WO2022101236A1

WO2022101236A1 - Method for providing a digital printing model and method for additively manufacturing a component

Info

Publication number: WO2022101236A1
Application number: PCT/EP2021/081181
Authority: WO
Inventors: Shahrooz Sadeghi Borujeni; Sven Crull
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2020-11-12
Filing date: 2021-11-10
Publication date: 2022-05-19
Also published as: DE102020214266A1

Abstract

The invention relates to a method (A) for providing a digital printing model (1, 1a) and to a method (B) for producing a component (2) on the basis of the printing model (1, 1a). A network model (3) of the component (2) to be produced is provided; heat treatment is simulated on the basis of the network model (3), whereby a digital deformation network model (11) is generated; the network model (3) is compared with the deformation network model (11), whereby local deviation amounts (13) between the models (3, 11) are determined; the models (3, 11) are oriented to each other such that a total deviation amount calculated from the local deviation amounts (13) between the network model (3) and the deformation network model (11) becomes minimal; and the printing model (1, 1a) is generated by modifying the network model (3) on the basis of the local deviation amounts (13). Before the simulated heat treatment on the network model (3), a deformation hot spot (6) is identified, and network elements (4) of the network model (3) away from the deformation hot spot (6) are enlarged.

Description

description

Method for providing a digital print model and method for additively manufacturing a component

According to patent claim 1, the invention relates to a method by means of which a digital printing model is provided, by means of which a component can be produced additively. Furthermore, the invention according to patent claim 10 relates to a method for the additive manufacturing of a component.

Nowadays, additive manufacturing methods are used in particular to produce components whose geometric shape cannot be realized or can only be realized with the use of particularly great effort using conventional, for example metal-cutting, manufacturing methods. Furthermore, the additive manufacturing processes are becoming more and more profitable with increasing further development, which is why more and more additive manufacturing processes are currently finding their way into the series production of components.

The additive manufacturing method can be embodied, for example, as 3D printing, in particular as a binder jetting printing method. Accordingly, a production unit is designed as a 3D printer, for example. The manufacturing unit, by means of which the binder jetting method is used to produce components, is configured to apply directly alternating layers of a binder and a material powder to a printing bed of the 3D printer in a printing process. In this case, it can be provided, for example, that a first layer, which is applied to the printing bed by means of the 3D printer, is a binder layer in order to prevent the object to be printed from slipping on the printing bed in an undesirable manner. The material powder is, for example, a metallic powder, the material powder and the binder, which can be a liquid, being designed or selected in such a way that two adjacent material powder layers are connected to one another by means of a binder layer. This means that in binder jet 3D printing, a first layer of material powder is applied to the print bed. A binder layer is then applied to this first layer of material powder at certain points (determined by a construction plan, such as a model of the component to be printed). This is followed by applying another material powder layer to the binder layer. The two material powder layers are connected to one another by means of the binder layer, glued for example. On the other hand, the first material powder layer consisting of individual powder elements is solidified by the individual powder elements being connected to one another by the application of the binder layer, since the binder penetrates into the material powder layer between the powder elements and connects the powder elements to one another.

Through such printing or production, an intermediate product - a so-called green part - "grows" with an increasing number of layers along a vertical printing axis, for example, with a digital print model including a geometry of the intermediate product or green part and in particular the points on which the binder is applied, characterized. In other words, the print model is made available to the production unit, with the production unit being designed to use the digital print model to arrange the alternating layers of binding agent or material powder on the print bed.

In order to improve the mechanical properties (such as strength, etc.) of the intermediate product or green part and to produce the corresponding component, the intermediate product or green part is heat-treated, for example sintered in a sintering furnace. The component is therefore produced in that the green part is heat-treated, in particular sintered. During the heat treatment or sintering, deformations occur on the component resulting from the green part, in particular due to gravitation, friction of the powder elements and anisotropic material properties of the material powder and/or the binder, etc. Furthermore, during heat treatment or sintering, the binder loses volume and/or partially escapes from the green part/component or is expelled. In addition, porosity reductions occur. Because of this, the green part shrinks and/or warps non-linearly during the heat treatment, so that a final geometry of the sintered component, which was produced by sintering the green part, is deformed in relation to a target geometry of the component due to the non-linear deformations. Due to the non-linearity of the deformations, deformation hotspots or "focal points" occur on/in the component, at which the deformations are particularly pronounced and/or the deformations cannot be compensated for.

Accordingly, for the additive manufacturing of the component, there is a need to take into account and at least partially compensate for the deformations that occur as a result of the heat treatment during the printing process. For this purpose, for example, a simulation of the heat treatment is known, whereby an expected shrinkage and/or warping behavior of the green part can be determined. Based on appropriate Simulation results are to be used to change or modify the geometry of the digital print model in such a way that the deformations caused by the heat treatment of the green part result in the component having the final geometry that corresponds as closely as possible to the target geometry.

However, such a simulation of the heat treatment is particularly complex, since the non-linear deformations are particularly difficult to efficiently compensate. This is because non-linear deformations occur on/in the component in some areas of the component, as a result of which deviations between the final geometry and the target geometry are particularly difficult to calculate or to compensate for.

A method for additively manufacturing an object is known from US Pat. No. 1,0456,833 B2, a support structure being provided to which a material powder and a binder are applied based on a digital mesh model in order to form the object. The support structure is designed in such a way that shrinkage of the support structure and shrinkage of the object are coordinated with one another during heat treatment of the object.

WO 2018/026 962 A1 discloses a method for producing a three-dimensional object, wherein a material powder and a binder substance are applied to a powder bed based on a model of the object. In this case, a computer is used to create the model, by means of which an original design of the model is scaled in order to compensate for sintering shrinkage of the object that occurs as a result of sintering of the object.

Furthermore, US 2019 001 658 A1 discloses a computer device by means of which at least one part of a model of the part can be dynamically generated or modified during additive construction of a part. To this end, the computer device is designed to determine a difference between a portion of the part that has already been manufactured and a portion of the part that is still to be manufactured, and to determine whether the difference would lead to a deviation of the part from the model of the part. Further, the computing device is configured to dynamically create or modify the portion of the model of the part based on the difference. However, none of these conventional methods offer a possibility of already taking into account and at least partially compensating for the deformations occurring on the green part as a result of the heat treatment during the printing process.

The object of the present invention is to reduce the effort involved in additively manufacturing a component with particularly advantageous dimensional tolerances.

This object is achieved by a method designed according to patent claim 1 for providing a digital printing model, based on which a component can be produced additively. Furthermore, this object is achieved by a method designed according to patent claim 10 for additively manufacturing the component. Features, advantages and advantageous configurations of the method according to the invention for providing the digital print model are to be regarded as features, advantages and advantageous configurations of the method for additively manufacturing the component and vice versa.

According to a first aspect of the invention, a method for providing a digital print model is proposed, wherein the print model provided or that can be provided by means of the method can be used or is used for the additive manufacturing of the component. The component is, for example, a metallic component that can be produced or is produced by means of an additive manufacturing method. The component is in particular a component that is used in the field of machine and/or plant construction, for example as a vehicle component, as a component of a production unit for vehicles, etc. The component can be used, for example, in motor vehicle construction, in particular series motor vehicle construction, or be used, for example as a tool, component, semi-finished product, etc. or as a component thereof.

In a (for example first) step S1 of the method, a mesh model of the component is provided. The mesh model is provided, for example, by creating a CAD (computer-aided design) model of the component and then meshing the CAD model (or a copy thereof) with mesh elements such as finite elements. Accordingly, the network model has a large number of network elements which adjoin one another at network nodes and are connected to one another. The respective network element is polygonal in shape, for example triangular. A maximum size of the network elements—ie a maximum network element size—is selected in particular on the basis of a desired print quality or print resolution for a method for producing the component. The CAD model has a target geometry of the component, ie information or data that characterizes a shape and dimensions or relative dimensional ratios of the component. The target geometry corresponds to a final geometry which the finished component should have—as precisely as possible. Ideally, the component has the target geometry of the CAD model as the end geometry when the component has been completely manufactured. In other words, it is provided that the finished component and the CAD model correspond to one another. Since the mesh model is based on the CAD model, the mesh model has the target geometry so that it is provided that the finished component and the mesh model correspond to one another. However, as explained at the outset, this is not readily possible, particularly if heat treatment is required to produce the component. This is the case, for example, when a green part is first produced in order to produce the component by means of a binder jetting printing process, and the green part is then subjected to a heat treatment in order to finish producing the component.

In a further (e.g. second) step S2 of the method, a heat treatment--e.g. sintering--is simulated using the network model (or a copy thereof), whereby a digital deformation network model of the component to be manufactured is generated. The deformation network model is thus generated or provided in that, for example by means of a computer simulation, the network model is simulatively exposed to thermal energy occurring during the heat treatment. The deformation network model has a deformation geometry or a deformed geometry of the component, in particular an intermediate product or the green part, with the deformation geometry and the target or end geometry differing from one another due to deformations. In this respect, the deformation geometry has information or data that characterizes a shape and dimensions or relative dimension ratios of the component that the component would have if it had been manufactured using the unchanged mesh model. However, the deformation geometry is undesirable as the target or end geometry of the component.

The method has a further (for example third) step S3, in which the network model is compared with the deformation network model, as a result of which local deviation amounts between the network model and the deformation network model are determined. The respective local deviation amount is, for example, an amount of a linear measure that defines the local deviation as a straight distance between corresponding geometric elements (e.g. points, edges, surfaces, etc.) of the network model and the deformation network model. Furthermore, the local deviation amount can be an amount of an angular measure that characterizes the local deviation as a deviation angle between the mutually corresponding geometric elements of the network model and the deformation network model. In this case, the straight distance and/or the angle of deviation between the corresponding geometric elements is/are related in particular to one or more of the three spatial directions x, y, z. In particular, the respective local deviation amount between a network node of the network model and a corresponding network node of the deformation network model can be determined.

The method also includes a further (e.g. fourth) step S4, in which the network model and the deformation network model are aligned with one another in such a way that a total deviation amount is minimal, which is formed from the local deviation amounts between the network model and the deformation network model. The process of aligning the mesh model and the deformation mesh model with one another is carried out by moving/translating the two models relative to one another or one of the two models relative to the other of the models in a translatory and/or rotary manner. In this case, local deviation amounts are determined/calculated at predetermined/predeterminable points of the models, which are then added to the total deviation amount. The two models are said to be most aligned when the total amount of deviation, i.e. a sum of all local amounts of deviation, has become minimal. Accordingly, the two models in question are aligned in particular by means of a so-called best-fit algorithm (best-fit: best possible adjustment). The alignment is carried out without actually producing the models involved, for example in a simulative manner.

In a further (e.g. fifth) step S5 of the method, the print model of the component to be manufactured is generated by changing or modifying the network model (or the copy thereof) using the local deviation amounts. The digital pressure model is therefore based on the network model, which was changed/modified using the local deviation amounts. This means that the print model generated according to the method is a networked print model, ie a print model which includes network elements, for example finite elements. If, for example, the deformation network model has a smaller dimension between two points than the network model, the pressure model is created on the basis of the network model, with the dimension of the network model being extended by the amount of deviation at the corresponding points of the pressure model. should that If the deformation network model has a larger dimension between the two points than the network model, the dimension of the network model is reduced by the amount of deviation when/for creating the pressure model at the corresponding points of the pressure model. The procedure for creating the print model is analogous in order to compensate for angular deviations: If, for example, a surface or edge of the deformation mesh model is tilted by a deviation angle against a corresponding surface or edge of the mesh model, the corresponding surface/edge of the print model is tilted when/for creating the print model tilted in opposite directions by the deviation angle amount.

Due to the non-linearity of deformations occurring on the green part during an actual heat treatment, a deformation hot spot or “focal point” or more than one deformation hot spot occurs on/in the component. The deformations are particularly pronounced and/or cannot be compensated for at the respective deformation hot spot. For example, a (human) designer commissioned with the construction of the component can identify the deformation hotspots as such when creating the CAD model and/or mesh model and mark them as such in the CAD model and/or mesh model, for example. In particular, a respective deformation hot spot and a sharp edge or corner of the component coincide. In other words, when the green part is heat treated, for example, one of the deformation hot spots occurs at the sharp edge/corner. Furthermore, a respective deformation hot spot and a flat or flat surface, in particular surface, of the component coincide, which means that due to the heat treatment of the green part on the flat or flat surface/surface, one of the deformation hot spots occurs. In other words, the respective deformation hot spot can be a corner, an edge and/or a surface and, alternatively or additionally, another element of the CAD model/green part (e.g. one of a surface, corner and/or edge independent point) act.

In order to reduce the effort involved in additively manufacturing the component with particularly advantageous dimensional tolerances, the invention provides for a deformation hot spot to be identified on the network model (or on its copy) before step S2 (for example in a step S1a). or a plurality of deformation hot spots are identified and network elements of the network model arranged away from the deformation hot spot(s) are enlarged. In other words, the network model for the steps of the method to be carried out after the enlargement of the network elements or after step S1a comprises the enlarged network elements located away from the respective deformation hotspot. When enlarging the mesh elements away from the deformation hot spots, a number of mesh nodes and a number of the mesh elements are reduced, with the mesh nodes remaining after the enlarging being positionally arranged at the same place as before the enlarging of the mesh elements. This means that in order to enlarge the mesh elements away from the deformation hot spots, a proportion of the mesh nodes are eliminated. Provision is then made, for example, for the network model reduced by the corresponding network nodes to be re-meshed at least away from the deformation hotspots between the remaining network nodes, which means that, due to the reduced number of network nodes, a respective size of the network elements involved in the re-meshing in relation to the Mesh model is larger before zooming. It is to be understood that the network elements which are arranged at the respective deformation hot spot remain unchanged, ie are neither enlarged nor reduced.

This reduces the effort involved in providing the print model, in particular the effort involved in simulating the heat treatment, the heat treatment being, for example, sintering. This is because the reduced number of network elements is accompanied by a reduced computing and simulation effort, which leads to an expedient and target-oriented simulation result. Consequently, with a given computing power of a simulation unit, by means of which the simulation is carried out, the duration for carrying out the simulation decreases significantly, for example by a factor by which the number of network elements is reduced. In this way, the printing model is created or provided more efficiently in the method, as a result of which the additive manufacturing of the component can be carried out particularly easily and/or with little effort.

In a preferred embodiment of the method, the at least one or the respective deformation hot spot is identified on the mesh model (or its copy) by using a short simulation to simulate a heat treatment using the mesh model. This short simulation, which is carried out in particular as part of step S1a of the method, differs from the simulative heat treatment in step S2 in particular in that the short simulation is terminated or aborted after a predetermined or specifiable simulation period. The simulation duration specified/specifiable for the short simulation is in particular less than one minute, in particular only a few seconds, for example ten seconds. The purpose of the short simulation, in which the mesh model is exposed to the thermal energy that occurs during the actual heat treatment of the green part for the duration of the simulation, is to identify the deformation hot spots. The particularly short simulation time is sufficient for this the short simulation, according to which the deformation hot spots are recognizable/identifiable as part of the deformation geometry due to the (simulative) deformations that have occurred. On the other hand, the short simulation duration of the short simulation is not sufficient to fully provide the deformation geometry. The complete simulation of the heat treatment or sintering and the compensation loops takes orders of magnitude longer than the short simulation, from around several hours to more than a day.

It is conceivable that the short simulation and the simulative heat treatment described in connection with step S2 (referred to below only as simulation) merge seamlessly into one another. In other words, it can be provided that the deformation hot spot is identified on the network model by using the short simulation to simulate a heat treatment based on the network model, with the end or termination of the short simulation after the short simulation duration being a continuation of the simulative heat treatment, so that the deformation model of the component is available after the continued simulative heat treatment. In other words, the short simulation can be designed or carried out as part of the simulation. The determination or identification of the deformation hot spots by means of the short simulation of the heat treatment of the mesh model offers the advantage that the deformation hot spots take place particularly quickly.

An alternative preferred embodiment of the method provides that—for example in step S1a—the at least one or the respective deformation hot spot is identified on the network model by creating a rough network model based on the network model (or the copy thereof). will. Compared to the network model, the coarse network model has fewer network nodes, but the same size and a corresponding geometry. Consequently, the coarse mesh model has a coarser mesh than the mesh model, so that - with the same size and corresponding geometry of the mesh model and the coarse mesh model - the coarse mesh model includes fewer but larger mesh elements than the mesh model. With regard to the network elements, the target geometry in the coarse network model has a lower resolution than in the network model. It can be provided, for example, that the rough network model has only a fraction of the number of network elements of the network model, approximately half, a third, a quarter, a fifth, a sixth, etc.

Then, the heat treatment is simulated using the coarse mesh model, thereby generating a deformed coarse mesh model. The deformed coarse mesh model is thus generated, by, for example by means of a computer simulation, simulating the coarse mesh model being exposed to thermal energy occurring during the heat treatment. The deformation hot spots are recognizable/identifiable on the deformed coarse mesh model, so that the deformation hot spots can be identified at corresponding points of the mesh model.

The advantage here is that a simulation effort is particularly low. Because a required computing power or simulation power to achieve useful and desired results is directly related to a number of mesh elements of the model that is subjected to the simulation. Since the coarse mesh model has far fewer mesh elements than the mesh model, there is advantageously less effort to identify the deformation hot spots. In particular, the computing or simulation power required decreases with the fraction by which the coarse network model has fewer network elements in relation to the network model. Furthermore, at least one source of human error is eliminated from the method, since the simulative heat treatment of the coarse mesh model takes place by means of a simulation unit, which is designed, for example, as a computer unit or at least has one.

According to an advantageous development of the method, after step S5 (for example in a step S5a), the network elements arranged away from the deformation hot spots are reduced in size, as a result of which the pressure model is refined. In other words—after the print model of the component to be manufactured has been generated—the print model is refined, with the previously enlarged network elements away from the deformation hot spot(s) being reduced (again). For example, it can be provided that the number of network elements and the number of network nodes that were reduced in step S1a are increased again in such a way that after the increase, the number of network elements or network nodes returns to the respective number before the increase in the number corresponds to network elements.

In this context, it has proven to be particularly advantageous if the network elements arranged away from the deformation hot spot are reduced in size according to a predetermined or predeterminable factor. This means that the number of network elements is increased according to the given factor. In particular, the factor is selected in such a way that the network elements arranged away from the deformation hotspot are reduced in size in accordance with the specified or specifiable print resolution. It is also conceivable that the size of the network element is specified as a boundary condition for the network model, wherein the mesh element size and a desired print resolution of the component correspond to each other. In other words, the network element size of the network model is selected or specified according to the specified or specifiable print resolution. After the reduction, the corresponding network elements thus have the network element size that corresponds to a desired print quality. In other words, after the reduction in size of the network elements, the network model has the number of network elements that corresponds to the desired or predetermined/predeterminable print quality. As a result, a print file corresponding to the desired print quality, for example an STL file (STL: Stereo-Lithography or Standard Triangulation/Tessellation Language), can be created particularly easily and/or with little effort on the basis of the print model.

Since the reduction in size of the network elements, which are arranged away from the deformation hot spots, only takes place after the simulation of the heat treatment, in particular after the preparation of the print model, the reduced network elements or the increased number of network elements away from the deformation hot spots are Spots not part of the simulative heat treatment. As a result, the simulation of the heat treatment requires particularly little effort, which in turn reduces the effort involved in providing the print model again.

When the component is actually manufactured, there is a point on/in the component at which a particularly low degree of deviation occurs during sintering or heat treatment. In particular, this point is free from deviation. So this point has zero deformation, which is why this point is called the zero deformation point. The zero deformation point and a footprint over which the green part for heat treatment is placed coincide. In other words, with respect to a height of the green part, the zero deformation point is at the same height as the footprint. In addition, the zero deformation point is congruent with the center of gravity of the green part in relation to a width and a length of the green part. The same applies when the finished sintered or heat-treated component is considered: the zero deformation point lies in a contact plane of the component and is congruent with the component's center of gravity in relation to a width and a length of the component. Accordingly, the green part at the zero deformation point is not deformed or only deformed to a particularly small extent by the heat treatment. To put it simply, the green part deforms or shrinks “towards the zero deformation point” or “around the zero deformation point” as a result of sintering. Since the zero deformation point of the corresponding model is known or can be easily determined, according to a further advantageous embodiment of the method, the network model and the deformation network model are aligned with one another with a condition according to which a zero deformation point of the network model and a corresponding zero deformation point of the deformation network model are arranged next to one another , such that the total deviation amount formed from the local deviation amounts is minimal when this condition is met. In the method, therefore, the zero deformation point of the network model and the zero deformation point of the deformation network model are first ascertained or determined. After this, the network model and the deformation network model are aligned with one another, so that the zero deformation point of the network model and the zero deformation point of the deformation network model coincide. Since a positional arrangement of the zero deformation point does not change by sintering, the zero deformation point can be regarded as a common point of the mesh model and the deformation mesh model. Step S4 of the method is therefore carried out on the premise that the zero deformation points which correspond to one another remain arranged on one another, in particular aligned with one another.

The condition according to which the zero deformation point of the network model and the corresponding zero deformation point of the deformation network model are arranged next to one another, in particular congruently, is met if the zero deformation point of the network model and the zero deformation point of the deformation network model coincide or coincide with or against one another at or during the alignment of the network model and the deformation network model remain congruently arranged. Furthermore, the condition is met if the zero deformation points coincide due to the alignment of the mesh model and the deformation mesh model. Due to the condition that the zero deformation points do not fall apart or differ from one another during alignment, an advantageously particularly low effort for providing the pressure model, in particular for aligning the network model and the deformation network model with one another, is achieved.

It has also proven to be advantageous in the method if a (first) permissible tolerance amount is specified for the respective local deviation amount. Furthermore, a loop comprising steps S2, S3 and S4 is carried out iteratively until the respective local deviation amount corresponds to the (first) permissible tolerance amount. In particular, this is before the generation of the pressure model - for example in a step S4a - a first iteration network model of the component generated by the network model based on the local Deviation amounts are changed or modified. The iteration network model is fed to step S2 for a first repetition and for further repetitions of the loop. This means that for the first iteration, the loop involves simulating the heat treatment against the iteration mesh model, which creates an iteration deformation mesh digital model of the part to be manufactured. The loop also includes comparing the network model with the iteration deformation network model, as a result of which the local deviation amounts between the iteration deformation network model and the network model are determined. In addition, the loop includes aligning the iteration deformation mesh model and the mesh model, thereby minimizing the total error amount formed from the local error amounts between the iteration deformation mesh model and the mesh model. The loop of the method also includes step S4a, in which a further iteration network model of the component is generated by changing or modifying the first iteration deformation network model based on the local deviation amounts. The further iteration network model is then fed to step S2 of the loop for a further repetition of the loop, etc.

In this case, it is provided in particular that the permissible tolerance amount is specified for the entire component, that is to say over the entire CAD model or network model. In other words, it is provided that the external dimensions of the component correspond to the permissible tolerance amount after its heat treatment. By executing the loop of the method iteratively until the local deviations correspond to the permissible tolerance amount, the mesh model and the deformation mesh model are aligned particularly precisely with one another.

In this context, it has been shown to be particularly advantageous for the method if a special geometry element of the network model and a special geometry element of the deformation network model are determined, and the loop S10 is carried out iteratively until the respective local deviation amount between the special geometry elements corresponds to the permissible tolerance amount, with offside of the special geometry elements, i.e. on other geometric elements of the mesh model or component, a larger tolerance amount is allowed. Here, the respective special geometry element is for example predetermined or specified, in particular by the designer identifying the special geometry elements as such when creating the corresponding CAD model or network model and marking them as such in the CAD model and/or network model, for example. The respective special geometry element is, for example, a respective point, a respective (even or odd) edge, around a respective corner around a mesh node and/or around a respective (flat or uneven) surface of the component or the mesh model.

This means that the special geometry elements are aligned with one another as precisely as possible, with a second permissible tolerance amount being provided on the other geometry elements (which are generally not special geometry elements here), the second permissible tolerance amount being greater than the first permissible tolerance amount. In other words, it is provided that, for example by means of the best-fit algorithm, a deviation amount that is formed from the local deviation amounts at the special geometry elements is primarily minimized by the alignment, with another deviation amount sum that is formed from the local deviation amounts at the other geometry elements is formed, is minimized subordinately when aligning. This results in the advantage that the special geometry elements, which are of particular importance for the functionality and/or for the production of the component, are arranged particularly accurately or precisely on one another, in particular congruently.

As a result, the effort involved in aligning the network model and the deformation network model and consequently in providing the pressure model is again reduced in an advantageous manner. Since the special geometry elements are of primary importance for the functionality and/or for the production of the component, a complex, particularly precise alignment of the models away from the special geometry elements - i.e. for example on the other geometric elements - can be dispensed with to increase the efficiency of the method. Nevertheless, simple manufacturability and the functionalities provided by the component remain guaranteed.

Furthermore, according to a further advantageous embodiment of the method, before step S2, in which the heat treatment is simulated using the mesh model and the deformation mesh model (or possibly the iteration deformation mesh model) of the component to be manufactured is generated, the mesh model is scaled based on an expected or probable deformation . In other words, before the simulative heat treatment or before the simulative sintering—before or after enlarging the mesh elements arranged away from the deformation hotspots—a scaled mesh model is generated (for example in a step S1b). In this case, the scaling takes place, in particular, translationally along one or more of the three spatial directions x, y, z. A scaling factor, by which the network model is scaled, emerges, for example, from internal experiments, for example from previous heat treatments of at least similarly designed ones Green parts with similar or the same parameters (temperature, duration, etc.) of the corresponding heat treatment. In this way, the outlay for the method is further reduced since, due to the scaling of the mesh model, the deformations are already roughly compensated for before the heat treatment is simulated.

As a result, the effort, in particular computing or simulation effort, for generating the deformation network model or the iteration deformation network model decreases. Furthermore, the effort involved in determining the deviation amounts and in aligning the network model and the deformation network model with one another is reduced. As a result, the effort involved in providing the print model is reduced overall. Step S1b can be part of the iteratively executable loop S10 (see above).

According to a second aspect of the invention, a method for additively manufacturing a component is proposed. In this manufacturing method, a material of the component to be manufactured is arranged according to a digital print model of the component. Furthermore, in the manufacturing process, the material is joined to form the component by means of heat treatment, in particular by means of sintering. The digital print model is provided using a method designed according to the above description. In particular, it is provided that an intermediate product or green part of the component is produced using binder jetting 3D printing. As explained above, the print model is provided as a networked print model. For this purpose, in particular an STL file is generated from the print model, which is made available to the production unit. This production unit (for example a binder jetting 3D printer) is designed to accept the networked print model as input data and to arrange the material of the component and a binder for the green part based on the networked print model.

The invention also includes developments of the production method according to the invention, which have features as have already been described in connection with the developments of the method according to the invention for providing the digital print model. For this reason, the corresponding developments of the manufacturing method according to the invention are not presented again here.

The invention also includes the combinations of features of the described embodiments. An exemplary embodiment of the invention is described below. For this shows:

1 shows a flowchart to clarify a method for providing a digital print model, wherein a component can be produced additively using the print model;

Fig. 2 shows a schematic and perspective view of a network model;

Figure 3 is a schematic view of a portion of the mesh model showing deformation hot spots;

4 shows a schematic view of an area of the network model, with a number of network elements networked with one another;

Figure 5 is a schematic view of the portion of the mesh model where the number of mesh elements has been reduced;

6 shows a schematic and perspective view of a deformation network model;

7 is a schematic view of a process in which the mesh model and the deformation mesh model are compared with each other;

Fig. 8 is a schematic view of the print model;

9 shows a schematic view of a refined pressure model; and

10 shows a schematic and perspective view of the component.

The exemplary embodiment explained below is a preferred embodiment of the invention. In the exemplary embodiment, the described components of the embodiment each represent individual features of the invention to be considered independently of one another, which also develop the invention independently of one another and are therefore also to be regarded as part of the invention individually or in a combination other than that shown. Furthermore, the embodiment described can also be supplemented by further features of the invention already described. Elements with the same function are each provided with the same reference symbols in the figures.

Fig. 1 shows a flow chart to illustrate a method A for providing a digital print model 1 (see Fig. 8 or Fig. 9), wherein the print model 1 is used to create a component 2 (see Fig. 10, in which a schematic and perspective view of the component is shown) can be produced additively. Furthermore, a method B for additively manufacturing the component 2 is illustrated by the flowchart in FIG. 1 . For the following, method A is considered a part of method B. Procedures A, B are described together. Where necessary, the differences between methods A and B are discussed. If reference is only made to “a method” or “the method” (without a reference number), the associated statements apply to methods A and B considered individually or in combination.

For the production of the component 2, for example by means of method B, it is provided that an intermediate product (not shown) is first produced by means of 3D printing, in particular by means of binder jetting 3D printing, using the digital printing model 1. The intermediate product, which can also be referred to as a green part, is produced by alternately arranging material powder layers and binder layers, with the respective binder layer being applied selectively to a respective material powder layer according to the printing model 1 . The result is the intermediate product or green part, which is made up of alternating layers of material powder or binder. Accordingly, the material powder layers are bonded to one another by means of the binder. Next, powder elements of the respective material powder layer are glued together.

It is preferred if a material connection is formed directly between the material powder layers and not only indirectly, that is to say not only by means of the binder. For this purpose it is provided that the intermediate product is heat-treated. For example, the green part is sintered, creating a direct material connection between the powder elements of the respective material powder layer and a direct material connection between the material powder layers. Furthermore, the binder is at least partially dissolved or expelled from the intermediate product by sintering, for example by liquefaction, evaporation or sublimation. As a result, the intermediate product is deformed / deformed. A component (not shown) produced in this way does not correspond, or insufficiently, to specifications that are specified by the target geometry or by a CAD model of the component 2 due to the deformations that have occurred as a result of the sintering. This is where procedure(s) A and/or B intervene. In a first step S1 of the (respective) method, a digital network model 3 is provided, which is shown in FIG. 2 in a schematic and perspective view. For example, a designer uses suitable software or a CAD program to create the CAD model, which has a target geometry of the component 2, i.e. information or data that indicates a desired shape and dimensions or relative dimensional ratios of the component 2 characterize. In order then to provide the network model 3, the CAD model is then networked with network elements 4, such as finite elements. For reasons of clarity, only a few of the network elements 4 are provided with the corresponding reference numbers in the figures. Accordingly, the network model 3 has a multiplicity of network elements 4 which adjoin one another at network nodes 5 (see FIG. 4 or FIG. 5) and are connected to one another. In the present case, the respective network element 4 is triangular in shape. A maximum size of the network elements 4--that is, a maximum network element size--is selected or specified on the basis of a desired print quality or print resolution for the production of the component 2.

In a step S1a, deformation hotspots 6 are identified on the network model 3 . For example, a respective deformation hot spot 6 and a sharp edge 7, a corner 8 or another point of the component 2 coincide. In other words, during the heat treatment of the green part designed as sintering, one of the deformation hot spots 6 occurs, for example, at the sharp edge 7, at the corner 8 and/or at the other point. Furthermore, a respective deformation hot spot 6 and a flat or flat surface, in particular surface 9, of the component 2 can coincide, which means that due to the heat treatment of the green part on the flat or flat surface/surface 9, one of the deformation Hot spots 6 occurs.

The deformation hot spots 6 can be identified as such by a (human) designer commissioned with the construction of the component 2 when creating the CAD model and/or network model 3 and as such, for example in the CAD model and/or network model 3 are marked. However, it is particularly preferred if the respective deformation hot spot 6 is identified by simulating a heat treatment using the mesh model 3 by means of a brief simulation. In this case, the short simulation is ended or aborted after a predetermined or predeterminable simulation period. The simulation duration for the short simulation is less than one minute, in particular only a few seconds, for example ten seconds. Purpose of the short simulation, in which the network model 3 for the duration of the simulation simulates the actual heat treatment of the Green part is exposed to thermal energy occurring, is to identify the deformation hot spots 6. The particularly short simulation time of just ten seconds is sufficient for this.

Alternatively, in step S1a, the deformation hotspots 6 are identified on the network model 3 by initially creating a coarse network model based on the network model 3. In comparison with the network model 3, the coarse network model has fewer network nodes 5, but the same size and a corresponding geometry. Consequently, the rough mesh model has a coarser mesh than mesh model 3, so that the rough mesh model includes fewer but larger mesh elements 4 than mesh model 3. With regard to network elements 4, the target geometry in the rough mesh model is less resolved than in mesh model 3. It can For example, it can be provided that the coarse mesh model has only a fraction of the number of mesh elements 4 of the mesh model 3, about half, a third, a quarter, a fifth, a sixth, etc. Then the heat treatment is simulated using the coarse mesh model, whereby a deformed coarse mesh model is generated. The deformed coarse mesh model is thus generated by, for example by means of a computer simulation, simulatively exposing the coarse mesh model to thermal energy occurring during the heat treatment. The deformation hot spots 6 are recognizable/identifiable on the deformed coarse mesh model, so that the deformation hot spots 6 can be identified at corresponding points of the mesh model 3 .

Step S1a also includes an enlargement of network elements 4 of the network model 3 that are arranged away from the deformation hotspots 6. The network model 3 then includes the steps of the method to be carried out after the enlargement of the network elements 4 or after step S1a respective deformation hot spots 6 and enlarged network elements 4. The network elements 4, which were enlarged in step S1a of the method, bear the reference numeral 10 in the figures Network elements 4 are enlarged is shown in FIG. 4 and in FIG.

6 shows the network model 3 in a schematic and perspective view, in which according to step S1a a proportion of the network elements 4 have been converted to the enlarged network elements 10 . It can be seen that the network elements 4, which are arranged at the respective deformation hot spot 6, remain unchanged, ie are neither enlarged nor reduced. The network elements 4 arranged or involved in the deformation hot spots 6 bear the reference number 22 in the figures around network elements 4 of the network model 3, through which, for example, one of the edges 7, corners 8 and/or surfaces 9 of the network model 3 or of the component 2 are characterized.

Fig. 4 shows a schematic view of an area of the network model 3 with a number of interconnected network elements 4, and Fig. 5 shows a schematic view of the area of the network model 3 in which the number of network elements 4 has been reduced or which apart from the Deformation hot spots 6 arranged network elements 4 to the network elements 10 were enlarged. This means that the enlarged network elements 10 are generated by enlarging the corresponding network elements 4 of the network model 3 .

When enlarging the mesh elements 4 away from the deformation hot spots 6, a number of mesh nodes 5 and consequently a number of the mesh elements 4 are reduced, with the mesh nodes 5 remaining after the enlargement being positionally arranged at the same place as before the enlargement of the mesh elements 10. This means that a portion of the network nodes 5 is eliminated in order to enlarge the network elements 4 away from the deformation hot spots 6 . It is then provided that the network model 3 reduced by the corresponding network nodes 5 is re-networked at least away from the deformation hot spots 6 between the remaining network nodes 5, whereby due to the reduced number of network nodes 5 a respective size of the network elements involved in the new network 10 is larger relative to the mesh model 3 before scaling up.

In a synopsis of FIG. 4 with FIG. 5 it can be seen that the area of the network model 3 is reduced in step S1a by the network elements 4 which are not involved in one of the deformation hot spots 6 . The corresponding network elements 4 are shown in broken lines in FIG.

In a further step S2, a heat treatment is simulated using the mesh model 3, as a result of which a digital deformation mesh model 11 of the component 2 to be manufactured is generated, which is shown in a schematic view in FIG. The deformation network model 11 has a deformation geometry or a deformed geometry of the component 2, the deformation geometry and the target geometry differing from one another due to deformations. The result of the simulative heat treatment are the deformations that would occur in reality on the component 2 if the intermediate product were printed using the mesh model 3 and then sintered. Before simulating the heat treatment or sintering (ie before step S2), the mesh model 3 is roughly scaled. In the present case, the network model 3 is scaled before step S1a, ie before the network elements 4 are enlarged. For this purpose, before step S2, in particular before step S1a, the method has a further step S1b, in which a scaled network model 3a is generated. A scaling factor, by which the mesh model 3 is scaled, emerges from internal experiments, by means of which previous heat treatments of at least similarly designed green parts were examined. In particular, the scaling factor is greater than one, ie the scaled network model 3a is larger than the network model 3. The scaling takes place, for example, translationally along one or more of the three spatial directions x, y, z. It is also conceivable that a separate scaling factor is determined for each of the spatial directions x, y, z, it being possible for the respective separate scaling factor to be greater than one, smaller than one or equal to one. In other words, the network model 3 can be scaled in the respective spatial direction x, y, z with a different scaling factor, as a result of which the scaled network model 3a is produced. Alternatively or additionally, the scaling or the generation of the scaled network model 3a includes a respective rotation about one or more of the three spatial directions x, y, z. The method steps described below can be used in relation to the network model 3 and/or in relation to the scaled network model 3a. For this reason, the following description no longer explicitly distinguishes between network model 3 and scaled network model 3a.

In a step S3 of the method, the network model 3 and the deformation network model 11 are compared with one another. 7 shows a schematic view of a process in which the network model 3 and the deformation network model 11 are compared with one another, as a result of which local deviations 12 between the network model 3 and the deformation network model 11 are determined. The respective local deviation 12 has a deviation amount 13, for example a straight deviation length 14 and/or a deviation angle 15. To determine or determine the deviation amounts 13, the network model 3 and the deformation network model 11 are, for example, roughly superimposed, for example using a CAD program or by means of an FEM program.

A further step S4 of the method follows, in which the network model 3 and the deformation network model 11 are aligned with one another in such a way that a total deviation amount, which is formed by adding the local deviation amounts 13, becomes minimal. For example, for this purpose the deformation network model 11 is moved translationally and/or rotationally in relation to the network model 3 until the total deviation amount has become as small as possible. In particular, a best-fit algorithm is used here. The two models 3, 11 are considered to be best aligned with one another when the total amount of deviation, ie a sum of all local amounts of deviation 13, is minimal. To this extent, the total amount of deviation is reduced or minimized by the models 3, 11 being aligned with one another—in particular using the best-fit algorithm.

A zero deformation point 16 of the component 2 or of the respective model 3, 11 is shown in FIG. A particularly low degree of deviation occurs at the zero deformation point 16 during sintering or heat treatment. In particular, the zero deformation point 16 is free from a deviation. The zero deformation point 16 and a contact plane 17, over which the green part is set up, for example for the heat treatment, coincide. The zero deformation point 16 is therefore at the same height as the contact plane 17 in relation to a height of the green part. In addition, the zero deformation point 16 is congruent with the center of gravity of the green part in relation to a width and a length of the green part. The same applies when the finished sintered or heat-treated component 2 is considered: The zero deformation point 16 lies in a contact plane 17 of the component 2 and is congruent with the center of gravity of the component 2 with regard to a width and a length of the component 2. This applies to the network model 3 as well as for the deformation network model 11. Accordingly, the zero deformation point 16 can be regarded as a common point of the component 2.

The zero deformation point 16 of the network model 3 and the corresponding zero deformation point 16 of the deformation network model 11 are arranged directly next to one another, in particular congruently, for or when aligning the models 3, 11 with one another and/or remain directly next to one another, in particular congruently, for or when aligning the models 3, 11 with one another. arranged. In other words, the alignment of the network model 3 and the deformation network model 11 takes place with a condition according to which the zero deformation point 16 of the network model 3 and the corresponding zero deformation point 16 of the deformation network model 11 are arranged next to one another, with the total deviation amount formed from the local deviation amounts being minimal while complying with this condition will.

It is also provided here that a first permissible tolerance amount is specified for the respective amount of deviation 13 at least between a respective special geometry element 18 of the network model 3 and a corresponding special geometry element 19 of the deformation network model 11 . Furthermore, for the respective local Deviation amount 13 between another geometry element 20 of the network model 3 and a corresponding other geometry element 21 of the deformation network model 11 is a second permissible tolerance amount. The second allowable tolerance amount is greater than the first allowable tolerance amount. Consequently, when aligning the models 3, 11, a higher tolerance is permitted on the other geometry elements 20, 21 (see FIG. 7) than on the special geometry elements 18, 19 (see FIG. 7).

In particular, the respective special geometry element 18 , 19 is designed as a functional element of component 2 . The special geometry element 18, 19 or the functional element is of particular importance for the functionality of the finished component 2. For example, the functional element is the area or surface 9. The functional element can also be a hole or other material recess, such as a through hole or blind hole without a thread and/or provided with an internal thread, a groove, etc. It is also conceivable that the functional element is designed as a cylindrical or polyhedral material projection, for example as a threadless and / or provided with an external thread pin or as another correspondingly designed material elevation. In particular, the functional element acts in the finished component 2 as a material, positive and/or non-positive connection element to form a material, positive and/or non-positive connection device, for example between the component 2 and another component, around the component 2 and the further component To be connected to one another in a material, positive and/or non-positive manner.

Furthermore, the respective special geometry element 18 , 19 can be an auxiliary production element of the component 2 . Such a special geometry element 18, 19 designed as the auxiliary production element is produced, for example, during 3D printing of the intermediate product, in particular layered from the same material powder as the intermediate product or the component 2 itself . The respective auxiliary production element is designed, for example, as an auxiliary printing element or as a so-called support structure, on the basis of which material projections, in particular material overhangs, can be produced by means of 3D printing. Furthermore, these material projections or overhangs do not break off during the sintering of the green part or intermediate product and/or do not deform when the material projections or overhangs are supported by the auxiliary production elements. The auxiliary production element is therefore of particular importance for the production of the component 2 . When the network model 3 and the deformation network model 11 are aligned with one another, the special geometry elements 18, 19 of the models 3, 11 in particular are arranged particularly close to one another--if possible congruently. Step S4 of the method is thus carried out, for example, in such a way that the special geometry elements 18, 19 which correspond to one another remain arranged close to one another, in particular remain aligned with one another. Consequently, the total deviation amount formed from the local deviation amounts 13 becomes minimum while satisfying such a condition.

The other geometry elements 20, 21 are elements of the network model 3 or component 2, which have very little, in particular no, importance for the production and for the functionality of the component 2. A deviation amount sum, which is formed from the local deviation amounts 13 on the special geometry elements 18, 19, is primarily minimized or reduced by the alignment, with another deviation amount sum, which is formed from the local deviation amounts 13 on the other geometry elements 20, 21, when Align is minimized or reduced subordinate.

The method also has a loop S10, which comprises at least steps S2, S3 and S4. Loop S10 is repeated at least once, that is to say it is run through at least once more after it has been run through for the first time, with step S2, for example, following step S4 for repeating steps S2, S3, S4. In particular, the loop S10 is carried out iteratively until the respective deviation amount 13 corresponds to the first permissible tolerance amount. For this purpose, the method has a step S4a before the generation of the pressure model 1, in which, after the first iteration of step S4, a first iteration network model of the component 2 is generated by the network model 3 being changed or modified using the deviation amounts 13. Then, when executing loop S10, the iteration network model is fed to step S2.

In particular, the loop S10 has the steps S1b, S1a, S2, S3, S4 and S4a, with the step S1b following the step S4a for the repetition of the steps S1b, S1a, S2, S3, S4 and S4a.

Accordingly, for the first repetition of loop S10, the first iteration network model generated in step S4a is fed to step S1b of loop S10, in which the first iteration network model is optionally scaled to form a scaled first iteration network model. The first iteration network model is then made available to step S1a, in which—analogously to the above description—the network elements 4 are enlarged. hereafter the first (possibly scaled) iteration network model with the enlarged network elements 10 is provided to step S2 of loop S10, in which the first iteration network model is heat-treated or sintered in a simulative manner. As a result, a first iteration deformation network model is generated, which is compared with network model 3 in step S3 of loop S10 and the deviation amounts between network model 3 and the iteration deformation network model are determined. In step S4 of loop S10, the iteration deformation network model and the network model 3 are aligned with one another, the total deviation amount formed from the deviation amounts 13 between the iteration deformation network model and the network model 3 becoming minimal. Loop S10 also includes step S4a, in which a second iteration network model is created for a second repetition of loop S10 by changing or modifying the first iteration deformation network model using local deviation values 13 . The second iteration network model is then fed to step S1b of the loop for a further iteration of the loop.

For the iterative execution of the loop S10, this includes--for example as a sub-step of step S4a--a test in which the respective deviation amount 13, in particular at the special geometry elements 18, 19, is checked for compliance with the first permissible tolerance amount. In particular, the test includes checking the respective amount of deviation 13 apart from the special geometry elements 18, 19 - ie for example on the other geometric elements 20, 21 - for compliance with the second permissible tolerance amount. If the check establishes that at least the first tolerance amount is complied with, loop S10 is terminated and the method, for example after step S4, is continued with a further step S5.

In step S5 of the method, the print model 1 of the component 2 to be manufactured is now generated. For this purpose, the network model 3 is changed or modified using the local deviation values 13 . The digital pressure model 1 is therefore based on the network model 3, which was changed/modified using the deviation amounts 13 that occurred between the network model 3 and the deformation network model 11. If loop S10 was run through or executed at least once before step S5, digital pressure model 1 is based on network model 3, which was changed/modified using deviation amounts 13 that occurred between network model 3 and the iteration deformation network model. Fig. 8 shows a schematic and perspective view of the print model 1, which is deformed in relation to the network model 3 and in relation to the deformation network model 11 in such a way that when the intermediate product that has been produced according to the print model 1 is sintered, the intermediate product or green part is deformed so that the shape of the component 2 is formed.

9 shows the print model 1 in a schematic and perspective view, which was refined into a fine print model 1a in a step S5a of the method. Step S5a takes place after the print model 1 has been provided, ie after step S5. The network elements 10 arranged away from the deformation hot spots 6 and enlarged in step S1a are reduced in size (again) in step S5a, as a result of which the print model 1 is refined or the fine print model 1a is generated. In other words—after the printed model 1 of the component 2 to be produced, which has the enlarged mesh elements 10, has been generated—the printed model 1 is refined, the previously enlarged mesh elements 10 away from the deformation hotspot(s) 6 being reduced in size. The number of network elements 10 and the number of network nodes 5, which were reduced in step S1a, are increased again in such a way that after the increase, the number of network elements 4 or network nodes 5 again corresponds to the respective number before the network elements 4 were increased.

The previously enlarged network elements 10 arranged away from the respective deformation hotspot 6 are (again) reduced in size according to a factor. This means that the number of network elements 10 is increased according to the predetermined factor. In particular, the factor is selected in such a way that the network elements 10 arranged away from the deformation hot spots are reduced in size in accordance with the desired print resolution. In addition, the size of the network element can be specified as a boundary condition for the network model 3 or for the print model 1, 1a, with the size of the network element and the desired print resolution for the component 2 corresponding to one another - the smaller the size of the network element or the more network elements 4, the higher is the print resolution. After the reduction, the corresponding network elements 10 have the same network element size as the network elements 4 have.

Method B differs from method A by a step S6, in which the digital print model 1, preferably the fine print model 1a, is provided to a production unit, in particular a binder jetting 3D printer. The intermediate product or green part of the component 2 is then actually produced by means of the 3D printer and using the print model 1 . Next, step S6 includes the (real) sintering of the green part, whereby the component 2 is produced. To this end, step S6 of method B follows method A, in particular step S5a of method A. In other words, method A has steps S1, S1b, S1a, S2, S3, S4, S4a, S5 and S5a , whereas method B comprises steps S1, S1b, S1a, S2, S3, S4, S4a, S5, S5a and S6.

Overall, the invention shows how particularly efficiently and precisely a deformation occurring on important elements of the component 2 as a result of a heat treatment is compensated for. Since the network model 3 is used to simulate the sintering, in which at least a portion of the network elements 4 has been enlarged to form the network elements 10, the effort involved in carrying out the simulation is advantageously particularly low.

In this case, the method concentrates on enlarging the network elements 4 and, in a further embodiment, on reducing the previously enlarged network elements 10 again.

Although method A, B has been described here in connection with an additive manufacturing method, it should be understood that method A, B can also be used for other manufacturing methods in which heat is introduced into the corresponding component or into a corresponding intermediate product, such as when welding.

Reference List

A Method of providing a digital print model

B Process for the additive manufacturing of a component

1 print model

1a fine print model

2 component

3 mesh model

3a scaled network model

4 mesh element

5 network node

6 deformation hot spot

7 edge

8 corner

9 surface

10 enlarged mesh element

11 Deformation Mesh Model

12 deviation

13 Deviation Amount

14 deviation length

15 deviation angles

16 zero deformation point

17 insurgency level

18 geometry special element

19 geometry special element

20 geometry element

21 geometry element

22 mesh element

S1 process step

S1a method step

S1b process step

S2 method step 53 process step

54 process step

S4a method step

55 process step

S5a method step

56 process step

S10 loop

Claims

Claims Method (A) for providing a digital print model (1, 1a) on the basis of which a component (2) can be produced additively, wherein

51 a mesh model (3) of the component (2) to be manufactured is provided;

52 a heat treatment is simulated using the mesh model (3), whereby a digital deformation mesh model (11) of the component (2) to be manufactured is generated;

53 the network model (3) is compared with the deformation network model (11), whereby local deviation amounts (13) between the network model (3) and the deformation network model (11) are determined;

54 the network model (3) and the deformation network model (11) are aligned with one another in such a way that a total deviation amount is minimal, which is formed from the local deviation amounts (13) between the network model (3) and the deformation network model (11); and

55 the pressure model (1, 1a) of the component (2) to be manufactured is generated by changing the network model (3) based on the local deviation amounts (13), characterized in that before step S2 on the network model (3) a deformation -Hot-Spot (6) is identified and network elements (4) of the network model (3) arranged away from the deformation hot-spot (6) are enlarged. Method (A) according to Claim 1, characterized in that the deformation hot spot (6) is identified on the network model (3) by using a short simulation to simulate a heat treatment using the network model (3). Method (A) according to Claim 1, characterized in that the deformation hot spot (6) is identified on the network model (3) by creating a coarse network model based on the network model (3) and simulating the heat treatment using the coarse network model is generated, creating a deformed coarse mesh model. Method (A) of one of the preceding claims, characterized in that after step S5 the pressure model (1) is refined by reducing the size of the network elements (4, 10) arranged away from the deformation hot spots (6); Method (A) according to Claim 4, characterized in that the network elements (4, 10) arranged away from the deformation hot spot (6) are reduced in size according to a predetermined factor. Method (A) according to one of the preceding claims, characterized in that the alignment of the network model (3) and the deformation network model (11) with one another takes place with a condition according to which a zero deformation point (16) of the network model (3) and a zero deformation point corresponding thereto ( 16) of the deformation network model (11) are arranged next to one another in such a way that the total deviation amount formed from the local deviation amounts (13) is minimal while complying with this condition. Method (A) according to one of the preceding claims, characterized in that a permissible tolerance amount is specified for the respective local deviation amount (13) between the network model (3) and the deformation network model (11) and a loop comprising steps S2, S3 and S4 S10 is carried out iteratively until the respective local deviation amount (13) corresponds to the permissible tolerance amount. Method (A) according to Claim 7, characterized in that a special geometry element (18) of the network model (3) and a special geometry element (19) of the deformation network model (11) are determined, and the loop S10 is carried out iteratively until the respective local deviation amount ( 13) between the special geometry elements (18, 19) corresponds to the allowable tolerance amount, with a larger tolerance amount being allowed away from the special geometry elements (18, 19). Method (A) according to one of the preceding claims, characterized in that before the brief simulation of the heat treatment, the network model (3) is scaled based on a deformation to be expected. Method (B) for the additive manufacturing of a component (2), wherein a material of the component (2) to be produced is arranged according to a digital printing model (1, 1a) of the component (2), and the material is formed into the component (2) by means of a heat treatment 2) is joined, characterized in that the digital print model (1, 1a) is provided by means of a method (A) designed according to one of Claims 1 to 9.