Sintering parameter determination method, electronic device, and storage medium

Abstract

This disclosure relates to a method for determining sintering parameters, an electronic device, and a storage medium, belonging to the field of process parameter optimization technology. The method includes: acquiring current sintering process parameters; determining, based on the current sintering process parameters and a sintering furnace model, the furnace environmental field data in the time domain for the spatial region within the sintering furnace; determining, based on the position of the target green billet in the sintering furnace, the environmental temperature time-history data and environmental pressure time-history data of the target green billet from the furnace environmental field data; determining the product performance prediction data of the target green billet based on the environmental temperature time-history data and environmental pressure time-history data; if the product performance prediction data does not meet the performance index requirements, correcting the current sintering process parameters, and re-executing the above steps based on the corrected sintering process parameters until the product performance prediction data meets the performance index requirements; and outputting the sintering process parameters that meet the performance index requirements. This method can shorten the sintering process development cycle.

Description

Technical Field

[0001] This disclosure relates to the field of process parameter optimization technology, and more specifically, to a method for determining sintering parameters, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Metal injection molding (MIM) is a near-net-shape forming technology that mixes and granulates metal powder with a binder and then injects it into the final product. It is widely used in manufacturing small metal parts with complex structures and high performance requirements. Its core process—sintering—is a crucial step that determines the dimensional accuracy, density, mechanical properties, and microstructure of the final product.

[0003] Currently, the determination of MIM sintering process parameters mainly relies on a trial-and-error development model: technicians repeatedly test-fire, measure, and adjust process parameters until the dimensional accuracy and performance of the sintered product meet the requirements. This model is highly dependent on engineering experience, the development cycle usually lasts for several months, product yield fluctuates greatly, R&D costs are high, and it is difficult to predict and control internal defects such as porosity, stress concentration, and abnormal grain growth, which seriously restricts the rapid development and large-scale production of new products. Summary of the Invention

[0004] One objective of this disclosure is to provide a technical solution for optimizing sintering parameters, thereby addressing the technical problems in related technologies, such as low development efficiency, high cost, and unstable quality due to reliance on trial-and-error methods, and the inability to balance prediction accuracy and industrial-grade computing efficiency due to the excessive computational load of traditional fully coupled simulation methods.

[0005] According to a first aspect of this disclosure, a method for determining sintering parameters is provided, comprising: Obtain the current sintering process parameters; Based on the current sintering process parameters and the sintering furnace model, the furnace environment field data in the time domain of the spatial region inside the sintering furnace is determined; wherein, the furnace environment field data includes temperature field data and pressure field data; Based on the position of the target green billet in the sintering furnace, the time history data of the ambient temperature and the time history data of the ambient pressure of the target green billet are determined from the furnace environmental field data. Based on the time history data of ambient temperature and ambient air pressure of the target green body, the product performance prediction data corresponding to the target green body is determined. If the product performance prediction data does not meet the performance index requirements, the current sintering process parameters are corrected to obtain corrected sintering process parameters. The above steps are then repeated based on the corrected sintering process parameters until the product performance prediction data meets the performance index requirements. Output the sintering process parameters that meet the performance requirements.

[0006] Optionally, determining the product performance prediction data corresponding to the target green compact based on the environmental temperature time history data and environmental pressure time history data includes: The ambient temperature time history data and the ambient air pressure time history data are used as the environmental driving force, and the thermal response prediction data of the target green body in the sintering process is determined according to the material state change model of the green body material in the sintering process. The material stress time history data of the target green body is determined based on the thermal response prediction data, the material temperature time history data of the target green body is determined based on the ambient temperature time history data, and the microscopic performance prediction data and macroscopic performance prediction data of the target green body are determined based on the material temperature time history data and the material stress time history data. The microscopic performance prediction data and the macroscopic performance prediction data are used as the product performance prediction data for the target green blank.

[0007] Optionally, the current sintering process parameters include heat source parameters and protective gas flow parameters. The step of determining the furnace environment field data in the time domain for the spatial region within the sintering furnace based on the current sintering process parameters and the sintering furnace model includes: Construct a sintering furnace model; wherein, the sintering furnace model includes a sintering furnace body structure model and multiple green billet structure models corresponding to each green billet; Based on the sintering furnace model, the heat source parameters of the sintering furnace, and the protective gas flow parameters, the furnace environment field data in the time domain of the spatial region inside the sintering furnace are determined.

[0008] Optionally, the position of the target green billet in the sintering furnace is determined by the following steps: The position of the target green billet model corresponding to the target green billet is registered with the spatial coordinate system of the sintering furnace model to obtain the position of the target green billet model in the furnace environment field data; The position of the target green body model in the furnace environment field data is taken as the position of the target green body in the sintering furnace.

[0009] Optionally, the material state change model is used to describe the dynamic evolution of the internal density, porosity and stress of the green material during the sintering process with temperature-pressure history.

[0010] Optionally, the thermodynamic response prediction data includes predicted stress time history data, predicted strain time history data, predicted final shrinkage deformation data, predicted density distribution data, and predicted displacement field distribution data of the target green blank throughout the sintering process.

[0011] Optionally, determining the predicted microstructure properties of the target green body and the predicted product performance of the target green body based on the material temperature time history data and the material stress time history data includes: The material temperature time history data and the material stress time history data are combined to form a joint load history. The combined load history is input into the performance prediction model to obtain the microscopic performance prediction data and the macroscopic performance prediction data.

[0012] Optionally, the performance prediction model is constructed through the following steps: The microscopic and macroscopic performance test data of the sample green body and the sample material test data corresponding to multiple sets of sintering process parameters are obtained to obtain a dataset; wherein, the sample material test data includes sample material temperature time history data and sample material stress time history data; The sample material temperature time history data and sample material stress time history data of each sample in the dataset are combined to form the sample joint load history, thus obtaining the training sample set; The performance prediction model is trained by using the joint load history of the samples in the training sample set as the input feature and the micro performance detection data and macro performance detection data in the training sample set as the output feature.

[0013] According to a second aspect of this disclosure, an electronic device is also provided, including a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method as described in the first aspect of this disclosure.

[0014] According to a third aspect of the present disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, the computer program implementing the method according to the first aspect of the present disclosure when executed by a processor.

[0015] This embodiment first constructs environmental field data (including temperature and pressure fields) of the sintering furnace. Then, it determines the environmental temperature and pressure time histories of the target green billet from this environmental field. Next, it determines the predicted product performance data of the target green billet under the corresponding process conditions. Finally, if the predicted product performance data does not meet the performance requirements, it corrects the process parameters and re-executes the above steps until sintering process parameters that meet the performance requirements are obtained. This enables virtual prediction of the product performance of the target green billet under different sintering process parameters on a computer, and automatically iteratively optimizes the sintering process parameters based on the prediction results. This avoids repeated trial firings in the actual sintering furnace to adjust process parameters, significantly reducing the number of trial firings, shortening the process development cycle, lowering R&D costs, and improving process development efficiency.

[0016] Other features and advantages of the embodiments of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the disclosure.

[0018] Figure 1 This is a schematic diagram of the structure of an electronic device to which the methods of the embodiments of this disclosure can be applied.

[0019] Figure 2 This is a flowchart illustrating a method for determining sintering parameters provided in an embodiment of this disclosure. Detailed Implementation

[0020] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0021] The following description of at least one exemplary embodiment is merely illustrative and is not intended to limit the scope of this disclosure or its application or use.

[0022] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0023] In all the 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.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0025] It should be noted that all data acquisition actions in this disclosure were carried out in compliance with the relevant data protection laws and policies of the country where the data is located, and with the authorization of the relevant equipment owner.

[0026] This disclosure relates to a method for determining sintering parameters. Figure 1 This is a schematic diagram of the structure of an electronic device to which the methods of the embodiments of this disclosure can be applied.

[0027] Electronic devices are essentially computers, which can be manifested as high-performance tower workstations, rack servers, or small computing clusters composed of interconnected servers.

[0028] like Figure 1 As shown, the electronic device 100 may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, etc. The processor 1100 may be a central processing unit (CPU), a microprocessor (MCU), etc. The memory 1200 may include, for example, ROM (Read-Only Memory), RAM (Random Access Memory), or non-volatile memory such as a hard disk. The interface device 1300 may include, for example, a USB interface, a headphone jack, etc. The communication device 1400 may be capable of wired or wireless communication. The display device 1500 may be, for example, a liquid crystal display (LCD), a touch screen, etc. The input device 1600 may include, for example, a touch screen, a keyboard, etc. Users can input / output voice information through the speaker 1700 and the microphone 1800.

[0029] exist Figure 1 In the electronic device shown, a typical implementation method for simulating the sintering process involves fully coupled, high-fidelity modeling of tens of thousands of green billets within the sintering furnace. A single solver simultaneously addresses multi-physics coupling issues such as furnace protective gas flow, thermal radiation and convection, green billet shrinkage and deformation at high temperatures, gravity and friction effects, and microscopic grain growth and phase transformation. Due to the massive scale of the model, the extremely high number of degrees of freedom, and the complex physical coupling, even on high-performance computing clusters, it is difficult to complete the calculation within an industrially acceptable timeframe, making the simulation method difficult to effectively implement in actual production. Furthermore, excessive simplification of the model to improve computational speed fails to accurately predict key performance indicators such as deformation, density distribution, and microstructure. In addition, related technologies often focus on single-scale physical field analysis, lacking a complete chain from process parameter input to product performance output and parameter optimization feedback.

[0030] To address these issues, this embodiment solves the problems as follows: First, environmental field data (including temperature and pressure fields) of the sintering furnace is constructed. Then, the environmental temperature and pressure time histories of the target green billet are determined from this environmental field. Next, product performance prediction data for the target green billet under corresponding process conditions is determined. Finally, if the product performance prediction data does not meet the performance requirements, the process parameters are corrected and the above steps are repeated until sintering process parameters that meet the performance requirements are obtained. This enables virtual prediction of the product performance of the target green billet under different sintering process parameters on a computer, and automatic iterative optimization of the sintering process parameters based on the prediction results. This avoids repeated trial firings in the actual sintering furnace to adjust process parameters, significantly reducing the number of trial firings, shortening the process development cycle, lowering R&D costs, and improving process development efficiency.

[0031] <First Embodiment> This embodiment provides a method for determining sintering parameters. Figure 2 This is a schematic flowchart illustrating a method for determining sintering parameters provided in an embodiment of this disclosure. Figure 2 As shown, the method may include the following steps S210 to S260.

[0032] Step S210: Obtain the current sintering process parameters.

[0033] In this embodiment, the current sintering process parameters can refer to the set of process parameters used to control the sintering process of metal powder green bodies. Correspondingly, obtaining the current sintering process parameters can be referred to as obtaining the current sintering process parameters used for sintering metal powder green bodies. These current sintering process parameters may include heat source parameters and protective gas flow parameters.

[0034] The heat source parameters may include heating power, temperature profile, etc. The heating power may be, for example, 5 kW to 50 kW, and the temperature profile may include the temperature-time relationship of the heating stage, the holding stage, and the cooling stage.

[0035] Protective gas flow parameters can include gas flow rate, gas type, and gas pressure. The gas flow rate can be, for example, from 1 cubic meter per hour to 10 cubic meters per hour, the gas type can be argon or nitrogen, and the gas pressure can be atmospheric pressure or negative pressure.

[0036] The current sintering process parameters can be obtained by the user through the input interface, read from historical sintering records, or by using the system's preset default values.

[0037] For example, a user can input the heating power as 20 kW, the temperature curve as heating from room temperature to 1350 degrees Celsius and holding for 2 hours, the protective gas as argon, and the gas flow rate as 5 cubic meters per hour through a graphical user interface.

[0038] Step S220: Based on the current sintering process parameters and the sintering furnace model, determine the furnace environment field data in the time domain for the spatial region inside the sintering furnace; wherein, the furnace environment field data includes temperature field data and pressure field data.

[0039] In this embodiment, the sintering furnace model refers to the computational model obtained by digitally modeling the sintering furnace. This sintering furnace model can include a furnace body structure model and multiple green billet structure models corresponding to each green billet. The furnace body structure model can include the geometric structure and material properties of components such as graphite heating tubes, ceramic plates, insulation layers, condenser tubes, protective gas inlets, protective gas outlets, and feeding supports. The green billet structure model can be a simplified representation of the green billets inside the sintering furnace; for example, tens of thousands of green billets can be simplified into equivalent blocks or representative units of equal volume, focusing only on the spatial distribution of each green billet within the furnace and ignoring its fine geometric structure.

[0040] Furnace environmental field data can include temperature field data and pressure field data. Temperature field data represents the temperature evolution over time at various spatial locations within the furnace. Pressure field data represents the gas pressure evolution over time at various spatial locations within the furnace. The time domain can refer to the entire time period from the start to the end of sintering.

[0041] The determination of furnace environmental field data can be achieved through computational fluid dynamics (CFD) simulation. Specifically, fluid dynamics governing equations can be established based on the sintering furnace model. These governing equations can include mass conservation equations, momentum conservation equations, and energy conservation equations. Then, the current sintering process parameters can be input as boundary and initial conditions into these governing equations. The governing equations can then be solved numerically to obtain the time-domain temperature and pressure field data of the spatial region within the sintering furnace.

[0042] For example, a model of the sintering furnace is established using computational fluid dynamics simulation software. The following process parameters are set in the model: heating tube power is 20 kW, protective gas flow rate is 5 cubic meters per hour, initial furnace temperature is room temperature, and the protective atmosphere is argon with its standard physical properties. Based on this, the forced and natural convection of the protective gas within the furnace, the thermal radiation between the heating tubes and the green body and the furnace wall, the interaction between gas flow and pressure field, and the heat dissipation effect of the furnace wall are simulated to obtain environmental field data (including temperature and pressure fields) within the sintering furnace.

[0043] In some embodiments, the current sintering process parameters include heat source parameters and protective gas flow parameters. Step S220, based on the current sintering process parameters and the sintering furnace model, determines the furnace environment field data in the time domain of the spatial region within the sintering furnace, which may include the following steps S2211 to S2212.

[0044] Step S2211: Construct a sintering furnace model.

[0045] In this embodiment, constructing the sintering furnace model may include constructing a sintering furnace body structure model and a green body structure model. Constructing the sintering furnace body structure model may include establishing a three-dimensional geometric model based on the actual structure of the sintering furnace, including components such as graphite heating tubes, ceramic plates, insulation layers, condenser tubes, protective gas inlets, protective gas outlets, and feeding supports, and setting corresponding material properties (such as density, specific heat capacity, thermal conductivity, emissivity, etc.) for each component.

[0046] Constructing a green structure model can include simplifying multiple green billets in the sintering furnace into equivalent blocks or representative units of equal volume, and setting their spatial distribution positions in the furnace.

[0047] For example, a three-dimensional geometric model of the sintering furnace structure can be created using computer-aided design software. The material properties of the graphite heating tubes can be set to the standard physical properties of graphite, and the material properties of the ceramic plates can be set to the standard physical properties of alumina ceramics. For the 10,000 green pieces in the sintering furnace, they can be simplified into equivalent blocks of equal volume, ignoring their fine geometric structure and focusing only on the spatial distribution of each equivalent block within the furnace. This significantly reduces the complexity of mesh generation and computational resource consumption while ensuring the accuracy of the global temperature and pressure field simulation within the furnace.

[0048] Step S2212: Based on the sintering furnace model, the heat source parameters of the sintering furnace, and the protective gas flow parameters, determine the furnace environment field data in the time domain for the spatial region inside the sintering furnace.

[0049] In this embodiment, fluid dynamics governing equations can be established based on a sintering furnace model. These governing equations can include mass conservation equations, momentum conservation equations, and energy conservation equations describing the flow of the protective gas. Then, heat source parameters can be input as heat source terms into the energy conservation equation, and protective gas flow parameters can be input as boundary conditions into the mass and momentum conservation equations. Numerical solutions, such as the finite volume method, are then used to solve these governing equations, thereby obtaining the time-domain temperature and pressure field data of the spatial region within the sintering furnace. In these examples, fluid dynamics simulation methods can accurately simulate the flow characteristics and temperature distribution of the protective gas within the furnace, providing accurate boundary conditions for subsequent local simulation of the green body.

[0050] Step S230: Based on the position of the target green billet in the sintering furnace, determine the time history data of the ambient temperature and the time history data of the ambient pressure of the target green billet from the furnace environmental field data.

[0051] In this embodiment, the target green billet can refer to a specific green billet that requires detailed simulation analysis. The target green billet can be any one or a group of green billets in the sintering furnace. For example, a green billet located at the center of the furnace, a green billet located at the edge of the furnace, or a green billet with a representative geometric structure can be selected as the target green billet.

[0052] The target green body can be any metal green body manufactured using metal injection molding and for which there are specific requirements for its dimensions and properties after sintering. The target green body can also be a target metal powder green body.

[0053] The target green blank can be made of materials such as stainless steel (e.g., 316L, 17-4PH), titanium alloy (e.g., Ti-6Al-4V), iron-based alloy, nickel-based alloy, copper-based alloy, hard alloy, and magnetic materials.

[0054] The position of the target green billet in the sintering furnace can refer to the three-dimensional coordinate position of the target green billet in the spatial coordinate system of the sintering furnace.

[0055] Ambient temperature time history data refers to a data sequence showing the temperature change over time at the spatial location of the target green body. This data sequence records the temperature value at each moment from the start to the end of sintering. Ambient gas pressure time history data refers to a data sequence showing the gas pressure change over time at the spatial location of the target green body. This data sequence records the gas pressure value at each moment from the start to the end of sintering.

[0056] The determination of the ambient temperature and ambient pressure time history data can be achieved through data extraction and interpolation mapping methods. Specifically, the position of the target green billet in the sintering furnace can be determined first, and then the temperature and pressure field data corresponding to that position can be extracted from the furnace environmental field data obtained in step S220. Since the furnace environmental field data is usually calculated on grid nodes, while the position of the target green billet may not be completely located on the grid nodes, interpolation methods, such as linear interpolation or cubic spline interpolation, can be used to interpolate the temperature and pressure data on the grid nodes to the position of the target green billet, thereby obtaining the ambient temperature and ambient pressure time history data.

[0057] For example, assume the target green billet is located at the center of the sintering furnace, with coordinates as follows: Temperature and pressure data of the grid nodes around the location can be extracted from the furnace environmental field data, and then the coordinates can be calculated using trilinear interpolation. The changes in temperature and air pressure over time are used to obtain time-history data of ambient temperature and ambient air pressure.

[0058] In some embodiments, the position of the green billet in the sintering furnace during step S230 is determined by the following steps S2311-S2312: Step S2311: Register the position of the target green billet model corresponding to the target green billet with the spatial coordinate system of the sintering furnace model to obtain the position of the target green billet model in the furnace environmental field data.

[0059] In this embodiment, the target green blank model can refer to a three-dimensional model obtained by performing detailed geometric modeling of the target green blank. The target green blank model can include the fine geometric structure of the target green blank, such as the outer contour of the green blank and the distribution of internal pores.

[0060] It should be noted that the target green billet model is different from the corresponding green billet structure model. The former is a three-dimensional model obtained by performing detailed geometric modeling on a single target green billet, including its fine geometric structure (such as external contour, internal pore distribution, etc.), used for accurate stress and strain solutions in local thermo-mechanical coupling simulations. The latter simplifies multiple green billets in the sintering furnace into equivalent blocks or representative elements of equal volume, and sets their spatial distribution positions in the furnace, used in global thermo-fluid field simulations of the furnace to reduce mesh complexity and computational scale. The two differ in modeling detail, geometric representation, and simulation applications.

[0061] The local coordinate system of the target green billet model is aligned with the global coordinate system of the sintering furnace model, ensuring that the position of the target green billet model within the sintering furnace model matches its actual position within the furnace during the sintering process. Registration methods can include geometric transformations such as translation, rotation, and scaling. Through registration, the position of the target green billet model within the global coordinate system of the sintering furnace model can be obtained, thereby determining its position within the furnace environmental field data.

[0062] Step S2312: The position of the target green billet model in the furnace environment field data is taken as the position of the target green billet in the sintering furnace.

[0063] In this embodiment, since the target green body model is a digital representation of the target green body, the position of the target green body model in the furnace environmental field data can be directly used as the position of the target green body in the sintering furnace. In this way, a spatial correspondence between the target green body model and the furnace environmental field data can be established, providing a positional basis for subsequently extracting the environmental temperature time history data and environmental pressure time history data of the target green body from the furnace environmental field data.

[0064] Step S240: Determine the product performance prediction data corresponding to the target green compact based on the time history data of ambient temperature and ambient air pressure of the target green compact.

[0065] In this embodiment, product performance prediction data refers to data obtained by predicting the performance of the product obtained after sintering the target green billet. This product performance prediction data may include microscopic performance prediction data and macroscopic performance prediction data.

[0066] In the example where the target green body is a target metal powder green body, the product performance prediction data can also be called metal product performance prediction data.

[0067] The microscopic performance prediction data can include predicted data such as grain size, porosity, and phase volume fraction. The grain size can be, for example, from 1 micrometer to 100 micrometers, the porosity can be, for example, from 0% to 10%, and the phase volume fraction can represent the volume proportion of different crystalline phases in the material.

[0068] Macroscopic performance prediction data can include predicted data for properties such as yield strength, elongation, and magnetic properties.

[0069] The yield strength can be, for example, 200 MPa to 1000 MPa, the elongation can be, for example, 1% to 50%, and the magnetic properties include magnetic domain density, adsorption force, etc.

[0070] In some embodiments, step S240, based on the time-history data of ambient temperature and ambient pressure of the target green compact, determines the product performance prediction data corresponding to the target green compact, including the following steps S2411 to S2413: Step S2411: Using the time history data of ambient temperature and ambient pressure as the environmental driving force, and based on the material state change model of the target green body material during sintering, determine the predicted data of the thermal response of the target green body during sintering.

[0071] In this embodiment, the environmental driving force can refer to the temperature and pressure loads acting on the outside of the target green body. This environmental driving force can be achieved by applying environmental temperature time history data and environmental pressure time history data as boundary conditions to the target green body model.

[0072] A material state change model can be a mathematical model that describes the dynamic evolution of the internal density, porosity, and stress of a target green body during sintering as a function of temperature and pressure. This material state change model can be established based on the thermodynamic properties of the material, the kinetics of sintering densification, and the mechanical constitutive relations.

[0073] Thermodynamic response prediction data refers to the structural mechanics and densification response results of the target green body under the action of environmental driving forces.

[0074] In some embodiments, the thermal response prediction data may include predicted stress time history data, predicted strain time history data, predicted final shrinkage deformation data, predicted density distribution data, and predicted displacement field distribution data of the target green compact throughout the sintering process.

[0075] The predicted stress time history data represents the cumulative evolution of stress over time at various points within the target green body caused by the combined effects of thermal and sintering stresses. The predicted strain time history data covers thermal expansion strain, elastic strain, creep strain, and viscoplastic shrinkage strain caused by sintering densification. The predicted final shrinkage deformation data represents the final shrinkage rate and warpage of the target green body in a specific direction. The predicted density distribution data characterizes the differences in relative density and densification uniformity within different regions of the target green body after sintering. The predicted displacement field distribution data represents the spatial displacement vectors of each node within the target green body, used to evaluate the dimensional accuracy and overall deformation mode of the final product.

[0076] In one embodiment, the determination of thermal response prediction data can be achieved through simulation using the finite element method (FEM).

[0077] Specifically, firstly, the target green blank model corresponding to the target green blank is obtained, and then the mesh is discretized based on the target green blank model to generate the finite element mesh model of the target green blank.

[0078] Then, material properties are assigned to the finite element mesh model and the material state change model is embedded to obtain the fused model. Specifically, this includes: setting the initial density and initial porosity distribution of the target green body, setting the high-temperature thermophysical parameters and high-temperature mechanical parameters of the green body material, and embedding the material state change model into the finite element solver in the form of a user-defined material constitutive model. This allows the finite element model to calculate the viscoplastic shrinkage strain increment driven by sintering densification in each increment step based on the current temperature and stress state, and to update the relative density and porosity of the elements synchronously.

[0079] Next, the ambient temperature time history data and ambient air pressure time history data are used as thermo-mechanical boundary conditions and applied to the corresponding boundaries of the fused model. The nonlinear transient finite element method is used to solve the problem step by step in the entire time domain of sintering heating, holding and cooling. The iteration is performed in each incremental step until the convergence criterion is met, thereby obtaining the stress tensor time history, strain tensor time history, displacement vector time history and element relative density time history of each node of the target green blank in the whole sintering process.

[0080] In this model, ambient temperature time-history data is applied to the outer surface through convection and thermal radiation. Ambient air pressure time-history data is applied as a pressure load to the outer surface. Simultaneously, a gravity load is applied to the model, and a contact constraint relationship is established between the target green body and the firing plate. A contact friction coefficient is set to simulate the tangential friction behavior between them.

[0081] Finally, the overall shrinkage rate and warpage of the target green blank are extracted from the displacement vector time histories of each node as data to predict the final shrinkage deformation. Predicted density distribution data are generated from the relative density time histories of each element, and the stress tensor time histories, strain tensor time histories, and displacement vector time histories of each node are used as predicted stress time histories, predicted strain time histories, and predicted displacement field distribution data, respectively. All of these data together constitute the thermodynamic response prediction data.

[0082] Step S2412: Determine the material stress time history data of the target green billet based on the thermal response prediction data, determine the material temperature time history data of the target green billet based on the ambient temperature time history data, and determine the microscopic performance prediction data and macroscopic performance prediction data of the target green billet based on the material temperature time history data and the material stress time history data.

[0083] In this embodiment, the material temperature time history data can refer to the data sequence of temperature changes of the material inside the target green body over time. This material temperature time history data can be obtained by interpolating the ambient temperature time history data and transmitting it to each node and element of the target green body model through a data mapping interface, thereby obtaining the temperature response at the corresponding location inside the target green body, and thus obtaining the material temperature time history data. The material temperature time history data can be the temperature response of a representative node inside the target green body, or the average value of the temperature responses of multiple feature points inside the target green body; no limitation is made here.

[0084] Material stress time history data can refer to the data sequence of stress changes over time within the material of a target green body. This material stress time history data can be extracted from the predicted stress time history data in the thermodynamic response prediction data. For example, the predicted stress time history data of a representative location within the target green body can be extracted, or the average predicted stress time history data of multiple locations within the target green body can be extracted.

[0085] The difference between material temperature time history data and material stress time history data and ambient temperature time history data and ambient pressure time history data is as follows: Material temperature time history data and material stress time history data are response data of the internal material of the target green body, reflecting the internal temperature and stress state of the target green body during sintering. Ambient temperature time history data and ambient pressure time history data are environmental input data from outside the target green body, reflecting the external temperature and pressure loads acting on the target green body.

[0086] Microscopic performance prediction data can refer to the predicted data on the microstructure characteristics of the target green body after sintering. This microscopic performance prediction data can include grain size, porosity, phase volume fraction, etc.

[0087] Macroscopic performance prediction data can refer to the predicted data of the macroscopic properties of the product after the target green body is sintered. This macroscopic performance prediction data can include yield strength, elongation, magnetic properties, etc.

[0088] The determination of micro-level and macro-level performance prediction data can be achieved through performance prediction models. These performance prediction models can be machine learning models built based on data-driven methods, such as Artificial Neural Networks (ANNs) and Gaussian Process Regression (GPRs), or constitutive models built based on physical mechanisms; no limitation is made here.

[0089] Since the methods for determining micro-performance prediction data and macro-performance prediction data will be explained in detail later, they will not be elaborated here.

[0090] In some embodiments, step S2412, based on material temperature time history data and material stress time history data, determining the predicted microstructure properties and macrostructure properties of the target green billet may include the following steps S2412.1 to S2412.2: Step S2412.1: Combine the material temperature time history data and the material stress time history data to form a joint load history.

[0091] In this embodiment, the combined load history can refer to multidimensional time series data formed by combining material temperature time history data and material stress time history data. This combined load history can reflect the combined effects of temperature load and stress load simultaneously experienced by the internal material of the target green compact during sintering.

[0092] The construction of the joint load history can include aligning the material temperature time history data and the material stress time history data according to time to form a two-dimensional or multi-dimensional time series data.

[0093] Step S2412.2: Input the joint load history into the performance prediction model to obtain micro-performance prediction data and macro-performance prediction data.

[0094] In this embodiment, the performance prediction model can refer to a mathematical model that establishes a mapping relationship between the joint load history and the microscopic and macroscopic performance prediction data. This performance prediction model can be a machine learning model based on data-driven methods, such as artificial neural networks, Gaussian process regression, or support vector machines; it can also be a constitutive model based on physical mechanisms, such as the Johnson-Mehl-Avrami-Kolmogorov (JMAK) grain growth model or the porosity evolution internal variable model. No limitation is made here. By inputting the joint load history into the performance prediction model, microscopic and macroscopic performance prediction data can be obtained.

[0095] For example, a trained artificial neural network model can be used as a performance prediction model. The input of this artificial neural network model is the joint load history, and the output is microscopic performance prediction data such as grain size of 10 micrometers, porosity of 2%, yield strength of 600 MPa, and elongation of 20%, as well as macroscopic performance prediction data.

[0096] In some embodiments, the performance prediction model can be constructed through the following steps S110 to S130: Step S110: Obtain the microscopic performance test data and macroscopic performance test data of the sample green blank under multiple sets of sintering process parameters, as well as the sample material test data, to obtain the dataset.

[0097] In this embodiment, to construct the performance prediction model, multiple sets of experimental data from the sample green blank are required. The sample green blank is made of the same material as the target green blank. This experimental data can be obtained by conducting sintering experiments on the sample green blank under different sintering process parameters, performing microscopic and macroscopic performance tests on the sintered sample green blank, and testing the sample green blank during the sintering process. Specifically, microscopic performance data can be obtained using techniques such as metallographic microscopy, scanning electron microscopy (SEM), and electron backscatter diffraction (EBSD). This microscopic performance data may include grain size, porosity, and phase volume fraction. Macroscopic performance data can be obtained through mechanical property testing and magnetic property testing. This product performance test data may include yield strength, elongation, and magnetic properties.

[0098] The sample material test data may include test data on the internal materials of the sample green blank during the sintering process. This sample material test data can be obtained by measuring by embedding thermocouples and stress sensors in the sample green blank, or by simulation methods similar to steps S2411 to S2412.

[0099] The sample material test data includes material temperature time history data and material stress time history data.

[0100] Step S120: Combine the sample material temperature time history data and sample material stress time history data of each sample in the dataset to form the sample joint load history, and obtain the training sample set.

[0101] In this embodiment, the microscopic performance test data and macroscopic performance test data of the sample green blank under a sintering process parameter, as well as the sample material test data, are used as a sample.

[0102] The method for constructing the sample joint load history is similar to that in step S2412.1. The sample material temperature time history data and sample material stress time history data can be aligned according to time to form a two-dimensional or multi-dimensional time series data.

[0103] Each training sample in the training sample set includes microscopic performance test data, macroscopic performance test data, and sample joint load history.

[0104] Step S130: The joint load history of the samples in the training sample set is used as the input feature of the performance prediction model, and the micro performance detection data and macro performance detection data in the training sample set are used as the output features to train the performance prediction model.

[0105] In this embodiment, machine learning methods can be used to train the performance prediction model. Specifically, multiple sets of samples with joint load histories can be used as input features of the training sample set, and the corresponding microscopic and macroscopic performance detection data can be used as output features of the training sample set. Then, a suitable machine learning algorithm, such as artificial neural networks, Gaussian process regression, or support vector machines, can be selected to construct the performance prediction model. Next, the performance prediction model can be trained using the training sample set, and the model parameters can be adjusted using optimization algorithms (such as gradient descent) to minimize the error between the predicted output and the actual output of the performance prediction model. After training, the performance prediction model can be obtained. In these examples, through orthogonal experimental design and experimental data, an accurate performance prediction model can be constructed, enabling rapid prediction from process parameters to product performance.

[0106] For example, orthogonal experiments can be designed, selecting different sintering process parameters such as sintering temperature, holding time, and protective gas flow rate to conduct sintering experiments. Metallographic, SEM, and EBSD analyses are performed on the sintered samples to obtain microscopic performance data such as grain size and porosity. Mechanical and magnetic property tests are then performed on the sintered samples to obtain macroscopic performance data such as yield strength, elongation, and magnetic properties. Simultaneously, temperature and stress time-history data of the green sample during the sintering process are obtained through simulation or measurement. These data are combined to form a training sample set, and an artificial neural network algorithm is used to train a performance prediction model.

[0107] Step S2413: Use the microscopic performance prediction data and the macroscopic performance prediction data as the product performance prediction data of the target green blank.

[0108] Step S250: If the product performance prediction data does not meet the performance index requirements, correct the current sintering process parameters to obtain the corrected sintering process parameters, and re-execute the above steps based on the corrected sintering process parameters until the product performance prediction data meets the performance index requirements.

[0109] In this embodiment, the performance index requirements refer to the target requirements for the performance of the sintered product. These performance index requirements can be set by the user according to the actual application needs of the product, and are not limited here. For example, performance index requirements can be set such as a yield strength of not less than 500 MPa, an elongation of not less than 15%, and a porosity of not more than 3%.

[0110] The performance requirements can be set by the user through an input interface, read from the product specification sheet, or by using industry standard default values.

[0111] Correcting the current sintering process parameters refers to adjusting the values ​​of these parameters to improve product performance. Correction methods can include manual adjustment and automatic optimization algorithm adjustments. Manual adjustment can refer to the user manually adjusting the sintering process parameters based on experience or simulation results.

[0112] Automatic optimization algorithm adjustment can refer to using optimization algorithms, such as genetic algorithms, particle swarm optimization algorithms, and Bayesian optimization algorithms, to automatically search for the optimal sintering process parameters.

[0113] The iterative loop logic can refer to the following: if the product performance prediction data does not meet the performance requirements, the current sintering process parameters are corrected to obtain corrected sintering process parameters. Then, based on the corrected sintering process parameters, steps S210 to S240 are re-executed to obtain new product performance prediction data. If the new product performance prediction data still does not meet the performance requirements, the sintering process parameters are corrected again, and steps S210 to S240 are re-executed. This cycle continues until the product performance prediction data meets the performance requirements.

[0114] Convergence conditions can refer to the product performance prediction data meeting the performance index requirements, or the number of iterations reaching the preset maximum number of iterations.

[0115] For example, suppose the performance requirements are a yield strength of not less than 600 MPa and an elongation of not less than 20%. In the first simulation, the predicted product performance data is a yield strength of 550 MPa and an elongation of 18%, which does not meet the performance requirements. At this time, a genetic algorithm can be used to correct the current sintering process parameters, for example, increasing the sintering temperature from 1350 degrees Celsius to 1400 degrees Celsius and extending the holding time from 2 hours to 2.5 hours. Then, based on the corrected sintering process parameters, steps S210 to S240 are re-executed to obtain new predicted product performance data of a yield strength of 620 MPa and an elongation of 22%, which meets the performance requirements, and the iteration ends.

[0116] Step S260: Output the sintering process parameters that meet the performance requirements.

[0117] In this embodiment, when the product performance prediction data meets the performance index requirements, the corresponding sintering process parameters can be output as the target sintering process parameters. These target sintering process parameters can serve as guidance for the actual sintering production process parameters.

[0118] The output can include various parameter values ​​of the target sintering process, such as heating power, temperature profile, protective gas flow rate, gas type, and gas pressure.

[0119] The output can take the form of a generated report document, stored in a database, or displayed through a visual interface. For example, a sintering process parameter report can be generated, which contains detailed information on the target sintering process parameters, such as a heating power of 25 kW, a temperature profile of heating from room temperature to 1400 degrees Celsius and holding at that temperature for 2.5 hours, argon as the protective gas, and a gas flow rate of 5 cubic meters per hour.

[0120] As described in steps S210-S250 above, the environmental field data (including temperature and pressure fields) of the sintering furnace is first constructed. Then, the environmental temperature and pressure time histories of the target green billet are determined from this environmental field. Next, the product performance prediction data of the target green billet under the corresponding process conditions is determined. Finally, if the product performance prediction data does not meet the performance requirements, the process parameters are corrected and the above steps are repeated until sintering process parameters that meet the performance requirements are obtained. This enables virtual prediction of the product performance of the target green billet under different sintering process parameters on a computer, and automatic iterative optimization of the sintering process parameters based on the prediction results. This avoids repeated trial firings in the actual sintering furnace to adjust process parameters, significantly reducing the number of trial firings, shortening the process development cycle, lowering R&D costs, and improving process development efficiency.

[0121] This disclosure also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the sintering parameter determination method in any embodiment of this disclosure.

[0122] Optionally, the computer-readable storage medium may be a non-transitory storage medium, but is not limited thereto; it may also be a temporary storage medium.

[0123] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0124] This disclosure may be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having a computer-readable program loaded thereon for causing a processor to implement any of the methods in the foregoing embodiments of this disclosure.

[0125] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media may include, for example, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), compact disc-read-only memory (CD-ROM), digital versatile disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any combination thereof. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0126] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an ambient computer or ambient storage device. The network may include one or more of copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to computer-readable storage media in the respective computing / processing device.

[0127] The computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object programs written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and conventional procedural programming languages ​​(such as the "C" language or similar programming languages). The computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as a standalone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network (e.g., a local area network or a wide area network), or it may be connected to an environment computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays, or programmable logic arrays, can execute computer-readable program instructions to implement various aspects of the embodiments of this disclosure by utilizing state information from the computer-readable program instructions.

[0128] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0129] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0130] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0131] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It should be noted that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are all equivalent.

[0132] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this disclosure is defined by the appended claims.

Claims

1. A sintering parameter determination method, characterized by, include: Obtain the current sintering process parameters; Based on the current sintering process parameters and the sintering furnace model, the furnace environment field data in the time domain of the spatial region inside the sintering furnace is determined; wherein, the furnace environment field data includes temperature field data and pressure field data; Based on the position of the target green billet in the sintering furnace, the time history data of the ambient temperature and the time history data of the ambient air pressure of the target green billet are determined from the furnace environmental field data; Based on the time history data of ambient temperature and ambient air pressure of the target green body, the product performance prediction data corresponding to the target green body is determined. If the product performance prediction data does not meet the performance index requirements, the current sintering process parameters are corrected to obtain corrected sintering process parameters. The above steps are then repeated based on the corrected sintering process parameters until the product performance prediction data meets the performance index requirements. Output the sintering process parameters that meet the performance requirements.

2. The method of claim 1, wherein, The step of determining the product performance prediction data corresponding to the target green compact based on the environmental temperature time history data and environmental air pressure time history data includes: The ambient temperature time history data and the ambient air pressure time history data are used as the environmental driving force, and the thermal response prediction data of the target green body in the sintering process is determined according to the material state change model of the green body material in the sintering process. The material stress time history data of the target green body is determined based on the thermal response prediction data, the material temperature time history data of the target green body is determined based on the ambient temperature time history data, and the microscopic performance prediction data and macroscopic performance prediction data of the target green body are determined based on the material temperature time history data and the material stress time history data. The microscopic performance prediction data and the macroscopic performance prediction data are used as the product performance prediction data for the target green blank.

3. The method of claim 1, wherein, The current sintering process parameters include heat source parameters and protective gas flow parameters. The step of determining the furnace environment field data in the time domain for the spatial region within the sintering furnace based on the current sintering process parameters and the sintering furnace model includes: Construct a sintering furnace model; wherein, the sintering furnace model includes a sintering furnace body structure model and multiple green billet structure models corresponding to each green billet; Based on the sintering furnace model, the heat source parameters of the sintering furnace, and the protective gas flow parameters, the furnace environment field data in the time domain of the spatial region inside the sintering furnace are determined.

4. The method of claim 1, wherein, The position of the target green billet in the sintering furnace is determined by the following steps: The position of the target green billet model corresponding to the target green billet is registered with the spatial coordinate system of the sintering furnace model to obtain the position of the target green billet model in the furnace environment field data; The position of the target green body model in the furnace environment field data is taken as the position of the target green body in the sintering furnace.

5. The method of claim 2, wherein, The material state change model is used to describe the dynamic evolution of the internal density, porosity and stress of the green material during the sintering process with temperature-pressure history.

6. The method of claim 2, wherein, The thermodynamic response prediction data includes predicted stress time history data, predicted strain time history data, predicted final shrinkage deformation data, predicted density distribution data, and predicted displacement field distribution data of the target green body throughout the sintering process.

7. The method of claim 2, wherein, The step of determining the predicted microscopic properties and macroscopic properties of the target green body based on the material temperature time history data and the material stress time history data includes: The material temperature time history data and the material stress time history data are combined to form a joint load history. The combined load history is input into the performance prediction model to obtain the microscopic performance prediction data and the macroscopic performance prediction data.

8. The method of claim 7, wherein, The performance prediction model is constructed through the following steps: The microscopic and macroscopic performance test data of the sample green body and the sample material test data corresponding to multiple sets of sintering process parameters are obtained to obtain a dataset; wherein, the sample material test data includes sample material temperature time history data and sample material stress time history data; The sample material temperature time history data and sample material stress time history data of each sample in the dataset are combined to form the sample joint load history, thus obtaining the training sample set; The performance prediction model is trained by using the joint load history of the samples in the training sample set as the input feature and the micro performance detection data and macro performance detection data in the training sample set as the output feature.

9. An electronic device, comprising: It includes a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the method according to any one of claims 1-8.

Images