Computer-implemented method for generating build job data for manufacturing components using a powder bed-based melting process, as well as a method for manufacturing components using a powder bed-based melting process, system and computer program.

Abstract

Computer-implemented method for generating build job data for manufacturing components (13a, 13b, 13c) using a powder bed-based melting process, comprising the following steps: a) Providing model data of components to be produced (1a, 1b, 1c); b) Providing a virtual build space (2) with boundary dimensions (3a, 3b, 3c) that define an outer edge of the virtual build space (2), wherein the virtual build space (2) is virtually filled with powder (5), wherein the virtual build space (2) represents a real volume space (4) in which the components (1a, 1b, 1c) to be produced are to be produced; c) Specifying a target value for a temperature for a virtual space (6) that arises between the components to be produced in the virtual build space and between a component to be produced (1a, 1b, 1c) and the outer edge of the virtual build space (2); d) Creating at least one arrangement pattern (7) by placing and orienting the components to be produced (1a, 1b, 1c) based on the model data in the virtual design space (2) depending on the target value; e) Determining a temperature distribution in the virtual build space (2) for the arrangement pattern (7) that characterizes an expected temperature distribution of the powder (5) in the real volume space (4); f) Evaluating the arrangement pattern (7) with regard to achieving the target value; g) Depending on the evaluation in step f), if necessary repeat steps d) to f) for further arrangement patterns until, for an arrangement pattern (7), the temperature distribution has reached the target value at which the quality of the remaining powder is maximized and the reject of powder which, in a manufacturing process for components, can no longer be used for further processes due to undesirable thermal influences, is reduced; h) Providing the temperature distribution for the virtual space (6) in the virtual installation space (2) for the arrangement pattern (7) in which the temperature distribution has reached the target value; i) Extracting the build job data from the layout pattern (7) in which the temperature distribution has reached the target value; and j) Providing the extracted construction job data.

Description

[0001] Aspects of the invention relate to a computer-implemented method for generating build job data for manufacturing components using a powder bed fusion process, and to a method for manufacturing components using a powder bed fusion process. Further aspects of the invention relate to a system and a computer program.

[0002] From WO 2021 / 086347 A1 and US 2022 / 0288862 A1, a method is known in which a training dataset is determined for an inference model to evaluate a spatial arrangement of objects to be produced in additive manufacturing. In this process, a temperature distribution is obtained for each of a plurality of spatial arrangements of objects to be produced in additive manufacturing using at least one processor. Based on the obtained temperature distribution, an indication of thermal interaction is obtained for each spatial arrangement using at least one processor.

[0003] WO 2022 / 005464A1 concerns a spatial arrangement for additive manufacturing.

[0004] In additive manufacturing of plastic components, for example in powder-based processes such as selective laser sintering (SLS) and multijet fusion (MJF), the surrounding, unsolidified powder is reused. However, this powder is subjected to thermal stress, as a result of which its properties no longer correspond to those of the original powder. This thermal stress depends on the packing density and component arrangement within the build chamber.

[0005] The invention is based on the objective of providing methods to improve the reusability of the powder. Furthermore, it aims to create a system and a computer program.

[0006] The problem is solved by a method, a system, and a computer program according to the independent claims. Advantageous embodiments of the invention are described by the dependent claims, the following description, and the figures.

[0007] A first aspect of the invention relates to a computer-implemented method for generating build job data for manufacturing components using a powder-based melting process, preferably comprising the following steps: a) In particular, providing model data of components to be produced; b) In particular, providing a virtual build space with boundary dimensions that measure an outer edge of the virtual build space, wherein the virtual build space is virtually filled with powder, and wherein the virtual build space represents a real volume space in which the components to be produced are to be produced; c) In particular, specifying a target value for a temperature for a virtual space that arises between the components to be produced in the virtual build space and between a component to be produced and the outer edge of the virtual build space; d) In particular, creating at least one arrangement pattern by placing and orienting the components to be produced based on the model data in the virtual design space depending on the target value; e) In particular, determining a temperature distribution in the virtual build space for the arrangement pattern that characterizes an expected temperature distribution of the powder in the real volume space; f) In particular, evaluating the arrangement pattern with regard to achieving the target value; g) In particular, depending on the evaluation in step f), if necessary, repeat steps d) to f) for further arrangement patterns until the temperature distribution of an arrangement pattern has reached the target value at which the quality of the remaining powder is maximized and the reject of powder which is no longer usable for further processes in a manufacturing process for components due to undesirable thermal influences is reduced; h) In particular, providing the temperature distribution for the virtual space in the virtual installation space for the arrangement pattern in which the temperature distribution has reached the target value; i) In particular, extracting the build job data from the layout pattern in which the temperature distribution has reached the target value; and j) In particular, providing the extracted construction job data.

[0008] This method now makes it possible to improve reusability, particularly through simulation. This also means that the amount of powder waste—which, due to undesirable thermal influences, is no longer usable for subsequent processes during component manufacturing—can be significantly reduced. In particular, the quality of the powder used in further processes is also improved. The proposed method allows for the generation of highly accurate and powder-efficient build job data. This, in turn, leads to improved analysis for the subsequent actual component manufacturing process.

[0009] The steps a) - j) do not necessarily have to be performed in the order given. At least some of the steps can be rearranged, omitted, and / or performed simultaneously, at least temporarily.

[0010] Powder bed fusion is a category of additive manufacturing processes. In additive manufacturing, material is built up, typically layer by layer, to create workpieces from 3D model data. Processes in this category include those in which thermal energy selectively bonds or fuses regions of a powder bed. The build job data specifies, in particular, where the powder is fused. The model data includes, for example, computer-aided design (CAD) data. The boundary dimensions of the virtual build space define the dimensions of the real volume. The component is produced within this real volume.

[0011] In particular, the target temperature of the powder in a real space within the real volume should be achieved. This temperature is specified, for example, by an operator. It is also possible for a machine learning algorithm to specify the temperature. Parameters such as powder material and the properties of a printing device can be taken into account. The target value could be, for example, a maximum temperature or a specific distribution, such as a homogeneous temperature distribution. If, for instance, a homogeneous temperature distribution is specified as the target value, then the temperature of the powder should be the same throughout the virtual build chamber and thus also throughout the real volume.

[0012] Creating at least one arrangement pattern can be referred to as nesting. Evaluating the arrangement pattern with regard to achieving the target value can be done using a computer-implemented method, such as software. However, it is also possible for the operator to perform the evaluation. This can be facilitated, for example, by displaying the arrangement pattern to the operator in a suitable manner. The expected temperature distribution can be represented using a color scheme.

[0013] If the target value is not reached in the arrangement pattern, another arrangement pattern can be created. Specifically, further arrangement patterns are created until the temperature distribution reaches the target value. It is possible to define a termination criterion. This termination criterion could, for example, be a limit on the number of further arrangement patterns. Once this limit is reached, the arrangement pattern that best achieves the target value can be selected. The temperature distribution for the virtual space within the virtual build volume is then provided for the arrangement pattern in which the temperature distribution reached the target value. This temperature distribution can be displayed as output on a screen. It is also possible to provide the temperature distribution for a subsequent algorithm in a later optimization step.

[0014] The build job data is extracted from the selected layout pattern and made available. Specifically, the extracted build job data is provided to a printing device, particularly a 3D printer.

[0015] A target value can also be set to ensure that the space between the powders experiences a homogeneous temperature load. In other words, it is possible to set not only a quantitative but also a qualitative target value. Specifically, the target value is set to maximize the quality of the remaining powder. This minimizes powder waste, resulting in a longer service life for the spent powder and controlled quality.

[0016] In one embodiment, the provided temperature distribution in the virtual installation space is calculated using a finite element method (FEM) algorithm.

[0017] This allows for improved simulation of the temperature distribution. The deviation between actual temperature values ​​and the simulated temperature values ​​is very small. For example, the deviation is less than 10 percent, in particular less than 5 percent, and in particular less than 1 percent.

[0018] In one embodiment, the temperature distribution is determined as a function of time for a simulated pressure sequence, and in particular, output.

[0019] In other words, the temperature distribution is determined not only as a function of location but also as a function of time for the simulated printing process. During a printing process, the temperature can change over time. For example, lower layers that have already been printed cool down over time. By taking the time dependence of the printing process into account, a realistic simulation is possible.

[0020] In one embodiment, a specific temperature value for the gap is specified as the target value.

[0021] It is possible to divide the virtual design space into voxels and / or voxel groups. In other words, the virtual design space is divided into grid points using a three-dimensional grid. A grid point can be referred to as a voxel. Several voxels can be grouped together to form a voxel group. A specific temperature value can be defined as the target value for a single voxel and / or a single voxel group.

[0022] This offers the advantage that individual voxels and / or voxel groups can be subjected to higher thermal stress to protect other voxels and / or voxel groups. This prevents, for example, a large quantity of powder from becoming unusable due to excessive thermal stress. This embodiment allows only a portion of the powder—namely, the voxels and / or voxel groups assigned a very high target value—to be subjected to excessive thermal stress.

[0023] In one embodiment, after the temperature distribution has been provided, the virtual space is divided into areas. Individual target values ​​for a temperature distribution are specified for each area. Steps d) to f) are repeated until all individual target values ​​are reached.

[0024] This is a further optimization step, performed primarily after verifying the provided temperature distribution. The individual target values ​​can be specified, for example, by the operator or by a machine learning algorithm. It is also possible to define a termination criterion, such as a maximum number of repetitions. Specifically, the arrangement pattern that best achieves the individual target values ​​is selected. The individual target values ​​are best achieved when the deviation between the calculated temperature distribution and the actual temperature distribution is minimal.

[0025] In one embodiment, depending on the target value, a virtual structure is placed and oriented in the space between the components after they have been positioned and oriented.

[0026] The virtual structure is, in particular, a bionic structure. The virtual structure corresponds to a real structure. Specifically, the real structure is printed with or from a different material than the components. For example, the material of the real structure has a different thermal conductivity coefficient than the material of the components.

[0027] This offers the advantage that heat can be dissipated and / or distributed homogeneously. Examples of bionic structures include honeycomb and tubular structures. Bionic structures have the advantage of requiring little material while maintaining high stability.

[0028] It is possible to monitor the actual temperature during the printing process. For example, the temperature can be measured using optical temperature sensors. The measured temperature values ​​can then be provided to the machine learning algorithm as input parameters. This allows for verification of whether the calculated temperature distribution corresponds to the measured temperature values. Furthermore, the definition of individual temperature targets can be improved. Additionally, a real-world structure can be introduced as a response to, for example, excessively high actual temperature values.

[0029] In one embodiment, design data, particularly CAD data, is provided as the build job data. This design data specifies at which points in the powder, for each applied powder layer, thermal energy is introduced by a directed energy source to selectively bond or fuse the powder at these points. The energy source is, for example, a laser.

[0030] Another aspect of the invention relates to a method for manufacturing components using a powder bed fusion process. The method preferably comprises the following steps: - In particular, applying a powder as a powder bed onto a support; - In particular, selective thermal action on the powder by means of a directed energy source depending on the build job data generated according to a method of the preceding claims; - In particular, removing the powder; - In particular, feeding reusable powder that is part of the removed powder into a next manufacturing cycle.

[0031] This enables a powder-saving and efficient manufacturing process. Powder waste can be significantly reduced. Components can also be manufactured with greater dimensional accuracy. In particular, this also increases the density of components produced in a single process.

[0032] The application of the powder as a powder bed onto a support is dependent on the build job data. The build job data is determined and provided, in particular, according to the procedure described above. The removal of the powder can be referred to as "depowdering." The removed powder can be reused. The quality and quantity of this reusable powder are to be improved by the process used to generate the build job data.

[0033] Another aspect of the invention relates to a system for carrying out a method comprising a storage unit, an evaluation unit and an output unit, wherein the system is configured to perform the steps of the method according to the invention.

[0034] In particular, the procedure is carried out using the system.

[0035] Another aspect of the invention relates to a computer program comprising instructions which, when the program is executed by a computer, cause it to execute the method according to the invention and which can be loaded into a storage unit and executed by a processor.

[0036] Another aspect of the invention relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause it to perform the steps of the method according to the invention.

[0037] For use cases or application situations that may arise during the procedure and are not explicitly described here, it may be provided that, according to the procedure, an error message and / or a request for user feedback is issued and / or a default setting and / or a predetermined initial state is set.

[0038] The invention also includes further developments of the system and computer program according to the invention, which have features already described in connection with the further developments of the method according to the invention. For this reason, the corresponding further developments of the system and computer program according to the invention are not described again here.

[0039] The invention also includes combinations of the features of the described embodiments.

[0040] The following describes exemplary embodiments of the invention. This is illustrated by: Fig. 1 a schematic flowchart of an embodiment of a method according to the invention; Fig. 2 a schematic representation of an exemplary embodiment of a virtual construction space; and Fig. 3 A schematic representation of a printing device for the production of components.

[0041] The embodiments described below are preferred embodiments of the invention. In these embodiments, the described components each represent individual features of the invention that can be considered independently of one another. Each of these features further develops the invention independently and can therefore be considered part of the invention individually or in a combination other than that shown. Furthermore, the described embodiments can also be supplemented by other features of the invention already described.

[0042] In the figures, functionally identical elements are each provided with the same reference symbols.

[0043] In Fig. Figure 1 shows a schematic flowchart of an embodiment of a simulation method according to the invention for generating build job data for manufacturing components using a powder bed-based melting process.

[0044] In step a) model data of components to be produced 1a, 1b, 1c ( Fig. 2) provided. For example, CAD data is provided.

[0045] In step b) a virtual construction space 2 ( Fig. 2) with limiting dimensions 3a, 3b, 3c ( Fig. 2) be provided. The boundary dimensions 3a, 3b, 3c can define an outer edge of the virtual construction space 2. The virtual construction space 2 can be a real volume space 4 ( Fig. 3) represent. The real volume space 4 can be virtually represented by powder 5 ( Fig. 3) be filled. The components to be produced can be created in and with the powder 5.

[0046] In step c), a target value for a temperature for a virtual space 6 ( Fig. 2) can be specified. The virtual space 6 can be created between the components 1a, 1b, 1c to be produced in the virtual build space 2 and between a component 1a, 1b, 1c to be produced and the outer edge of the virtual build space 2.

[0047] In step d) at least one arrangement pattern 7 can be created by placing and orienting the components 1a, 1b, 1c to be produced based on the model data in the virtual construction space 2 depending on the target value.

[0048] In step e), a temperature distribution in the virtual build space 2 for the arrangement pattern 7 can be determined. The temperature distribution can characterize an expected temperature distribution of the powder 5 in the real volume space 4.

[0049] In step f), the arrangement pattern 7 can be evaluated with regard to achieving the target value.

[0050] Depending on the evaluation in step f), steps d) to f) may be repeated for further arrangement patterns, in particular until the temperature distribution of an arrangement pattern has reached the target value.

[0051] In step h), the temperature distribution for the virtual space 6 in the virtual installation space 2 can be provided for the arrangement pattern 7, in which the temperature distribution has reached the target value. In one embodiment, the arrangement pattern 7 can be provided for a further adaptation or improvement step. For example, the virtual space 2 can be subdivided into areas, in particular voxel groups 10a, 10b. Individual target values ​​for a temperature distribution can be specified for the areas, and steps d) to f) can be repeated until all individual target values ​​are reached.

[0052] In step i), the build job data can be extracted from the arrangement pattern 7, in which the temperature distribution has reached the target value.

[0053] In step j), the extracted construction job data is provided.

[0054] Fig. Figure 2 shows an example of the virtual construction space 2. The virtual construction space 2 can contain the components 1a, 1b, 1c to be produced. The virtual construction space 2 can be limited by its external dimensions 3a, 3b, 3c. The external dimensions 3a, 3b, 3c correspond, in particular when scaled and especially at a reduction scale, to the external dimensions 8a, 8b, 8c ( Fig. 3) of the real volume space 4 ( Fig. 3).

[0055] It is possible that the virtual design space 2 is subdivided into voxels 9 and / or voxel groups 10a, 10b. For example, voxels 9 are cuboid, or more specifically, cube-shaped, volume units within the virtual design space 2. In particular, voxels 9 are of equal size. Several voxels 9 can, for example, form a voxel group 10a, 10b.

[0056] Arrangement pattern 7 is one way in which the components 1a, 1b, 1c to be generated can be placed and oriented within the virtual build space 2. In particular, it may be necessary to meet certain boundary conditions. Possible boundary conditions include, for example, that the components 1a, 1b, 1c to be generated must not overlap, and / or that a minimum distance must be maintained between the components 1a, 1b, 1c and / or between them and the outer edge of the virtual build space 2, and / or that a minimum number of components 1a, 1b, 1c to be generated must be placed within the virtual build space 2.

[0057] After an initial arrangement pattern 7 has been generated, an initial temperature distribution for this arrangement pattern 7 can be determined, for example, using an FEM simulation. In determining the temperature distribution, pressure parameters can be taken into account in addition to the arrangement pattern 7, for example, properties of a real thermal energy source 11 ( Fig. 3), in particular a laser, properties and material of the powder 5 and a printing speed.

[0058] The determined temperature distribution can specify an expected temperature value for each voxel 9 and / or voxel group 10a, 10b. After, for example, an initial temperature distribution has been determined, it can be checked, particularly automatically, whether and / or for which voxels 9 the target value has not been met. If voxels 9 and / or voxel groups 10a, 10b are identified that exceed the target value, a further arrangement pattern 7 is generated.

[0059] The target value can be, for example, a specific temperature value and / or a desired temperature distribution, such as a homogeneous distribution. If, for instance, a homogeneous distribution is desired, an absolute or percentage mean deviation value can be specified as the target value to be adhered to. The mean deviation value, for example, denotes a maximum permissible temperature deviation for individual voxels 9 and / or voxel groups 10a, 10b from the mean of the expected temperature values ​​of all voxels 9.

[0060] It is possible that, for example, in a further adjustment step, particularly automatically, for instance by means of an algorithm, it is recognized that with the provided arrangement pattern 7, it may be advantageous to subject a certain first voxel group 10a to a higher thermal load than specified by the target value, in order to allow a lower thermal load for other voxel groups 10b. In other words, the first voxel group 10a is sacrificed. A higher thermal load occurs particularly when the components 1a, b, 1c to be produced are arranged closer together. The voxels 9 and / or voxel groups 10a that are sacrificed can be marked during a printing process, particularly with color. This can enable sorting, especially automated sorting.In this way, the thermally highly stressed powder 5 can be separated from the less stressed powder 5, so that, for example, only the less stressed powder 5 is reused for a further printing process.

[0061] Once an arrangement pattern 7 is determined whose associated temperature distribution, in particular all, meets or meets with minimal deviations, build job data of this arrangement pattern 7 can be extracted and in particular sent to a printing device 12 ( Fig. 3) be provided.

[0062] In Fig. Figure 3 shows a schematic representation of the printing device 12 for the actual production of components 13a, 13b, 13c. This also represents a system. The components 13a, 13b, 13c can be produced by powder bed fusion. In particular, at least one powder layer 14 can be applied. This layer can then be heated locally by means of the thermal energy source 11, for example, a laser, of the printing device 12. The heated powder 15 melts. Subsequently, the next powder layer can be applied. For example, a support 16 of the printing device 12 is moved downwards, i.e., in the direction of arrow P1. A powder unit, for example, a doctor blade 17, of the printing device 12 applies the powder 5 for the next powder layer onto the previous powder layer 14.

[0063] In Fig. Figure 3 shows an example of a point in time during the printing process where components 12a and 12b have already been printed based on the extracted build job data, and component 12c has not yet been completely printed. The area above layer 16 is already filled with powder 5.

[0064] The processes generally make it possible to reuse more of the powder 5 that was not affected by the energy source 11 for the production of components 13a, 13b, and 13c in subsequent processes. This results in less thermally degraded and unusable powder 5, particularly at the transition zones between a thermally treated powder 5 used for a component 13a, 13b, or 13c and an untreated powder 5.

[0065] In one embodiment, a current real temperature distribution in the real volume space 4 can be measured by means of a temperature measuring unit 19 of the pressure device 12. The measured real temperature distributions can be stored and, in particular, made available to an algorithm. The algorithm can determine the temperature distributions in step e) ( Fig. 1) to improve the expected temperature distribution. In particular, the algorithm is a machine learning algorithm.

[0066] In one embodiment, structures can be printed, particularly during the printing process, especially if excessively high temperature values ​​are detected by the temperature measuring unit 19. For example, bionic structures are printed from a material that has a different thermal conductivity than the powder 5. This allows heat to be dissipated and / or distributed.

[0067] The printing process is completed, in particular, when the print bed 16 reaches a lower stop 18 of the printing device 12 and / or all components 1a, 1b, 1c to be produced have been printed as components 13a, 13b, 13c. The components 13a, 13b, 13c can be removed from the actual volume space 4 and depowdered. The remaining powder 5 can then be sorted according to quality, i.e., according to thermal stress, particularly automatically.

Claims

[1] Computer-implemented method for generating build job data for manufacturing components (13a, 13b, 13c) using a powder bed-based melting process, comprising the following steps: a) Providing model data of components to be produced (1a, 1b, 1c); b) Providing a virtual build space (2) with boundary dimensions (3a, 3b, 3c) that define an outer edge of the virtual build space (2), wherein the virtual build space (2) is virtually filled with powder (5), wherein the virtual build space (2) represents a real volume space (4) in which the components (1a, 1b, 1c) to be produced are to be produced; c) Specifying a target value for a temperature for a virtual space (6) that arises between the components to be produced in the virtual build space and between a component to be produced (1a, 1b, 1c) and the outer edge of the virtual build space (2); d) Creating at least one arrangement pattern (7) by placing and orienting the components to be produced (1a, 1b, 1c) based on the model data in the virtual design space (2) depending on the target value; e) Determining a temperature distribution in the virtual build space (2) for the arrangement pattern (7) that characterizes an expected temperature distribution of the powder (5) in the real volume space (4); f) Evaluating the arrangement pattern (7) with regard to achieving the target value; g) Depending on the evaluation in step f), if necessary repeat steps d) to f) for further arrangement patterns until, for an arrangement pattern (7), the temperature distribution has reached the target value at which the quality of the remaining powder is maximized and the reject of powder which, in a manufacturing process for components, can no longer be used for further processes due to undesirable thermal influences, is reduced; h) Providing the temperature distribution for the virtual space (6) in the virtual installation space (2) for the arrangement pattern (7) in which the temperature distribution has reached the target value; i) Extracting the build job data from the layout pattern (7) in which the temperature distribution has reached the target value; and j) Providing the extracted construction job data. [2] Method according to claim 1, wherein the provided temperature distribution in the virtual installation space (2) is calculated using a finite element method (FEM) algorithm. [3] Method according to one of the preceding claims, wherein the temperature distribution is determined and, in particular, output as a function of time for a simulated pressure sequence. [4] Method according to one of the preceding claims, wherein a specific temperature value for the space (6) is specified as the target value. [5] Method according to one of the preceding claims, wherein after providing the temperature distribution the virtual space (6) is divided into areas (9, 10a, 10b) for which individual target values ​​of a temperature distribution are specified and steps d) to f) are repeated until all individual target values ​​are achieved. [6] Method according to one of the preceding claims, wherein, depending on the target value, a virtual structure is placed and oriented in the space (6) after the components have been placed and oriented. [7] Method according to one of the preceding claims, wherein construction data, in particular CAD data, are provided as build job data, which indicate at which points in the powder (5) thermal energy is introduced by a directed energy source (11) for each applied powder layer in order to selectively bond or fuse the powder (5) at these points. [8] Method for manufacturing components using a powder bed fusion process, comprising the following steps: - Applying a powder (5) as a powder bed onto a support (16); - Selective thermal action on the powder by means of a directed energy source (11) depending on the build job data generated according to a method of the preceding claims; - Removal of the powder (5); and - Feeding reusable powder (5) that is part of the removed powder (5) for a next manufacturing cycle; [9] System for manufacturing components using a powder bed-based melting process, comprising a storage unit, an evaluation unit and an output unit, wherein the system is configured to perform the steps of the method according to claim 8. [10] Computer program comprising instructions which, when the program is executed by a computer, cause it to execute the method according to any one of claims 1 to 7, and which is loadable into a storage unit and executable by a processor.

Images