Computer-implemented method for transferring a component geometry into a CAD application, method for manufacturing a component, component and vehicle

Abstract

The invention relates to a computer-implemented method for transferring a component geometry mapped onto a discrete grid to a freeform surface model in a CAD application. The method according to the invention is characterized by the following process steps: - Changing the wall thickness (W) of a first freeform surface model (FM1) in accordance with the optimized shape of the component geometry represented by a second discrete grid; - Checking the first freeform surface model (FM1) to see if at least one undercut (1) exists in a given demolding direction (2) of the shell component, and if so: - in the area of ​​the undercut (1): move a first surface (O1) of the first freeform surface model (FM1) in the direction of or against the normal vector (N) of the first surface (O1) until the undercut (1) is compensated; - Parallel displacement of a second surface (O2) of the first freeform surface model (FM1) opposite the first surface (O1) in the same direction such that the wall thickness (W) in the area of ​​the former undercut (1) remains constant; and - Save the second freeform surface model (FM2) generated in this way as a second CAD model (CAD2).

Description

[0001] The invention relates to a computer-implemented method for transferring a component geometry mapped on a discrete grid to a freeform surface model in a CAD application according to the type defined in more detail in the preamble of claim 1, a method for manufacturing a component, a component and a vehicle.

[0002] Vehicles should be as light as possible to provide good acceleration and minimize fuel consumption during use. Vehicle manufacturers are therefore striving to reduce the weight of vehicle components. At the same time, sufficient crash safety must be ensured. Individual vehicle components, especially structural parts, must not fail when subjected to a specified load. In particular, load-bearing structural elements must not tear or deform excessively. Computer-aided engineering (CAE) can support engineers in the development of components.

[0003] It is well-established in the art to use finite element methods in structural optimization. For this purpose, the geometry of a component is first modeled using computer-aided design (CAD) and saved as a CAD model. The component geometry is defined by freeform surfaces, the contours of which can be mathematically described by various parameters. These freeform surfaces define the outer skin of the component, with the volume enclosed by the freeform surfaces being virtually filled with material. This allows for a comparatively accurate representation of the actual component geometry. To perform finite element simulations, the component geometry must be discretized. For this purpose, a computational mesh is generated based on the CAD model. Due to the mesh structure, steps or jumps can occur in the component geometry, meaning that the computational mesh only provides an approximation of the exact component geometry.For shell-like components, so-called "mid-surface models" are often used, in which the nodes of the computational grid are distributed along the shell-like walls, with the wall thickness assigned to the nodes in the form of thickness information. Such shell-like components are usually initially designed with a constant wall thickness.

[0004] During structural optimization, relevant load collectives for the planned application are applied to the underlying finite element model, and the resulting deformations and stresses of the component under the given material parameters are analyzed. As the optimization progresses, the component geometry is modified manually or automatically from iteration to iteration, and the impact on deformation and failure behavior is analyzed. This allows for the identification of an improved component geometry that leads to better deformation and load-bearing behavior. Based on this optimized geometry, corresponding components can then be manufactured, incorporating the improved physical properties resulting from the optimized shape. For these shell-like components, this typically means optimizing the initially constant wall thickness, resulting in a shell component with an inhomogeneously distributed wall thickness.

[0005] This requires transferring the optimized component geometry resulting from the structural optimization into a CAD model, which can then be used as an input file for a suitable machining or manufacturing machine. In this way, such a machine is able to derive control commands from a CAD file or CAD model.

[0006] One difficulty that arises in this context is that the component geometry in structural optimization is represented by a discrete mesh, while in a corresponding CAD program it is described by parameter-dependent freeform surfaces. Various methods exist for converting the results of a structural optimization into a CAD model. For example, the thickness of the freeform surfaces can be varied in the CAD application. However, due to the smooth transitions, this only allows for an approximation of the component geometry described by the finite element mesh. Furthermore, the manual effort is considerable.

[0007] Another method for transferring the results of a structural optimization for a shell component into a CAD model is also known from the applicant's patent application, German file number 10 2024 000 190.5, which was not yet published at the time of filing. This method involves using the thickness information of the component geometry resulting from the structural optimization, which is assigned to the respective nodes or grid elements of the computational mesh of the FEM model, to shift the nodes of the mesh accordingly. The computational mesh then assumes the shape of the optimized component geometry. The freeform surfaces of the original CAD model are then subdivided into a multitude of polygons, and the polygons of the freeform surfaces are shifted according to the geometry of the computational mesh. However, this approach has some disadvantages.Segmenting the freeform surfaces of a CAD model results in a comparatively high number of polygons. Consequently, the file size of the underlying CAD file increases, wasting valuable storage space. Furthermore, loading the model into a computer system's RAM takes longer. Adjusting the CAD model also requires significant manual effort. Subsequent modifications are difficult because the design history within the CAD application is extensively updated. Due to the high polygon count, visualizing the component geometry within the CAD application can also be challenging, leading to jerky movements and / or waiting times before the CAD program updates the view when moving or rotating the virtual component.

[0008] The transfer of an optimized component geometry into a CAD geometry is also known, for example, from DE 10 2021 003 565 A1. Methods and means for finding a topology-optimized component geometry are also known from US 2022 / 0284153 A1.

[0009] Various manufacturing processes, such as injection molding, require the removal of a component from a mold. For this, the mold is opened and the component is removed from the holder in a demolding direction. This necessitates a component geometry free of undercuts. This means that the wall thickness of the component must increase or decrease continuously and not alternate between thin and thick. However, when using the methods described above for transferring the component geometry into a CAD application, such undercuts can occur. Therefore, it is necessary to provide methods and means to avoid such undercuts.

[0010] It is known that if undercuts occur, they can be compensated for by thickening the material in that area. However, this increases the wall thickness again, so the resulting component geometry deviates from the optimized shape. This means that weight-saving potential cannot be fully exploited.

[0011] The present invention therefore aims to provide an improved method for transferring a component geometry into a CAD model, which allows the optimized component geometry developed through structural optimization to be represented in the CAD model with the greatest possible agreement and thereby ensures manufacturability.

[0012] According to the invention, this problem is solved by a computer-implemented method for transferring a component geometry into a CAD application with the features of claim 1, and by a method for manufacturing a component with the features of claim 3. Advantageous embodiments and further developments, as well as a correspondingly manufactured component and a vehicle with such a component, are described in the dependent claims.

[0013] A generic computer-implemented method for transferring a component geometry mapped onto a discrete grid to a freeform surface model in a CAD application, wherein a first CAD model, comprising a first freeform surface model describing the component geometry of a shell component in an initial form, is imported into a structural optimizer, the component geometry in its initial form is mapped onto a first discrete grid, a second discrete grid is generated by means of a structural optimization performed by the structural optimizer based on the first discrete grid, representing the component geometry in an optimized form, wherein, in the course of the optimization, a wall thickness of the shell component is changed, and thickness information is assigned to the nodes of the second discrete grid to describe the wall thickness.and the component geometry in the optimized form is transferred to a second freeform surface model to generate a second CAD model, is further developed according to the invention by the following process steps: - Changing the wall thickness of the first freeform surface model in accordance with the optimized shape of the component geometry represented by the second discrete grid; - Check the first freeform surface model to see if at least one undercut exists in a given demolding direction of the shell component, and if so: - in the area of ​​the undercut: Shifting a first surface of the first freeform surface model in the direction of or against the normal vector of the first surface until the undercut is compensated; - Parallel translation of a second surface of the first freeform surface model opposite the first surface in the same direction such that the wall thickness in the area of ​​the former undercut remains constant; and - Saving the second freeform surface model generated in this way as a second CAD model.

[0014] The method according to the invention makes it possible to transfer the optimized shape of the component geometry developed by the structural optimizer to the second CAD model with the greatest possible accuracy, while still compensating for any undercuts that may occur. The method according to the invention is based on the idea of ​​applying the wall thickness thickenings carried out to compensate for the undercut on the opposite side of the respective shell section of the shell component in the same direction, so that the wall thickness remains constant. "Constant" in this context means that the respective shell section has the same wall thickness after the undercut has been compensated as originally determined by the structural optimization. The advantageous physical properties imposed on the component by the structural optimization can thus be retained.

[0015] Undercuts are smoothed out so that the shell component can be manufactured. This could be, for example, a cast component, preferably an injection-molded component. Such a component is removed from the mold in the demolding direction. Since there are no longer any undercuts, removal in the demolding direction is ensured.

[0016] There are four different ways in which undercuts can be compensated for. For example, the first surface can be the outward-facing surface, the one extending away from the shell component, which in this context can also be referred to as the outer surface. To compensate for the undercut, it may be necessary to move the outer surface away from the shell component, thus increasing the wall thickness. The second surface in this context can be referred to as the inner surface. If the outer surface is moved away from the component, the wall thickness on the inner surface is reduced accordingly. The second surface therefore moves into the wall of the shell component.

[0017] It would also be possible that, to compensate for the undercut, the outer surface would need to be moved inwards into the component. To then maintain a constant wall thickness, the inner surface would need to be correspondingly convex.

[0018] It is also possible that the first surface represents the inner surface and the second surface the outer surface. In this case, to compensate for the undercut, it might be necessary to thicken the inner surface and, to maintain a constant wall thickness, to shift the outer surface into the wall or shell.

[0019] It is also conceivable that, to compensate for the undercut, the inner surface is shifted into the wall of the shell component, and that, to maintain a constant wall thickness, the outer surface has an outward curvature. These four different cases will be explained in detail later in the figures.

[0020] An advantageous further development of the method according to the invention provides that, to change the wall thickness of the first freeform surface model, the nodes of the second structured grid are shifted according to their respective assigned thickness information, the first freeform surface model is segmented into a plurality of polygons, wherein the vertices of the polygons in a top view of a respective freeform surface correlate with the arrangement position of the nodes of the second discrete grid, and the vertices of the polygons are shifted according to the arrangement position of the nodes of the second discrete grid. With the aid of this particular embodiment of the method according to the invention, the greatest possible agreement with the optimized shape of the component geometry could be ensured while simultaneously maintaining manufacturability.

[0021] Advantageously, the density or number of nodes or vertices of the polygons is varied depending on the change in material thickness, i.e., the extent of the displacement of the respective surfaces. If the first and second surfaces are displaced further, the number of nodes is also increased. This allows for smoother transitions along the component surface.

[0022] An inventive method for manufacturing a component involves imprinting the component geometry described by a second CAD model onto the component, wherein the second CAD model was generated using a method described above. Thanks to the component geometry imprinted on the second CAD model by the inventive method, the manufacturability of the component is ensured while simultaneously achieving the greatest possible conformity to the optimized shape developed through structural optimization. A component manufactured in this way is therefore lighter and / or more resilient than a component manufactured using conventional manufacturing methods. Despite the compensation of undercuts, a weight reduction can be achieved.

[0023] Preferably, the second CAD model is loaded into a machining center, and the machining center imprints the component geometry described by the second CAD model onto a workpiece. The machining center derives control commands by processing the second CAD model. The machining center can be any common machine for primary forming, forming, or machining workpieces. For example, it could be a 3D printer, a bending machine, a deep-drawing machine, a milling machine, or the like. By processing the second CAD model, the machining center is able to generate control commands for a corresponding control program, enabling the machining center to imprint the component geometry described by the second CAD model onto the workpiece.It is also possible for several machining steps to be performed sequentially, even by multiple machining centers, or for intermediate steps to be carried out. For example, a milling operation could be performed on a first machining center, followed by a forming process on a second machining center, and then a roughing operation on a third machining center. Alternatively, for instance, an injection mold could be milled from a solid block using a milling machine, and corresponding components could then be manufactured using injection molding based on the resulting mold.

[0024] Preferably, the component is pre-molded using an injection molding machine. As already mentioned, injection molding requires a component geometry free of undercuts in the demolding direction. Thanks to the use of the method according to the invention, the component geometry is free of undercuts, thus ensuring successful manufacturing. For this purpose, corresponding injection molds are produced based on the optimized shape of the component geometry contained in the second CAD model. For example, such an injection mold can be produced automatically by a CNC-controlled milling machine based on the second CAD model. Subsequently, the injection mold is inserted into a suitable injection molding machine and the injection molding process is started.

[0025] According to the invention, a correspondingly manufactured component also falls within the scope of protection of the invention.

[0026] A vehicle according to the invention preferably comprises at least one such component. Since, thanks to its manufacture using the inventive method, the component exhibits the optimized shape of the component geometry developed through structural optimization to the greatest possible extent, the improved physical properties resulting from structural optimization can also be transferred to the component. Such components are characterized by constant or even increased load-bearing capacity with a simultaneously reduced weight. Accordingly, the overall weight of the vehicle can also be reduced, or the load-bearing capacity of individual components can be increased.

[0027] Further advantageous embodiments of the computer-implemented method according to the invention for transferring a component geometry mapped on a discrete grid to a freeform surface model in a CAD application, as well as a method according to the invention for manufacturing a component, also result from the exemplary embodiments which are described in more detail below with reference to the figures.

[0028] This shows: Fig. 1 a schematic representation of a section of a CAD model of a shell component with an undercut; Fig. 2. A schematic representation of various possibilities for manipulating the wall thickness of the CAD model to compensate for undercutting while maintaining manufacturability; and Fig. 3 a schematic representation of a in Fig. 2. Manipulation shown in greater detail.

[0029] Shell components are used in vehicles to perform a wide variety of functions. Such a shell component is characterized by several differently oriented shell elements that are joined together to form a cohesive structure. The shell elements can be monolithic. Compared to components made from a solid material, shell components offer outstanding load-bearing capacity combined with low weight. A consistent wall thickness across the entire component is particularly advantageous. These components can be made of plastic, such as injection-molded parts.

[0030] To further reduce the overall weight of the vehicle, one development goal is to lower the weight or mass of such shell components even further. This is possible by saving material. For this purpose, the wall thickness of the shell elements can be made inhomogeneous across sections or even the entire component. In areas of lower stress, the wall thickness is thus reduced, and in areas of high stress, it is increased.

[0031] Fig. Figure 1 shows a section of a first CAD model CAD1 representing the component geometry of such a shell component. The component geometry is represented by a so-called first freeform surface model FM1. The wall thickness W is constant, but changes in the area of ​​an undercut 1. This is preferably an injection-molded component, which is to be removed from a corresponding material mold in a demolding direction 2. In the Fig. In the embodiment shown in 1, the structure of the shell component is depicted as approximately L-shaped. The in Fig. One vertical leg corresponds to a surface parallel to the demolding direction. The alternating decrease and increase in wall thickness W can make demolding difficult or even impossible.

[0032] Conventional methods for compensating for undercuts 1 involve increasing the wall thickness W in the area of ​​the undercut 1, so that the wall thickness W changes continuously, either only increasing or only decreasing. However, this leads to a change in the optimized shape of the component geometry developed through structural optimization, so that the improved physical properties actually imposed on the component by the structural optimization are lost or at least can no longer be fully utilized.

[0033] To solve this problem, the component geometry is revised using a computer-implemented method according to the invention. In this process, the geometry can be modified in Fig. Set the 2 shown cases. Fig. Figure 2 is subdivided into sub-figures a) to d), each representing a different application. Each sub-figure shows a section of a respective shell segment 3 of the shell component with wall thickness W. The surface facing an outer surface A can also be referred to as the outer surface AF. Due to the shell structure, the shell component at least partially encloses an interior space I. The surface of the shell component facing the interior space I can also be referred to as the inner surface IF.

[0034] To compensate for the undercut 1, it is necessary to shift a first surface O1 in the direction or opposite to its normal vector N. This is done in Fig. 2 indicated by a white arrow. In the Fig. 2a) and Fig. 2c) The displacement of the first surface O1 occurs in the direction of the normal vector N, i.e. out of the wall, whereas the displacement in the Fig. 2b) and Fig. 2d) opposite to the normal vector N, i.e. into the wall. Into the Fig. 2a) and Fig. 2c) therefore, it is a thickening of the wall and in the Fig. 2b) and Fig. 2d) by reducing the wall thickness W.

[0035] According to the invention, however, in order to maintain the improved component properties resulting from the structural optimization, the wall thickness W resulting from the optimization should also be retained in the area of ​​the undercut 1. According to the inventive method, the second surface O2, which is opposite the first surface O1 on the first freeform surface model FM1, is now shifted parallel to itself. In doing so, the second surface O2 is moved behind the first surface O1 by such an amount that the wall thickness W resulting from the optimization is established. This is achieved in Fig. 2 indicated by a hatched arrow. In the Fig. 2a) and Fig. 2b) The first surface O1 is the outer surface AF and the second surface O2 is the inner surface IF. In the Fig. In 2c) and 2d), however, the first surface O1 is the inner surface IF and the second surface O2 is the outer surface AF.

[0036] After the corresponding operations have been carried out, the modified freeform surface model is saved as a second freeform surface model, FM2, and then as a second CAD file, CAD2. This is exemplified in Fig. 3 shown.

[0037] Fig. Figure 3 shows a corresponding operation for adapting the first freeform surface model FM1 to generate the second freeform surface model FM2 in greater detail. This is the one described in Fig. 2a) the case shown. A median surface MF projecting through the wall is also depicted. The resulting wall thickness W between the first and second surfaces O1, O2, between the first surface O1 and the median surface MF, and between the second surface O2 and the median surface MF is illustrated by various arrows. These are dimensionless distances. Fig.Figure 3 shows that thickening the first surface O1, or the outer surface AF, would increase the wall thickness and thus deviate from the target value. By adjusting the opposite surface, in this case the second surface O2, or the inner surface IF, the wall thickness W can be kept constant, so that the target value is achieved.

[0038] Using the method according to the invention, the optimized shape of the component geometry developed through structural optimization can be transferred more precisely to a second CAD model, CAD2. The improved properties resulting from the structural optimization can thus be more reliably transferred to a workpiece to be manufactured in reality. Manufacturability is ensured. Components manufactured in this way are significantly lighter than those manufactured using conventional methods, which reduces manufacturing costs and lowers energy consumption when integrated into a vehicle. This typically results in a reduction of CO2 emissions. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2021 003 565 A1

[0008] US 2022 / 0284153 A1

[0008]

Claims

[1] Computer-implemented method for transferring a component geometry mapped onto a discrete grid to a freeform surface model in a CAD application, wherein a first CAD model (CAD1), comprising a first freeform surface model (FM1) describing the component geometry of a shell component in an initial form, is imported into a structural optimizer, the component geometry in the initial form is mapped onto a first discrete grid, a second discrete grid is generated by means of a structural optimization carried out by the structural optimizer based on the first discrete grid, mapping the component geometry in an optimized form, wherein a wall thickness (W) of the shell component is changed in the course of the optimization, wherein thickness information is assigned to the nodes of the second discrete grid to describe the wall thickness (W),and the component geometry in its optimized form is transferred to a second freeform surface model (FM2) to generate a second CAD model (CAD2), , characterized by the following procedural steps: - Changing the wall thickness (W) of the first freeform surface model (FM1) in accordance with the optimized shape of the component geometry represented by the second discrete grid; - Checking the first freeform surface model (FM1) to see if at least one undercut (1) exists in a given demolding direction (2) of the shell component, and if so: - in the area of ​​the undercut (1): Moving a first surface (O1) of the first freeform surface model (FM1) in the direction of or against the normal vector (N) of the first surface (O1) until the undercut (1) is compensated; - Parallel displacement of a second surface (O2) of the first freeform surface model (FM1) opposite the first surface (O1) in the same direction such that the wall thickness (W) in the area of ​​the former undercut (1) remains constant; and - Save the second freeform surface model (FM2) generated in this way as a second CAD model (CAD2). [2] Method according to claim 1, characterized by, that to change the wall thickness (W) of the first freeform surface model (FM1), the nodes of the second structured grid are moved according to their respective assigned thickness information, the first freeform surface model (FM1) is segmented into a plurality of polygons, wherein the vertices of the polygons in a top view of a respective freeform surface correlate with the arrangement position of the nodes of the second discrete grid and the vertices of the polygons are moved according to the arrangement position of the nodes of the second discrete grid. [3] Method for manufacturing a component, characterized by , that the component is imprinted with the component geometry described by a second CAD model (CAD2), wherein the second CAD model (CAD2) was generated using a method according to claim 1 or 2. [4] Method according to claim 3, characterized by, that the second CAD model (CAD2) is loaded into a machining center and the machining center imprints the component geometry described by the second CAD model (CAD2) onto a workpiece, whereby the machining center derives control commands by processing the second CAD model (CAD2). [5] Method according to claim 3 or 4, characterized by that the component is formed using an injection molding machine. [6] component, characterized by a manufacturing process according to any one of claims 3 to 5. [7] vehicle, characterized by at least one component according to claim 6.

Images