Method for predicting the microstructure of an additively manufactured component

Abstract

Method for predicting the microstructure of an additively manufactured component, comprising the following steps: a) Determining a thermal history for at least one section (4, 5, 6) of the component from a thermal simulation of an additive manufacturing process of the component, b) Determine at least one characteristic value from the thermal history, the magnitude of which depends on the geometry of the section, c) Comparing at least one characteristic value with a predetermined threshold value, and d) Predictions of the microstructure of the section based on the comparison result from step c).

Description

[0001] The present invention relates to a method for predicting the microstructure of a component to be additively manufactured and a method for additively manufacturing a component taking the predicted microstructure into account. Furthermore, a data processing system, a computer program, and a computer-readable storage medium are provided. The invention can be used, in particular, for the additive manufacturing of components with geometrically complex, weight-saving, yet highly resilient structures.

[0002] Additive manufacturing processes, also known as 3D printing, such as powder bed fusion of metal using a laser beam (PBF-LB / M), are characterized by a high degree of geometric freedom and are advantageous for the production of geometrically complex structures that cannot be realized technically and / or economically with conventional manufacturing processes such as milling or casting.

[0003] However, additively manufactured components often exhibit an inhomogeneous microstructure and, consequently, inhomogeneous mechanical properties. These inhomogeneous mechanical properties can lead to problems in the use of additively manufactured components. For example, problems, particularly deterioration, can arise in failure behavior, structural integrity, fatigue resistance, and / or corrosion susceptibility.

[0004] The object of the invention is to improve the handling of microstructures in additively manufactured components.

[0005] The problem is solved by the features of the independent claims. Advantageous embodiments are specified in the dependent claims. It should be noted that the features specified in the claims can be combined individually with each other and / or with the subject matter of the general description, thereby revealing further embodiments of the invention.

[0006] The inventive method for predicting the microstructure of an additively manufactured component comprises the following steps: a) Determining a thermal history for at least one section, preferably several, particularly preferably all, sections of the component from a thermal simulation of an additive manufacturing process of the component, b) Determine at least one characteristic value from the thermal history, the magnitude of which depends in particular on the geometry of the section, c) Comparing at least one characteristic value with a predetermined threshold value, and d) Predictions of the microstructure of the section based on the comparison result from step c).

[0007] An additive manufacturing process involves the layer-by-layer application of material, for example, by repeatedly melting and solidifying the material, thereby building up a component layer by layer. Preferably, the additive manufacturing process is 3D printing, particularly metal 3D printing. It is preferred that the additive manufacturing process is powder bed fusion, for example, using laser beam melting (LPBF) and / or electron beam melting (EBM).

[0008] One particularly favored additive manufacturing process is, for example, powder bed fusion (PBF-LB / M), in which a metallic material in powder form is applied in a thin layer to a base plate and locally melted completely by laser radiation, forming a solid layer after solidification. The base plate is then lowered by one layer thickness, and the material is applied again. This cycle is repeated until all layers have been melted.

[0009] Preferably, the raw material, e.g., the powder, for the additive manufacturing process, and thus in particular the material from which the component to be manufactured is made, is metal. The metal, e.g., the metal powder, preferably comprises or consists of titanium, particularly preferably Ti-6Al-4V.

[0010] An additively manufactured component is a component produced using an additive manufacturing process, such as the PBF-LB / M. An additively manufactured component can be a lightweight component featuring lightweight-optimized structures characterized by a load-adapted material distribution based on bionic design principles. In particular, such a component can be additively manufactured from the titanium alloy Ti-6Al-4V using the PBF-LB / M.

[0011] The key concept of the invention is that the microstructure formation of a component is the result of the thermal history in the additive manufacturing process, i.e., the thermal cycles of repeated melting and solidification of a material for the layer-by-layer construction of the component, and that the geometry of the component, in turn, influences the thermal history. It is therefore possible to predict the microstructure using thermal simulation. In this way, the prediction can be carried out in a time- and computationally efficient manner.

[0012] According to step a), a thermal history for a section, preferably several sections of the component, and particularly preferably all sections of the component, is determined from a thermal simulation of an additive manufacturing process for the production of this component. The method thus preferably comprises the step of performing a simulation, in particular a thermal simulation, of the additive manufacturing of the component. The simulation is performed, for example, on the basis of raw data for the additive manufacturing of the component, in particular on the basis of a 3D model of the component, preferably a 3D CAD model, information about the base material, preferably the powder material, information about process parameters, preferably laser parameters, and / or information about support structures for the production of the component. In particular, the thermal history of the entire component is simulated.

[0013] It is particularly advantageous in step a) to simulate the component section by section for different geometries in order to determine the thermal history from the simulated thermal cycles of repeated melting and solidification of the material. In this way, the thermal history can be determined for sections with different geometries. That is, if a section with a specific geometry is simulated, a thermal history corresponding to that geometry is determined; if another section with a different geometry is simulated, a different thermal history is determined. Determining the thermal history here can mean acquiring temperature data over a time window in which the material is cooled, for example, from its melting point to ambient temperature. The temperature data can include data on the temperature distribution within the simulated section.

[0014] It is particularly advantageous to perform thermal simulation using the finite element method (FEM), whereby the component is divided into elements, especially finite ones, which correspond, for example, to sections. Preferably, the elements form individual element layers that are larger in size than the actual layers. This enables efficient simulation at the component level. The component is virtually modeled and simulated, particularly with regard to different geometries, in order to determine, for example, the interaction between the component geometry and the thermal behavior caused by the component geometry section by section. The advantages of finite element analysis are the reduced modeling effort and the shorter computation times.

[0015] Alternatively or cumulatively, thermal simulation can be transient thermal simulation. In particular, thermal simulation takes into account the temporal evolution of the thermal history. Preferably, the heating and / or cooling process is simulated in transient thermal simulation. For example, thermal inertia, especially heat capacity, mass, and / or density, can be considered. In transient thermal simulation, for example, the cooling process from the melting temperature of the material in question to a predefined cooling temperature, e.g., room temperature, is simulated. The result of the simulation is, for example, the temperature distribution as a function of time, thus, in particular, a time-temperature profile.

[0016] It is preferred that the section for which the thermal history is determined is a point, a layer, or several layers, particularly those directly bonded together, of the component to be manufactured. Alternatively, the section may be an area consisting of partial parts or several layers, particularly those directly bonded together. A layer corresponds, in particular, to a distributed powder layer that is applied and then selectively melted by the laser to produce the component layer by layer.

[0017] Alternatively or cumulatively, thermal simulation can be performed using a super-layer approach. In this approach, several, preferably five to twenty, and particularly preferably ten sections, especially layers, of the additively manufactured component are combined for the simulation. The combined sections, especially layers, are preferably consecutive and / or directly connected to one another. Preferably, the thermal history for these multiple sections of the component is thus determined jointly by the thermal simulation. Simulation using the super-layer approach is particularly advantageous in terms of time and / or computational efficiency.

[0018] According to step b), at least one characteristic value is determined from the thermal history determined in step a), the size of which depends on the geometry of the section.

[0019] As already mentioned, microstructure formation depends on the thermal history, and the thermal history in turn depends on the component geometry. Therefore, it is possible to define at least one parameter in advance to quantitatively indicate the relationship between microstructure formation and the thermal history, and to derive this parameter from the thermal history determined in step a) with respect to the simulated section with a specific geometry.

[0020] Such a parameter can relate to the cooling rate. This is based on the phenomenon that a certain geometry can lead to heat buildup, thus reducing the cooling rate after layer formation. This reduced cooling rate can, in turn, influence microstructure formation.

[0021] Such a parameter can also relate to temperature fluctuations during the cooling process. This is based on the phenomenon of "intrinsic heat treatment," where an additional volume is built up above a section under consideration, causing heat input into that section. This means that during the cooling process, the temperature does not decrease gradually but can rise again, resulting in temperature fluctuations. These temperature fluctuations can, in turn, influence microstructure formation.

[0022] According to step c), the at least one characteristic value determined in step b) is compared with a predefined threshold. The threshold is an empirically determined limit value for the characteristic value determined in step b), below or above which the resulting microstructure may deviate from the standard microstructure. The threshold can be determined experimentally beforehand through simulation.

[0023] According to step c), the at least one characteristic value determined in step b) is compared with a predefined threshold. The threshold is an empirical limit for the characteristic value, below or above which the resulting microstructure may deviate from the standard microstructure. The threshold can be determined experimentally beforehand through simulation. Using the previously determined empirical limit, the evaluation of the newly determined characteristic value can be carried out simply and efficiently.

[0024] According to step d), the microstructure of the section is predicted based on the comparison result from step c).

[0025] Steps a) to d) can be repeated so that the microstructure for each section of the component, i.e. the local microstructure for the entire component, can be predicted.

[0026] Although steps a) to d) are given here in a specific order, this order does not always have to be followed. For example, the individual steps can be repeated independently of each other and / or partially omitted during a repetition. It is possible that the steps may overlap at least partially.

[0027] The method described here allows the microstructure of an additively manufactured component, even with larger dimensions, to be predicted efficiently, computationally, and user-friendly using the thermal history from a simulation. The microstructure (e.g., the occurrence of the β-phase as a characteristic feature of titanium alloy components) of entire components, for example, with dimensions of 200 × 200 × 200 mm, can be predicted. 3 can be predicted, whereas with the state of the art, microstructure prediction is only possible for a few mm. 3This is possible. Furthermore, the time required is significantly less compared to the state of the art, amounting to just a few hours per component compared to several years for conventional simulation approaches. In addition, the described method uses commercially available simulation software, the application of which requires less expert knowledge than more complex simulation methods such as phase-field simulations or ICME approaches.

[0028] A particular advantage of the method according to the invention is that the local microstructure at various sections of a component to be additively manufactured can be predicted and the associated mechanical properties can be determined from this, in order to, for example, optimize the component geometry during design and avoid component failure in operation.

[0029] In contrast to the invention, a prediction would not be possible or only possible to a very limited extent; for example, the prediction would either be limited to a size scale of a few mm. 3 The inventive prediction method is limited and / or the computation time can take several years with current computing resources. The prediction method is particularly advantageous in the additive manufacturing of lightweight components with geometrically complex, weight-saving, and / or high-strength structures, since the mechanical properties of an additively manufactured component can be determined from its microstructure. Without this knowledge of the microstructure and the associated mechanical properties, high safety factors would otherwise be necessary, leaving a significant potential for lightweight construction untapped.

[0030] The inventive method for predicting the microstructure of an additively manufactured component particularly allows the microstructure to be predicted in a time- and / or computationally efficient manner, and preferably also enables the mechanical properties of the component to be determined from this prediction. In particular, components with geometrically complex, weight-reduced and / or high-strength structures can thus be manufactured in a material-saving manner, and their operational reliability can be ensured despite the material savings.

[0031] It is preferred that in step a) the thermal simulation is performed using the FEM and / or the super-layer approach and / or transiently, from which the thermal history is determined in the form of a time-temperature profile. In step a), the thermal history in the form of a time-temperature profile (1, 2, 3) is determined in particular by means of transient thermal finite element simulation. With the FEM, it is advantageous to investigate sections with different geometries separately in order to unambiguously determine the influence of the geometry on the thermal history and the microstructure resulting from the thermal history. The super-layer approach allows the investigation of a cross-section with multiple layers. This is advantageous for a time-efficient simulation.The transient simulation allows the determination of the thermal history in the form of a time-temperature profile, which is particularly advantageous for characterizing the relationship between microstructure formation and thermal history.

[0032] It is preferred, when in step a) the thermal simulation is performed using the super-layer approach, that two to 100, preferably five to twenty, and particularly preferably ten layers are combined for the simulation. The layers have a thickness of 5 to 300 µm, preferably 20 to 200 µm, and particularly preferably 30 to 120 µm. In the particularly preferred case with ten layers and a thickness of 60 µm, the combined layers, and thus the super-layer, have a total thickness of 600 µm. With this number and layer thickness, the simulation can be performed efficiently, particularly without significant loss of accuracy.

[0033] It is preferred that in step a) the time-temperature profile depends on the duration during which the material of the section is cooled from its melting temperature to room temperature. This period is advantageous for investigating most common materials with guaranteed accuracy.

[0034] It is preferred that in step b) the at least one characteristic value is a cooling constant and / or a holding time of a, in particular intrinsic, heat treatment. The use of a cooling constant and / or a holding time of a heat treatment as a characteristic value has the advantage that the cooling constant and / or the holding time can be both intuitively recognized from a time-temperature profile and mathematically calculated.

[0035] It is preferred that in step c) the threshold value is specified depending on the material type of the section. In principle, the microstructure of an additively manufactured component made of different materials can be predicted using the method described here. For this purpose, the threshold value with respect to a target material can be determined experimentally by simulation.

[0036] It is preferred that the material is a titanium alloy. It is particularly preferred that in step d), the microstructure is predicted taking into account the presence of a β-phase fraction when the material is a titanium alloy. In particular, the presence of the β-phase fraction is predicted in step d). Preferably, when predicting the presence of the β-phase fraction, it is predicted whether the β-phase is present, e.g., "yes" or "no," and / or the quantitative magnitude of the β-phase fraction is predicted. This is particularly advantageous in the additive manufacturing of mechanically highly stressed components made of the titanium alloy Ti-6Al-4V, since this alloy is characterized by high specific strength and specific stiffness compared to other metals. The microstructure of the components additively manufactured from Ti-6Al-4V using PBF-LB / M is generally characterized by the martensitic α'-phase.However, differences in thermal history due to geometry can also lead to an α + β phase distribution, which ultimately results in different mechanical properties. Therefore, the β-phase fraction is a suitable microstructural parameter for quantifying the influence of component geometry on material properties. For example, the β-phase fraction can vary between approximately 0 and 20% depending on the component geometry.

[0037] It is preferred that in step d) it is predicted that the microstructure exhibits a β-phase fraction when the holding time of the, in particular intrinsic, heat treatment reaches the holding time threshold. The holding time of the heat treatment describes the time between the initial melting and the point in time at which the temperature last exceeds a temperature value, in particular a material-specific one. This temperature value is, for example, between 100°C and 600°C, preferably between 300°C and 500°C, and particularly preferably at 400°C, wherein this temperature value preferably applies to a titanium alloy, e.g., Ti-6Al-4V. It is particularly preferred that the temperature value corresponds to the temperature at which thermally induced transformation processes first take place in the respective alloy. In particular, in the case of Ti-6Al-4V, this is the so-called martensite decomposition, which is observed from 400°C upwards and leads to the formation of the β-phase.In particular, the threshold for the holding time, preferably for a titanium alloy, especially preferably Ti-6Al-4V, can be, for example, 2000 seconds to 5000 seconds, or 3000 seconds to 4000 seconds, or particularly 3600 seconds. Above this threshold, it can be predicted that a β-phase fraction is present.

[0038] It is preferred that in step d) it is predicted that the microstructure has a β-phase fraction if the cooling constant does not reach the threshold value. The cooling constant characterizes the slope of the cooling rate after initial solidification and can be calculated as the slope of an exponential function. For a titanium alloy, e.g., Ti-6Al-4V, the threshold value of the cooling constant can be approximately 1.12 × 10⁻²⁻²; below this threshold, a β-phase fraction may be present.

[0039] It is preferred that steps a) to d) are repeated so that the microstructure of the entire component is predicted section by section. It is particularly preferred if, after step d), a mechanical property of the component is determined from the microstructure of the entire component. It is also preferred if the mechanical property is the maximum load-bearing capacity of the component. In this way, components with geometrically complex, weight-reduced, but highly resilient structures can be additively manufactured in a material-saving manner, and their operational reliability can be ensured despite the material savings.

[0040] The invention further comprises a data processing system that includes means for carrying out the method according to the invention.

[0041] The invention further comprises a computer program for carrying out the method according to the invention. In other words, this relates in particular to a computer program (product) comprising instructions which, when the program is executed by a computer, cause it to execute a method described herein.

[0042] The invention further comprises a machine-readable storage medium on which the proposed computer program is stored. The machine-readable storage medium is typically a computer-readable data carrier.

[0043] Furthermore, the invention comprises a method for the additive manufacturing of a component, comprising the following steps: i) Carrying out the above-described procedure for predicting the microstructure of the additively manufactured component, and ii) Additive manufacturing of the component, taking into account the predicted microstructure.

[0044] In additive manufacturing, the component is preferably 3D printed, particularly preferably using metal 3D printing. Specifically, additive manufacturing is carried out as powder bed fusion, e.g., using laser beam melting (LPBF) and / or electron beam melting (EBM). Additive manufacturing is particularly preferably carried out as PBF-LB / M. The raw material, e.g., the powder, for additive manufacturing, and thus particularly the material from which the component to be manufactured is made, is preferably metal. The metal, e.g., the metal powder, preferably comprises or consists of titanium, particularly Ti-6Al-4V.

[0045] It is preferred that in step ii) the component is manufactured taking into account the predicted microstructure property, such that additive process parameters, component geometry and / or post-treatment are adapted accordingly. In particular, the following process parameters can be adapted taking into account the predicted microstructure property: - the power and / or the speed of movement of the laser in LPBF and / or the electron beam in EBM; - Break times, especially between the production of shifts; - a geometry, e.g., the component geometry, support structures and / or heat sinks; and / or - a post-treatment, e.g. a heat treatment.

[0046] The method proposed here allows the microstructure of an additively manufactured component to be predicted through simulation, without the need for a prototype. Based on the predicted microstructure, the component's microstructure-dependent mechanical properties can be determined, particularly to ascertain whether topology optimization is required. This is especially advantageous for additively manufactured lightweight components that are intended to have weight-saving yet highly resilient structures, enabling the appropriate application of safety factors and thus better exploiting lightweight design potential.

[0047] Predicting the microstructure of an additively manufactured component allows for improvements to its design and topology at an early stage of development. The proposed method helps save time and costs while simultaneously increasing safety.

[0048] The solution presented here and its technical context are explained in more detail below with reference to the figures. It should be noted that the invention is not intended to be limited by the exemplary embodiments shown. In particular, unless explicitly stated otherwise, it is also possible to extract partial aspects of the situations explained in the figures and combine them with other components and / or findings from other figures and / or the present description. The figure schematically illustrates: Fig. 1. Thermal histories in the form of time-temperature profiles for three exemplary sections, Fig. 2. Definition of a first characteristic value taking into account an intrinsic heat treatment, Fig. 3. Definition of a second characteristic value taking into account a reduced cooling rate, Fig. 4. Definition of the threshold for the first parameter, and Fig. 5. Definition of the threshold value for the second parameter.

[0049] Fig. 1, Fig. 2, Fig. 3, Fig. 4 and Fig. Figure 5 jointly demonstrates a specific implementation variant of a method proposed here, in which the prediction of the microstructure of a component to be additively manufactured using PBF-LB / M from the titanium alloy Ti-6AI-4V is shown.

[0050] The microstructure of such a component is typically characterized by the martensitic α' phase. However, geometrically induced differences in the thermal history can also lead to an α + β phase distribution, which ultimately results in different mechanical properties. Therefore, in this design variant, the β phase fraction is a suitable microstructural parameter for quantifying the influence of the component geometry on the material properties. Scientific studies show that the β phase fraction can vary between approximately 0 and 20%, depending on the component geometry.

[0051] Fig. Figure 1 shows thermal histories in the form of time-temperature profiles for three exemplary investigated sections 4, 5, and 6 with exemplary sample geometries "C11" (geometry 9) and "C25" (geometries 10 and 11). Geometries 10 and 11 are identical. A first time-temperature profile 1 from a thermal simulation for a first section 4, a second time-temperature profile 2 from a thermal simulation for a second section 5, and a third time-temperature profile 3 from a thermal simulation for a third section 6 are shown.

[0052] The first section 4, the second section 5 and the third section 6 are chosen as test samples such that the reduced cooling rate and / or the intrinsic heat treatment caused by the component geometry can occur, which in the PBF-LB / M process lead to a deviation from the standard microstructure, i.e. an a + β phase distribution as opposed to a martensitic α' phase in the microstructure.

[0053] In Fig. Figure 1 shows that the first time-temperature curve is a continuous curve in which the temperature drops from 1605 °C to 100 °C in approximately 1000 seconds. The cooling time of 1000 seconds is longer than the cooling time of approximately 200 seconds (see Figure 1). Fig. 3) Under normal circumstances. From the first time-temperature profile 1, it can be seen that due to a geometrically induced heat build-up in the first section 4, the cooling rate after visualization is reduced. The thermal influence is evident in the microstructure of the first section 4 by an increased proportion of the β-phase. The selected test position for section 4 is shown as an example in sample geometry 9 with P0.

[0054] In Fig. It can be seen that, in contrast to the first time-temperature curve 1, which is a continuous curve, the second time-temperature curve 2 exhibits temperature fluctuations. This means that after an initial cooling, a secondary temperature increase occurs to a temperature level in the range of 400°C to 600°C. From the second time-temperature curve 2, it can be seen that intrinsic heat treatment occurs when an additional volume is built up above the second section 5, which causes heat input into the second section 5, also leading to a β-phase fraction in the microstructure of the second section 5. The selected investigation point for section 5 is shown as an example in the sample geometry 10 with P. u .

[0055] In Fig. It can also be seen that the time-temperature profile 3 behaves like a superposition of the first time-temperature profile 1 and the second time-temperature profile 2. This could mean that both the reduced cooling rate and the intrinsic heat treatment occur in the third section 6, which would also reveal a β-phase fraction in the microstructure of the third section 6. The selected test position for section 6 is shown as an example in sample geometry 11 with P2.

[0056] The time-temperature profiles 1, 2, 3 resulting from the thermal simulation of the three sections 4, 5, 6 each show characteristic profiles indicating a reduced cooling rate and / or intrinsic heat treatment. Compared to the second time-temperature profile 2, the first time-temperature profile 1 is characterized by a reduced cooling rate from approximately 700 °C to 100 °C, while the second time-temperature profile 2 exhibits a more pronounced temperature drop. After initial cooling, the second time-temperature profile 2 shows a secondary temperature increase to a level between 400 °C and 600 °C, which is maintained almost continuously until the sample is finished. The intrinsic heat treatment is thus also evident in the simulated thermal history.The third time-temperature curve 3 shows the superposition of the reduced cooling rate at the beginning and the intrinsic heat treatment in the further course.

[0057] Fig. Figure 2 shows an example of how a first characteristic value is defined taking into account an intrinsic heat treatment.

[0058] Fig. Figure 2 shows a holding time Δt IHT 7 of an intrinsic heat treatment. The holding time Δt IHT 7 describes the time between the initial melting and the time when the temperature last exceeds 400 °C.

[0059] In Fig. 2. It can be seen that the holding time Δt IHT 7 is approximately 2250 seconds and is therefore significantly shorter than the holding time of the second time-temperature curve 2 in Fig. 1 of 5500 seconds and the holding time of the third time-temperature curve 3 in Fig. 1 in approximately 4000 seconds.

[0060] The holding time of an intrinsic heat treatment is therefore suitable as a parameter for predicting the microstructure.

[0061] Fig. Figure 3 shows an example of how a second characteristic value is defined taking into account a reduced cooling rate.

[0062] Fig. Figure 3 shows a time-temperature curve in the form of a smooth curve. This contrasts with the first time-temperature curve (Figure 1). Fig. 1, which is also a smooth curve, shows the time-temperature profile in Fig. 3 a greater slope, so that the cooling time is reduced to 250 seconds, while the cooling time of the first time-temperature curve 1 in Fig. 1 is approximately 1000 seconds.

[0063] The slope of a time-temperature curve in the form of a smooth curve is therefore suitable as a characteristic value for predicting the microstructure.

[0064] Mathematically, this time-temperature curve can be represented by an exponential function, and the slope of this time-temperature curve can therefore be represented by a cooling constant. The cooling constant can be expressed using the formula... T(t)=100°C+(1605°C−100°C)∗e−k∗t The cooling constant k can be calculated, where k is the cooling constant, T(t) is the temperature at time t, e is the Euler number, 1605 °C is the initial temperature, and 100 °C is the final temperature. Here, the cooling constant k can be approximated from the first six data points (8) of the time-temperature curve from the initial temperature to the final temperature.

[0065] Fig. Figure 4 shows, by way of example, the critical holding time Δt of the intrinsic heat treatment. IHT,krit , which represents the time period after which the occurrence of the β-phase can be experimentally detected.

[0066] In Fig. 4 shows that the critical holding time Δt IHT,kritapproximately 3600 seconds and below the critical holding time only a reduced cooling rate occurs, resulting in a β-phase fraction of a maximum of 6% (see Fig. 5), and from the critical holding time onwards, both a reduced cooling rate and an intrinsic heat treatment can occur, and thus a β-phase fraction of up to 20% may be present.

[0067] Fig. Figure 5 shows, by way of example, the critical cooling constant k. krit , which characterizes the slope of the cooling rate after initial solidification and serves as the threshold of the cooling constant below which the occurrence of the β-phase can be experimentally demonstrated.

[0068] Fig. 5 shows that the critical cooling constant k krit has a value of approximately 1.12E-02. Is a cooling constant smaller than the critical cooling constant k? krit, a β-phase fraction may be present, and a cooling constant is greater than the critical cooling constant k. krit (e.g. in the range of 1.5E-O2 or 2.25E-O2), there is no β-phase component.

[0069] The in Fig. 5. Cooling constant shown, which is smaller than the critical cooling constant k. krit , is the cooling constant derived from the first time-temperature curve 1 in Fig. 1 was determined according to formula (1), in which a β-phase fraction of a maximum of approximately 6% may be present. Reference symbol list 1. First time-temperature curve 2 Second time-temperature curve 3 Third time-temperature curve 4 First Section 5 Second Section 6 Third Section 7 Holding time Δt IHT 8 data points 9-11 Geometries

Claims

[1] Method for predicting the microstructure of an additively manufactured component, comprising the following steps: a) Determining a thermal history for at least one section (4, 5, 6) of the component from a thermal simulation of an additive manufacturing process of the component, b) Determine at least one characteristic value from the thermal history, the magnitude of which depends in particular on the geometry of the section, c) Comparing at least one characteristic value with a predetermined threshold value, and d) Predictions of the microstructure of the section based on the comparison result from step c). [2] Method according to claim 1, wherein in step a) the thermal simulation is carried out using an FEM. [3] Method according to claim 2, wherein in step a) the thermal simulation is carried out using a super-layer approach, such that several, e.g. five to twenty, sections, in particular layers of the additively manufactured component, are combined for the simulation. [4] Method according to one of the preceding claims, wherein in step a) the thermal history in the form of a time-temperature profile (1, 2, 3) is determined by means of transient thermal finite element simulation. [5] Method according to one of the preceding claims, wherein in step a) the time-temperature profile (1, 2, 3) depends on the duration during which the material of the section is cooled from the melting temperature to a cooling temperature, in particular room temperature. [6] Method according to one of the preceding claims, wherein in step b) the at least one characteristic value is a cooling constant and / or a holding time of a, in particular intrinsic, heat treatment. [7] Method according to one of the preceding claims, wherein in step c) the threshold value is specified depending on the type of material of the section. [8] Method according to claim 7, wherein the material is a titanium alloy. [9] Method according to claim 8, wherein in step d) the microstructure is predicted taking into account the presence of a β-phase fraction. [10] Method according to claim 9, wherein in step d) it is predicted that the microstructure has a β-phase fraction when the holding time of the intrinsic heat treatment reaches the holding time threshold. [11] Method according to claim 9 or 10, wherein in step d) it is predicted that the microstructure has a β-phase fraction if the cooling constant does not reach the threshold of the cooling constant. [12] Method according to any of the preceding claims, wherein steps a) to d) are repeated so that the microstructure of the entire component is predicted section by section. [13] Method according to claim 12, wherein, after step d), a mechanical property of the component is determined from the microstructure of the entire component. [14] Method according to claim 13, wherein the mechanical property is the maximum load-bearing capacity of the component. [15] Data processing system comprising means for carrying out the method according to any one of claims 1 to 14. [16] 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 14. [17] Computer-readable storage medium, wherein the storage medium comprises instructions which, when executed by a computer, cause it to execute the method according to claim 1 or claim 14, and / or the computer program according to claim 16 is stored on the storage medium. [18] Method for the additive manufacturing of a component, comprising the steps: i) Carrying out the method according to any one of claims 1 to 14 for predicting the microstructure of the additively manufactured component, and ii) Additive manufacturing of the component, taking into account the predicted microstructure. [19] Method according to claim 18, wherein in step ii) the component is manufactured taking into account the predicted microstructure property such that additive process parameters, component geometry and / or post-treatment are adapted according to the predicted microstructure property.

Images