3D printing microstructure and performance accurate regulation method fed back by thermal field behavior monitoring

Abstract

The application discloses a kind of 3D printing microstructure and performance accurate regulation and control method by thermal field behavior monitoring feedback, belong to the technical field of additive manufacturing, including: step one, obtain the process parameters in the process of additive manufacturing and real-time monitoring thermal field behavior;Step two, optimize 3D thermal field model and generate a group of selected process parameters under temperature field;Step three, utilize cellular automaton coupling through the temperature field under selected process parameters generated by 3D thermal field model optimized in step two and its corresponding multi-physical field, to simulate the microstructure topography under corresponding process parameters;Step four, train microstructure accurate regulation and control model;Step five, target part printing and its microstructure distribution real-time accurate regulation and control.It can be seen from this that the application can accurately regulate the generation of microstructure, improve the overall mechanical properties of target printed parts.

Description

Technical Field

[0001] This invention relates to a method for precise control of the microstructure and properties of 3D printing through thermal field behavior monitoring feedback, belonging to the technical field of additive manufacturing. Background Technology

[0002] Additive manufacturing (AM), also known as 3D printing, is a technology based on the discrete-stacking principle, which manufactures solid parts by depositing materials layer by layer. Unlike traditional equal-material manufacturing (such as casting and forging) or subtractive manufacturing (such as machining), this technology enables the direct forming of complex structures, greatly improving design freedom and production efficiency. However, during additive manufacturing, the repeated remelting and rapid cooling of powder layers makes it difficult to precisely control the microstructure of the formed components, leading to problems such as anisotropy.

[0003] Currently, methods for controlling the microstructure in metal additive manufacturing are based on simulating the thermal field and microstructure growth during the additive manufacturing process using the relevant properties of powder materials. Then, process parameters are optimized, such as laser power and scanning speed, to adjust the temperature gradient and promote grain refinement. However, the simulated temperature field cannot accurately represent the complex changes in the temperature field during additive manufacturing. Furthermore, this method cannot control the formation of microstructures in real time.

[0004] Therefore, in order to accurately control the morphology of microstructures in real time, it is necessary to study a method for precise control of 3D printed microstructures based on thermal field behavior monitoring and machine learning. Summary of the Invention

[0005] To address the shortcomings and improvement needs of existing technologies, this invention provides a method for precise control of the microstructure and properties of 3D printed parts through thermal field behavior monitoring and feedback. It optimizes a 3D thermal field model by real-time monitoring of the thermal field behavior during additive manufacturing, enabling the optimized model to output the temperature field corresponding to any process parameter within a given process parameter range. Then, based on cellular automata coupled with multiphysics simulation, it generates the microstructure morphology under the corresponding process parameters, thereby constructing a dataset to train a microstructure control model based on a BP neural network. The trained microstructure control model is then used to intelligently control the microstructure morphology during additive manufacturing, suppressing crack propagation and improving the overall mechanical properties of the target printed part.

[0006] To achieve the above-mentioned technical objectives, the present invention will adopt the following technical solution:

[0007] A method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback includes the following steps:

[0008] For a specific printing powder, a range of process parameters is given, and multiple printing schemes are set up in pairs based on different process parameters within the given range;

[0009] By acquiring the process parameters corresponding to each printing scheme and by acquiring the slice data of the printed part model in the additive manufacturing process corresponding to each printing scheme, and monitoring the thermal field behavior of the molten pool under the corresponding process parameters based on the slice layer height, the thermal field behavior monitoring data corresponding to the corresponding process parameters can be obtained, and the physical field related to the material properties of the printing powder can be acquired simultaneously.

[0010] The 3D thermal field model is optimized based on the obtained thermal field behavior monitoring data to numerically simulate the temperature field of the molten pool during additive manufacturing. Based on the optimized 3D thermal field model, a set of temperature fields under a given range of process parameters is output.

[0011] The dataset is constructed by coupling the temperature field under various process parameters output by the 3D thermal field model with cellular automata, the laser flux calculated based on the corresponding process parameters, and the physical field related to the material properties of the printing powder. The microstructure morphology under the corresponding process parameters is simulated and generated according to the material properties of the printing powder, and then a dataset is constructed. Each sample in the dataset includes the process parameters, the temperature field under the corresponding process parameters output by the 3D thermal field model, and the microstructure morphology under the corresponding process parameters generated by the cellular automata simulation.

[0012] A microstructure control model is constructed based on a BP neural network model. The input layer of the microstructure control model is the process parameters in the dataset and the temperature field under the corresponding process parameters output by the 3D thermal field model. The output layer is the microstructure morphology under the corresponding process parameters generated by cellular automata simulation. The microstructure control model is trained by this method, and each sample in the training set is stored as historical data in the historical database of the microstructure control model. The input of the trained microstructure control model is the process parameters and the thermal field behavior monitoring data under the corresponding process parameters. The output is the microstructure morphology prediction under the corresponding process parameters.

[0013] During the additive manufacturing of the target part, a trained microstructure control model is used to monitor the microstructure morphology of each slice layer, and the process parameters of each slice layer are intelligently controlled based on historical data in the historical database to ensure that the microstructure morphology of each slice layer matches the microstructure morphology selected based on the performance requirements of the target part.

[0014] Preferably, during the additive manufacturing of the target part, the microstructure morphology of each slice layer is ensured to match the microstructure morphology selected based on the performance requirements of the target part. This is achieved in the following ways:

[0015] During the additive manufacturing of the target part, real-time monitoring data of thermal field behavior and corresponding process parameters are input into a trained microstructure control model to monitor the microstructure morphology of the target part in the current slice layer in real time. The model evaluates whether the microstructure morphology of the current slice layer output by the microstructure control model matches the microstructure morphology selected based on the performance requirements of the target part. If the evaluation results show that the microstructure morphology of the current slice layer output by the microstructure control model does not match the microstructure morphology selected based on the performance requirements of the target part, then suitable process parameters are intelligently matched from the historical database based on the microstructure morphology selected based on the performance requirements of the target part and input into the control program of the additive manufacturing forming equipment. This guides the additive manufacturing forming equipment to form the next slice layer according to the matched process parameters until the microstructure morphology of the current slice layer output by the microstructure control model matches the microstructure morphology selected based on the performance requirements of the target part.

[0016] Preferably, the process parameters include laser power, laser beam shape, spot size, laser scanning speed, powder layer thickness, and scanning spacing.

[0017] Preferably, the thermal behavior monitoring data includes molten pool depth, molten pool width, temperature gradient, cooling rate, and subcooling.

[0018] The thermal behavior monitoring data is acquired using an infrared thermal imager, a high-speed visible spectrum camera, a high-speed color camera, and / or a photoelectric detector.

[0019] Preferably, the microstructure morphology includes microstructure classification and its distribution; the microstructure classification is based on the influence of grain size and mechanical properties of the part, including equiaxed crystals and columnar crystals; the distribution of the microstructure classification includes the proportion of equiaxed crystals, the proportion of columnar crystals, and the average grain size of equiaxed crystals.

[0020] Preferably, when optimizing the 3D thermal field model, the obtained thermal field behavior monitoring data is continuously compared with the output results of the 3D thermal field model, and then the setting parameters of the 3D thermal field model are changed to continuously optimize the 3D thermal field model until the output results of the 3D thermal field model match the obtained thermal field behavior monitoring data.

[0021] Preferably, the physical field obtained that relates to the material properties of the printing powder includes the powder flow velocity and the protective gas flow velocity.

[0022] Preferably, when the target part has multiple regions corresponding to different microstructures, a microstructure transition region is provided between two adjacent regions corresponding to different microstructures.

[0023] During the additive manufacturing of target parts, when printing the slices corresponding to the transition regions of the microstructure, the laser energy density changes in a gradient with the increase of the corresponding slices, so as to achieve a natural transition between adjacent regions with different microstructure morphologies.

[0024] Preferably, the printed microstructure transition region corresponds to any slice layer. Corresponding laser energy density satisfy:

[0025] , ;

[0026] In the above formula: , This indicates the laser energy density used when printing two adjacent different microstructures; This indicates the total number of slice layers corresponding to the transition region of the microstructure; This represents the slice layer count index corresponding to the transition region of the microstructure. It represents the energy regulation index.

[0027] Another technical objective of this invention is to provide a computer program, including a memory, a processor, and a computer program stored in the memory and running on the processor, which executes the above-described method for precise control of 3D printed microstructures and properties through thermal field behavior monitoring feedback.

[0028] Based on the above-mentioned technical objectives, the present invention has the following advantages compared with the prior art:

[0029] 1. The 3D printing microstructure precision control method of this invention optimizes the 3D thermal field model by real-time monitoring of the thermal field behavior during additive manufacturing, and outputs the temperature field corresponding to any process parameter within a given process parameter range from the optimized 3D thermal field model. Then, based on cellular automata coupled with multiphysics simulation, the microstructure morphology under the corresponding process parameters is generated, thereby constructing a dataset to train the microstructure control model based on BP neural network. The trained microstructure control model is then used to intelligently control the microstructure morphology during additive manufacturing. Thus, this invention combines real-time monitoring technology of thermal field behavior during additive manufacturing with machine learning, enabling real-time monitoring of the microstructure morphology of local areas when single or multiple layers of powder melt during additive manufacturing, and fine control of the microstructure morphology according to the required component performance. This allows the performance of the part to be controlled at the microscale, suppressing crack propagation, making additive manufacturing more effective and reliable in microscale processing, and improving the overall mechanical properties of the target printed part.

[0030] 2. Based on real-time monitoring of thermal field behavior, this invention uses parameters such as temperature gradient, cooling rate, and supercooling to construct a temperature field model (3D thermal field model). Then, based on cellular automata coupled with multiphysics, it simulates the microstructure growth during additive manufacturing. The generated microstructure morphology is closer to the actual microstructure than that generated by relying solely on temperature field simulation.

[0031] 3. Based on machine learning of thermal field behavior, this invention can be applied to the microstructure control of various metal materials in the additive manufacturing process. It can also be further used to optimize process parameters, reduce the need for a large amount of experimental data, reduce the cost of front-end development and research for enterprises, and enable intelligent manufacturing to be implemented in the industrial production process.

[0032] 4. This invention aims at microscale control, which can greatly reduce the problems of low dimensional accuracy, surface roughness and anisotropy of mechanical properties faced by formed parts in the additive manufacturing process. Attached Figure Description

[0033] Figure 1 This is a flowchart of the method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in this invention;

[0034] Figure 2 This is a structural diagram of the micro-organism regulation model based on BP neural network described in this invention;

[0035] Figure 3 This is a flowchart of the dataset used to train, test, and validate the micro-organism regulation model according to the present invention;

[0036] Figure 4 This is a flowchart of the microstructure simulation based on cellular automata described in this invention;

[0037] Figure 5 This is a diagram illustrating the two types of microstructures as defined in this invention;

[0038] Figure 6 It is a fitting graph after the micro-organism regulation model is regulated using the training set.

[0039] Figure 7 The figures are microstructure generation diagrams of different layers of the target printed part of the present invention. In the figures, (a) represents the microstructure morphology diagram of the previous printed layer of the target printed part, (b) represents the microstructure morphology diagram of the current printed layer of the target printed part, and (c) represents the microstructure morphology diagram of the next printed layer of the target printed part. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. Unless otherwise specifically stated, the relative arrangement, expressions, and values ​​of components and steps set forth in these embodiments do not limit the scope of the present invention. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0041] Example 1

[0042] like Figure 1 As shown, this embodiment provides a method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback, including the following steps:

[0043] Step 1: Obtain process parameters and monitor thermal behavior in real time during additive manufacturing.

[0044] For a specific printing powder, a range of process parameters is given, and multiple printing schemes are set up in pairs based on different process parameters within the given range.

[0045] Each printing scheme is additively manufactured to form corresponding test pieces. By acquiring slice data of each test piece (printed part model) during the additive manufacturing process, and using the slice layer height as a basis for monitoring the thermal field behavior of the molten pool, thermal field behavior monitoring data corresponding to the corresponding process parameters can be obtained. Simultaneously, the physical field related to the material properties of the printing powder can also be acquired. The thermal field behavior monitoring data includes molten pool depth, molten pool width, temperature gradient, cooling rate, and undercooling.

[0046] In this step, the process parameters include laser power, laser beam shape, spot size, laser scanning speed, powder layer thickness, and scanning spacing. These process parameters are obtained as follows:

[0047] The laser beam shape and spot size parameters are collected on the laser, and the process parameters are collected by acquiring the slice data of each specimen on the additive manufacturing equipment.

[0048] In this step, the method for obtaining the thermal field behavior is as follows:

[0049] Infrared thermal imagers, high-speed visible spectrum cameras, high-speed color cameras, and photodetectors can all be used to measure the temperature field during additive manufacturing and obtain thermal behavior data. Infrared thermal imaging technology can be used to observe the changes in the molten pool temperature field at different layer heights during printing, and to record in detail the temperature gradient and cooling rate of the molten pool.

[0050] In this step, the physical fields obtained that are related to the material properties of the printing powder include the powder flow velocity and the protective gas flow velocity.

[0051] Step 2: Optimize the 3D thermal field model

[0052] The 3D thermal field model is optimized by continuously comparing the obtained thermal field behavior monitoring data with the output results of the 3D thermal field model, and then changing the setting parameters of the 3D thermal field model to continuously optimize the 3D thermal field model. This ensures that the output results of the optimized 3D thermal field model match the actual monitored thermal field behavior data, so as to numerically simulate the temperature field of the molten pool during the additive manufacturing process. This ensures that the temperature field changes of the multi-layer molten powder and the internal temperature field of the molten pool output by the optimized 3D thermal field model are more accurate and reliable.

[0053] Based on the optimized 3D thermal field model, a set of temperature fields under specified process parameters is output within a given range of process parameters, providing input data for the subsequent cellular automata coupled simulation to output the corresponding microstructure morphology.

[0054] Step 3: Simulate microstructure morphology using cellular automata coupled with multiphysics.

[0055] like Figures 1-3 As shown, the temperature field under each process parameter, the laser flux calculated based on the corresponding process parameters, and the physical field related to the material properties of the printing powder are coupled by cellular automata through the 3D thermal field model optimized in step two. The microstructure morphology under the corresponding process parameters is simulated and generated according to the material properties of the printing powder, and then a dataset is formed.

[0056] Each sample in the dataset includes process parameters, the temperature field under the corresponding process parameters output by the 3D thermal field model, and the microstructure morphology under the corresponding process parameters generated by cellular automata simulation.

[0057] Each sample in the dataset has undergone data preprocessing operations such as normalization.

[0058] Step 4: Training the micro-organism regulation model

[0059] A microstructure control model is constructed based on a BP neural network model, and the constructed microstructure control model is trained using the dataset established in step three, so that the trained microstructure control model has the function of precise control of the microstructure of local areas of single-layer or multi-layer molten powder.

[0060] During training, the temperature field under the corresponding process parameters output by the 3D thermal field model and the process parameters in the dataset are used as the input layer of the microstructure control model, while the microstructure morphology under the corresponding process parameters generated by cellular automata simulation is used as the output layer of the microstructure control model. Each sample in the dataset is stored as historical data in the historical database of the microstructure control model.

[0061] The input to the trained microstructure control model is the process parameters and the thermal behavior monitoring data under the corresponding process parameters, and the output is the microstructure morphology prediction under the corresponding process parameters.

[0062] In this step, the microstructure morphology, such as Figure 5 As shown, the microstructure classification includes its distribution; the microstructure classification is based on the influence of grain size and the mechanical properties of the part, including equiaxed crystals and columnar crystals; the distribution of the microstructure classification includes the volume ratio of equiaxed crystals, the volume ratio of columnar crystals, and the average grain size of equiaxed crystals.

[0063] Step 5: Print the target part and precisely control its microstructure distribution in real time.

[0064] In the additive manufacturing process of the target part, a trained microstructure control model is used to monitor the microstructure morphology of each slice layer. Based on historical data from a historical database, the process parameters of each slice layer are intelligently adjusted to ensure that the microstructure morphology of each slice layer matches the microstructure morphology selected based on the performance requirements of the target part. Specifically, during the additive manufacturing process of the target part, real-time monitored thermal behavior data and corresponding process parameters are input into the trained microstructure control model to monitor the microstructure morphology of the target part in the current slice layer in real time. The model then evaluates whether the microstructure morphology output by the microstructure control model for the current slice layer matches the microstructure morphology selected based on the performance requirements of the target part. If the evaluation result indicates that the microstructure morphology output by the microstructure control model for the current slice layer does not match the microstructure morphology selected based on the performance requirements of the target part, then... Based on the microstructure morphology selected according to the performance requirements of the target part, suitable process parameters are intelligently matched from the historical database and input into the control program of the additive manufacturing equipment. This guides the additive manufacturing equipment to form the next slice layer according to the matched process parameters, thereby controlling the local microstructure morphology of the current slice layer that has been melted and solidified, in order to form according to the preset microstructure morphology. Then, the generated microstructure morphology is monitored until the microstructure morphology of the current slice layer output by the microstructure control model matches the microstructure morphology selected based on the performance requirements of the target part.

[0065] The micro-organism control model described in this embodiment also has a self-updating capability. After updating the historical data in the historical database, it can be trained using a certain number (e.g., 20) of the latest historical data, thereby adjusting and optimizing the micro-organism control model and improving its ability to control micro-organism generation during the printing process.

[0066] Accuracy adjustment and verification of the micro-organism regulation model: The model was debugged using the printing parameters shown in Table 1.

[0067] Table 1

[0068]

[0069] Following the steps described in the embodiment, firstly, the process parameters listed in Table 1 are selected for the aluminum alloy parts (the only changes are the laser power and scanning speed; other printing parameters, such as laser beam shape, spot size, laser scanning speed, and scanning spacing, remain unchanged and are therefore not listed in Table 1). Then, the optimized 3D thermal field model outputs the temperature field corresponding to each process parameter. Based on the material properties of the aluminum alloy parts, a cellular automaton is used to couple the selected process parameters and the corresponding temperature field output by the optimized 3D thermal field model to generate the corresponding microstructure morphology, including equiaxed crystals, columnar crystals, and their proportions. Finally, the obtained process parameters and the temperature field under each process parameter are used as the input layer of the microstructure control model, and the microstructure morphology under each process parameter is used as the output layer of the microstructure control model. The accuracy of the microstructure control model is optimized to obtain the following result: Figure 6 The simulation results of tissue regulation shown demonstrate that this model has high accuracy in regulating tissue generation.

[0070] In some implementations, step five also includes verifying the actual printing effect of the microstructure precision control model, including:

[0071] By modifying the microstructure type monitored during the printing process using the microstructure control model, the target part is obtained. Then, the microstructure morphology of the target part is verified to observe whether the actual generated microstructure conforms to the set structure type.

[0072] Further verification of the printing results can effectively validate the accuracy of the organization's precise control model in practical applications.

[0073] like Figure 7 As shown, the microstructure distribution of the printed aluminum alloy parts in the model is set to 90% equiaxed grains and 10% columnar grains. From bottom to top, the proportion of equiaxed grains gradually increases in different layers of the molten pool, and the grain size of the equiaxed grains becomes finer, which is consistent with the set microstructure distribution.

[0074] Example 2

[0075] To meet the various mechanical performance requirements of different parts of complex metal parts (such as gears), it is necessary to match the corresponding microstructures in the database according to the performance of each part. However, a transition layer is needed between adjacent microstructure regions to allow for gradual transformation. This is mainly achieved by adjusting the process parameters and then gradually adjusting the laser energy density.

[0076] During layer-by-layer printing, sudden and drastic adjustments to energy density can have serious consequences. Such abrupt changes disrupt stable thermal cycling, generating significant residual stress within the material, leading to warping, cracking, and even peeling of parts from the substrate. Simultaneously, it directly creates fatal metallurgical defects, severely impairing the structural integrity and mechanical properties of the parts. Furthermore, microstructure inhomogeneity occurs in the abrupt change regions, forming weak interfaces. Therefore, sudden changes in energy density pose a primary threat to process stability and part yield, and must be avoided through gradual adjustments.

[0077] In other words, when the target part has multiple regions corresponding to different microstructures, this invention, when planning the printing file, needs to set a microstructure transition region between two adjacent regions corresponding to different microstructures. Therefore, during the additive manufacturing of the target part, when printing each slice layer corresponding to the microstructure transition region, the laser energy density changes in a gradient with the increase of the corresponding slice layer, achieving a natural transition between two adjacent regions corresponding to different microstructures. This allows for the printing of any slice layer corresponding to the microstructure transition region. Corresponding laser energy density satisfy:

[0078] , ;

[0079] In the above formula: , This indicates the laser energy density used when printing two adjacent different microstructures; This indicates the total number of slice layers corresponding to the transition region of the microstructure; This represents the slice layer count index corresponding to the transition region of the microstructure. The energy regulation index, which can be quickly obtained based on preliminary estimation methods using theory and experience, is typically represented by slice layers. Approaching laser energy density In the corresponding region, the selected energy modulation index p < 1 is close to the laser energy density. When the corresponding region is selected, the energy regulation index p>1 is chosen. When it is in the middle of the micro-tissue transition region, the energy regulation index p=1 is chosen. The total number of slice layers n is generally between 20 and 30 layers.

[0080] In the process of 3D printing aluminum alloy powder to manufacture metal gears, in order to meet the performance requirements of "internal strength and toughness, and surface hardness and wear resistance," it is necessary to control the existence of a microstructure transition layer between the inside and the surface of the gear. First, the core of the metal gear is printed. This part needs to achieve high tensile strength and good ductility. A bimodal structure with coexisting columnar and equiaxed crystals is often the better choice, with columnar and equiaxed crystals accounting for approximately 80% and 20% of the volume, respectively, and the equiaxed crystal grain size being approximately 2μm. Therefore, process parameters matching this type of microstructure are selected from the database, and then a microstructure control model is used to predict and match whether the microstructure meets the required performance. Since the surface of the metal gear requires extremely high hardness and wear resistance, efforts should be made to obtain a high proportion of fine equiaxed crystals. Therefore, process parameters with an equiaxed crystal distribution of over 90% and a grain size of approximately 1μm are selected for adjustment. Based on the slice data, the energy density is gradually changed starting from the position on the surface of the metal gear, gradually transitioning the microstructure from a bimodal structure with coexisting columnar and equiaxed crystals to ultrafine equiaxed crystals. In the microstructure transition layer, thermal field behavior is monitored in real time and the formation of microstructures is predicted to determine whether the microstructure meets the actual requirements. If not, the range of energy density variation is reduced until the microstructures of adjacent layers can reach a stable bonding state. The microstructure transition region is divided into three parts from bottom to top according to the slice layers: the first region (i.e., the part near the metal gear core), the middle region, and the second region (i.e., the part near the metal gear surface). The slice layers constituting the first region... The energy regulation index p is set to 0.6 for each slice layer constituting the second region. The energy regulation index p is set to 1.8.

[0081] Example 3

[0082] The present invention also provides a storage medium, wherein the computer program stored in the storage medium executes the above-described method for precise control of 3D printing microstructure and performance through thermal field behavior monitoring feedback.

[0083] Example 4

[0084] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for precise control of 3D printed microstructures and properties through thermal field behavior monitoring feedback via the computer program.

[0085] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0086] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0087] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0088] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0090] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback, characterized in that, Includes the following steps: For a specific printing powder, a range of process parameters is given, and multiple printing schemes are set up in pairs based on different process parameters within the given range; By acquiring the process parameters corresponding to each printing scheme and by acquiring the slice data of the printed part model in the additive manufacturing process corresponding to each printing scheme, and monitoring the thermal field behavior of the molten pool under the corresponding process parameters based on the slice layer height, the thermal field behavior monitoring data corresponding to the corresponding process parameters can be obtained, and the physical field related to the material properties of the printing powder can be acquired simultaneously. The 3D thermal field model is optimized based on the obtained thermal field behavior monitoring data to numerically simulate the temperature field of the molten pool during additive manufacturing. Based on the optimized 3D thermal field model, a set of temperature fields under a given range of process parameters is output. The dataset is constructed by coupling the temperature field under various process parameters output by the 3D thermal field model with cellular automata, the laser flux calculated based on the corresponding process parameters, and the physical field related to the material properties of the printing powder. The microstructure morphology under the corresponding process parameters is simulated and generated according to the material properties of the printing powder, and then a dataset is constructed. Each sample in the dataset includes the process parameters, the temperature field under the corresponding process parameters output by the 3D thermal field model, and the microstructure morphology under the corresponding process parameters generated by the cellular automata simulation. A microstructure regulation model was constructed based on a BP neural network model. The temperature field under the corresponding process parameters output by the 3D thermal field model was used as the input layer of the microstructure regulation model, and the microstructure morphology under the corresponding process parameters was generated by cellular automata simulation as the output layer of the microstructure regulation model. The microstructure regulation model was trained, and each sample in the training set was stored as historical data in the historical database of the microstructure regulation model. The input to the trained microstructure control model is the process parameters and the thermal field behavior monitoring data under the corresponding process parameters, and the output is the microstructure morphology prediction under the corresponding process parameters. During the additive manufacturing of the target part, a trained microstructure control model is used to monitor the microstructure morphology of each slice layer, and the process parameters of each slice layer are intelligently controlled based on historical data in the historical database to ensure that the microstructure morphology of each slice layer matches the microstructure morphology selected based on the performance requirements of the target part.

2. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, In the process of additive manufacturing of the target part, ensuring that the microstructure morphology of each slice layer matches the microstructure morphology selected based on the performance requirements of the target part is achieved in the following ways: During the additive manufacturing of the target part, real-time monitoring data of thermal field behavior and corresponding process parameters are input into a trained microstructure control model to monitor the microstructure morphology of the target part in the current slice layer in real time. The model evaluates whether the microstructure morphology of the current slice layer output by the microstructure control model matches the microstructure morphology selected based on the performance requirements of the target part. If the evaluation results show that the microstructure morphology of the current slice layer output by the microstructure control model does not match the microstructure morphology selected based on the performance requirements of the target part, then suitable process parameters are intelligently matched from the historical database based on the microstructure morphology selected based on the performance requirements of the target part and input into the control program of the additive manufacturing forming equipment. This guides the additive manufacturing forming equipment to form the next slice layer according to the matched process parameters until the microstructure morphology of the current slice layer output by the microstructure control model matches the microstructure morphology selected based on the performance requirements of the target part.

3. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, The process parameters include laser power, laser beam shape, spot size, laser scanning speed, powder layer thickness, and scanning spacing.

4. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, The thermal behavior monitoring data includes molten pool depth, molten pool width, temperature gradient, cooling rate, and subcooling. The thermal behavior monitoring data is acquired using an infrared thermal imager, a high-speed visible spectrum camera, a high-speed color camera, and / or a photoelectric detector.

5. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, The microstructure morphology includes microstructure classification and its distribution; the microstructure classification is based on the influence of grain size and mechanical properties of the part, including equiaxed crystals and columnar crystals; the distribution of the microstructure classification includes the proportion of equiaxed crystals, the proportion of columnar crystals, and the average grain size of equiaxed crystals.

6. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, When optimizing the 3D thermal field model, the obtained thermal field behavior monitoring data is continuously compared with the output results of the 3D thermal field model. Then, the setting parameters of the 3D thermal field model are changed to continuously optimize the 3D thermal field model until the output results of the 3D thermal field model match the obtained thermal field behavior monitoring data.

7. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, The physical fields obtained are related to the material properties of the printing powder, including powder flow velocity and protective gas flow rate.

8. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 1, characterized in that, When the target part has multiple regions corresponding to different microstructures, a microstructure transition region is set between two adjacent regions corresponding to different microstructures. In the process of additive manufacturing of target parts, when printing the slices corresponding to the transition regions of microstructures, the laser energy density changes in a gradient with the increase of the corresponding slices, so as to achieve a natural transition between adjacent regions with different microstructure morphologies.

9. The method for precise control of 3D printing microstructure and properties through thermal field behavior monitoring feedback as described in claim 8, characterized in that, Arbitrary slices corresponding to the transition region of the printed microstructure Corresponding laser energy density satisfy: ， ； In the above formula: , This indicates the laser energy density used when printing two adjacent different microstructures; This indicates the total number of slice layers corresponding to the transition region of the microstructure; This represents the slice layer count index corresponding to the transition region of the microstructure. It represents the energy regulation index.

10. A computer program, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The computer program executes the method for precise control of 3D printed microstructures and properties through thermal field behavior monitoring feedback as described in any one of claims 1 to 9.

Images