Method of calibrating a powder bed fusion manufacturing process simulation

By calibrating powder bed fusion simulations with in-situ measurement data, the method addresses the inaccuracies in predicting recoater interactions and thermal expansion, enhancing the manufacturing process efficiency and reducing material waste.

GB2629784APending Publication Date: 2024-11-13SIEMENS ENERGY GLOBAL GMBH & CO KG
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
GB2023006815
Authority / Receiving Office
GB · GB
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-09
Publication Date
2024-11-13

AI Technical Summary

Technical Problem

Current powder bed fusion manufacturing simulations, particularly for complex shapes and high-performance materials like nickel-based superalloys, fail to accurately predict and prevent issues such as hotspots, warpage, and recoater collisions due to insufficient modeling of recoater types, directions, and process parameters, leading to material waste and inefficiencies.

Method used

A method for calibrating powder bed fusion simulations by recording in-situ measurement data during the manufacturing process, including recoater interactions, and using this data to derive actual displacements and calibrate simulations, which can assess the risk of recoater interactions and optimize the manufacturing process, incorporating factors like recoater type, orientation, and cooldown times.

Benefits of technology

This approach enables more accurate and versatile simulations, reducing the need for extensive testing and material waste by predicting potential recoater collisions and thermal expansion issues, thereby optimizing the manufacturing process in terms of time and material usage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

A method of calibrating a powder bed fusion manufacturing process simulation comprises i) recording measurement data, particularly data concerning a recoater interaction with a build-up, during the manufacturing process 1, ii) measuring or deriving a value of an actual z displacement of the build-up during the process by way of the measurement data, iii) calibrating a simulation of the manufacturing process based on the measured value of the z displacement, and iv) running a calibrated simulation that outputs an indication (W) comprising a risk and / or a probability assessment of a possible recoater interaction with the build-up (1). The risk may be graded by indicating an estimated risk of recoater contact without any damage, recoater damage and / or risk of a build crash. The calibration step may be based on a calibration build to assess build aspects such as overhang angle, recoater direction, feature sizes, and cooldown time. Moreover, a related computer program product is subject to the present invention.
Need to check novelty before this filing date? Find Prior Art

Description

Method of calibrating a powder bed fusion manufacturing process simulation The present invention relates to a method of calibrating a simulation of a powder bed fusion manufacturing process, a related data processing apparatus and a related computer program product. The powder bed fusion process preferably pertains to a laser powder bed fusion (LPBF). The calibration method may as well related to an optimization of the mentioned process simulation. Additive manufacturing (AM) or 3D-printing techniques comprise e. g. powder-bed-fusion methods, such as selective laser melting (SLM), selective laser sintering (SLS) and electron beam melting (EBM). Additive manufacturing, particularly powder-bed methods have proven to be useful and advantageous in the fabrication of prototypes or complex components, such as components with filigree structure or functionally cooled components. Further, the additive manufacture stands out for its short chain of process steps which in turn enables material economization and a particularly low lead time. By the mentioned manufacturing approaches, articles of components with ever more complex shapes are developed, such as components of a turbo machine, e. g. applied in the flow path hardware of a gas turbine. The component at issue might even be established from high-performance materials like nickel-or cobalt-based superalloys. Also, the related component may be foreseen or applied elsewhere, like as heat exchangers in the energy technology or power conversion or generation, and e. g. in the aviation or the automotive industry. Currently, despite lots of progress in material and / or parameter development in AM, there is the demand to validate a given component design in terns of the required parameters which need to be optimized individually to avoid e. g. hotspots, excessive heat, poor heat dissipation, warpage or structured that delaminate from or build platform, or other adverse effects during the process. For such validations, state-of-the-art (finite element) simulations are employed to partly predict the thermo-mechanical behavior and optimize the manufacture accordingly. Current (simulation) tools particularly enable to simulate thermal deformation or warpage of a component's features along the vertical direction (Z) and, in this way, to assess risks of a possible contact of a recoater which may lead to damage of the hardware or even to abortion of the whole fabrication process. The mentioned recoater is essential to the mentioned processes to distribute any new powdery material layer, once the previous one has been selectively irradiated by an energy beam, such as an electron beam or laser. The conventional prediction of potential process issues and incidents does not provide sufficient capacity to model a realistic situation and / or account for all relevant circumstances. Particularly, state-of-the-art simulations do not account for the recoater type used, such as if it is a blade, a brush or silicon lip recoater, the recoater direction (as compared to edges of the build-up), and further process parameters, like cooldown times or the like. Hence, there is still an extensive need e. g. to perform test buildups for developing a design- and material-dependent set of process parameters . A person skilled in the art is aware of the fact that particularly the additive buildup of complex shapes and high-performance materials, such as nickel base superalloys, are very difficult to establish, due to the frequent occurrence of hotspots and weldability issues. Moreover, there is a strong tendency of poor heat dissipation, particularly when the features at issue are thin, and / or overhanging, and the related heat output is difficult through the underlying (qua-si-insulating) powder bed. It is an object of the present invention to provide means which improve the mentioned parameters base of the simulations and thereby to provide a versatile and accurate simula-5 tion result. Consequently, validation or test processes can largely be dispensed with to the benefit of a simpler and more efficient process. Particularly industrialized additive manufacturing processes can be significantly optimized, and rendered much more effective in terms of time and material 10 used. The mentioned object is achieved by the subject-matters of the independent claims. Advantageous embodiments are subjectmatter of the dependent claims. 15 An aspect of the present invention relates to a method, such as a computer-implemented method, of calibrating or optimizing a (laser) powder bed fusion manufacturing process simulation. The method comprises recording, preferably in-situ, 20 measurement data during the manufacturing process, particularly measurements data concerning a recoater interaction, such a recoater contact, collision and / or damage, with the build-up . 25 Said term "build-up" shall denote any given structure as part of a component which is currently validated or manufactured. The method comprises measuring or deriving a value of an actual displacement of the build-up along a (vertical) build 30 direction during the process by way of the recorded measurements data. The method further comprises calibrating a simulation of the manufacturing process based on the measured or derived value 35 of the displacement. The method further comprises running a calibrated simulation, wherein the simulation program or process outputs an indica tion comprising a risk and / or the probability assessment of a possible recoater interaction with or contacting the buildup. The mentioned indication nay be given via a graphical user interface, e. g. to a user, engineer, or operator of the 3D-printer or process . The presented approach advantageously enables a customized, versatile and accurate simulation of possible recoater interferences. The inventive calibration and / or simulation even enables to validate the AM manufacturing route to a huge number of components or design features and material classes (quasi-independent from material and design), wherein complex testing and calibration builds can almost completely be avoided. The presented approach is advantageously capable of considering and indicating major issues and incidents relating to the thermal expansion of the build-up and related recoater collisions or damages early on. In an embodiment, a risk and / or a probability assessment of the output is, depending on a severity of the interaction, graded by indicating an estimated risk of a contact of the recoater (but without any damages), an estimated risk of the recorded damage, and / or if applicable (in addition) an estimated risk of a build crash. The latter incident, i. e. a true build crash, not indicated to the user early-on, means to abort the whole manufacturing process. Thereby, valuable raw material lots held available are probably likewise wasted and need to be rejected. It is expected that damage of the recoater may as well lead to abortion of the process, and vice versa. It may as well occur, that a damaged recoater blade does not necessarily lead to an aborted process, and vice versa. Rather, it depends on the actual circumstances, parameters and z-displacement modelled for a given build setup and the recoater type used, which of the outlined incidents needs to be indicated . In an embodiment the process simulation of the manufacturing process is a thermos-mechanical simulation. Additionally or alternatively said simulation may be a finite element or qua-si-finite element simulation. Preferably, the simulation uses the so-called inherent strain approach. In this approach, the program employs strains that are inherited from the thermomechanical process and allow for a calculation of the deformed shape of the build-up. As a merit, the inherent strain approach is capable of reducing a complex thermo-mechanical simulation to a sole mechanical one, which renders the simu-lation much faster and more efficient. In an embodiment, the measurement data comprises optical, such as optical tomography, monitoring data and / or data recorded from a force sensor or an acoustic sensor during the process. By these means, a valuable measurement input can be generated that is essential for the simulation, and particularly for the simulation to be calibrated and run anew, to provide for the inventive advantages. In an embodiment, the calibration step is based on a calibration build or build process, the measurement data recorded of which are fed into the simulation as input. In an embodiment the calibration build is set up such that at least one of the following build or measurement aspects are assessed during the recording of the related measurement data in view of the related recoater interaction: overhang angle of the build-up, recoater direction / orientation e. g. relative to build or structure contours, feature sizes (or geometry resolution), and cool-down times. As a special capability of the present invention, it can account for the indicated aspects as main boundary conditions. By contrast, the conventional state-of-the-art simulation approaches are insufficient for a reliable prediction of the described recoater collision or interaction issues. In an embodiment, the calibration step is based on an input of an information on the recoater type used, e. g. a brush recoater, a solid or blade-type type recoater, or a lip, such as silicon lip, recoater. Another aspect of the present invention relates to a data processing apparatus comprising means for carrying out the calibration of the simulation according to the described method. A further aspect of the present invention relates to a computer program product comprising instructions which, when a related program is executed by a computer or data processing device, to carry out the calibration of the simulation according to the method as described herein. A computer program product as referred to herein may relate to a computer program or media constituting or comprising a computer-readable storage medium (be it volatile and nonvolatile) like a memory card, a USB stick, a CD-ROM, a DVD or a file downloaded or downloadable from a server or network. Such product may be provided by a wireless communication network or via transfer of the corresponding information by the given computer program, computer program product or the like. A computer program product may be, include or be included by a (non-transitory) computer-readable storage medium storing applications, programs, program modules, scripts, source code, program code, object code, byte code, compiled code, interpreted code, machine code, executable instructions, and / or the like. Advantages and embodiments relating to the described calibration method are valid or pertain likewise to the computer program product, and vice versa. Further, features and advantageous embodiments become apparent from the following description of the exemplary embodiment in connection with the Figures. Figure 1 shows an example of the state-of-the-art finite element simulation result of a structure to be manufactured additively . Figure 2 indicates method steps of the present invention by way of a schematic flow chart. Figure 3 shows a perspective view of a build platform and a design of a complex designed calibration build-up according to the present invention. Figures 4 to 6 show different examples of calibration samples aa part of the calibration build-up, particularly a trunklimb structure in greater detail, respectively. Figure 7 shows a somatic top view of a build platform showing the various structures forming the additive calibration build. Figure 8 shows a schematic side view of the calibration build-up . Like elements, elements of the same kind and identically acting elements may be provided with the same reference numerals in the figures. The Figures are not necessarily depicted true to scale and may be scaled up or down to allow for a better understanding of the illustrated principles. Rather, the described Figures are to be construed in a broad sense and as a qualitative base which allows a person skilled in the art to apply the presented teaching in a versatile way. The term „and / or" as used herein shall mean that each of the listed elements may be taken alone or in conjunction with two or more of further listed elements. Figure 1 shows an example of the state-of-the-art finite element simulation result of the complex structure to be established by additive manufacturing, particularly laser powder bed fusion. Related machine hardware or setups for such methods usually comprise a manufacturing or build platform on which the component is built layer-by-layer after the feeding (recoating) of a layer of base material which may then be melted, e. g. by an energy beam, such as a laser, and subsequently solidified. The layer thickness is usually determined by a recoater that moves, e.g. automatically, over the powder bed and removes excess material from a manufacturing plane or build space. Typical layer thicknesses amount to between 20 pm and 40 pm, for instance. During the manufacture, said energy beam scans over the surface and melts the powder on selected areas which may be predetermined by a CAD-file according to the geometry of the component to be manufactured. Said scanning or irradiation is preferably carried out in a computer-assisted way, such as by Computer-Aided-Manufacturing (CAM) or related instructions, which may be present in the form of a dataset. Said dataset or CAM-file may be or refer to a computer program or computer program product. Metal-powder-bed-fusion additive manufacturing excels in great design freedom for the related component and structure variety. However, it poses significant challenges in terms of weldability, dimensional accuracy and resistance to cracks and rupture. The mentioned drawbacks stem from this very complex and fragile additive welding (solidification) process, employing very localized heat input and rapid cooling behavior. A reliable prediction of the process in order to gain full reproducibility has, thus, not yet been achieved in the development of this technology. This is in spite of the application of complex thermo-mechanical process simulations in general. Though, in a layer-by-layer fashion a moving heat source could already be modeled to melt a powdery base material, which then immediately solidifies once the heat source moves on. Around the melt pool of the beam spot, material usually expands locally during heating and contracts afterwards in a cool down (solidification) process. On the right in Figure 1, it is shown that the depicted component design shows the greatest distortions in the process, which is e. g. a consequence of the poor heat output due to the overhanging character of that region. In an ideal theoretical situation, the shape changes caused by the expansions are reversed by the contractions. However, in a realistic situation, the expansions in the heat-affected zone around the melt pool tend to push the surrounding material away so that these regions reversibly remain in a compressed state even though parts of the material shrink as well as the heat source moves on. As mentioned, there have been attempts to simulate this complex thermos-mechanical behavior with finite element type simulations making use of the inherent strain approach. Due to the reduction of the complex thermo-mechanical problem into a mere mechanical one, this method can partly produce realistic predictions even for this rapid solidification process . The inherent strain simulation is, however, insufficient to predict the thermos-mechanical behavior of the whole additive manufacturing process in its entirety. This is because, the related modeling approaches suffer from taking into account the topology and properties of the recoater, e. g. its position as compared to an edge of the build-up. Further, heat input of the energy beam can only be predicted insufficiently and without regard of the specific irradiation (heat input) and hatching strategy. Still further, the modelling cannot at all account for the rapid solidification behavior of hard-to-weld or high-performance materials. Figure 2 shows a schematic flow chart indicating method steps according to the present invention. The method refers to a calibration or improvement of a powder bed fusion manufacturing process simulation. In step i) the method comprises recording measurement data during the manufacturing process, particularly data concerning a recoater interaction with a build-up 1. In step ii) the method comprises measuring or deriving of a value of an actual displacement z of the build-up 1 along a build direction Z during the process by way of the recorded measurements data. In step iii) the method comprises calibrating or adjusting a simulation of the manufacturing process based on the measured value of the displacement z. In step iv) the method finally comprises running a calibrated simulation, wherein the simulation outputs an indication like a warning W or any other signal to an operator of the system comprising a risk and / or a probability assessment of a possible recoater interaction with the build-up 1. Such indication or output of the simulation is a direct result of the inventive method and may constitute a computer program product C, as subject to the present invention. The risk and / or a probability assessment of the output is, depending on a severity of the interaction, preferably graded by indicating an estimated risk of a contact of the recoater 2 (without any damage), an estimated risk of a recoater damage, and / or an estimated risk of a build crash. The calibrating step is based on a calibration build 10 process, the measurement data of which are fed into the simulation as input (cf. Figure 3 below). Upfront, it shown on the left in Figure 2, that said calibration build-up is set up such that at least one of the following build aspects are assessed during the recording of related measurement data in view of the related recoater interaction: overhang angle of the build-up 1, recoater direction, feature sizes, and cooldown time of the build-up 1. This particularly allows for generating valuable input for the calibration of the software and the subsequent performance of the simulation in an improved or calibrated state . Figure 3 indicates a design of a test build (calibration build) setup, indicating by way of an example only, several different tower-like or trunk structures which inter alia differ in the final Z height. These structures are all arranged and supposed to be established on a common build platform 6 in a regular pattern. A long a lateral X direction, the Z-height of the structure increases from one "tree" to the next. Along the other lateral direction (Y direction), at least one other parameter is varied, such as shown in one of the following figures in greater detail. In the depicted example, three rows of samples are shown along the Y direction. Without loss of generality, or any other number of rows, structures of features could be used within the context of the invention . In the context of the invention, also further (process) parameters, like a layer thickness, a melt pool geometry and dimension, heat impact per volume or area, laser wavelength, hatching distance, i. e. distance of adjacent scanning lines, a beam offset, a beam speed, a geometry of beam spot, a beam angle, may be investigated by a dedicated or pre-defined calibration build setup. The calibration build-up advantageously allows to account for a plurality of potential recoater issues, preferably for all practically occurring incidents, during the process as mentioned above. Figures 4 to 6 indicate by way of examples only, possible sample structures for the calcbration in order to improve the simulation according to the present invention. Figure 4 particularly shows in sample SI an elongate vertical trunk 11 or tree with limbs 12 arranged pointing to four lateral directions around the trunk 11, wherein each limb 12 has the feature size or width a. The limbs 12 hence face four different directions representing different orientations to the recoater instrument during the process. It is further shown that the limbs are formed with or constituting an overhanging portion which is due to the fact that a fully horizontal features can hardly be implemented via additive powder bed methods due to issues with mechanical support and proper heat dissipation. With increasing Z-height and progress of the build-up it is shown that each set of limbs is overlapping stronger and further as compared to the below indicated set of limbs 12. For instance, the bottom limb set protrudes according to an angle ol, being slightly greater than the angle «2, of the overlying limb 12. It is further expected that the uppermost limb protruding at the angle «3 is most critical for the build-up due to its protruding distance and strongly overhanging extension . A similar situation applies for sample S2 shown in Figure 5 which may be arranged in another row or line of the overall calibration build 10. As compared to sample SI, sample S2 comprises a smaller feature size (though indicated with the same reference numerals a). The same applies mutatis mutandis for the situation shown in Figure 6 and sample S3, providing a still decreased feature size a. As shown by the dashed lines in the upper right part of the figure, it shall be indicated that particularly small or thin limbs or other filigree structures show a certain tendency to be deformed by thermo-mechanical stresses in the process and particularly deform upwards along the vertical direction as indicated by the displacement z. This in turn leads to the outlined recoater interactions, e. g. a colli sion of friction of the recoater hardware over warped or thermally deformed parts of the builds. In other words, Figures 4 to 6 show different calibration sample designs. The proposed structures differ in dimension a, i. e. the thickness of the single overhangs. A range between 5 mm to 1 mm is a reasonable value for the outlined samples. With a decrease of a, overheating is provoked and by that an effect on the z-displacement z. Each of the tree samples feature three different overhang angles. The overhang decreases example given from 45° to 35° in the figures, respectively. The risk of a recoater collision, damage or crash hence increases in z-height. This is also due to the fact that the likelihood of hotspots increases with increased build height due to the relatively poor heat dissipation. Figure 7 indicates a top view sketch of the whole calibration build setup in the regular pattern distributed over the built platform 6. Different rows of trunk-limb samples or structures are indicated (from bottom to the top) by numerals RI, R2 etc. to R6. The rows hence develop increasingly along a lateral X direction. It is particularly shown that a further parameter is accounted for in the calibration, that is e. g. that the sample structures are rotated in the second row R2 (as compared to the first RI) in order to have different potential recoater impacts on the structures. The same strategy applies for rows R3 and R4 onwards. The rotation parameter assessment goes with equal samples in the z-height (cf. also the previous Figures and Figure 8 below). This particularly enables to assess the representation of different cool down times per layer during the additive process, which may also affect overheating (hotspots) on the part. The inventive method comprises recording measurement data concerning the recoater interaction with the build-up 1 during the process and measuring a value of the said displace ment z of the build-up 1 along a build direction Z by way of the recorded measurements data, as described above. To this effect, the measurement data nay comprise optical monitoring data and / or data recorded from a force sensor 4 or an acoustic sensor 5 as also schematically depicted in Figure 7. Moreover, the inventive calibrating step is preferably based on an input of information on the recoater type used, e. g. a brush recoater, a solid blade-type recoater, a lip recoater, or any other recoating instrument applied. As indicated by the arrows in increasing X and Y direc-tion(s), there is generally an increased risk of a recoater interactions along the X or Y direction (s), when one or more parameters are varied according to this example. Concluding in other words, the presented calibration build represents the situation, wherein the risk of recoater interaction increases from left to right, from front to back and from bottom to top. In total at least four different main effects are hence considered by the inventive calibration build, i.e.: overhang geometry, recoater (movement) direction, feature resolution or geometry, and cool down time or overheating issue. For the sake of a better overview, side view image of the calibration build design (perspective corresponds to view along the Y direction in Figure 7) is further shown in Figure 8 indicating the Z-extensions, rotations and as well some details on the overhanging limbs in greater detail. The component as referred to herein may particularly relate to a part or an article of complex shape, such as with filigree portions of structures. Preferably, said component is made of a high-performance material, such as a material of great strength and / or thermal resistivity. Particularly, said part may constitute a part of a steam or gas turbine component, such as a blade, vane, shroud, shield, such as heat shield, tip, segment, insert, injector, seal, transition, burner, nozzle, strainer, orifice, liner, distributor, dome, boost, cone, lance, plate, resonator, piston or any corresponding retrofit kit. Alternatively, said component may re- 5 late to a heat exchanger component or another or similar component .

Claims

Patent claims1. A Method of calibrating a powder bed fusion manufacturing process simulation comprising the following steps:- i) recording measurement data during the manufacturing process, particularly data concerning a recoater interaction with a build-up (1),- ii) measuring or deriving of a value of an actual displacement (z) of the build-up (1) along a build direction (Z) during the process by way of the recorded measurements data,- iii) calibrating a simulation of the manufacturing process based on the measured value of the displacement (z), and- iv) running a calibrated simulation, wherein the simulation outputs an indication (W) comprising a risk and / or a probability assessment of a possible recoater interaction with the build-up (1).

2. The method according to claim 1, wherein the risk and / or a probability assessment of the output is, depending on a severity of the interaction, graded by indicating an estimated risk of a contact of the recoater (2) without any damage, an estimated risk of a recoater damage, and / or an estimated risk of a build crash.

3. The method according to claim 1 or 2, wherein the process simulation of the manufacturing process is a thermomechanical finite-element-simulation using the Inherent-Strain approach.

4. The method according to one of in the measurement data comprises and / or data recorded from a force sensor (5) during the process.the previous claims, where-optical monitoring data (3) sensor (4) or an acoustic5. The method according to one of the previous claims, wherein the calibrating step is based on a calibration build (10)process, the measurement data of which are fed into the simulation as input.

6. The method according to one of the previous claims, wherein the calibration build is set up such that at least one of the following build aspects are assessed during the recording of related measurement data in view of the related recoater interaction: overhang angle of the build-up (1), recoater direction, feature sizes, and cooldown time of the build-up (1) •7. The method according to one of the previous claims, wherein the calibrating step is based on an input of information on the recoater type used, e. g. a brush recoater, a solid recoater or a lip recoater.

8. A data processing apparatus (20) comprising means for carrying out the calibration of the simulation according to the method of claim 1.

9. A computer program product (C) comprising instructions which, when a related program is executed by a computer, cause the computer (20) to carry out the calibration of the simulation according to the method of claim 1.

Citation Information

Patent Citations

  • Recoater collision prediction and correction method for additive manufacturing and system thereof

    US20220163943A1

  • Apparatuses, systems, and methods for providing enhanced recoater event prediction for DMLM additive manufacturing

    US20230121090A1

  • System and method for build error detection in an additive manufacturing environment

    WO2019094472A1