A ceramic sintering optimization method and system based on dynamic monitoring

Abstract

The present application discloses a ceramic sintering optimization method and system based on dynamic monitoring, which relates to the field of sintering technology. The method includes: synchronously acquiring in-situ a set of sintering data of a ceramic sample during the sintering process under multiple sintering process conditions; based on the sintering data set, establishing a densification kinetic model corresponding to multiple sintering process conditions; based on the densification kinetic model corresponding to multiple sintering process conditions, determining multiple iso-density lines, and constructing a kinetic prediction model based on multiple densification kinetic models and multiple iso-density lines, so as to optimize the sintering process parameters based on the kinetic prediction model. The present application relies on dynamic monitoring means, and can synchronously acquire the mass change rate and size change rate during the sintering process in-situ, and provides comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of the sintering process parameters can be improved, thereby improving the performance of ceramic materials.

Description

Technical Field

[0001] The present application relates to the field of sintering technology, and in particular to a ceramic sintering optimization method and system based on dynamic monitoring. Background Art

[0002] Ceramic materials are widely used in aerospace, electronics, medical devices, new energy and other high-tech fields due to their excellent mechanical properties, thermal stability and corrosion resistance. These characteristics make ceramic materials inaccessible in extreme environments and high-demand applications. Sintering is a key step in the ceramic preparation process, which has a direct impact on the performance of the final ceramic product (such as density).

[0003] Sintering is a process that heats powdered materials to a temperature below their melting point to allow particles to adhere to each other and form a solid whole. The sintering process can significantly increase the strength, hardness and density of the material and reduce porosity.

[0004] In current technology, the monitoring of ceramic sintering process relies on static measurement methods, such as X-ray diffraction (XRD), thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC), etc. These methods can usually only provide data at a specific time point. For example, the ceramic material is sintered to 100°C and the quality data of the current material is measured; after the ceramic material is cooled, it is sintered to 200°C and the quality data of the current material is measured; and so on, relying on static measurement methods to monitor the sintering process.

[0005] It can be seen from this that the monitoring of the ceramic sintering process in current technology relies on static measurement methods, which cannot accurately understand or capture the densification mechanism during the sintering process, affecting the optimization effect of the sintering process parameters, thereby limiting the performance of ceramic materials. Summary of the invention

[0006] Based on the above problems, the present application provides a ceramic sintering optimization method and system based on dynamic monitoring. Relying on dynamic monitoring means, it can synchronously measure the mass change rate and size change rate during the sintering process in real time, and provide comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of sintering process parameters can be improved, thereby improving the performance of ceramic materials.

[0007] The embodiments of the present application disclose the following technical solutions:

[0008] In a first aspect, the present application discloses a ceramic sintering optimization method based on dynamic monitoring, comprising:

[0009] Synchronously acquiring in situ a sintering data set of a ceramic sample during a sintering process under a plurality of sintering process conditions; wherein the sintering process conditions are a set of sintering process parameters; the sintering data set corresponds to the plurality of sintering process conditions respectively, and the sintering data set comprises: a plurality of sets of sintering data, a set of sintering data comprises: temperature, and a mass change rate and a size change rate corresponding to the temperature;

[0010] Based on the corresponding sintering data set, a densification kinetic model corresponding to each of the plurality of sintering process conditions is established; wherein the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions;

[0011] Based on the densification kinetic models corresponding to the multiple sintering process conditions, multiple isodensity lines are determined, and based on the densification kinetic models corresponding to the multiple sintering process conditions and the multiple isodensity lines, a kinetic prediction model is constructed to optimize the sintering process parameters based on the kinetic prediction model; wherein the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions.

[0012] Optionally, the sintering data set corresponding to the sintering process conditions is obtained in the following manner:

[0013] Synchronously acquiring in situ the mass change, size change and consumption time corresponding to multiple preset temperatures during a sintering process of the ceramic sample under the sintering process conditions; wherein the consumption time is the difference between the start time of the sintering process and the time of sintering to the preset temperature;

[0014] Based on the mass change amount and the consumed time corresponding to the preset temperature, determine the mass change rate corresponding to the preset temperature, and based on the size change amount and the consumed time corresponding to the preset temperature, determine the size change rate corresponding to the preset temperature;

[0015] Based on a plurality of preset temperatures and mass change rates and dimensional change rates corresponding to the plurality of preset temperatures, a sintering data set corresponding to the sintering process conditions is obtained.

[0016] Optionally, the establishing, based on the corresponding sintering data set, densification kinetic models corresponding to the plurality of sintering process conditions respectively comprises:

[0017] Based on the multiple groups of sintering data in the corresponding sintering data set, constructing a relationship curve between temperature and mass change rate and a relationship curve between temperature and size change rate corresponding to the multiple sintering process conditions respectively;

[0018] Based on the relationship curves between temperature and mass change rate and the relationship curves between temperature and size change rate respectively corresponding to the plurality of sintering process conditions, densification kinetic models respectively corresponding to the plurality of sintering process conditions are constructed.

[0019] Optionally, the densification kinetic model is: a curve of the relationship between temperature and densification rate; wherein, the curve of the relationship between temperature and densification rate is: a curve of the relationship between temperature and densification rate with the inverse of temperature as the horizontal axis and the densification rate in logarithmic form as the vertical axis; the densification rate is obtained based on the mass change rate and the size change rate.

[0020] Optionally, constructing a kinetic prediction model based on the densification kinetic models corresponding to the multiple process conditions, respectively, and the multiple isopycnic lines, comprises:

[0021] Based on the slopes of the multiple isopycnals, obtaining diffusion activation energies corresponding to the multiple isopycnals;

[0022] According to the diffusion activation energies corresponding to the multiple isopycnals, the sintering mechanisms that play a leading role in different stages are determined, and based on the sintering mechanisms that play a leading role in different stages, the densification mechanism is determined; wherein the sintering mechanisms include: surface diffusion mechanism, grain boundary diffusion mechanism or volume diffusion mechanism;

[0023] Based on the densification mechanism, an initial kinetic equation corresponding to the densification mechanism is determined, and a kinetic prediction model is constructed based on the densification kinetic models corresponding to the multiple sintering process conditions and the initial kinetic equation.

[0024] Optionally, the kinetic prediction model is represented by the following formula:

[0025]

[0026] in, Z ( t ) is the time t The corresponding density is t 1 is the start time of the temperature cycle, t 2 is the end time of the temperature cycle, Z ( t 1) is the density corresponding to the start time, Z ( t 2) is the density corresponding to the end time, Z m and Z n are adjacent isodense lines, for Z mThe densification rate corresponding to the isopycnal lines is for Z n Densification rates corresponding to isopycnals.

[0027] Optionally, the optimizing the sintering process parameters based on the kinetic prediction model includes:

[0028] Based on the kinetic prediction model, a mapping model training set is constructed; wherein the mapping model training set includes: multiple groups of training data, one group of training data includes: corresponding sintering process conditions and density;

[0029] Based on the mapping model training set, a mapping model is trained through machine learning; wherein the mapping model is used to output corresponding sintering process conditions based on density;

[0030] Based on the mapping model, the sintering process parameters are optimized.

[0031] Optionally, the kinetic prediction model is used to output a corresponding predicted density based on the input sintering process parameters; and optimizing the sintering process parameters based on the kinetic prediction model includes:

[0032] During the sintering process of ceramics, the initial sintering process parameters are obtained in real time;

[0033] Based on the kinetic prediction model, a predicted density corresponding to the initial sintering process parameters is obtained, and actual sintering data is simultaneously acquired to obtain a corresponding actual density based on the actual sintering data;

[0034] Based on the predicted density and the actual density, the initial sintering process parameters are optimized.

[0035] Optionally, the optimizing the initial sintering process parameters based on the predicted density and the actual density includes:

[0036] When the difference between the predicted density and the actual density is greater than a preset difference, the initial sintering process parameters are optimized through PID control based on the difference between the predicted density and the actual density.

[0037] In a second aspect, the present application discloses a ceramic sintering optimization system based on dynamic monitoring, comprising:

[0038] An in-situ synchronous acquisition module is used to synchronously acquire in-situ a sintering data set of a ceramic sample during a sintering process under multiple sintering process conditions; wherein the sintering process conditions are a set of sintering process parameters; the sintering data set corresponds to the multiple sintering process conditions respectively, and the sintering data set includes: multiple sets of sintering data, one set of sintering data includes: temperature, and mass change rate and size change rate corresponding to the temperature;

[0039] A kinetic analysis module, for establishing, based on the corresponding sintering data set, densification kinetic models corresponding to the plurality of sintering process conditions, respectively; wherein the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions;

[0040] A kinetic prediction module, for determining a plurality of isopycnals based on the densification kinetic models respectively corresponding to the plurality of sintering process conditions, and constructing a kinetic prediction model based on the densification kinetic models respectively corresponding to the plurality of sintering process conditions and the plurality of isopycnals; wherein the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions;

[0041] The process parameter optimization module is used to optimize the sintering process parameters based on the kinetic prediction model.

[0042] Compared with the prior art, the present application has the following beneficial effects: the present application relies on dynamic monitoring means, and can measure the mass change rate and size change rate during the sintering process in real time and synchronously (i.e., synchronously in situ), providing comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of the sintering process parameters can be improved, thereby improving the performance of ceramic materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 A schematic diagram of a process flow of a ceramic sintering optimization method based on dynamic monitoring provided in an embodiment of the present application;

[0045] Figure 2 An example diagram of the densification kinetic model and isopycnic lines provided in the embodiments of the present application;

[0046] Figure 3A schematic diagram of a process for constructing a kinetic prediction model provided in an embodiment of the present application;

[0047] Figure 4 A schematic flow chart of a method for optimizing sintering process parameters provided in an embodiment of the present application;

[0048] Figure 5 A schematic flow chart of another method for optimizing sintering process parameters provided in an embodiment of the present application;

[0049] Figure 6 A schematic diagram of the structure of a ceramic sintering optimization system based on dynamic monitoring provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] As described above, the current technology relies on static measurement methods to monitor the sintering process, which can only provide data at a specific time point, that is, only the final quality data and / or size data can be obtained for a single sintering process. Exemplarily, the current technology needs to obtain data from the same multiple sintering processes, such as sintering a ceramic material to 100°C, measuring the quality data and size data of the current material; sintering the ceramic material to 200°C after it cools down, measuring the quality data and size data of the current material; sintering the ceramic material to 300°C after it cools down, measuring the quality data and size data of the current material, and so on. Based on the statically measured quality data and size data in the same multiple sintering processes, the sintering process can be monitored.

[0051] The static measurement methods in the current technology can only obtain / measure data at specific locations (time, temperature), lacking real-time monitoring of dynamic changes in the sintering process, and obtaining data at multiple specific locations in a sintering process requires repeated heating and cooling, resulting in inaccurate data at multiple specific locations in a sintering process. For example, the data of the ceramic material obtained by the first sintering to 200°C is different from the data of the ceramic material obtained by first sintering to 100°C, cooling, and then sintering to 200°C, resulting in inaccurate data from static monitoring. Therefore, the monitoring of the ceramic sintering process in the current technology relies on static measurement methods, which cannot accurately understand or capture the densification mechanism in the sintering process, affecting the optimization effect of the sintering process parameters, thereby limiting the performance of the ceramic material.

[0052] Furthermore, in the current technology, static measurement methods can only obtain / measure data at specific locations, but cannot capture the mass change data and dimensional change data of the sintering process. In addition, the data obtained by static monitoring methods has poor continuity, and further cannot accurately understand or capture the densification mechanism of the sintering process.

[0053] The present application provides a ceramic sintering optimization method based on dynamic monitoring, including: synchronously acquiring in situ a sintering data set of a ceramic sample during the sintering process under multiple sintering process conditions; based on the corresponding sintering data set, establishing a densification kinetic model corresponding to the multiple sintering process conditions; based on the densification kinetic model corresponding to the multiple sintering process conditions, determining multiple iso-density lines, and based on the densification kinetic model corresponding to the multiple sintering process conditions and the multiple iso-density lines, constructing a kinetic prediction model, so as to optimize the sintering process parameters based on the kinetic prediction model. The present application relies on dynamic monitoring means, and can measure the mass change rate and size change rate during the sintering process in real time and synchronously (i.e., synchronous in situ), and provides comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of the sintering process parameters can be improved, thereby improving the performance of ceramic materials.

[0054] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] Embodiment 1:

[0056] Combine the following Figure 1-Figure 5 , a ceramic sintering optimization method based on dynamic monitoring provided in an embodiment of the present application is introduced in detail.

[0057] like Figure 1 As shown, a ceramic sintering optimization method based on dynamic monitoring provided in an embodiment of the present application includes the following steps:

[0058] S101. Synchronously and in situ obtain a sintering data set of a ceramic sample during the sintering process under multiple sintering process conditions.

[0059] Among them, the sintering process conditions are a set of sintering process parameters. Sintering process parameters refer to various conditions and variables that can significantly affect the performance of the final ceramic product during the sintering process of ceramics. Sintering process parameters are directly important for obtaining microstructures and macroscopic properties that meet the requirements.

[0060] The sintering process parameters include but are not limited to: sintering temperature, that is, the highest temperature experienced by the ceramic sample during the sintering process; heating rate, that is, the speed from room temperature or preheating temperature to sintering temperature, such as: 1℃ / min, 10℃ / min, etc.; holding time, that is, the length of time the ceramic sample is kept at the sintering temperature.

[0061] It should be noted that in the embodiments of the present application, the heating rate can be selected as a variable in different sintering process conditions to obtain a sintering data set of the ceramic sample during the sintering process at multiple heating rates. Of course, other sintering process parameters can also be selected as variables to obtain a sintering data set of the ceramic sample under different sintering conditions. In general, the sintering data set of the sintering process under multiple sintering process conditions obtained synchronously in situ should cover multiple sintering process conditions (multiple sets of sintering process parameters) to ensure the comprehensiveness and reliability of the sintering data set corresponding to multiple sintering process conditions.

[0062] The sintering data set includes multiple groups of sintering data, and one group of sintering data includes: a preset temperature and a mass change rate and a size change rate corresponding to the preset temperature.

[0063] The mass change rate refers to the change in the mass of the ceramic sample per unit time, which reflects the dynamic change in the mass of the ceramic material during the sintering process. The mass change rate can also be called mass change data.

[0064] The dimensional change rate (volume change rate) refers to the change in the geometric dimensions of a ceramic sample per unit time, reflecting the dynamic changes in the dimensions of the ceramic material during the sintering process. The dimensional change rate can also be called dimensional change data.

[0065] A set of sintering data includes a preset temperature, and a mass change rate and a size change rate when the ceramic sample is sintered to the preset temperature.

[0066] Here, synchronization refers to multiple time acquisition processes occurring at the same time, or being carried out according to a fixed time relationship. In the embodiment of the present application, synchronization refers to the synchronous acquisition of the temperature, mass change rate and size change rate of the ceramic sample during the sintering process, that is, during the sintering process of the ceramic sample, the temperature, size change rate and mass change rate are monitored simultaneously and continuously.

[0067] Here, in-situ refers to observing or measuring the sample without changing the sample environment. In the embodiment of the present application, in-situ means directly obtaining the temperature, dimensional change rate and mass change rate of the ceramic sample during the sintering process under the sintering process conditions.

[0068] In the embodiment of the present application, a sintering data set of a ceramic sample is obtained in situ during the sintering process under multiple sintering process conditions. Specifically, during the sintering process under one sintering process condition, the sintering process is not interrupted and the ceramic sample is not removed, and the temperature, size change rate and mass change rate are measured / obtained at the same time to obtain the corresponding sintering data set. Acquiring the sintering data set in situ simultaneously, that is, acquiring the sintering data set without interrupting the sintering process or removing the ceramic sample, can obtain true and accurate sintering data.

[0069] In a possible implementation, a set of sintering data includes: a preset temperature, and a mass change rate and a size change rate corresponding to the preset temperature. That is, multiple preset temperatures are preset, and when a preset temperature is reached during the sintering process of the ceramic sample, the mass change rate and the size change rate corresponding to the preset temperature are synchronously acquired in situ.

[0070] Specifically, for ease of understanding, the following is a detailed introduction: (a) The sintering data set corresponding to the sintering process conditions is obtained in the following way:

[0071] The mass change, size change and consumed time corresponding to multiple preset temperatures of the ceramic sample during a sintering process under the sintering process conditions are synchronously obtained in situ; wherein the consumed time is the difference between the start time of the sintering process and the time of sintering to the preset temperature, that is, the time consumed for sintering to the preset temperature; based on the mass change and consumed time corresponding to the preset temperature, the mass change rate corresponding to the preset temperature is determined, and based on the size change and consumed time corresponding to the preset temperature, the size change rate corresponding to the preset temperature is determined; based on multiple preset temperatures and the mass change rates and size change rates corresponding to the multiple preset temperatures, a sintering data set corresponding to the sintering process conditions is obtained.

[0072] Specifically, based on the mass change amount and the consumption time, the mass change rate is obtained through formula (1).

[0073] (1)

[0074] in, dm / dt is the first-order derivative of mass with respect to time, i.e., the rate of mass change; Δ m is the mass change; Δ t is the consumption change.

[0075] Specifically, based on the size change amount and the consumed time, the size change rate is obtained through formula (2).

[0076] (2)

[0077] in, dV / dt is the first-order derivative of size (volume) with respect to time, i.e., the rate of size change; Δ V is the size change (i.e. volume change), Δ t To consume time; dε x / dt is the rate of change of size in the x direction, dε y / dt The dimensional change rate in the y-axis direction, dε z / dt is the rate of change of size in the z-axis direction.

[0078] S102. Based on the corresponding sintering data set, a densification kinetic model corresponding to a plurality of sintering process conditions is established.

[0079] Among them, the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions. That is, the densification kinetic model can describe the interaction between internal particles and the energy change during the sintering process under the corresponding sintering process conditions.

[0080] Densification refers to the process in which the internal pores of the material are reduced, the volume is shrunk, and the density is increased through the interaction between particles during the sintering process.

[0081] Densification rate (also known as density change rate, relative density change rate) refers to the change in relative density (density) of a ceramic sample per unit time. The densification rate reflects the speed at which the relative density (density) of a ceramic sample changes over time during the sintering process. It is a dynamic feature that emphasizes the reduction of pores and increase of density in the ceramic sample during the densification process.

[0082] Density (Density of Sintering or Degree of Densification) can also be understood as a percentage of relative density. Density refers to the density percentage of the ceramic sample after sintering relative to its theoretical density.

[0083] For ease of understanding, the density is introduced below in conjunction with formula (3).

[0084] (3)

[0085] in, Z For density, ρ actual is the actual density of the ceramic sample, ρ theoretical is the theoretical density of the ceramic sample, that is, the density under ideal conditions without pores.

[0086] In one possible implementation, based on the corresponding sintering data set, relationship curves between temperature and mass change rate and relationship curves between temperature and size change rate corresponding to multiple sintering process conditions are constructed; based on the relationship curves between temperature and mass change rate and relationship curves between temperature and size change rate corresponding to multiple sintering process conditions, densification kinetic models corresponding to multiple sintering process conditions are constructed.

[0087] Specifically, the densification kinetic model is a relationship curve between temperature and densification rate.

[0088] Furthermore, the relationship curve between temperature and densification rate is a relationship curve between temperature and densification rate with the inverse of temperature as the horizontal axis and the densification rate in logarithmic form as the vertical axis. And the densification rate is obtained based on the mass change rate and the size change rate.

[0089] Exemplarily, for the same sintering process conditions, the density change rate is obtained based on the mass change rate and size change rate corresponding to the same temperature; the densification rate corresponding to the temperature is obtained based on the density change rate and the theoretical density; based on multiple sets of temperatures and their corresponding densification rates, the corresponding densification kinetic model (i.e., the relationship curve between temperature and densification rate) is obtained.

[0090] From the above, it can be seen that based on the relationship curve between temperature and mass change rate and the relationship curve between temperature and size change rate corresponding to the sintering process conditions, the corresponding relationship curve between temperature and densification rate is constructed.

[0091] For ease of understanding, the densification kinetic model is introduced in detail below in combination with formula (4), formula (5) and formula (6).

[0092] First, we will introduce the relationship between the densification rate and the diffusion coefficient based on formula (4).

[0093] (4)

[0094] in, dρ / ρdt is the densification rate, C is a constant; γ sv is the interfacial energy between the melt and the gas phase, G is the grain size, n depends on the exponential of the rate control mechanism, i.e. G n It can represent temperature gradient; D is the diffusion coefficient; T is the absolute temperature. It should be noted that C , γ sv andG n are predetermined parameters, which can be understood as preset constant term parameters.

[0095] Absolute temperature ( T ) refers to the temperature measured using the absolute temperature scale, the most commonly used of which is the Kelvin (K) scale. The characteristic of absolute temperature is that its zero point corresponds to the lowest possible temperature in theory, absolute zero, at which all thermal motion in the classical sense ceases.

[0096] Diffusion coefficient ( D ) describes the ability of atoms or ions to move inside a solid material. During the ceramic sintering process, the diffusion coefficient reflects the rate of material migration between particles, which has a direct impact on the elimination of pores (i.e., the densification process).

[0097] The diffusion coefficient usually follows the form of the Arrhenius equation. For ease of understanding, the diffusion coefficient is introduced below in conjunction with formula (5).

[0098] (5)

[0099] in, D is the diffusion coefficient; D 0 is the frequency factor (or pre-exponential factor); Q is the diffusion activation energy; R is the ideal gas constant, which is generally 8.314 J / (mol·k); T is the absolute temperature.

[0100] Frequency factor (pre-exponential factor, D 0 ) refers to the theoretical value of the diffusion coefficient at infinitely high temperature (i.e. when the effect of activation energy can be ignored), reflecting the maximum possible rate of internal atomic or molecular movement, which is not limited by the energy barrier in the thermal activation process. D 0 It is affected by the properties of the material itself, such as crystal structure, defect concentration, etc., and is also related to the material's peripheral structure, such as particle size, shape and distribution.

[0101] As shown in formula (5), the diffusion coefficient in the form of the Arrhenius equation is D With temperature (absolute temperature T ) increases significantly with the increase of temperature, because the high temperature provides enough energy to overcome the energy barriers encountered when atoms or ions move.

[0102] To simplify data analysis, the diffusion coefficient shown in formula (5) is substituted into formula (4), and the densification rate is logarithmically transformed as shown in formula (6).

[0103] (6)

[0104] in, dρ / ρdt is the densification rate; C is a constant; sv The interfacial energy between the melt and the gas phase can also be understood as a preset constant parameter; D 0 is the frequency factor (or pre-exponential factor), which can also be understood as a preset constant parameter; n Depending on the exponential of the rate control mechanism, G is the grain size, which can also be understood as a preset constant parameter; R is the ideal gas constant; Q is the diffusion activation energy; T is the absolute temperature.

[0105] From formula (6), we can see that the densification kinetic model (i.e., the relationship curve between temperature and densification rate) is based on the inverse of temperature (i.e., 1 / T ) is the horizontal axis, and the densification rate is expressed in logarithmic form (i.e., ln( dρ / ρdt )) is the relationship curve between temperature and densification rate on the vertical axis.

[0106] In one possible implementation, for a sintering process condition, multiple groups of sintering data in the corresponding sintering data set are processed using Matlab programming, and the temperature-mass change rate relationship curve and the temperature-size change rate relationship curve are constructed and plotted; then Matlab programming is used to process the temperature-mass change rate relationship curve and the temperature-size change rate relationship curve, and a densification kinetic model (i.e., the temperature-densification rate relationship curve) is constructed and plotted.

[0107] S103, determining a plurality of isopycnals based on densification kinetic models corresponding to a plurality of sintering process conditions, and constructing a kinetic prediction model based on the densification kinetic models corresponding to a plurality of sintering process conditions and the plurality of isopycnals.

[0108] Among them, the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions.

[0109] Among them, the isodensity line refers to the isovalue line of state points with the same density (relative density) in the densification kinetic model (i.e., the temperature-densification rate relationship curve) corresponding to multiple sintering process conditions.

[0110] Exemplarily, a series of representative densities (relative densities) are pre-set, for example: 70%, 80%, 90%, etc.; for each pre-set density (relative density), the state point of the pre-set density (relative density) is determined in the densification kinetic model corresponding to multiple sintering process conditions; a linear fit is performed on the state point of the same pre-set density (relative density) to obtain an isodensity line.

[0111] For ease of understanding, the following Figure 2 The densification kinetic models and isodensity lines corresponding to multiple sintering process conditions are introduced with examples.

[0112] like Figure 2 As shown, four sintering process conditions are taken as an example, and the four sintering process conditions correspond to a set of sintering process parameters with only different heating rates, so the four sintering process conditions are directly expressed by four heating rates. Exemplarily, the four sintering process conditions are 1°C / min, 2°C / min, 5°C / min and 10°C / min, respectively.

[0113] like Figure 2 As shown in the figure, the densification kinetic model uses the inverse of the (absolute) temperature as the horizontal axis (x-axis), i.e. 1 / T is the x-axis, and its corresponding unit is °C -1 ; The vertical axis (y axis) is the exponential form of the densification rate, i.e. ln( dρ / ρdt ) is the y-axis.

[0114] like Figure 2 As shown, the temperature-densification rate relationship curves (densification kinetic model) corresponding to 1°C / min, 2°C / min, 5°C / min and 10°C / min were plotted respectively, as well as 11 isopycnic lines with relative density (density) of 0.81 (81%), 0.83 (83%), 0.85 (85%), 0.87 (87%), 0.89 (89%), 0.91 (91%), 0.93 (93%), 0.95 (95%), 0.97 (97%), 0.98 (98%) and 0.99 (99%).

[0115] For ease of understanding, the following Figure 3 This paper introduces in detail how to construct a kinetic prediction model based on the densification kinetic models corresponding to multiple sintering process conditions and multiple isopycnic lines.

[0116] S301. Based on the slopes of the multiple isodensity lines, diffusion activation energies corresponding to the multiple isodensity lines are obtained.

[0117] Specifically, multiple isopycnals are obtained based on the densification kinetics model corresponding to multiple sintering process conditions, and the densification kinetics model is a temperature-densification rate relationship curve with the inverse of (absolute) temperature as the horizontal axis and the densification rate in exponential form as the vertical axis. Therefore, the slope of the isopycnal line is - Q / R (-diffusion activation energy / ideal gas constant), then the corresponding diffusion activation energy can be calculated based on the slope of the isopycnal line.

[0118] For example, assume that the slope of the isopycnal line is slope ,but slope =- Q / R ; Ideal gas constant R =8.3145 J / (mol·K), then Q (Activation Energy of Diffusion) = - slope * R =- slope *8.3145 (unit: J / mol), further, the diffusion activation energy Q The unit is generally kJ / mol, then Q =- slope * R / 1000 (unit: kJ / mol).

[0119] S302. Determine the sintering mechanism that plays a dominant role in different stages according to the diffusion activation energy corresponding to the multiple isodensity lines, and determine the densification mechanism based on the sintering mechanism that plays a dominant role in different stages.

[0120] Among them, the sintering mechanism includes: surface diffusion mechanism, grain boundary diffusion mechanism or volume diffusion mechanism.

[0121] The surface diffusion mechanism refers to the process of atoms or molecules (particles) moving along the surface of the material. The surface diffusion mechanism mainly occurs near the contact point of the particles. In low-temperature sintering or when a small grain size needs to be maintained, the surface diffusion mechanism is the main sintering mechanism.

[0122] Grain boundary diffusion (GBD) is the diffusion of atoms through grain boundaries (i.e., grain boundaries). In the case of Chinese sintering of polycrystalline materials, the GBD mechanism may be the main sintering mechanism, which helps to achieve rapid densification without causing excessive grain growth.

[0123] Volume or Bulk Diffusion refers to the diffusion of atoms through the crystal structure (bulk phase) of the material. During high temperature sintering and later densification, when other diffusion mechanisms are insufficient to continue densification, the volume diffusion mechanism is the main sintering mechanism.

[0124] Among them, multiple isodensity lines divide the sintering process into multiple stages. For example, the three isodensity lines of 0.83 (83%), 0.91 (91%) and 0.99 (99%) divide the sintering process into three stages, namely: the density stage of 0-0.83, the stage of 0.83-0.91, and the stage of 0.91-0.99.

[0125] By analyzing the diffusion activation energy corresponding to multiple isodensities, when the diffusion activation energy corresponding to the isodensities is within the first preset interval, it is determined that the sintering mechanism that dominates the stage corresponding to the isodensities is the surface diffusion mechanism; when the diffusion activation energy corresponding to the isodensities is within the second preset interval, it is determined that the sintering mechanism that dominates the stage corresponding to the isodensities is the grain boundary diffusion mechanism; when the diffusion activation energy corresponding to the isodensities is within the third preset interval, it is determined that the sintering mechanism that dominates the stage corresponding to the isodensities is the volume diffusion mechanism.

[0126] For example, there are three isopycnals with relative densities (density) of 0.83 (83%), 0.91 (91%) and 0.99 (99%). The diffusion activation energy corresponding to the isopycnal of relative density 0.83 is 150 kJ / mol, which belongs to the first preset interval (0,200). Therefore, the sintering mechanism that plays a dominant role in the relative density stage of 0-0.83 is the surface diffusion mechanism; the diffusion activation energy corresponding to the isopycnal of relative density 0.91 is 300 kJ / mol, which belongs to the second preset interval [200,400). Therefore, the sintering mechanism that plays a dominant role in the relative density stage of 0.83-0.91 is the grain boundary diffusion mechanism; the diffusion activation energy corresponding to the isopycnal of relative density 0.99 is 450 kJ / mol, which belongs to the third preset interval [400,600). Therefore, the sintering mechanism that plays a dominant role in the relative density stage of 0.91-0.99 is the volume diffusion mechanism.

[0127] Furthermore, when the dominant sintering mechanisms at different stages are determined, the densification mechanism is determined based on the dominant sintering mechanisms at different stages.

[0128] The density mechanism includes: the dominant sintering mechanism in each stage and the transition point of the dominant sintering mechanism.

[0129] Specifically, when the dominant sintering mechanism in the stage is the surface diffusion mechanism, the corresponding densification mechanism is: particle rearrangement and pore migration. Particle rearrangement is: contact points between particles begin to form, and surface atoms move to the contact points between particles through surface diffusion to form necks; pore migration is: surface diffusion causes small pores to migrate to large pores, initially reducing the number of pores.

[0130] When the dominant sintering mechanism in this stage is the grain boundary diffusion mechanism, the corresponding densification mechanism is: pore closure and particle growth. Pore closure is: the grain boundary region becomes the main diffusion path, atoms move through the grain boundary diffusion, and promote the closure of pores; particle growth is: the grain boundaries between particles move, causing the particles to grow and further reduce the pores.

[0131] When the dominant sintering mechanism in this stage is the volume diffusion mechanism, the corresponding densification mechanism is: pore elimination and structural stabilization. Pore elimination means that atoms move through the interior of the lattice to further reduce the remaining pores and improve the overall density of the material; structural stabilization means that as the pores are completely closed, the material structure tends to be stable and reaches a state close to the theoretical density.

[0132] Further, based on the sintering mechanisms that play a dominant role in different stages, the densification mechanism is determined, which can be understood as: integrating and analyzing the sintering mechanisms that play a dominant role in different stages to obtain the corresponding densification mechanism. Exemplary: the sintering mechanism of the stage corresponding to the density line with a relative density of 0.81 is the surface diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.83 is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.85 is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.87 is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.89 is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.91 is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line with a relative density of 0.93 is the grain boundary diffusion mechanism, and the sintering mechanism of the stage corresponding to the density line with a relative density of 0.95 is The sintering mechanism is the grain boundary diffusion mechanism, the sintering mechanism of the stage corresponding to the density line of relative density 0.97 is the volume diffusion mechanism, and the sintering mechanism of the stage corresponding to the density line of relative density 0.99 is the volume diffusion mechanism; based on the sintering mechanisms that play a dominant role in different stages, the densification mechanism is obtained as follows: the stage of relative density 0-0.81 is dominated by the surface diffusion mechanism, the stage of relative density 0.81-0.95 is dominated by the grain boundary diffusion mechanism, and the stage of relative density 0.95-0.99 is dominated by the volume diffusion mechanism, and the transition points of the sintering mechanisms that play a dominant role in different stages are identified: the transition point between the surface diffusion mechanism and the grain boundary diffusion mechanism is 0.81, and the transition point between the grain boundary diffusion mechanism and the volume diffusion mechanism is 0.95.

[0133] S303, based on the densification mechanism, determine the initial kinetic equation corresponding to the densification mechanism, and construct a kinetic prediction model based on the densification kinetic models and the initial kinetic equation corresponding to multiple sintering process conditions.

[0134] Among them, the kinetic prediction model is the basis for optimizing the sintering process parameters, that is, the sintering process parameters are optimized based on the kinetic prediction model.

[0135] Specifically, based on the densification mechanism, the initial kinetic equation corresponding to the densification mechanism is determined; the kinetic equation describes the change law of density (relative density) over time. And using the temperature-densification rate relationship curve corresponding to multiple sintering process conditions and the initial kinetic equation, a kinetic prediction model is constructed.

[0136] In a possible implementation, the kinetic prediction model can be in an integral form, as shown in formula (7):

[0137] (7)

[0138] in, Z ( t ) is the time t The corresponding density is t 1 is the start time of the temperature cycle, t 2 is the end time of the temperature cycle, Z ( t 1) is the density corresponding to the start time, Z ( t 2) is the density corresponding to the end time, Z m and Z n are adjacent isodense lines, for Z m The densification rate corresponding to the isopycnal lines is for Z n Densification rates corresponding to isopycnals.

[0139] The above combination Figure 3 This paper introduces in detail how to obtain the dynamic prediction model. Figure 4 and Figure 5 Two methods of optimizing sintering process parameters based on kinetic prediction models are introduced respectively.

[0140] Combine the following Figure 4 , a method for optimizing sintering process parameters is introduced in detail.

[0141] S401. Construct a mapping model training set based on the kinetic prediction model.

[0142] The mapping model training set includes: multiple groups of training data, one group of training data includes corresponding sintering process conditions and density. Density is one of the important sintering properties of ceramic products.

[0143] Specifically, the kinetic prediction model is used to output the corresponding predicted density based on the input sintering process parameters. Therefore, based on the kinetic prediction model, multiple sets of corresponding sintering process conditions (i.e., a set of sintering process parameters) and density can be obtained.

[0144] In a possible implementation, a set of training data includes: corresponding sintering process conditions and product performance, and the product performance includes the sintering performance and usage performance of the ceramic product.

[0145] Sintering performance refers to the characteristics exhibited by ceramics during the sintering process. The sintering performance of ceramic products can be directly evaluated by density.

[0146] Performance refers to the characteristics of ceramic products in their intended application environment, such as wear resistance, fatigue life, thermal conductivity, corrosion resistance, etc.

[0147] Specifically, multiple densities corresponding to multiple sintering process conditions (multiple groups of sintering process parameters) are obtained based on the kinetic prediction model, and the density is used to characterize the sintering performance corresponding to the corresponding sintering process conditions; at the same time, multiple usage performances corresponding to the multiple sintering process conditions (multiple groups of sintering process parameters) input manually are received, thereby obtaining multiple groups of mutually corresponding sintering process conditions and product performances.

[0148] Furthermore, in a possible implementation, each set of training data in the mapping model training set includes: components, sintering process conditions and density corresponding to each other, that is, each set of training data is: components-sintering process conditions-density.

[0149] Composition refers to the chemical composition and phase structure of ceramic materials. Ceramics are usually composed of one or more inorganic non-metallic materials such as oxides, nitrides, carbides, etc. Depending on the application requirements, ceramics can contain a variety of different elements and compounds to optimize their specific performance characteristics.

[0150] S402: Based on the mapping model training set, a mapping model is trained through machine learning.

[0151] When each set of training data included in the mapping model training set includes: corresponding sintering process conditions and density, that is, sintering process conditions-density, the mapping model obtained through machine learning training is used to output the corresponding sintering process conditions (a set of sintering process parameters) based on the density, and of course can also be used to output the corresponding density based on the sintering process conditions (a set of sintering process parameters).

[0152] In one possible implementation, when each set of training data included in the mapping model training set includes: corresponding sintering process conditions and performance, that is, sintering process conditions-performance, the mapping model obtained through machine learning training is used to output the corresponding sintering process conditions based on the performance, and of course can also be used to output the corresponding performance based on the sintering process conditions.

[0153] In one possible implementation, when each set of training data included in the mapping model training set includes: corresponding components, sintering process conditions and density, that is, components-sintering process conditions-density, the mapping model obtained through machine learning training is used to output the corresponding sintering process conditions based on the components and density; similarly, the mapping model is used to output the corresponding density based on the components and sintering process conditions.

[0154] For example, the mapping model may be a feedforward neural network (FNN) model, as shown in formula (8):

[0155] Y= f (W1×X+b1)×W2+b2 (8)

[0156] Among them, Y is the output of the mapping model; X is the input of the mapping model; W1 and W2 are weight matrices; b1 and b2 are bias vectors; f It is an activation function used to introduce nonlinear features. Common activation functions include ReLU, sigmoid, and tanh.

[0157] S403. Optimize sintering process parameters based on the mapping model.

[0158] Specifically, before sintering the ceramic, the target density of the ceramic sintering is known in advance, that is, there is a preset requirement for the density of the ceramic product, and the sintering process purpose corresponding to the preset sintering process parameters (sintering process conditions) is to obtain the target density. Therefore, the target density is input into the mapping model, and the mapping model outputs the sintering process conditions (a set of sintering process parameters) corresponding to the target density. Based on the corresponding sintering process conditions (a set of sintering process parameters) output by the mapping model, the preset sintering process parameters are optimized so that the density of the ceramic product obtained by sintering meets the preset requirements.

[0159] The mapping model training set constructed based on the kinetic prediction model is trained by machine learning to obtain the mapping model, and the sintering process parameters are optimized based on the mapping model. Through machine learning technology, the mapping model can learn the complex relationship between sintering process conditions (a set of sintering process parameters) and density, and the sintering process parameters can be optimized based on the mapping model, thereby improving the optimization speed of the sintering process parameters.

[0160] Combine the following Figure 5 , another method to optimize the sintering process parameters is introduced in detail.

[0161] S501. During the sintering process of the ceramic, initial sintering process parameters are obtained in real time.

[0162] Specifically, during the sintering process of the ceramic, the initial sintering process parameters (sintering process conditions) at the current moment are obtained in real time.

[0163] It should be noted that the initial sintering process parameters are the sintering process parameters at the current moment during the sintering process of the ceramic, that is, the initial sintering process parameters can be the sintering process parameters preset before the sintering process, or the sintering process parameters that have been optimized during the sintering process.

[0164] S502: Based on the kinetic prediction model, a predicted density corresponding to the initial sintering process parameters is obtained, and actual sintering data is simultaneously acquired to obtain the corresponding actual density based on the actual sintering data.

[0165] The predicted density corresponding to the initial sintering process parameters is the density at the current moment predicted by the kinetic prediction model.

[0166] Specifically, based on the kinetic prediction model, the predicted density corresponding to the initial sintering process parameters at the current moment is obtained, and the actual sintering data during the sintering process is obtained synchronously (i.e. at the same current moment) to obtain the corresponding actual density based on the actual sintering data.

[0167] S503. Optimize initial sintering process parameters based on the predicted density and the actual density.

[0168] In one possible implementation, it is determined whether the difference between the predicted density and the actual density is greater than a preset difference; when the difference between the predicted density and the actual density is not greater than the preset difference, it indicates that there is no or almost no deviation in the sintering process, and there is no need to adjust the initial sintering parameters obtained in real time; when the difference between the predicted density and the actual density is greater than the preset difference, it indicates that there is a deviation in the sintering process, and it is necessary to adjust the initial sintering process parameters obtained in real time.

[0169] In a possible implementation, when the difference between the predicted density and the actual density is greater than a preset difference, the initial sintering process parameters are optimized through PID control based on the difference between the predicted density and the actual density.

[0170] PID control (Proportional-Integral-Derivative Control) is a feedback control system widely used in the field of industrial automation and process control. It minimizes the deviation in system response by calculating the error between the set value (expected value) and the actual measured value, and adjusting the control signal according to this error. In the embodiment of the present application, the PID control method is: calculate the difference between the predicted density and the actual density, and adjust / optimize the sintering process parameters according to this difference to reduce the difference between the predicted density and the actual density.

[0171] For easier understanding, the following example illustrates the PID control principle using formula (9).

[0172] (9)

[0173] in, u ( t ) is the optimized sintering process parameter, e ( t ) is the difference between the predicted density and the actual density, K p is the proportional gain parameter, K i is the integral gain parameter, K d is the differential gain parameter, is from time 0 to time t The difference e ( t ), de ( t ) / dt Is the difference e ( t ) is the derivative of .

[0174] Through the PID control method, the sintering process parameters can be dynamically optimized during the sintering process, and the optimized sintering process parameters can be further fed back to the sintering equipment to achieve closed-loop control and ensure that the sintering process continues according to the optimized sintering process parameters (sintering process conditions).

[0175] Above Figure 4One method of optimizing sintering process parameters is to optimize the sintering process parameters according to the target density in the preset requirements before sintering the ceramic, so that the density of the ceramic after sintering in the ideal state can meet the target density. Figure 5 Another optimization method of sintering process parameters introduced is to optimize the sintering process parameters in real time based on the difference between the predicted density and the actual density during the ceramic sintering process, so as to continue to sinter the ceramic material based on the optimized sintering process parameters, realize closed-loop control, and ensure that the density at each stage of the ceramic sintering process meets the predicted requirements.

[0176] The embodiment of the present application provides a ceramic sintering optimization method based on dynamic monitoring, including: synchronously acquiring in situ a sintering data set of a ceramic sample during the sintering process under multiple sintering process conditions; based on the corresponding sintering data set, establishing a densification kinetic model corresponding to the multiple sintering process conditions; based on the densification kinetic model corresponding to the multiple sintering process conditions, determining multiple iso-density lines, and based on the densification kinetic model corresponding to the multiple sintering process conditions and the multiple iso-density lines, constructing a kinetic prediction model, so as to optimize the sintering process parameters based on the kinetic prediction model. The embodiment of the present application relies on dynamic monitoring means, and can measure the mass change rate and size change rate during the sintering process in real time and synchronously (i.e., synchronous in situ), providing comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of the sintering process parameters can be improved, thereby improving the performance of the ceramic material.

[0177] Furthermore, the sintering data during the sintering process can be obtained in situ simultaneously, and repeated sintering is not required, which improves the accuracy of the sintering data, thereby further improving the optimization effect of the sintering process parameters. In addition, the sintering data during the sintering process can be obtained in situ, and the continuity of the obtained sintering data is good, which can further accurately understand or capture the densification mechanism during the sintering process.

[0178] Embodiment 2:

[0179] Combine the following Figure 6 , a ceramic sintering optimization system based on dynamic monitoring provided in an embodiment of the present application is introduced in detail.

[0180] like Figure 6 As shown, a ceramic sintering optimization system based on dynamic monitoring provided in an embodiment of the present application includes the following modules:

[0181] The in-situ synchronous acquisition module 601 is used to synchronously acquire in-situ a sintering data set of a ceramic sample during a sintering process under multiple sintering process conditions; wherein the sintering process conditions are a set of sintering process parameters; the sintering data set corresponds to the multiple sintering process conditions respectively, and the sintering data set includes: multiple sets of sintering data, and one set of sintering data includes: temperature, and mass change rate and size change rate corresponding to the temperature;

[0182] A kinetic analysis module 602 is used to establish a densification kinetic model corresponding to a plurality of sintering process conditions based on a corresponding sintering data set; wherein the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions;

[0183] A kinetic prediction module 603 is used to determine a plurality of iso-density lines based on the densification kinetic models corresponding to the plurality of sintering process conditions, and to construct a kinetic prediction model based on the densification kinetic models corresponding to the plurality of sintering process conditions and the plurality of iso-density lines; wherein the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions;

[0184] The process parameter optimization module 604 is used to optimize the sintering process parameters based on the kinetic prediction model.

[0185] In one possible implementation, the in-situ synchronous acquisition module 601 is specifically used to synchronously in-situ acquire the mass change, size change and consumed time corresponding to multiple preset temperatures during a sintering process of the ceramic sample under the sintering process conditions; determine the mass change rate corresponding to the preset temperature based on the mass change and consumed time corresponding to the preset temperature, and determine the size change rate corresponding to the preset temperature based on the size change and consumed time corresponding to the preset temperature; based on multiple preset temperatures, and the mass change rates and size change rates corresponding to the multiple preset temperatures, obtain a sintering data set corresponding to the sintering process conditions.

[0186] In one possible implementation, the kinetic analysis module 602 is specifically used to construct, based on the multiple groups of sintering data in the corresponding sintering data set, a relationship curve between temperature and mass change rate and a relationship curve between temperature and size change rate corresponding to the multiple sintering process conditions respectively; based on the relationship curve between temperature and mass change rate and the relationship curve between temperature and size change rate corresponding to the multiple sintering process conditions respectively, construct a densification kinetic model corresponding to the multiple sintering process conditions respectively.

[0187] In one possible implementation, the kinetic prediction module 602 is specifically used to obtain the diffusion activation energy corresponding to the multiple isopycnals based on the slopes of the multiple isopycnals; determine the sintering mechanisms that play a dominant role in different stages according to the diffusion activation energy corresponding to the multiple isopycnals, and determine the densification mechanism based on the sintering mechanisms that play a dominant role in different stages; determine the initial kinetic equation corresponding to the densification mechanism based on the densification mechanism, and construct a kinetic prediction model based on the densification kinetic models corresponding to the multiple sintering process conditions and the initial kinetic equation.

[0188] In one possible implementation, the process parameter optimization module 604 is specifically used to construct a mapping model training set based on a kinetic prediction model; wherein the mapping model training set includes: multiple groups of training data, one group of training data includes: mutually corresponding sintering process conditions and density; based on the mapping model training set, a mapping model is trained through machine learning; wherein the mapping model is used to output corresponding sintering process conditions based on density; based on the mapping model, the sintering process parameters are optimized.

[0189] In one possible implementation, the process parameter optimization module 604 is also used to obtain the initial sintering process parameters in real time during the sintering process of the ceramic; based on the kinetic prediction model, obtain the predicted density corresponding to the initial sintering process parameters, and simultaneously obtain the actual sintering data to obtain the corresponding actual density based on the actual sintering data; based on the predicted density and the actual density, optimize the initial sintering process parameters.

[0190] In one possible implementation, the process parameter optimization module 604 is specifically used to optimize the initial sintering process parameters through PID control based on the difference between the predicted density and the actual density when the difference between the predicted density and the actual density is greater than a preset difference.

[0191] The embodiment of the present application provides a ceramic sintering optimization system based on dynamic monitoring, including: an in-situ synchronous acquisition module 601, which is used to synchronously acquire in-situ the sintering data set of the ceramic sample during the sintering process under multiple sintering process conditions; a dynamic analysis module 602, which is used to establish a densification dynamic model corresponding to multiple sintering process conditions based on the corresponding sintering data set; a dynamic prediction module 603, which is used to determine multiple iso-density lines based on the densification dynamic model corresponding to multiple sintering process conditions, and to construct a dynamic prediction model based on the densification dynamic model corresponding to multiple sintering process conditions and multiple iso-density lines; a process parameter optimization module 604, which is used to optimize the sintering process parameters based on the dynamic prediction model. The embodiment of the present application relies on dynamic monitoring means, which can measure the mass change rate and size change rate during the sintering process in real time and synchronously (i.e., synchronous in-situ), and provides comprehensive dynamic monitoring data, so that the densification mechanism during the sintering process can be accurately understood or captured, and the optimization effect of the sintering process parameters can be improved, thereby improving the performance of the ceramic material.

[0192] Furthermore, the sintering data during the sintering process can be obtained in situ simultaneously, and repeated sintering is not required, which improves the accuracy of the sintering data, thereby further improving the optimization effect of the sintering process parameters. In addition, the sintering data during the sintering process can be obtained in situ, and the continuity of the obtained sintering data is good, which can further accurately understand or capture the densification mechanism during the sintering process.

[0193] It should be noted that each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the method and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments. The method and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0194] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A ceramic sintering optimization method based on dynamic monitoring, characterized in that: include: Synchronously acquiring in situ a sintering data set of a ceramic sample during a sintering process under a plurality of sintering process conditions; wherein the sintering process conditions are a set of sintering process parameters; the sintering data set corresponds to the plurality of sintering process conditions respectively, and the sintering data set comprises: a plurality of sets of sintering data, a set of sintering data comprises: temperature, and a mass change rate and a size change rate corresponding to the temperature; Based on the corresponding sintering data set, a densification kinetic model corresponding to each of the multiple sintering process conditions is established; wherein the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions; the densification kinetic model is: a relationship curve between temperature and densification rate; wherein the relationship curve between temperature and densification rate is: a relationship curve between temperature and densification rate with the inverse of temperature as the horizontal axis and the densification rate in logarithmic form as the vertical axis; the densification rate is obtained based on the mass change rate and the size change rate; Based on the densification kinetic models corresponding to the multiple sintering process conditions, a plurality of iso-density lines are determined, and based on the densification kinetic models corresponding to the multiple sintering process conditions and the multiple iso-density lines, a kinetic prediction model is constructed, so as to optimize the sintering process parameters based on the kinetic prediction model; wherein the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions; The method of constructing a kinetic prediction model based on the densification kinetic models corresponding to the multiple process conditions and the multiple isopycnic lines comprises: obtaining the diffusion activation energy corresponding to the multiple isopycnic lines based on the slopes of the multiple isopycnic lines; determining the sintering mechanisms that play a dominant role in different stages according to the diffusion activation energy corresponding to the multiple isopycnic lines, and determining the densification mechanism based on the sintering mechanisms that play a dominant role in different stages; wherein the sintering mechanism comprises: surface diffusion mechanism, grain boundary diffusion mechanism or volume diffusion mechanism; determining the initial kinetic equation corresponding to the densification mechanism based on the densification mechanism, and constructing a kinetic prediction model based on the densification kinetic models corresponding to the multiple sintering process conditions and the initial kinetic equation.

2. The method according to claim 1, characterized in that The sintering data set corresponding to the sintering process conditions is obtained in the following manner: Synchronously acquiring in situ the mass change, size change and consumption time corresponding to multiple preset temperatures during a sintering process of the ceramic sample under the sintering process conditions; wherein the consumption time is the difference between the start time of the sintering process and the time of sintering to the preset temperature; Based on the mass change amount and the consumed time corresponding to the preset temperature, determine the mass change rate corresponding to the preset temperature, and based on the size change amount and the consumed time corresponding to the preset temperature, determine the size change rate corresponding to the preset temperature; Based on a plurality of preset temperatures and mass change rates and dimensional change rates corresponding to the plurality of preset temperatures, a sintering data set corresponding to the sintering process conditions is obtained.

3. The method according to claim 1, characterized in that The densification kinetic models corresponding to the plurality of sintering process conditions are established based on the corresponding sintering data set, including: Based on the multiple groups of sintering data in the corresponding sintering data set, constructing the relationship curves between temperature and mass change rate and the relationship curves between temperature and size change rate respectively corresponding to the multiple sintering process conditions; Based on the relationship curves between temperature and mass change rate and the relationship curves between temperature and size change rate respectively corresponding to the plurality of sintering process conditions, densification kinetic models respectively corresponding to the plurality of sintering process conditions are constructed.

4. The method according to claim 1, characterized in that The kinetic prediction model is expressed by the following formula: ； in, Z ( t ) is the time t The corresponding density is t 1 is the start time of the temperature cycle, t 2 is the end time of the temperature cycle, Z ( t 1) is the density corresponding to the start time, Z ( t 2) is the density corresponding to the end time, Z m and Z n are adjacent isodensities, for Z m The densification rate corresponding to the isopycnal lines is for Z n Densification rates corresponding to isopycnals.

5. The method according to claim 1, characterized in that: The step of optimizing the sintering process parameters based on the kinetic prediction model includes: Based on the kinetic prediction model, a mapping model training set is constructed; wherein the mapping model training set includes: multiple groups of training data, one group of training data includes: corresponding sintering process conditions and density; Based on the mapping model training set, a mapping model is trained through machine learning; wherein the mapping model is used to output corresponding sintering process conditions based on density; Based on the mapping model, the sintering process parameters are optimized.

6. The method according to claim 1, characterized in that The kinetic prediction model is used to output the corresponding predicted density based on the input sintering process parameters; The step of optimizing the sintering process parameters based on the kinetic prediction model includes: During the sintering process of ceramics, the initial sintering process parameters are obtained in real time; Based on the kinetic prediction model, a predicted density corresponding to the initial sintering process parameters is obtained, and actual sintering data is simultaneously acquired to obtain a corresponding actual density based on the actual sintering data; Based on the predicted density and the actual density, the initial sintering process parameters are optimized.

7. The method according to claim 6, characterized in that The optimizing the initial sintering process parameters based on the predicted density and the actual density includes: When the difference between the predicted density and the actual density is greater than a preset difference, the initial sintering process parameters are optimized through PID control based on the difference between the predicted density and the actual density.

8. A ceramic sintering optimization system based on dynamic monitoring, characterized in that: include: An in-situ synchronous acquisition module is used to synchronously acquire in-situ a sintering data set of a ceramic sample during a sintering process under multiple sintering process conditions; wherein the sintering process conditions are a set of sintering process parameters; the sintering data set corresponds to the multiple sintering process conditions respectively, and the sintering data set includes: multiple sets of sintering data, one set of sintering data includes: temperature, and mass change rate and size change rate corresponding to the temperature; A kinetic analysis module, for establishing, based on the corresponding sintering data set, a densification kinetic model corresponding to each of the plurality of sintering process conditions; wherein the densification kinetic model is used to describe the relationship between the temperature and the densification rate of the ceramic sample during the sintering process under the corresponding sintering process conditions; the densification kinetic model is: a relationship curve between temperature and densification rate; wherein the relationship curve between temperature and densification rate is: a relationship curve between temperature and densification rate with the inverse of temperature as the horizontal axis and the densification rate in logarithmic form as the vertical axis; the densification rate is obtained based on the mass change rate and the size change rate; A kinetic prediction module, for determining a plurality of isopycnals based on the densification kinetic models respectively corresponding to the plurality of sintering process conditions, and constructing a kinetic prediction model based on the densification kinetic models respectively corresponding to the plurality of sintering process conditions and the plurality of isopycnals; wherein the kinetic prediction model is used to predict the densification behavior during the sintering process under different sintering process conditions; A process parameter optimization module, used for optimizing the sintering process parameters based on the kinetic prediction model; The kinetic prediction module is specifically used to obtain the diffusion activation energy corresponding to the multiple isopycnals based on the slopes of the multiple isopycnals; determine the sintering mechanisms that play a dominant role in different stages according to the diffusion activation energy corresponding to the multiple isopycnals, and determine the densification mechanism based on the sintering mechanisms that play a dominant role in different stages; wherein the sintering mechanism includes: surface diffusion mechanism, grain boundary diffusion mechanism or volume diffusion mechanism; based on the densification mechanism, determine the initial kinetic equation corresponding to the densification mechanism, and construct a kinetic prediction model based on the densification kinetic models corresponding to the multiple sintering process conditions and the initial kinetic equation.

Images