Design method and preparation method of functionally graded materials based on molten pool flow field control

Abstract

This invention discloses a design method for the preparation process of functionally graded materials (FJTs) based on molten pool flow field control, and a preparation method for FJTs. The design method includes the following steps: designing the composition of the FJT; designing and conducting orthogonal experiments based on the FJT to obtain the molten pool dilution rate and composition gradient under different process parameters; collecting data, establishing a database, constructing an initial model, and determining quantitative relationships; combining the actual microstructure and performance requirements of the target FJT to infer the composition gradient of the FJT that meets the actual microstructure and performance requirements, and then giving the corresponding molten pool dilution rate and process parameter range based on the initial model; preparing additive manufacturing samples of the FJT; characterizing the composition gradient, microstructure, and mechanical properties, and feeding the data back to the initial model for continuous calibration and correction to form a corrected model; and obtaining the optimal process parameters for the target FJT based on the corrected model.

Description

Technical Field

[0001] This invention belongs to the field of functionally graded materials design and preparation technology, specifically relating to a design method for a functionally graded materials preparation process based on molten pool flow field control, and a preparation method for functionally graded materials including the design method. Background Technology

[0002] Functionally graded materials (FJCTs), with their gradient composition and structure, can achieve a gradient distribution of material properties and functions. This offers unique advantages for achieving synergistic enhancement of multiple functions, strengthening interfacial reliability, and alleviating thermal stress. Therefore, they have significant application needs and broad development prospects in many core fields such as aerospace, biomedicine, energy and nuclear engineering, high-end manufacturing, and electronic packaging. Thus, developing efficient and high-quality fabrication technologies for FJCTs is currently one of the core research directions.

[0003] Traditional techniques for preparing functionally graded materials include vapor deposition, thermal spraying, powder metallurgy, additive manufacturing, centrifugal casting, and self-propagating high-temperature synthesis. Among these, additive manufacturing, also known as 3D printing, is a novel technology based on the discrete-stacking principle. It combines energy sources such as lasers, plasmas, and electron beams with three-dimensional path planning technology to fabricate components by layer-by-layer material accumulation. It offers advantages such as high control precision, a small heat-affected zone, and good flexibility in processing complex structures, making it highly promising for the preparation of functionally graded materials.

[0004] The current basic approach to preparing functionally graded materials (FJTs) mainly involves altering the mixing ratio, deposition rate, or arrangement of different materials to achieve heterogeneous materials with gradual changes in composition, microstructure, and structure. This approach either requires the preparation and formulation of multiple raw materials or demands high-performance equipment with multi-channel delivery capabilities, relying heavily on high-performance equipment platforms. There is still room for optimization in terms of operability and economics. Furthermore, the difficulty in establishing a quantitative correlation between compositional and microstructure gradients and processing parameters leads to significant time and manpower costs in the experimental trial-and-error phase for different FJTs, severely limiting the R&D efficiency and engineering applications of FJTs.

[0005] The above background information is disclosed only to assist in understanding the inventive concept and technical solution of this invention, and does not necessarily belong to the prior art of this invention. In the absence of clear evidence that the above information was disclosed before the filing date of this invention, the above background information should not be used to evaluate the novelty and inventiveness of this invention. Summary of the Invention

[0006] In view of this, the present invention provides a design method for the preparation process of functionally graded materials based on the control of molten pool flow field, so as to overcome the defects of the prior art.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A design method for the fabrication process of functionally graded materials based on molten pool flow field regulation includes the following steps:

[0009] Design the composition of functionally graded materials;

[0010] Based on the composition design of functionally graded materials, orthogonal experiments were conducted to obtain the melt pool dilution rate and composition gradient under different preparation process parameters;

[0011] We collected different preparation process parameters and corresponding melt pool dilution rate and composition gradient data, established a database, and used machine learning methods to build an initial model to determine the relationship between preparation process parameters, melt pool dilution rate and composition gradient.

[0012] Based on the actual tissue performance requirements of the target functional graded material, the composition gradient of the functional graded material that meets the actual tissue performance requirements is deduced from the relationship, and then the corresponding melt pool dilution rate and preparation process parameter range are given based on the initial model.

[0013] Additive manufacturing samples of functionally graded materials were prepared based on the range of preparation process parameters given by the initial model.

[0014] The compositional gradient, microstructure and mechanical properties of the prepared additive manufacturing samples were characterized, and the data were fed back into the initial model. The initial model was continuously calibrated and corrected, and finally a corrected model was formed.

[0015] The optimal fabrication process parameters for the target functionally graded material are obtained based on the modified model.

[0016] Preferably, the melt pool dilution rate of the orthogonal experiment corresponds to the melt pool dilution rate of each layer, that is, the orthogonal experiment is used to obtain the melt pool dilution rate of each layer of functionally graded materials under different preparation process parameters.

[0017] According to some preferred embodiments of the present invention, the composition of the design functionally graded material is:

[0018] The functionally graded material comprises a substrate and a surface material, with the gradient transition layer formed by the substrate and surface material in different mass proportions; or...

[0019] The functionally graded material comprises three materials: a substrate, a surface material, and a third transition component. The gradient transition layer is formed by the third transition component, the substrate, and the surface material in different mass percentage gradients.

[0020] That is, when the functionally graded material includes two materials, one of which is used as the substrate and the other as the surface material, the intermediate gradient transition layer is formed by the substrate and the surface material; when the functionally graded material includes three materials, one of which is used as the substrate and the other as the surface material, the intermediate gradient transition layer is formed by the third transition component and the substrate, or by the third transition component and the surface material, or by the third transition component, the substrate, and the surface material together. In some embodiments, preferably, the transition begins with the third transition component and the substrate, with the proportion of the substrate gradually decreasing and the proportion of the third transition component gradually increasing. After reaching a certain level, the surface material is introduced to continue the transition until the proportion of the surface material is 100%.

[0021] According to some preferred embodiments of the present invention, the process parameters include one or more of the following: heat source parameters, scanning speed, feeding speed, overlap rate, and defocusing amount. For example, when the heat source is a laser, the heat source parameter is power; when the heat source is an electric arc or an electron beam, the heat source parameter is current.

[0022] According to some preferred embodiments of the present invention, the molten pool dilution rate is calculated by measuring the cross-sectional area of ​​the molten pool after characterizing the molten pool morphology using an optical microscope or a scanning electron microscope.

[0023] According to some preferred embodiments of the present invention, the molten pool dilution rate is calculated according to the following formula:

[0024]

[0025] In the formula, S1 and S2 are the cross-sectional areas of the molten pool on the substrate side and the deposition layer side of the interface, respectively. The substrate side is the side closer to the substrate, and the deposition layer side is the side farther from the substrate.

[0026] According to some preferred embodiments of the invention, the composition gradient is obtained by measuring along the deposition direction using an energy dispersive spectrometer or an electron probe.

[0027] According to some preferred embodiments of the present invention, the machine learning method is selected from decision trees, support vector machines, K-nearest neighbors, random forests, artificial neural networks, and adaptive reinforcement models.

[0028] According to some preferred embodiments of the present invention, the heat source for the preparation of the functionally graded material is selected from laser, electric arc, electron beam or ion beam.

[0029] According to some preferred embodiments of the present invention, during the preparation of the functionally graded material, the greater the required melt pool dilution rate, the greater the proportion of the deposited component integrated into the matrix, and the smaller the component gradient. Here, the matrix refers to the matrix component corresponding to the lower layer during deposition. The matrix composition may change accordingly with each deposition. The deposited component is formed by a third transition component, the substrate, and / or the surface material. When the top layer of the functionally graded material is deposited, the deposited component is 100% surface material.

[0030] Specifically, when preparing the functionally graded material, if a melt pool dilution rate of 60%-90% is required, the preparation process parameters are adjusted to allow a higher proportion of additive manufacturing raw materials to be melted into the substrate, thereby obtaining a smaller compositional gradient; if a melt pool dilution rate of 30%-60% is required, the preparation process parameters are adjusted to ensure that the proportion of additive manufacturing raw materials and the substrate in the melt pool is approximately equal, resulting in a moderate compositional gradient; if a melt pool dilution rate of <30% is required, the preparation process parameters are adjusted to allow a lower proportion of additive manufacturing raw materials to be melted into the substrate, thereby obtaining a larger compositional gradient.

[0031] In some embodiments of the present invention, the design method for the fabrication process of functionally graded materials based on molten pool flow field regulation specifically includes the following steps:

[0032] Step 1: Select appropriate heat sources and additive manufacturing equipment based on the physical properties of the materials (melting point, thermal conductivity, reflectivity, etc.) and the requirements of functionally graded materials, and design the composition of functionally graded materials.

[0033] Step 2: Design the composition based on functionally graded materials and conduct orthogonal experiments under different preparation process parameters. Characterize the molten pool morphology using optical microscopy or scanning electron microscopy, measure the cross-sectional area of ​​the molten pool, calculate the molten pool dilution rate, and measure the composition gradient along the deposition direction using energy dispersive spectroscopy or electron probe microanalysis.

[0034] Step 3: Collect different preparation process parameters and their corresponding output results (melt pool dilution rate and composition gradient) and establish a database. Use machine learning methods to construct an initial model of "process parameters-melt pool dilution rate-composition gradient" and determine the relationship between several factors such as "process parameters-melt pool dilution rate-composition gradient".

[0035] Step 4: Based on the actual microstructure and performance requirements of functionally graded materials, the composition gradient that meets the conditions is deduced from the relationships obtained above. Then, based on the initial model, a melt pool dilution rate scheme and the corresponding processing window (preferred preparation process parameter range) are given. Based on the model output results, additive manufacturing samples of functionally graded materials are prepared.

[0036] Step 5: Characterize the composition gradient, microstructure and mechanical properties of the prepared additive manufacturing sample, conduct a comprehensive evaluation of the functionally graded material, and feed the data back to the initial model. Continuously calibrate and correct the initial model to finally form the corrected model.

[0037] Step 6: Obtain the optimal process parameters for the target functionally graded material based on the modified model. The target functionally graded material can then be prepared based on the modified model.

[0038] The present invention also provides a method for preparing functionally graded materials, the method comprising the design method of functionally graded material preparation process based on molten pool flow field control as described above.

[0039] Due to the application of the above technical solutions, the present invention has the following advantages compared with the prior art. The design method of functional graded material preparation process based on melt pool flow field control of the present invention establishes the relationship between "process parameters - melt pool dilution rate - composition gradient", realizes the ability to back-calculate the melt pool dilution rate through the target composition gradient, and then back-calculate the preparation process parameters through the melt pool dilution rate, and establishes a database and model to improve accuracy and simplify the process. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the functionally graded materials in an embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram of the molten pool dilution rate measurement according to an embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram illustrating the relationship between different melt pool dilution rates and compositional gradients in embodiments of the present invention;

[0044] Figure 4 This is a schematic diagram of the model construction and optimization process in an embodiment of the present invention;

[0045] Figure 5 These are macroscopic morphology images of the functionally graded materials prepared in Example 2 of this invention;

[0046] Figure 6 This is a graph showing the performance characterization results of the first layer during the preparation of the functionally graded material in Embodiment 2 of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0048] This invention provides a design method for the preparation process of functionally graded materials based on the control of the molten pool flow field. By understanding the relationship between the molten pool dilution behavior, the composition gradient, and the preparation process parameters, the molten pool dilution rate is derived from the target composition gradient, and then the preparation process parameters are derived from the molten pool dilution rate. This achieves high-efficiency, high-precision, and low-cost control of the composition gradient, thereby improving the heavy dependence of existing preparation technologies on raw material preparation or equipment platforms.

[0049] This invention primarily utilizes a layer-by-layer dilution process in the molten pool to create a compositional gradient. Therefore, the selection of heat source and material form is relatively flexible. Heat sources can include lasers, electric arcs, electron beams, or ion beams, and the additive manufacturing raw material can be in the form of powder or filaments. For example... Figure 1 As shown, the design method of the present invention can be used to prepare a compositional gradient between two materials, or a third transitional component can be introduced to combine with the two materials to form a gradient transition layer.

[0050] like Figure 3 As shown, different compositional gradient distributions can be obtained by employing different melt pool dilution ratio strategies, and the melt pool dilution ratio can be different for each layer. The substrate is 100% A, and the melt pool dilution ratio for the first layer is... η 1. Then the composition of A in the first layer is η 1×A; the dilution rate of the molten pool in the second layer is... η 2. Then the composition of A in the second layer is η 1× η 2×A, and so on, until the nth layer, the molten pool dilution rate of the nth layer is... η n Then the composition of A in the nth layer is η 1× η 2×…… η n ×A. When only two materials, substrate A and surface material B, exist, the proportion of surface material B can be derived from the proportion of substrate A. The top layer of the entire functionally graded material is 100% surface material B. When a third transitional component C also exists, the proportions of C and B can be calculated based on the matrix composition using the same principle.

[0051] The present invention provides a design method for a high-efficiency, high-precision, and low-cost functional graded material preparation process based on molten pool flow field control. This method has low requirements for raw material preparation and equipment platforms, and specifically includes the following steps:

[0052] Step 1: Select appropriate heat sources and additive manufacturing equipment based on the physical properties of the materials (melting point, thermal conductivity, reflectivity, etc.) and the requirements of functionally graded materials, and design the composition of functionally graded materials.

[0053] The composition of the designed functionally graded material is as follows:

[0054] Functionally graded materials (FJTs) consist of two materials, one used as the substrate and the other as the surface material. The gradient transition layer is formed by the substrate and the surface material in different mass ratios; or,

[0055] Functionally graded materials consist of three materials: a substrate, a surface material, and a third transitional component. The gradient transition layer is formed by the third transitional component, the substrate, and / or the surface material in different mass proportions.

[0056] That is, when the functionally graded material includes two materials, one of which is used as the substrate and the other as the surface material, the intermediate gradient transition layer is formed by the substrate and the surface material; when the functionally graded material includes three materials, one of which is used as the substrate and the other as the surface material, the intermediate gradient transition layer is formed by the third transition component and the substrate, or by the third transition component and the surface material, or by the third transition component, the substrate, and the surface material together. In some embodiments, preferably, the transition begins with the third transition component and the substrate, with the proportion of the substrate gradually decreasing and the proportion of the third transition component gradually increasing. After reaching a certain level, the surface material is introduced to continue the transition until the proportion of the surface material is 100%.

[0057] The heat source for preparing functionally graded materials (FGMs) is selected from lasers, electric arcs, electron beams, or ion beams. The energy input of the selected heat source must meet the melting conditions of the material to be additively manufactured. For materials with high thermal conductivity and reflectivity (such as aluminum and copper), the energy absorption rate of the heat source must be considered. To obtain a smaller compositional gradient, heat sources with good coherence and small molten pools and heat-affected zones, such as lasers, are preferred. Regarding the selection of additive manufacturing equipment, atmospheric protection is required for reactive and easily oxidized metals; otherwise, ordinary additive manufacturing equipment can be used.

[0058] For additive manufacturing, both powders and filaments are suitable for material selection. Generally, materials with higher melting points are chosen as the base material, and materials with lower melting points are chosen as the surface material.

[0059] Step 2: Design the composition of functionally graded materials and conduct orthogonal experiments under different preparation process parameters. Characterize the molten pool morphology using optical microscopy or scanning electron microscopy, measure the cross-sectional area of ​​the molten pool and calculate the molten pool dilution rate, and use energy dispersive spectroscopy or electron probe microanalysis to measure the compositional gradient of the functionally graded materials along the deposition direction.

[0060] Orthogonal experiments are used to obtain the melt pool dilution rate of each layer of functionally graded materials under different preparation process parameters, so as to more accurately control the composition gradient and corresponding preparation process parameters of functionally graded materials in the future, and obtain the target functionally graded materials that meet the requirements.

[0061] Process parameters include one or more of the following: heat source parameters, scanning speed, feeding speed, overlap rate, and defocusing amount. For example, when the heat source is a laser, the heat source parameter is power; when the heat source is an electric arc or electron beam, the heat source parameter is current.

[0062] The molten pool dilution rate is measured as follows: The molten pool is cut along the overlap direction using an electrical discharge wire cutter. The molten pool morphology is characterized using an optical microscope or scanning electron microscope to obtain the cross-sectional area of ​​the molten pool. The molten pool dilution rate is calculated as the percentage of the matrix alloy in the fusion zone relative to the total cross-sectional area of ​​the molten pool. Specifically, it is calculated using the following formula:

[0063]

[0064] In the formula, S1 and S2 are the cross-sectional areas of the molten pool on the matrix side and the deposition layer side at the interface, respectively. That is, the molten pool is divided into two parts by the interface, with S1 being the cross-sectional area of ​​the molten pool closer to the matrix and S2 being the cross-sectional area of ​​the molten pool farther from the matrix. The molten pool dilution rate is calculated using the area ratio method, which is more representative and accurate. Figure 2 As shown.

[0065] For orthogonal experiments in additive manufacturing, a multi-factor, multi-level orthogonal table is designed by combining process parameters such as heat source parameters, scanning speed, feeding speed, overlap rate, and decoking amount. The cross-section of the molten pool is taken by the wire EDM machine along the direction perpendicular to the heat source movement, and the molten pool morphology is observed after sanding, polishing, and etching.

[0066] Composition gradient is determined by taking samples at intervals along the deposition direction using an energy dispersive spectrometer or electron probe. In areas with large composition gradients, the sampling interval can be appropriately reduced to form a composition gradient curve.

[0067] In the preparation of functionally graded materials (FGMs), the higher the required melt pool dilution rate, the greater the proportion of additive manufacturing materials incorporated into the substrate, and the smaller the compositional gradient. Specifically, when a melt pool dilution rate of 60%-90% is required, adjusting process parameters allows a higher proportion of additive manufacturing materials to be melted into the substrate, resulting in a smaller compositional gradient, which is more suitable for situations without abrupt changes in performance gradient. When a melt pool dilution rate of 30%-60% is required, adjusting process parameters ensures a relatively even ratio of additive manufacturing materials to the substrate in the melt pool, resulting in a moderate compositional gradient. When a melt pool dilution rate of <30% is required, adjusting process parameters allows a lower proportion of additive manufacturing materials to be melted into the substrate, resulting in a larger compositional gradient, suitable for traversing hazardous component ranges (such as hard and brittle phases, low-melting-point phases, and abrupt changes in performance).

[0068] More specifically, the range of laser powder feeding process parameters is as follows:

[0069] High dilution rate (60%-90%): Laser power 2200-2500W, scanning speed 5-6mm / s, powder feeding speed 18-22g / min, defocusing amount -1 to -2mm;

[0070] Medium dilution rate (30%-60%): use laser power of 1800-2200W, scanning speed of 6-8mm / s, powder feeding speed of 16-20g / min, and defocusing amount of -2 to -3mm;

[0071] Low dilution rate (<30%): use laser power of 1500-1800W, scanning speed of 8-10mm / s, powder feeding speed of 12-16g / min, and defocusing amount of -3 to -4mm.

[0072] The range of laser wire feeding process parameters is as follows:

[0073] High dilution rate (60%-90%): Laser power 2000-2200W, scanning speed 5-6mm / s, wire feed speed 6-8mm / s, defocusing amount -1 to -2mm;

[0074] Medium dilution rate (30%-60%): use laser power of 1600-2000W, scanning speed of 6-7mm / s, wire feed speed of 4-6mm / s, and defocusing amount of -2 to -3mm;

[0075] Low dilution rate (<30%): use laser power of 1400-1600W, scanning speed of 7-9mm / s, wire feed speed of 2-4mm / s, and defocusing amount of -3 to -4mm.

[0076] The range of parameters for the electric arc powder feeding process is as follows:

[0077] High dilution rate (60%-90%): Use current of 150-180A, scanning speed of 6-8mm / s, and powder feeding speed of 18-22g / min;

[0078] Medium dilution rate (30%-60%): use current of 130-150A, scanning speed of 8-12mm / s, and powder feeding speed of 14-18g / min;

[0079] Low dilution rate (<30%): Use a current of 110-130A, a scanning speed of 12-14mm / s, and a powder feeding speed of 10-14g / min.

[0080] The range of arc wire feeding process parameters is as follows:

[0081] High dilution rate (60%-90%): Use current of 150-180A, scanning speed of 8-10mm / s, and wire feed speed of 5-6mm / s;

[0082] Medium dilution rate (30%-60%): Use current of 130-150A, scanning speed of 10-14mm / s, and wire feed speed of 3-5mm / s;

[0083] Low dilution rate (<30%): Use current of 100-130A, scanning speed of 14-16mm / s, and wire feed speed of 2-3mm / s.

[0084] When preparing functionally graded materials using different heat sources, the process parameters can be adjusted within the range mentioned above.

[0085] Step 3: Collect process parameters and corresponding output results (melt pool dilution rate and composition gradient) and establish a database. Use machine learning methods to construct an initial model of "process parameters-melt pool dilution rate-composition gradient" and determine the relationship between several factors such as "process parameters-melt pool dilution rate-composition gradient".

[0086] The machine learning method is selected from one of the following: decision tree, support vector machine, k-nearest neighbors, random forest, artificial neural network, and adaptive reinforcement model.

[0087] The main process of building a machine learning model is as follows: Figure 4 As shown, the optimal machine learning algorithm was determined and a quantitative relationship between "process parameters - melt pool dilution rate - composition gradient" was constructed through steps such as data acquisition and preprocessing, feature selection and dimensionality reduction, design of different machine learning models, model training and loss function design, regularization and hyperparameter tuning, model validation and iterative optimization, and construction of quantitative relationships between multiple parameters.

[0088] Step 4: Based on the actual microstructure and performance requirements of functionally graded materials, the composition gradient that meets the conditions is deduced, and then the melt pool dilution rate scheme and the corresponding processing window (preferred process parameter range) are given based on the initial model. Additive manufacturing samples of functionally graded materials are prepared according to the model output results.

[0089] Step 5: Characterize the composition gradient, microstructure and mechanical properties of the prepared additive manufacturing sample, conduct a comprehensive evaluation of the functionally graded material, and feed the data back to the initial model. Continuously calibrate and correct the initial model to finally form the corrected model.

[0090] Step 6: Obtain the optimal process parameters for the target functionally graded material based on the modified model. The target functionally graded material can then be prepared based on the modified model.

[0091] Step four is used to provide a design scheme, and step five is used to verify the scheme and improve the accuracy of the machine learning model.

[0092] Composition gradient characterization methods can employ energy dispersive spectroscopy (EDS) or electron probe microanalysis (EPMA) to sample at intervals along the deposition direction. In areas with large composition gradients, the sampling interval can be appropriately reduced. Microstructure characterization mainly includes optical microscopy (OM) and scanning electron microscopy (SEM) to analyze tissue characteristics and phase types, and to summarize the distribution patterns of any microscopic defects. Mechanical property testing mainly involves microhardness and tensile strength. Depending on the specific application requirements of functionally graded materials, friction and wear tests, corrosion tests, etc., can be added.

[0093] The technical solution of the present invention will be further described below with reference to three specific embodiments:

[0094] Example 1: This example uses Inconel 625 nickel-based alloy and 316 stainless steel as examples to design a process for preparing functionally graded materials, specifically including the following steps:

[0095] Step 1: Select Inconel 625 nickel-based alloy sheet as the base material and 316 stainless steel as the surface material, without involving any third transition component. 316 stainless steel is drawn into wires with a diameter of 0.8-1.2 mm and a surface roughness Ra ≤ 1.6 μm.

[0096] Step 2: Functionally graded materials are additively manufactured using an electric arc wire feeding process. Orthogonal experiments are designed and conducted under different process parameters. The cross-sectional area of ​​the molten pool of one or more layers is measured and the molten pool dilution rate is calculated. The compositional gradient of the functionally graded materials is characterized.

[0097] Step 3: Establish a database and use machine learning methods to build an initial model of "process parameters-melt pool dilution rate-composition gradient" to determine the relationship between several factors such as "process parameters-melt pool dilution rate-composition gradient" and realize the model-guided design of melt pool dilution rate scheme.

[0098] Step 4: Based on the actual microstructure and performance requirements of functionally graded materials, the composition gradient that meets the conditions is deduced, and then the melt pool dilution rate scheme and the corresponding processing window (preferred process parameter range) are given based on the initial model. Additive manufacturing samples of functionally graded materials are prepared according to the model output results.

[0099] The optimal melt pool dilution rate scheme given by the model is as follows: 20%-30% dilution rate for the first layer, and 50%-60% dilution rate for subsequent layers. The optimal process window is: 120-130A current, 14-15mm / s scanning speed, and 2-3mm / s wire feed speed for the first layer. For subsequent layers, maintain 140-150A current, 10-12mm / s scanning speed, and 3-4mm / s wire feed speed.

[0100] Step 5: Characterize the composition gradient, microstructure and mechanical properties of a series of additive manufacturing samples from Step 4, comprehensively evaluate the functionally graded materials, and feed the data back to the initial model. Continuously calibrate and correct the initial model to finally form the corrected model.

[0101] Step 6: Obtain the optimal process parameters for the target functionally graded material based on the modified model, prepare the target functionally graded material according to the optimal process parameters, and test the target functionally graded material.

[0102] The optimal dilution ratio combination determined in this embodiment is: 25% dilution ratio for the first layer, and 60% dilution ratio for subsequent layers. The optimal process parameters are: 125A current, 14mm / s scanning speed, and 2mm / s wire feed speed for the first layer; and 145A current, 10mm / s scanning speed, and 3mm / s wire feed speed for subsequent layers.

[0103] The microstructure characterization results of the target functionally graded material (FJT) show that a compositional gradient of Ni, Cr, Nb, and Mo is formed within the FJT. This gradient is gradual and without abrupt changes, and the microstructure and precipitates exhibit a clear gradient transition. Furthermore, no large-sized pores or cracks were found, indicating good forming quality. Performance testing shows that the Vickers hardness of the gradient layer reaches 180–230 HV (average 206 HV), and the matrix-side hardness matches that of the base material (Inconel 625 nickel-based alloy has a hardness of approximately 220–240 HV), forming a smooth gradient transition. The yield strength of the FJT along the additive manufacturing direction is 525 MPa, comparable to that of Inconel 625 nickel-based alloy (approximately 540–580 MPa, vacuum-cast) and 316 stainless steel (approximately 480–580 MPa, additively manufactured).

[0104] Example 2: This example uses 304F stainless steel and Norem02 alloy as an example to design a fabrication process for functionally graded materials, specifically including the following steps:

[0105] Step 1: 304F stainless steel sheet is selected as the base material, and Norem02 alloy is used as the additive manufacturing surface material. Simultaneously, a 1:1 mixture of 304F steel and Norem02 alloy is used as the initial gradient transition component. Both the Norem02 alloy and the gradient transition component powder are prepared using an atomization powdering process at an argon pressure of 0.75 MPa, producing 80-270 mesh spherical powders with a particle size of 53-177 μm and an oxygen content ≤0.03% to avoid oxidation inclusions.

[0106] Step 2: Functionally graded materials are additively manufactured using a ring laser powder feeding process. Orthogonal experiments are designed and conducted under different process parameters such as laser power, scanning speed, powder feeding speed, and defocusing amount. The cross-sectional area of ​​the molten pool is measured and the molten pool dilution rate is calculated to characterize the compositional gradient of the functionally graded materials.

[0107] Step 3: Establish a database and use machine learning methods to construct a numerical model of "process parameters-melt pool dilution rate-composition gradient" to determine the relationship between several factors such as "process parameters-melt pool dilution rate-composition gradient" and realize the model-guided design of melt pool dilution rate scheme.

[0108] Step 4: Based on the actual microstructure and performance requirements of functionally graded materials, the composition gradient that meets the conditions is deduced, and then the melt pool dilution rate scheme and the corresponding processing window (preferred process parameter range) are given based on the initial model. Additive manufacturing samples of functionally graded materials are prepared according to the model output results.

[0109] The optimal pool dilution ratio scheme given by the model is as follows: 50%-60% dilution ratio for the first layer, and 30%-40% dilution ratio for subsequent layers. The optimized process window is as follows: for the first layer, use a laser power of 2000-2200W, a scanning speed of 6-7mm / s, a powder feeding speed of 18-20g / min, and a defocusing amount of -2 to -3mm. For subsequent layers, maintain a laser power of 1800-2000W, a scanning speed of 7-8mm / s, a powder feeding speed of 16-18g / min, and a defocusing amount of -2 to -3mm.

[0110] Step 5: Characterize the composition gradient, microstructure and mechanical properties of a series of additive manufacturing samples from Step 4, comprehensively evaluate the functionally graded materials, and feed the data back to the initial model. Continuously calibrate and correct the initial model to finally form the corrected model.

[0111] Step Six: Obtain the optimal process parameters for the target functionally graded material based on the modified model, and prepare the target functionally graded material according to the optimal process parameters, such as... Figure 5 As shown, the target functionally graded material is tested.

[0112] The optimal dilution ratio combination determined in this embodiment is: 55% dilution ratio for the first layer, and 35% dilution ratio for subsequent layers. The optimal process parameters are: 2000W laser power, 6mm / s scanning speed, 18g / min powder feed rate, and -2mm defocusing amount for the first layer. Subsequent layers maintain a laser power of 1900W, a scanning speed of 7mm / s, a powder feed rate of 17g / min, and -3mm defocusing amount.

[0113] The microstructure characterization results show that the functionally graded material (FJT) forms a compositional gradient of C, Cr, and Mo elements. This gradient is gradual and without abrupt changes, and the microstructure and precipitates exhibit a clear gradient pattern. Furthermore, no large-sized pores or cracks were found, indicating good molding quality. Performance testing shows that the Vickers hardness of the gradient layer exhibits good gradual characteristics, ranging from 380 to 450 HV. XRD measurements indicate a residual stress of approximately 320 MPa, effectively buffering the differences in mechanical properties between dissimilar materials and reducing the risk of cracking. In this embodiment, the mechanical property data corresponding to the melt pool dilution rates of the first layer at 15%, 35%, and 55% are as follows: Figure 6 As shown. Figure 6 This indicates that the functionally graded material prepared with a 55% dilution rate in the first layer has a shear strength of up to 460 MPa with the substrate, demonstrating good bonding strength between dissimilar materials.

[0114] Example 3: This example uses 617B nickel-based alloy and G115 heat-resistant steel as examples to design a process for preparing functionally graded materials, specifically including the following steps:

[0115] Step 1: 617B nickel-based alloy sheet is selected as the base material, G115 heat-resistant steel is used as the additive manufacturing surface material, and FeNiCr medium-entropy alloy is used as the gradient transition composition. Both G115 heat-resistant steel and FeNiCr medium-entropy alloy are drawn into wires with a diameter of 0.8-1.2 mm and a surface roughness Ra≤1.6 μm.

[0116] Step 2: Functionally graded materials are additively manufactured using a ring laser wire feeding process. Orthogonal experiments are designed and conducted under different process parameters such as laser power, scanning speed, wire feeding speed, and defocusing amount. The cross-sectional area of ​​the molten pool is measured and the molten pool dilution rate is calculated to characterize the compositional gradient of the functionally graded materials.

[0117] Step 3: Establish a database and use machine learning methods to construct a numerical model of "process parameters-melt pool dilution rate-composition gradient" to determine the relationship between several factors such as "process parameters-melt pool dilution rate-composition gradient" and realize the model-guided design of melt pool dilution rate scheme.

[0118] Step 4: Based on the actual microstructure and performance requirements of functionally graded materials, the composition gradient that meets the conditions is deduced, and then the melt pool dilution rate scheme and the corresponding processing window (preferred process parameter range) are given based on the initial model. Additive manufacturing samples of functionally graded materials are prepared according to the model output results.

[0119] The optimal molten pool dilution ratio scheme given by the model is as follows: 40%-50% dilution ratio for FeNiCr medium-entropy alloy and 20%-30% dilution ratio for G115 heat-resistant steel. For FeNiCr medium-entropy alloy, the laser power is 1800-2000W, scanning speed is 6-7mm / s, wire feed speed is 5-6mm / s, and defocusing amount is -2 to -3mm. For G115 heat-resistant steel, the laser power is 1500-1600W, scanning speed is 7-8mm / s, wire feed speed is 3-4mm / s, and defocusing amount is -3 to -4mm.

[0120] Step 5: Characterize the composition gradient, microstructure and mechanical properties of a series of additive manufacturing samples from Step 4, comprehensively evaluate the functionally graded materials, and feed the data back to the initial model. Continuously calibrate and correct the initial model to finally form the corrected model.

[0121] Step 6: Obtain the optimal process parameters for the target functionally graded material based on the modified model, prepare the target functionally graded material according to the optimal process parameters, and test the target functionally graded material.

[0122] The optimal dilution ratio combination determined in this embodiment is as follows: 45% dilution for FeNiCr medium-entropy alloy and 30% dilution for G115 heat-resistant steel. The optimal process parameters are: 1900W laser power, 6mm / s scanning speed, 5mm / s wire feed speed, and -2mm defocusing amount for FeNiCr medium-entropy alloy; and 1600W laser power, 8mm / s scanning speed, 4mm / s wire feed speed, and -3mm defocusing amount for G115 heat-resistant steel.

[0123] Microstructural characterization results show that the functionally graded material (FJD) exhibits a compositional gradient of Ni, Cr, Mo, W, and Co elements. This gradient is gradual and without abrupt changes, and the microstructure and precipitates show a clear gradient pattern. Furthermore, no large-sized pores or cracks were found, indicating good molding quality. Performance testing shows that the Vickers hardness of the gradient layer exhibits good gradual characteristics, ranging from 200 to 240 HV. XRD measurements indicate a residual stress of approximately 360 MPa, effectively mitigating the differences in mechanical properties between dissimilar materials and residual stress. The yield strength of the FJD along the additive manufacturing direction is 584 MPa, comparable to that of 617B nickel-based alloy (approximately 480-520 MPa, vacuum-cast) and G115 heat-resistant steel (approximately 620-650 MPa, additively manufactured).

[0124] This invention relates to a functionally graded material (FJT) additive manufacturing technology based on molten pool dilution rate control, aiming to improve the operability and economy of FJT preparation. The main principle is based on the relationship between "process parameters, molten pool dilution rate, and compositional gradient." The molten pool dilution rate is derived from the target compositional gradient, and then the preparation process parameters are derived from the molten pool dilution rate. A database and model are established to improve accuracy and simplify the process. The specific implementation process is as follows: Heat sources and additive manufacturing raw materials are selected according to material properties and construction conditions. Orthogonal experiments are used to understand the influence of process parameters on molten pool dilution rate and compositional gradient. A database is established and a machine learning model is constructed. Based on the model, dilution rate combination schemes and processing windows are provided, and sample preparation is completed. The model and process are continuously calibrated and corrected based on the microstructure performance evaluation results, ultimately forming an optimized FJT additive manufacturing technology. This technology allows for flexible adjustment of the heat source type according to actual needs and equipment conditions, eliminates the need for mixing and blending multiple powders, and is compatible with powder feeding and wire feeding processes, significantly reducing the preparation cost of FJTs. This invention presents a functionally graded material additive manufacturing technology based on molten pool dilution rate control. It eliminates the need to separately powder two matrix materials and then mix them in a specific ratio; instead, it only requires preparing one material into powder or filament, significantly reducing manufacturing costs. Existing functionally graded material equipment demands high precision in multi-channel raw material transport and powder control. In contrast, this invention controls the compositional gradient based on layer-by-layer dilution in the molten pool. The manufacturing precision is primarily related to the size and control of the heat source. Since heat sources such as lasers have high parallelism, the control difficulty is significantly reduced compared to powder processing. Furthermore, the type of heat source can be flexibly adjusted according to actual needs and equipment conditions, achieving high-quality functionally graded material preparation while maintaining cost-effectiveness. Moreover, this invention belongs to additive manufacturing technology, enabling three-dimensional layer-by-layer deposition with computer programming assistance, resulting in high manufacturing efficiency and processing precision. Therefore, this technology can be widely applied to the preparation of functionally graded materials in different systems and under different operating conditions, and can also be used to prepare gradient transition layers for bonding dissimilar metal materials.

[0125] The above embodiments are only for illustrating the technical concept and features of the present invention. Their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be used to limit the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

[0126] The endpoints and any values ​​of the ranges disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed herein.

Claims

1. A design method of a functional gradient material preparation process based on molten pool flow field regulation, characterized in that, The method comprises the following steps: designing the composition of the functionally graded material; designing and performing an orthogonal experiment based on the composition of the functionally graded material to obtain the corresponding melt pool dilution rate and composition gradient under different preparation process parameters; the orthogonal experiment is used to obtain the melt pool dilution rate of each layer of the functionally graded material under different preparation process parameters and the composition gradient for characterizing the functionally graded material; collecting different preparation process parameters and the corresponding melt pool dilution rate and composition gradient data, establishing a database, and constructing an initial model of the preparation process parameter-melt pool dilution rate-composition gradient based on the database by using a machine learning method to determine the relationship between the preparation process parameter, the melt pool dilution rate and the composition gradient; the greater the melt pool dilution rate required during preparation of the functionally graded material, the greater the proportion of the deposited composition incorporated into the matrix, and the smaller the composition gradient; the process parameters include one or more of the heat source parameter, the scanning speed, the feeding speed, the overlap rate and the defocusing amount; the melt pool dilution rate is calculated after the melt pool cross-sectional area is measured according to the characterization of the melt pool morphology by an optical microscope or a scanning electron microscope; the composition gradient is measured by interval sampling along the deposition direction by using an energy spectrometer or an electron probe; combined with the actual organizational performance requirements of the target functionally graded material, the composition gradient of the functionally graded material meeting the actual organizational performance requirements is inversely deduced based on the relationship, and then the melt pool dilution rate corresponding to the composition gradient of the target functionally graded material and the corresponding preparation process parameter range are given based on the initial model; that is, the melt pool dilution rate is inversely deduced through the composition gradient of the target functionally graded material, and then the preparation process parameter is inversely deduced through the melt pool dilution rate; an additive manufacturing sample of the functionally graded material is prepared based on the preparation process parameter range given by the initial model; the prepared additive manufacturing sample is characterized in terms of the composition gradient, the microstructure and the mechanical performance, and the data is fed back to the initial model for continuous calibration and correction of the initial model, and finally a corrected correction model is formed; the microstructure characterization includes optical microscopy and scanning electron microscopy to analyze the organizational characteristics and phase types; the mechanical performance test is a microhardness and tensile strength test; the optimal preparation process parameter corresponding to the target functionally graded material is obtained based on the corrected model.

2. The design method of claim 1, wherein The composition of the functionally graded material comprises two materials, i.e., a base material and a surface material, and a gradient transition layer is formed by the base material and the surface material in different mass fraction gradients.

3. The method of claim 1, wherein, The composition of the functionally graded material comprises three materials, i.e., a base material, a surface material and a third transition component, and a gradient transition layer is formed by the third transition component, the base material and / or the surface material in different mass fraction gradients.

4. The method of claim 1, wherein, The melt pool dilution rate is calculated according to the following formula: ； In the formula, S1 and S2 are the cross-sectional areas of the melt pool on the interface base material side and the deposition layer side, respectively.

5. The method of claim 1, wherein, The machine learning method is one selected from the group consisting of a decision tree, a support vector machine, K-nearest neighbor, random forest, artificial neural network and adaptive boosting model.

6. The method of claim 1, wherein, The heat source during preparation of the functionally graded material is one selected from the group consisting of a laser, an electric arc, an electron beam and an ion beam.

7. A method of producing a functionally graded material, characterized by, The preparation method comprises a design method of the functional gradient material preparation process based on the molten pool flow field regulation according to any one of claims 1-6.

Images