of the solid region. For example, continuously differentiable morphological constraints can be used to weaken local wall thickness abrupt changes in the structure.

[0173] Wall thickness requirement verification is a testing process that checks the thickness of each local area of ​​the structure item by item, referring to the reasonable wall thickness range specified in the injection molding process. For example, wall thickness requirement verification can be used to check whether the minimum wall thickness of the shell base meets the lower limit requirement of molding, or to check whether the wall thickness of the dense reinforcing rib area exceeds the upper limit allowed by the process.

[0174] Density field deformation mapping is used to complete the verification without undercut constraints, and undesirable features in the structural geometry that are not conducive to injection molding demolding are screened in advance. The wall thickness requirement of the whole area is verified by continuous differentiable morphological constraints, and the reasonable range of structural thickness distribution is standardized. Each step of the solution process is continuously limited to the manufacturable feasible domain, so that parameter iteration and structural optimization always fit the process conditions of injection molding production, and the engineering feasibility of the optimization scheme is improved.

[0175] S450 employs a global optimization algorithm based on an adaptive surrogate model to perform optimization on a combined surrogate model, thereby obtaining the globally optimal design parameters for the rear housing of the headlight to be optimized.

[0176] It is understandable that the global optimization algorithm for adapting to the surrogate model is an optimization calculation method that combines the fitting characteristics and response patterns of the combined surrogate model to make targeted improvements. It can perform parameter search in accordance with the data characteristics of the surrogate model, thereby improving the adaptability and computational efficiency of the optimization process. For example, the global optimization algorithm for adapting to the surrogate model can be an improved genetic algorithm adapted to the Kriging surrogate model, or an adaptive particle swarm optimization algorithm adapted to the multivariate coupled model.

[0177] Optimization solution is a numerical calculation process that iteratively selects and compares various parameter combinations within the feasible range of two-level design variables, based on a preset optimization objective and multiple constraints. For example, optimization solution can be a parameter solution that combines structural performance indicators and injection molding process constraints, or a multi-condition optimization solution that takes into account both lightweighting and vibration requirements.

[0178] By selecting a global optimization algorithm that adapts to the data characteristics of the surrogate model, iterative calculations are carried out in accordance with the parameter response law of the combined surrogate model. By quickly completing the optimization solution under multiple constraints at the level of the combined surrogate model, the computational consumption caused by repeated simulations is reduced. Within a reasonable design range, the parameter combination with better comprehensive performance is selected. Finally, the globally optimal design parameters of the rear shell of the headlight that take into account mechanical performance, lightweight and injection molding manufacturing requirements are obtained, providing a reliable parameter basis for the refined structural design.

[0179] S500, based on globally optimal design parameters, reconstructs and outputs the final three-dimensional digital model of the rear housing of the headlight to be optimized.

[0180] It can be understood that model reconstruction is a modeling process that rebuilds the structural geometry based on defined design parameters and parametric modeling logic. For example, model reconstruction may involve reconstructing the stiffener arrangement based on the optimal stiffener position parameters, or regenerating the local structural outline based on the optimal cross-sectional size parameters.

[0181] The final three-dimensional digital model is a finalized digital structural model that integrates all optimization indicators and meets mechanical constraints and injection molding manufacturing requirements.

[0182] Based on globally optimal design parameters, the final three-dimensional digital model of the rear housing of the headlight to be optimized can be reconstructed and output in one way or by relying on a pre-built standardized parametric modeling template, directly reading the globally optimal design parameters, and having the modeling program automatically complete the unified reconstruction of the overall structure, batch generating and outputting the final three-dimensional digital model. This method is suitable for components with regular structures and simple shapes. Alternatively, the rear housing of the headlight can be divided into independent modules such as the base, mounting base, and reinforcing ribs. Each module can be reconstructed separately according to its corresponding optimal parameters. Then, the connection points between the modules can be adjusted and spliced ​​together to generate a complete final three-dimensional digital model.

[0183] By using globally optimal design parameters as the modeling basis, the optimization values ​​of the stiffener topology layout and cross-sectional dimensions are implemented; the overall structure is fully constructed and its shape is corrected through parametric model reconstruction, and a standardized and complete final three-dimensional digital model is output. The abstract optimization parameters are transformed into a visualized and reusable digital structural solution, providing a stable and reliable digital foundation for subsequent performance verification, process validation and product structure implementation.

[0184] In one possible implementation, the S500, based on globally optimal design parameters, reconstructs and outputs the final three-dimensional digital model of the rear housing of the headlight to be optimized, including: S510 determines the final stiffener layout based on the topology layout variables in the globally optimal design parameters.

[0185] It can be understood that the layout of stiffeners refers to the distribution, orientation, connection method, and overall arrangement of stiffeners on the surface of the shell structure.

[0186] Based on the topology layout variables in the global optimal design parameters, the final stiffener layout can be determined by directly reading the node offset coordinates and axis angle parameters in the topology layout variables, mapping them point-to-point to generate the stiffener direction and intersection position, and quickly determining the overall layout scheme; or by using the topology layout variables as the basic framework, combining the boundary constraints of injection molding demolding and structural assembly to make local fine-tuning corrections, optimizing the stiffener connection transition form, and finally determining a compliant and reasonable stiffener layout.

[0187] By extracting the topology layout variables within the global optimal design parameters, the key node positions and orientation adjustment parameters of the stiffeners are locked; by relying on standardized layout rules to complete parameter matching, the final stiffener layout is determined, allowing the optimized force transmission path to be fixed and ensuring the stable presentation of the optimized structural mechanical performance.

[0188] S520 assigns corresponding cross-sectional parameters to each stiffener based on the cross-sectional dimension variables in the global optimal design parameters and automatically adds injection molding process features; among which, the injection molding process features include at least one of the following features: draft feature and root fillet.

[0189] It can be understood that the cross-sectional parameters are the specific quantitative values ​​corresponding to the cross-sectional size variables, and are the data that determines the geometric shape of the cross-section of a single reinforcing rib; for example, the cross-sectional parameters of the main reinforcing rib of the rear housing of the headlight can be 6mm in height and 3mm in width, and the cross-sectional parameters of the auxiliary reinforcing rib can be 4mm in height and 2mm in width, etc.

[0190] Injection molding process features are process-related structural features added to the 3D model to ensure that the reinforcing ribs and the rear shell of the headlights can be successfully injection molded and to reduce injection defects. For example, injection molding process features can be draft features on the side of the reinforcing ribs, root fillets at the junction of the reinforcing ribs and the shell base, etc.

[0191] Draft features are inclined angle structures added to the surface of reinforcing ribs or shell structures to eliminate demolding resistance after injection molding and ensure that the product can be smoothly removed from the mold; for example, draft features may be a 3° draft feature added to the side of the reinforcing rib of the rear shell of a car headlight, or a 5° draft feature added to the reinforcing rib around the mounting base, etc.

[0192] The root fillet is an arc transition structure added at the junction of the reinforcing rib and the rear housing of the headlight. It is used to reduce stress concentration, avoid injection molding shrinkage marks, and improve the structural connection strength. For example, the root fillet can be an R2mm fillet added to the root of the main reinforcing rib of the rear housing of the headlight, or an R1.5mm fillet added to the root of the auxiliary reinforcing rib.

[0193] By extracting the cross-sectional dimension variables from the globally optimal design parameters, each reinforcing rib is assigned corresponding cross-sectional parameters to ensure that the cross-sectional shape of the reinforcing rib meets the optimized mechanical performance requirements. By automatically adding at least one injection molding process feature such as draft features and root fillets, the manufacturability of the reinforcing rib is improved, avoiding problems such as injection molding demolding difficulties, shrinkage marks, and stress concentration. This ensures that the reinforcing rib has both good load-bearing capacity and can adapt to actual injection molding production needs, providing a process-level guarantee for the subsequent output of the final three-dimensional digital model and engineering implementation.

[0194] The S530 generates and outputs a 3D digital model of the rear housing of the headlight with complete process features through the reconstruction of the 3D design software kernel.

[0195] As can be understood, the kernel of a 3D design software is its core operating module, responsible for parsing design parameters, building geometric models, adding process features, and supporting the reconstruction and output of the entire model. Examples of 3D design software include Siemens NX (UG), Creo, and SolidWorks.

[0196] Complete process features refer to various process structures that meet the requirements of injection molding production, including draft angle, root fillet, wall thickness control, etc., which can avoid problems such as shrinkage marks and demolding difficulties during injection molding.

[0197] The three-dimensional digital model of the rear housing of the headlight is a digital model containing complete process features, generated by software kernel reconstruction based on optimized design parameters.

[0198] The method of reconstructing and generating a 3D digital model of the taillight rear shell with complete process features using the kernel of a 3D design software can be as follows: First, the software can be automated. The optimized design parameters (topology layout parameters, cross-sectional dimension parameters) can be directly imported, and the software kernel can automatically reconstruct the model, adding complete process features such as draft angles and root fillets, generating a 3D digital model of the taillight rear shell with complete process features, which can then be directly output. Alternatively, the optimized design parameters can be imported through the software kernel first, and model details (such as the transition shape at the stiffener joints and the position of process features) can be manually adjusted. Process features such as draft angles and root fillets that meet the requirements can be manually added. After confirming that the model has no process defects, the final reconstruction can be completed and output through the software kernel.

[0199] By leveraging the core functionality of 3D design software, the structural form and technological features of the rear housing of the headlights are reconstructed. At the same time, by adding complete technological features that meet the requirements of injection molding production, a manufacturable 3D digital model is generated and output. This not only transforms the optimized design into a visual and implementable digital model, but also provides a digital basis for subsequent mold design and production processing, adapting to the actual production process.

[0200] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0201] Corresponding to the method for optimizing the rear housing structure of the headlight described in the above embodiments, this application also provides a car headlight, which includes a rear housing. The structure of the rear housing can be optimized by the method described in any of the above embodiments.

[0202] Corresponding to the headlight rear housing structure optimization method described in the above embodiments, this application also provides a headlight rear housing structure optimization system, wherein each unit of the system can implement each step of the headlight rear housing structure optimization method.

[0203] The system includes: The optimization iteration unit is used to perform topology optimization iteration with manufacturability constraints on the design space of the rear housing of the headlight to be optimized. In each iteration step, a differentiable density field deformation mapping is applied to the test density field to obtain the physical density field that meets the requirements of undercut-free injection molding. The density field deformation mapping is used to force the topology optimization process to always search within the manufacturable feasible region, which is the range of the design space that meets the requirements of undercut-free injection molding.

[0204] Topological configuration elements are used to solve mechanical response and update design variables based on physical density field. After iterative convergence, the macroscopic stiffener topological configuration of the rear shell of the headlight to be optimized is obtained.

[0205] The spatial construction unit is used to identify the macroscopic stiffener topology and construct a two-layer design variable space containing topological layout variables and stiffener cross-sectional size variables.

[0206] The parameter design unit is used to construct a combined surrogate model that synchronously fits the coupled influence of two types of variables based on a two-layer design variable space, and obtains the global optimal design parameters of the rear shell of the headlight to be optimized through a global optimization algorithm. The combined surrogate model is used to characterize the influence of the coordinated changes in topology and cross-sectional dimensions on structural performance.

[0207] The output unit is used to reconstruct and output the final three-dimensional digital model of the rear housing of the headlight to be optimized based on the globally optimal design parameters.

[0208] It should be noted that the information interaction and execution process between the above-mentioned units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0209] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0210] This application also provides a device for optimizing the structure of a vehicle headlight rear housing. The device includes at least one processor, at least one memory, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it enables the device to implement the steps in any of the above-described embodiments of the vehicle headlight rear housing structure optimization method, or enables the device to implement the functions of each unit in the above-described system embodiments.

[0211] The headlight rear housing structure optimization device can be a computing device such as an industrial design workstation, desktop computer, or industrial server. This headlight rear housing structure optimization device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are merely examples of headlight rear housing structure optimization devices and do not constitute a limitation on the headlight rear housing structure optimization device. It may include more or fewer components than described above, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0212] In the embodiments provided in this application, it should be understood that the disclosed vehicle headlight rear housing structure optimization system, apparatus, and method can be implemented in other ways. For example, the vehicle headlight rear housing structure optimization system and apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms.

[0213] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for optimizing the structure of a vehicle headlight rear housing, characterized in that, include: A topology optimization iteration with manufacturability constraints is performed on the design space of the rear housing of the headlight to be optimized. In each iteration step, a differentiable density field deformation mapping is applied to the test density field to obtain a physical density field that meets the requirements of undercut-free injection molding. The density field deformation mapping is used to force the topology optimization process to always search within the manufacturable feasible region, which is the range of the design space that meets the requirements of undercut-free injection molding. Based on the physical density field, the mechanical response is solved and the design variables are updated. After iterative convergence, the macroscopic reinforcing rib topology of the rear housing of the headlight to be optimized is obtained. The macroscopic stiffener topology is identified by skeletonization, and a two-layer design variable space containing topology layout variables and stiffener cross-sectional size variables is constructed. Based on the aforementioned dual-layer design variable space, a combined surrogate model is constructed to simultaneously fit the coupled influence of two types of variables, and the global optimal design parameters of the rear housing of the headlight to be optimized are obtained through a global optimization algorithm; wherein, the combined surrogate model is used to characterize the influence law of the coordinated change of topology layout and cross-sectional dimensions on structural performance; Based on the globally optimal design parameters, the final three-dimensional digital model of the rear housing of the headlight to be optimized is reconstructed and output.

2. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, The process of applying a differentiable density field deformation mapping to the test density field to obtain a physical density field that meets the requirements for undercut-free injection molding includes: Along the preset main demolding direction, the design space of the rear housing of the headlight to be optimized is divided into multiple continuous layers perpendicular to the main demolding direction; The maximum value projection calculation is performed layer by layer in the order from the parting surface toward the main demolding direction; wherein, the physical density value of the current layer is the larger of the current layer test density value and the physical density value of the adjacent layer directly below. Based on the calculation results of all layers, a physical density field without suspended structures and without inverted loops is generated.

3. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 2, characterized in that, Before performing the maximum value projection operation layer by layer, the following is also included: Based on the preset minimum draft angle, determine the lateral outward offset corresponding to each layer; For the physical density field of the adjacent layer directly below, perform a continuously differentiable morphological dilation process according to the lateral outward offset, and use the processed physical density value for the maximum value projection calculation of the current layer.

4. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, Before completing the mechanical response solution and design variable update based on the physical density field, the process also includes: The corrosion radius is determined based on a preset minimum wall thickness value, and a continuously differentiable morphological corrosion operation is performed on the physical density field to obtain the post-corrosion density field. Based on the volume difference ratio between the physical density field and the density field after corrosion, a quantitative assessment of the minimum wall thickness constraint is completed, wherein the volume difference ratio is used to characterize the proportion of structures with a minimum wall thickness in the physical density field. The expansion radius is determined based on the preset maximum wall thickness value, and a continuously differentiable morphological expansion operation is performed on the physical density field to obtain the expanded density field. Based on the proportion of the volume difference between the expanded density field and the physical density field, a quantitative assessment of the maximum wall thickness constraint is completed, wherein the proportion of the volume difference is used to characterize the proportion of structures in the physical density field that are greater than the maximum wall thickness. The wall thickness fluctuation range of the physical density field is determined based on the volume difference between the density field after corrosion and the density field after expansion. By constraining the upper limit of the wall thickness fluctuation range, quantitative control of wall thickness uniformity is achieved.

5. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, The process of solving for the mechanical response and updating the design variables based on the physical density field includes: A finite element model is constructed based on the physical density field, and the mechanical response quantities of structural compliance, volume, and modal frequencies in the current iteration step are obtained by solving the model. The sensitivity of each mechanical response quantity and process constraint quantity to the physical density field is calculated by the adjoint method, and then the sensitivity is fed back to the design variable corresponding to the experimental density field by the chain rule. The returned sensitivity is input into the gradient-based optimizer to update the design variables.

6. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, The construction of a two-layer design variable space, including topology layout variables and stiffener cross-sectional dimension variables, includes: For the stiffener network obtained by the skeletonization identification, the cross-sectional height and cross-sectional width are defined as cross-sectional dimension variables for each stiffener; For the topology layout of the connection points of the stiffeners and the key areas for load transfer, define fine-tunable topology layout variables; The cross-sectional dimension variables are combined with the topology layout variables to form a two-layer design variable space.

7. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, The process of obtaining the globally optimal design parameters of the rear housing of the headlight to be optimized through a global optimization algorithm includes: With minimizing structural flexibility as the optimization objective and lightweighting objective, modal frequency requirements, allowable stress requirements, and injection molding manufacturability requirements as constraints, a global collaborative optimization model is constructed. A global optimization algorithm based on an adaptive surrogate model is used to perform optimization on the combined surrogate model, thereby obtaining the globally optimal design parameters of the rear housing of the headlight to be optimized.

8. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 7, characterized in that, The verification of the injection molding manufacturability requirements includes: For each set of design parameters to be solved, the non-overhead constraint verification is first completed through density field deformation mapping, and then the wall thickness requirement verification is completed through continuously differentiable morphological constraints, ensuring that all solution processes are performed within the stated manufacturable feasible domain.

9. The method for optimizing the rear housing structure of a vehicle headlight as described in claim 1, characterized in that, The process of reconstructing and outputting the final three-dimensional digital model of the rear housing of the headlight to be optimized based on the globally optimal design parameters includes: Based on the topology layout variables in the global optimal design parameters, the final stiffener layout is determined; Based on the cross-sectional dimension variables in the global optimal design parameters, each stiffener is assigned corresponding cross-sectional parameters, and injection molding process features are automatically added; wherein, the injection molding process features include at least one of the following features: draft feature and root fillet. The 3D digital model of the rear housing of the headlight is generated and output by reconstructing the kernel of the 3D design software with complete process features.

10. A type of automotive headlight, characterized in that, The vehicle headlight includes a rear headlight housing, the structure of which is optimized by the method described in any one of claims 1 to 9.

Images