Methods and systems for testing the rheological properties of sanitary ceramic slurry

By acquiring microscopic morphology images and rheological response data through intelligent sensing and acquisition components, establishing probability mapping, and generating a comprehensive anomaly index, the problem of anomaly judgment stability in the detection of rheological properties of sanitary ceramic slurry is solved, and high-precision molding quality control is achieved.

CN123111885APending Publication Date: 2026-11-17JIANGXI VOCATIONAL & TECH COLLEGE OF CERAMIC ARTS & CRAFTS +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611166449.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-03
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing methods for testing the rheological properties of sanitary ceramic slurry lack stability in determining anomalies under different batches, temperatures, and shear histories, making it difficult to meet the high-precision molding quality control requirements of diverse ceramic sanitary ware product designs.

Method used

By acquiring microscopic morphology images through intelligent sensing and acquisition components, extracting microscopic structural feature parameters, and combining them with measured macroscopic rheological response data, a probability mapping from microscopic structural feature parameters to macroscopic rheological feature parameters is established, a comprehensive anomaly index is generated, and rheological performance anomalies are determined.

Benefits of technology

This improves the stability of determining abnormal rheological properties of sanitary ceramic slurry under different temperature and shear history conditions, ensuring the reliability and consistency of molding quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111885A_ABST
    Figure CN123111885A_ABST
Patent Text Reader

Abstract

This application provides a method and system for testing the rheological properties of sanitary ceramic slurry. The method involves acquiring microscopic morphology images of a batch of slurry to be tested using an intelligent sensing acquisition component, and extracting particle dispersion state and particle boundary neighborhood texture parameters. The batch of slurry is divided into independent equal samples, and the same shear test is performed at different target temperatures. Measured macroscopic rheological characteristic parameters are obtained through the intelligent sensing acquisition component. A probabilistic mapping from microscopic structural characteristic parameters to macroscopic rheological characteristic parameters is established to obtain the expected value and prediction uncertainty of macroscopic rheological properties. A comprehensive anomaly index is generated based on the dimensionless deviation of the measured macroscopic rheological characteristic parameters relative to the expected value and the process baseline. An anomaly is determined when the index exceeds the upper limit for anomaly determination determined by independent verification samples. This application can improve the stability of anomaly determination in the rheological properties of sanitary ceramic slurry under different temperature and shear history conditions, and can meet the high-precision molding quality control requirements of diverse ceramic sanitary ware product designs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of rheological property testing technology, and more specifically, to a method and system for testing the rheological properties of sanitary ceramic slurry. Background Technology

[0002] Rheological property testing is an analytical method used to characterize the flow and deformation behavior of materials under external forces. By measuring key parameters such as viscosity, storage modulus, loss modulus, and yield stress using equipment such as rotational rheometers and capillary rheometers, the response of the material's internal structure to conditions such as temperature, shear rate, or time can be revealed. This testing technology is widely used in fields such as polymer melts, coatings, food, cosmetics, and biomaterials.

[0003] Sanitary ceramic products have complex curved surfaces, and the slip casting process for sanitary ceramics requires stability in the fluidity, thixotropic recovery, and viscoelasticity of the slurry. These factors directly determine core indicators for ceramic appearance design and mass production, such as the fullness of the formed body, demolding effect, fidelity of curved surface texture, uniformity of wall thickness in thin-walled large pieces, and the smoothness of the finished glaze. Current testing methods typically measure individual parameters such as viscosity and modulus at a single temperature and compare the measured values ​​with empirical limits. However, due to the thixotropic nature of sanitary ceramic slurries, test results are affected by sampling differences, sample temperature, and shear history. When the particle dispersion state of the same batch of slurry changes, individual macroscopic parameters may not have exceeded empirical limits, easily leading to design defects such as material shortages in complex curved components, air bubbles at corners, and incomplete relief patterns. Normal fluctuations caused by changes in testing conditions may also be judged as abnormal. Therefore, current sanitary ceramic slurry testing methods suffer from insufficient stability in anomaly detection under different batches, temperatures, and shear histories, making it difficult to meet the high-precision molding quality control requirements of diverse ceramic sanitary ware product designs. Summary of the Invention

[0004] This application provides a method and system for testing the rheological properties of sanitary ceramic slurry, which improves the stability of the determination of abnormal rheological properties of sanitary ceramic slurry under different temperature and shear history conditions.

[0005] In a first aspect, this application provides a method for testing the rheological properties of sanitary ceramic slurry, applied to a sanitary ceramic slurry testing system equipped with intelligent sensing and acquisition components, comprising the following steps: The microscopic morphology images of the same batch of sanitary ceramic slurry to be tested are obtained by the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and the microscopic structural feature parameters characterizing the dispersion state of solid particles and the texture of the particle boundary neighborhood are extracted. The same batch of sanitary ceramic slurry to be tested was divided into independent equal samples, and the same shear test was performed at different target temperatures. The measured macroscopic rheological response data was obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component, and the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate were determined to form the measured macroscopic rheological characteristic parameters. Based on paired samples consisting of microstructural characteristic parameters of historical batches of mud, measured macrorheological characteristic parameters, and molding quality labels, a probability mapping from microstructural characteristic parameters to macrorheological characteristic parameters is established to obtain the expected macrorheological value and prediction uncertainty corresponding to the current microstructural characteristic parameters. Based on the dimensionless deviation of the measured macrorheological characteristic parameters relative to the expected macrorheological value and the current molding process benchmark, a comprehensive anomaly index that increases with the increase of the dimensionless deviation is generated. When the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process, the rheological properties of the sanitary ceramic slurry are determined to be abnormal.

[0006] In some embodiments, acquiring microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing acquisition component specifically includes: Low-temperature scanning electron microscopy was used to acquire cross-sectional images of rapidly frozen and fixed samples, or confocal laser scanning microscopy was used to acquire images of sealed wet samples. The sample image is enhanced and the background is corrected according to the calibration parameters corresponding to the microscopic imaging method used, so as to obtain the microscopic morphology image.

[0007] In some embodiments, extracting microstructural feature parameters characterizing the dispersion state of solid particles and the texture of particle boundary neighborhoods specifically includes: The microscopic morphology image is decomposed into multiple scales and segmented by thresholding to identify solid particle regions, and the neighborhood with a preset width outside the particle boundary is determined as the particle boundary neighborhood. Based on the solid-phase particle region, the particle area fraction and average proximity distance are calculated to form particle dispersion parameters; Gray-level co-occurrence matrix texture analysis is performed on the particle boundary neighborhood to extract contrast, energy and homogeneity, forming particle boundary neighborhood texture parameters. The particle dispersion parameters and the particle boundary neighborhood texture parameters are combined to form the microstructure feature parameters.

[0008] In some embodiments, the same batch of sanitary ceramic slurry to be tested is divided into independent equal samples, and the same shear test is performed on each sample at different target temperatures. The measured macroscopic rheological response data is obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component, specifically including: After uniformly mixing the sanitary ceramic slurry from the same batch, take at least three independent equal portions of the sample and load them into the measuring fixture of the rotational rheometer. Each independent sample was stabilized at a target temperature, and the preset temperature stabilization condition was confirmed by a temperature sensor in contact with the sample. The same pre-shear, static recovery, and shear test sequence was performed on each independent aliquot of the sample. The response data generated by the shear test sequence at each target temperature are summarized to form the measured macroscopic rheological response data.

[0009] In some embodiments, determining the structural recovery parameters, storage modulus, loss modulus, and apparent viscosity at the process shear rate specifically includes: Based on the baseline rheological response before high shear failure and the recovered rheological response after a preset recovery time following the cessation of high shear failure, the ratio of the recovered value to the baseline value for the same rheological variable is calculated to obtain the dimensionless structural recovery parameters. After the preset recovery time, a frequency scan is performed in the linear viscoelastic region to determine the energy storage modulus and the loss modulus; The apparent viscosity at the process shear rate is determined by scanning the steady-state shear rate, and the structural recovery parameter, the storage modulus, the loss modulus, and the apparent viscosity are combined to form the measured macroscopic rheological characteristic parameters.

[0010] In some embodiments, establishing a probabilistic mapping from microstructural characteristic parameters to macroscopic rheological characteristic parameters specifically includes: The paired samples were acquired, and the microstructural characteristic parameters and measured macroscopic rheological characteristic parameters of each paired sample were determined under the same image acquisition conditions and rheological testing conditions. The paired samples are divided into training samples, calibration samples, and independent validation samples; The probabilistic regression model is trained using the microstructural feature parameters of the training samples as input and the measured macroscopic rheological feature parameters as output, and the prediction uncertainty of the probabilistic regression model is calibrated using calibration samples. By fixing the calibrated model parameters, the current microstructural characteristic parameters are input into the probabilistic regression model to obtain the macrorheological expectation value and the prediction uncertainty.

[0011] In some embodiments, generating a comprehensive anomaly index based on the dimensionless deviation of the measured macroscopic rheological characteristic parameters relative to the expected macroscopic rheological value and the current molding process baseline specifically includes: The standardized deviation between the measured macroscopic rheological characteristic parameters and the expected macroscopic rheological values ​​is calculated to obtain the cross-scale consistency deviation. The structural response deviation is obtained by calculating the deviations of the structural recovery parameters and the ratio of the loss modulus to the energy storage modulus relative to the current molding process reference. The process matching deviation is obtained by calculating the distance by which the apparent viscosity at the process shear rate exceeds the allowable range of the current molding process. The cross-scale consistency deviation, the structural response deviation, and the process matching deviation are converted into dimensionless values, where a larger value indicates a larger deviation. The values ​​are then weighted using weights determined and fixed by historical paired samples of the current molding process to obtain the comprehensive anomaly index.

[0012] In some embodiments, determining the upper limit of the anomaly determination and determining the normal state of the rheological properties of the sanitary ceramic slurry specifically includes: Select the corresponding sample from the independent verification samples according to the current molding process type, and determine the anomaly judgment upper limit based on the comprehensive anomaly index distribution of normal samples and the preset false alarm upper limit; When the comprehensive anomaly index does not exceed the upper limit of the anomaly judgment, the rheological properties of the sanitary ceramic slurry are determined to be normal.

[0013] In some embodiments, the measuring fixture is a coaxial cylindrical structure or a conical plate structure.

[0014] Secondly, this application provides a system for testing the rheological properties of sanitary ceramic slurry, comprising: The extraction module is used to acquire microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and extract microscopic structural feature parameters that characterize the dispersion state of solid particles and the texture of particle boundary neighborhoods. The processing module is used to divide the same batch of sanitary ceramic slurry into independent equal samples, perform the same shear test at different target temperatures, acquire measured macroscopic rheological response data through the rheological response data acquisition unit in the intelligent sensing acquisition component, and determine the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate to form measured macroscopic rheological characteristic parameters. The processing module is further configured to establish a probability mapping from microstructural feature parameters to macroscopic rheological feature parameters based on paired samples consisting of microstructural feature parameters of historical batches of mud, measured macroscopic rheological feature parameters, and molding quality labels, to obtain the expected macroscopic rheological value and prediction uncertainty corresponding to the current microstructural feature parameter; and to generate a comprehensive anomaly index that increases with the increase of the dimensionless deviation based on the dimensionless deviation of the measured macroscopic rheological feature parameter relative to the expected macroscopic rheological value and the current molding process benchmark. The execution module is used to determine that the rheological properties of the sanitary ceramic slurry are abnormal when the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The method and system for testing the rheological properties of sanitary ceramic slurry provided in this application acquires microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing and acquisition component, and extracts microstructural feature parameters characterizing the dispersion state of solid particles and the texture of particle boundary neighborhoods; the same batch of sanitary ceramic slurry is divided into independent equal samples, and the same shear test is performed on each sample at different target temperatures; the measured macroscopic rheological response data is acquired through the rheological response data acquisition unit in the intelligent sensing and acquisition component, and the structural recovery parameter, storage modulus, loss modulus, and apparent viscosity at the process shear rate are determined to form measured macroscopic rheological characteristics. The parameters are as follows: Based on paired samples consisting of microstructural characteristic parameters of historical batches of slurry, measured macroscopic rheological characteristic parameters, and molding quality labels, a probability mapping from microstructural characteristic parameters to macroscopic rheological characteristic parameters is established to obtain the expected macroscopic rheological value and prediction uncertainty corresponding to the current microstructural characteristic parameters; Based on the dimensionless deviation of the measured macroscopic rheological characteristic parameters relative to the expected macroscopic rheological value and the current molding process benchmark, a comprehensive anomaly index is generated that increases with the increase of the dimensionless deviation; When the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by the independent verification samples of the current molding process, it is determined that the rheological properties of the sanitary ceramic slurry are abnormal.

[0016] Independent sampling within the same batch and identical shear tests ensure comparability of responses at different target temperatures and reduce the influence of previous temperatures and shear processes on subsequent test results. Structural recovery parameters, modulus, and apparent viscosity are all determined by the measured response of the current sample. Probability mapping only provides the macroscopic rheological expectation value and prediction uncertainty corresponding to the microstructure and does not replace measured results. Dimensionless deviations are weighted according to the direction where larger values ​​indicate higher anomalies, avoiding cancellation of evaluation directions. The upper limit for anomaly judgment is determined by independently verified samples of the current molding process, and the judgment results can be verified under preset false alarm constraints. Therefore, this application can improve the stability of anomaly judgment in the rheological properties of sanitary ceramic slurry under different temperature and shear history conditions. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a method for testing the rheological properties of sanitary ceramic slurry according to some embodiments of this application; Figure 2 This is an exemplary flowchart illustrating the determination of microstructural feature parameters according to some embodiments of this application; Figure 3 This is an exemplary flowchart illustrating the determination of probability mappings according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a sanitary ceramic slurry rheological property testing system according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a method for testing the rheological properties of sanitary ceramic slurry, according to some embodiments of this application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a method for testing the rheological properties of sanitary ceramic slurry according to some embodiments of this application. The method for testing the rheological properties of sanitary ceramic slurry mainly includes the following steps: The method for detecting the rheological properties of sanitary ceramic slurry in this embodiment is applied to a sanitary ceramic slurry detection system equipped with an intelligent sensing and acquisition component. The intelligent sensing and acquisition component includes a microscopic imaging data acquisition unit and a rheological response data acquisition unit; the microscopic imaging data acquisition unit is used to acquire microscopic morphological images output by a microscopic imaging device, and the rheological response data acquisition unit is used to acquire measured macroscopic rheological response data output by a rotating rheometer.

[0020] In step 101, the microscopic morphology image of the same batch of sanitary ceramic slurry to be tested is acquired by the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and the microscopic structural feature parameters characterizing the dispersion state of solid particles and the texture of the particle boundary neighborhood are extracted. In some embodiments, acquiring microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing acquisition component can be achieved by the following steps: Low-temperature scanning electron microscopy was used to acquire cross-sectional images of rapidly frozen and fixed samples, or confocal laser scanning microscopy was used to acquire images of sealed wet samples. The sample image is enhanced and the background is corrected according to the calibration parameters corresponding to the microscopic imaging method used, so as to obtain the microscopic morphology image.

[0021] In practice, a sample preparation and imaging route can be selected based on the microscopic imaging equipment. When using a cryogenic scanning electron microscope, samples are taken from the same batch of mud and rapidly frozen for fixation. Sample cross-sections are prepared and images are acquired under cryogenic conditions. Low vacuum conditions are used only for this electron microscopy imaging route; pores or cracks formed during drying are not considered as wet liquid phase interface features. When using a confocal laser scanning microscope, wet mud is placed in a sealed sample cell. Fluorescence or reflection channels are set according to the imaging differences between solid particles and the liquid phase, and images are acquired under normal pressure and sealed conditions. The two routes use corresponding pixel scales, illumination backgrounds, and grayscale response calibration parameters, respectively. The same route is used for the same batch of samples, and multiple fields of view are randomly acquired at a fixed sampling depth.

[0022] It should be noted that the microscopic morphology images are used to characterize the area ratio, proximity distance, and grayscale texture of the particle boundary neighborhood of solid particles. Texture parameters obtained from dry cross-sections or wet optical images do not directly represent the bonding strength between particles and the liquid phase, the thickness of the adsorbed layer, or the strength of the hydration film.

[0023] In some embodiments, reference Figure 2 The figure is an exemplary flowchart for determining microstructural feature parameters in some embodiments of this application. In this embodiment, the extraction of microstructural feature parameters characterizing the dispersion state of solid particles and the texture of the particle boundary neighborhood can be achieved by the following steps: In step 1011, the microscopic morphology image is decomposed into multiple scales and segmented by thresholding to identify solid particle regions, and the neighborhood with a preset width outside the particle boundary is determined as the particle boundary neighborhood. In step 1012, the particle area fraction and average proximity distance are calculated based on the solid particle region to form particle dispersion parameters; In step 1013, gray-level co-occurrence matrix texture analysis is performed on the particle boundary neighborhood to extract contrast, energy and homogeneity, forming particle boundary neighborhood texture parameters. In step 1014, the particle dispersion parameters and the particle boundary neighborhood texture parameters are combined into the microstructure feature parameters.

[0024] In practice, random noise is first suppressed and slowly changing backgrounds are corrected according to the calibration parameters of the corresponding microscopic imaging mode. Then, a segmentation threshold is determined based on the grayscale distribution of the entire image and the local neighborhood. Connected regions that meet the particle grayscale conditions and whose area is not less than the minimum area corresponding to the device resolution are identified as solid particle regions. The boundaries of each solid particle region are expanded outward by a preset pixel width, and the solid particle region is subtracted to obtain the particle boundary neighborhood. The preset pixel width is set according to the pixel scale calibration results and remains unchanged within the same detection batch.

[0025] In addition, in specific implementation, the calculation of the area fraction and average neighbor distance of the particle group based on the solid particle region, as the first parameter set characterizing the uniformity of solid particle dispersion in the mud, can be achieved in the following way: The Thiessen polygon graph analysis method is used to quantify the spatial distribution heterogeneity among particles. The centroid coordinates of each particle are extracted as generators of the Thiessen polygons, constructing a Thiessen polygon mesh covering the entire image region. The coefficient of variation of the area of ​​each polygon unit and the standard deviation of the polygon side length distribution are calculated to quantify the spatial distribution uniformity of the particle group. Simultaneously, the proportion of the total area of ​​solid particles to the total area of ​​the image field of view is statistically analyzed to obtain the area fraction of the particle group; the centroid distance between each particle and its nearest neighbor particle is calculated, and the arithmetic mean of this distance value for all particles is taken to obtain the average neighbor distance of the particles. The area fraction and average neighbor distance are used as the first parameter set characterizing the uniformity of solid particle dispersion in the mud, where the area fraction reflects the particle filling density and the average neighbor distance reflects the uniformity of particle spacing.

[0026] In addition, in the specific implementation, the gray-level co-occurrence matrix of the particle boundary neighborhood is statistically analyzed in multiple preset directions, and the contrast, energy, and homogeneity of each direction are averaged to form the particle boundary neighborhood texture parameter. This parameter only characterizes the gray-level variation and texture uniformity near the particle boundary and is not used to directly infer the bonding strength between the particles and the liquid phase.

[0027] In other embodiments, methods such as radial distribution function or local binary mode can also be used to extract micro-features, and this application does not limit this to these methods.

[0028] It should be noted that the particle dispersion parameters include particle area fraction and average neighbor distance, used to characterize the proportion and spatial spacing of solid particles within the current imaging field of view. The particle boundary neighborhood texture parameters include contrast, energy, and homogeneity, used to characterize grayscale changes near the particle boundaries. The microstructure feature parameters consist of the two types of parameters mentioned above that can be directly calculated from the calibration image.

[0029] In step 102, the batch of sanitary ceramic slurry to be tested is divided into independent equal samples, and the same shear test is performed at different target temperatures. The measured macroscopic rheological response data is obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component, and the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate are determined to form measured macroscopic rheological characteristic parameters. In some embodiments, the same batch of sanitary ceramic slurry to be tested is divided into independent equal samples, and the same shear test is performed on each sample at different target temperatures. The measured macroscopic rheological response data can be obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component by the following steps: After uniformly mixing the sanitary ceramic slurry from the same batch, take at least three independent equal portions of the sample and load them into the measuring fixture of the rotational rheometer. Each independent sample was stabilized at a target temperature, and the preset temperature stabilization condition was confirmed by a temperature sensor in contact with the sample. The same pre-shear, static recovery, and shear test sequence was performed on each independent aliquot of the sample. The response data generated by the shear test sequence at each target temperature are summarized to form the measured macroscopic rheological response data.

[0030] In practice, the sanitary ceramic slurry from the same batch is first thoroughly and uniformly mixed. Then, at least three independent aliquots of the same mass or volume are taken, with each target temperature using an independent sample that has not undergone testing at other target temperatures. The measuring fixture is a coaxial cylindrical structure or a conical plate structure. The appropriate structure is selected based on the particle size and solid content, and a rough surface or anti-slip fixture is used to reduce the influence of wall slippage. The temperature sensor is positioned close to the measuring gap and in contact with the sample. When the difference between the measured sample temperature and the target temperature remains within a preset tolerance for a preset duration, the temperature stability condition is confirmed.

[0031] It should be noted that each independent aliquot of sample uses the same loading method, pre-shear rate, pre-shear time, and settling recovery time. The target temperature is set according to the ambient temperature range allowed by the sanitary ceramic grouting process; different target temperatures are not continuously increased or decreased on the same sample to avoid the temperature history and previous shear history from being mixed with the target temperature effect.

[0032] In practice, after each independent aliquot of samples reaches the target temperature, the baseline rheological response before high-shear failure is first recorded under specified low-shear conditions, and then specified high-shear conditions are applied. After the high-shear condition is stopped, the sample is restored to the low-shear condition, and the recovered rheological response is recorded after a preset recovery time. Frequency scanning is performed within the pre-measured linear viscoelastic region, and steady-state shear rate scanning covers the shear rate corresponding to the current grouting process. If the high-shear process affects subsequent frequency scanning or steady-state scanning, parallel subsamples are taken from the aliquots corresponding to the same target temperature for separate testing, while maintaining consistent pre-shear and resting conditions.

[0033] It should be noted that the measured macroscopic rheological response data includes shear stress, shear rate, apparent viscosity, storage modulus, loss modulus, and response data as a function of recovery time, which are directly collected under specified temperature, sample loading, pre-shearing, and static conditions. The sample temperature and test time are also saved.

[0034] In some embodiments, determining the structural recovery parameters, storage modulus, loss modulus, and apparent viscosity at the process shear rate can be achieved by the following steps: Based on the baseline rheological response before high shear failure and the recovered rheological response after a preset recovery time following the cessation of high shear failure, the ratio of the recovered value to the baseline value for the same rheological variable is calculated to obtain the dimensionless structural recovery parameters. After the preset recovery time, a frequency scan is performed in the linear viscoelastic region to determine the energy storage modulus and the loss modulus; The apparent viscosity at the process shear rate is determined by scanning the steady-state shear rate, and the structural recovery parameter, the storage modulus, the loss modulus, and the apparent viscosity are combined to form the measured macroscopic rheological characteristic parameters.

[0035] In practice, the structural recovery parameters are directly determined based on the measured rheological responses before and after high-shear failure. A thixotropic rheological constitutive model incorporating structural parameters can also be used to simultaneously fit steady-state flow data and transient failure and recovery data to check the consistency of the test curves. The fitting results of the constitutive model do not replace the measured macroscopic rheological characteristic parameters of the current sample, nor do they independently invert the structural recovery rate and structural failure rate based on the steady-state flow curve.

[0036] In practice, under the same low-shear measurement conditions before and after high-shear failure, the same flow variable is selected as the comparison quantity. The recovery value after a preset recovery time is divided by the baseline value before high-shear failure to obtain the dimensionless structural recovery parameter. The flow variable can be apparent viscosity or shear stress, but the flow variable, low-shear conditions, and recovery time are consistent for the same test batch.

[0037] In addition, in specific implementation, after a preset recovery time, a frequency scan is performed within the linear viscoelastic region. The storage modulus and loss modulus are determined based on the in-phase and quadrature components of the stress response relative to the input strain, respectively. Based on the measured results of the steady-state shear rate scan, the apparent viscosity at the process shear rate is read or interpolated. The structural recovery parameters, the storage modulus and loss modulus at the reference frequency, and the apparent viscosity at the process shear rate are combined in a fixed order to form the measured macroscopic rheological characteristic parameters.

[0038] It should be noted that the structural recovery parameter is the ratio of the recovered value to the baseline value of the same rheological variable, and is dimensionless, used to characterize the thixotropic recovery capacity of the mud. The storage modulus and loss modulus use the same pressure unit, and their ratio is dimensionless, used to characterize the elastic and viscous response of the mud. The apparent viscosity uses the uniform viscosity unit output by the rheometer. All parameters are derived from direct testing of the current sample. The output of the probability mapping does not replace the above measured values, but serves as a measured benchmark for comparison with the macroscopic rheological expectation value corresponding to the microstructure and for anomaly detection.

[0039] In step 103, a probability mapping from microstructural feature parameters to macroscopic rheological feature parameters is established based on paired samples consisting of microstructural feature parameters of historical batches of mud, measured macroscopic rheological feature parameters, and molding quality labels. This obtains the expected macroscopic rheological value and prediction uncertainty corresponding to the current microstructural feature parameter. Based on the dimensionless deviation of the measured macroscopic rheological feature parameter relative to the expected macroscopic rheological value and the current molding process benchmark, a comprehensive anomaly index that increases with the increase of the dimensionless deviation is generated. In some embodiments, reference Figure 3 The figure is an exemplary flowchart of determining the probability mapping in some embodiments of this application. In this embodiment, the probability mapping from microstructural feature parameters to macroscopic rheological feature parameters can be achieved by the following steps: In step 1031, the paired samples are acquired, and the microstructural characteristic parameters and measured macroscopic rheological characteristic parameters of each paired sample are determined according to the same image acquisition conditions and rheological testing conditions. In step 1032, the paired samples are divided into training samples, calibration samples, and independent validation samples; In step 1033, the microstructural feature parameters of the training samples are used as input and the measured macroscopic rheological feature parameters are used as output to train the probabilistic regression model, and the prediction uncertainty of the probabilistic regression model is calibrated using calibration samples. In step 1034, the calibrated model parameters are fixed, and the current microstructural characteristic parameters are input into the probabilistic regression model to obtain the macroscopic rheological expectation value and the prediction uncertainty.

[0040] In practice, multiple batches of slurry using the same raw material system and similar molding process are selected from the historical production database. For each historical batch, microstructural characteristic parameters and measured macroscopic rheological characteristic parameters are obtained under the aforementioned imaging and rheological testing conditions. Molding quality labels are then generated based on the cracking, deformation, thickness uniformity, or demolding quality of the corresponding billet, thus forming a one-to-one paired sample. The microscopic image data and macroscopic rheological data must come from the same historical batch; paired samples are not created by piecing together statistical means from different batches.

[0041] In addition, in specific implementation, training samples, calibration samples, and independent validation samples are divided according to production batches, and data from the same production batch only enters one of these sets. Each feature is standardized using the center value and scale value determined by the training samples; the calibration samples and independent validation samples use the same center value and scale value without recalculation.

[0042] In practical implementation, Gaussian process regression can be used as a probabilistic regression model. The microstructural characteristic parameters of the training samples are used as input, and the corresponding structural recovery parameters, storage modulus, loss modulus, and apparent viscosity are used as outputs. The model learns the nonlinear mapping between the input and output. The model hyperparameters are determined only using the training samples, and the prediction interval is calibrated based on the distribution of prediction errors in the calibration samples, ensuring that the prediction uncertainty reflects the residual range in unseen batches.

[0043] In practice, after uncertainty calibration, the standardized parameters of the features and the parameters of the probabilistic regression model are fixed. The microstructural feature parameters of the current sample are input into the fixed model to obtain the expected values ​​and prediction uncertainties of each macroscopic rheological feature. The length scale of the model kernel function can be used to check the input sensitivity, but it is not used again as weights to embed into the already trained kernel function, nor is it used as a conclusion on the causal contribution of microscopic features to the macroscopic response.

[0044] It should be noted that the probability mapping is a conditional statistical relationship supported by historical paired samples, used to provide the expected macro-rheological value and prediction uncertainty corresponding to the current microstructural characteristics. The expected macro-rheological value is used to measure whether the current measured macro-rheological response deviates from the normal level corresponding to the microstructure, and the prediction uncertainty characterizes the statistically reliable fluctuation range of the expected value. This mapping does not replace the rheological testing of the current sample and is not used to reverse-determine the formulation adjustment range based on the target macroscopic parameters.

[0045] In some embodiments, generating a comprehensive anomaly index based on the dimensionless deviation of the measured macroscopic rheological characteristic parameters relative to the expected macroscopic rheological value and the current molding process baseline can be achieved through the following steps: The standardized deviation between the measured macroscopic rheological characteristic parameters and the expected macroscopic rheological values ​​is calculated to obtain the cross-scale consistency deviation. The structural response deviation is obtained by calculating the deviations of the structural recovery parameters and the ratio of the loss modulus to the energy storage modulus relative to the current molding process reference. The process matching deviation is obtained by calculating the distance by which the apparent viscosity at the process shear rate exceeds the allowable range of the current molding process. The cross-scale consistency deviation, the structural response deviation, and the process matching deviation are converted into dimensionless values, where a larger value indicates a larger deviation. The values ​​are then weighted using weights determined and fixed by historical paired samples of the current molding process to obtain the comprehensive anomaly index.

[0046] In practice, cross-scale consistency bias is used to represent the degree of inconsistency between the measured macroscopic rheological characteristics of the current sample and the expected values ​​corresponding to its microstructure; structural response bias is used to represent the degree to which the structural recovery parameter and modulus ratio deviate from the normal range of the current molding process; and process matching bias is used to represent the degree to which the apparent viscosity exceeds the allowable range of the process at the process shear rate. All three biases are represented by zero to indicate that no corresponding bias was observed, and increase with the degree of bias.

[0047] In practice, for each measured macroscopic rheological characteristic, the absolute difference between it and the corresponding expected value is calculated, and standardized using a calibrated prediction uncertainty. When the prediction uncertainty is below the lower limit determined by equipment repeatability testing, this lower limit is used to avoid amplifying deviations due to small denominators. The excess distances for the structural recovery parameter and the ratio of loss modulus to storage modulus are calculated relative to the allowable range of normal samples in the current process. When the apparent viscosity is within the allowable range, the process matching deviation is zero; when it exceeds the range, it is dimensionless based on the magnitude of the closest boundary.

[0048] In addition, in practical implementation, for each molding process, the weights of cross-scale consistency deviation, structural response deviation, and process matching deviation are pre-determined using training samples, and the sum of each weight is made equal to one. These weights are fixed after model calibration; the eigenvalues ​​of the current test sample are not used to temporarily change the weights. Different molding processes use their own fixed weights and process benchmarks.

[0049] In practice, the three dimensionless deviations are multiplied by their respective fixed weights and then summed to obtain the comprehensive anomaly index. Since the directions of the deviations are consistent, a larger comprehensive anomaly index indicates a greater degree of deviation of the current sample from the expected microstructure and the molding process baseline.

[0050] It should be noted that cross-scale consistency deviation characterizes the standardized deviation between the measured macroscopic rheological characteristics of the current sample and the expected macroscopic rheological values ​​corresponding to its microstructure. Structural response deviation characterizes the degree to which the structural recovery parameters and the ratio of loss modulus to storage modulus deviate from the normal baseline of the current molding process. Process matching deviation characterizes the distance by which the apparent viscosity at the process shear rate exceeds the allowable range of the current molding process. The comprehensive anomaly index indicates the degree to which the current slurry deviates overall from the expected microstructure and the molding process baseline. The comprehensive anomaly index is used to rank and threshold the deviation of the current sample relative to historical paired samples and the current molding process baseline; it does not represent the absolute quality of the slurry, nor is it directly equivalent to the probability of green body defects. The deviations of each component can be used as subsequent manual verification items.

[0051] In step 104, when the comprehensive anomaly index exceeds the anomaly determination upper limit determined by independent verification samples of the current molding process, it is determined that the rheological properties of the sanitary ceramic slurry are abnormal.

[0052] In some embodiments, the determination of the upper limit of the anomaly assessment and the determination of the normal state of the rheological properties of the sanitary ceramic slurry can be achieved by the following steps: Select the corresponding sample from the independent verification samples according to the current molding process type, and determine the anomaly judgment upper limit based on the comprehensive anomaly index distribution of normal samples and the preset false alarm upper limit; When the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process, the rheological properties of the sanitary ceramic slurry are determined to be abnormal.

[0053] When the comprehensive anomaly index does not exceed the upper limit of the anomaly judgment, the rheological properties of the sanitary ceramic slurry are determined to be normal.

[0054] In practice, corresponding samples are selected from the independent validation samples according to the molding process type, the comprehensive anomaly index of each sample is calculated, and normal samples and abnormal samples are distinguished according to the molding quality label. First, the candidate upper limit is determined based on the acceptable false alarm ratio of normal samples, and then the detection of abnormal samples is checked; if the preset false alarm upper limit and minimum detection requirements cannot be met at the same time, the paired samples are expanded or the detection conditions are adjusted and retraining is performed, without using the current test sample to reverse the adjustment of the anomaly judgment upper limit.

[0055] It should be noted that the anomaly detection upper limit is a one-sided threshold determined for the current molding process, microscopic imaging route, and rheological testing conditions. If there are substantial changes to the microscopic imaging route, measuring fixture, temperature nodes, or shear test sequence, this upper limit should be reconfirmed using new, independent validation samples.

[0056] In addition, in practical implementation, the comprehensive anomaly index of the sample under test is directly compared with the upper limit of the anomaly judgment corresponding to the current molding process. If the comprehensive anomaly index exceeds the upper limit, a rheological performance anomaly warning is output; if it does not exceed the upper limit, a normal rheological performance warning is output. The system also saves the deviation of each component, test conditions, model version, and threshold version for verification.

[0057] It should be noted that when an anomaly is detected, priority review items can be determined based on the magnitude of cross-scale consistency deviation, structural response deviation, and process matching deviation. The component deviations mentioned are only used to prompt for retesting of microscopic images, structural recovery tests, modulus tests, or apparent viscosity tests; they will not automatically infer the cause of the anomaly or generate formula adjustment amounts when independent verification evidence is lacking.

[0058] In another aspect, in some embodiments, this application provides a sanitary ceramic slurry rheological property testing system, which includes an intelligent sensing and acquisition component, comprising a microscopic imaging data acquisition unit and a rheological response data acquisition unit; Reference Figure 4 The figure is a schematic diagram of the structure of a sanitary ceramic slurry rheological property testing system according to some embodiments of this application. The sanitary ceramic slurry rheological property testing system 400 includes: an extraction module 401, a processing module 402, and an execution module 403, which are described below: The extraction module is used to acquire microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and extract microscopic structural feature parameters that characterize the dispersion state of solid particles and the texture of particle boundary neighborhoods. The processing module is used to divide the same batch of sanitary ceramic slurry into independent equal samples, perform the same shear test at different target temperatures, acquire measured macroscopic rheological response data through the rheological response data acquisition unit in the intelligent sensing acquisition component, and determine the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate to form measured macroscopic rheological characteristic parameters. The processing module is further configured to establish a probability mapping from microstructural feature parameters to macroscopic rheological feature parameters based on paired samples consisting of microstructural feature parameters of historical batches of mud, measured macroscopic rheological feature parameters, and molding quality labels, to obtain the expected macroscopic rheological value and prediction uncertainty corresponding to the current microstructural feature parameter; and to generate a comprehensive anomaly index that increases with the increase of the dimensionless deviation based on the dimensionless deviation of the measured macroscopic rheological feature parameter relative to the expected macroscopic rheological value and the current molding process benchmark. The execution module is used to determine that the rheological properties of the sanitary ceramic slurry are abnormal when the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process.

[0059] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for detecting the rheological properties of sanitary ceramic slurry.

[0060] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing a method for testing the rheological properties of sanitary ceramic slurry according to some embodiments of this application. The method for testing the rheological properties of sanitary ceramic slurry in the above embodiments can be achieved through... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0061] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0062] The communication bus 502 can be used to transmit information between the aforementioned components.

[0063] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0064] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0065] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0066] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0067] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0068] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for detecting the rheological properties of sanitary ceramic slurry.

[0069] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0070] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for testing the rheological properties of sanitary ceramic slurry, applied to a sanitary ceramic slurry testing system equipped with intelligent sensing and acquisition components, characterized in that, Includes the following steps: The microscopic morphology images of the same batch of sanitary ceramic slurry to be tested are obtained by the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and the microscopic structural feature parameters characterizing the dispersion state of solid particles and the texture of the particle boundary neighborhood are extracted. The same batch of sanitary ceramic slurry to be tested was divided into independent equal samples, and the same shear test was performed at different target temperatures. The measured macroscopic rheological response data was obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component, and the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate were determined to form the measured macroscopic rheological characteristic parameters. Based on paired samples consisting of microstructural characteristic parameters of historical batches of mud, measured macrorheological characteristic parameters, and molding quality labels, a probability mapping from microstructural characteristic parameters to macrorheological characteristic parameters is established to obtain the expected macrorheological value and prediction uncertainty corresponding to the current microstructural characteristic parameters. Based on the dimensionless deviation of the measured macrorheological characteristic parameters relative to the expected macrorheological value and the current molding process benchmark, a comprehensive anomaly index that increases with the increase of the dimensionless deviation is generated. When the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process, the rheological properties of the sanitary ceramic slurry are determined to be abnormal.

2. The method as described in claim 1, characterized in that, The acquisition of microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing and acquisition component specifically includes: Low-temperature scanning electron microscopy was used to acquire cross-sectional images of rapidly frozen and fixed samples, or confocal laser scanning microscopy was used to acquire images of sealed wet samples. The sample image is enhanced and the background is corrected according to the calibration parameters corresponding to the microscopic imaging method used, so as to obtain the microscopic morphology image.

3. The method as described in claim 1, characterized in that, The extraction of microstructural feature parameters characterizing the dispersion state of solid particles and the texture of particle boundary neighborhoods specifically includes: The microscopic morphology image is decomposed into multiple scales and segmented by thresholding to identify solid particle regions, and the neighborhood with a preset width outside the particle boundary is determined as the particle boundary neighborhood. Based on the solid-phase particle region, the particle area fraction and average proximity distance are calculated to form particle dispersion parameters; Gray-level co-occurrence matrix texture analysis is performed on the particle boundary neighborhood to extract contrast, energy and homogeneity, forming particle boundary neighborhood texture parameters. The particle dispersion parameters and the particle boundary neighborhood texture parameters are combined to form the microstructure feature parameters.

4. The method as described in claim 1, characterized in that, The same batch of sanitary ceramic slurry to be tested was divided into independent equal samples, and the same shear test was performed on each sample at different target temperatures. The measured macroscopic rheological response data was obtained through the rheological response data acquisition unit in the intelligent sensing acquisition component, specifically including: After uniformly mixing the sanitary ceramic slurry from the same batch, take at least three independent equal portions of the sample and load them into the measuring fixture of the rotational rheometer. Each independent sample was stabilized at a target temperature, and the preset temperature stabilization condition was confirmed by a temperature sensor in contact with the sample. The same pre-shear, static recovery, and shear test sequence was performed on each independent aliquot of the sample. The response data generated by the shear test sequence at each target temperature are summarized to form the measured macroscopic rheological response data.

5. The method as described in claim 1, characterized in that, Determining the structural recovery parameters, storage modulus, loss modulus, and apparent viscosity at the process shear rate specifically includes: Based on the baseline rheological response before high shear failure and the recovered rheological response after a preset recovery time following the cessation of high shear failure, the ratio of the recovered value to the baseline value for the same rheological variable is calculated to obtain the dimensionless structural recovery parameters. After the preset recovery time, a frequency scan is performed in the linear viscoelastic region to determine the energy storage modulus and the loss modulus; The apparent viscosity at the process shear rate is determined by scanning the steady-state shear rate, and the structural recovery parameter, the storage modulus, the loss modulus, and the apparent viscosity are combined to form the measured macroscopic rheological characteristic parameters.

6. The method as described in claim 1, characterized in that, Establishing a probabilistic mapping from microstructural characteristic parameters to macroscopic rheological characteristic parameters specifically includes: The paired samples were acquired, and the microstructural characteristic parameters and measured macroscopic rheological characteristic parameters of each paired sample were determined under the same image acquisition conditions and rheological testing conditions. The paired samples are divided into training samples, calibration samples, and independent validation samples; The probabilistic regression model is trained using the microstructural feature parameters of the training samples as input and the measured macroscopic rheological feature parameters as output, and the prediction uncertainty of the probabilistic regression model is calibrated using calibration samples. By fixing the calibrated model parameters, the current microstructural characteristic parameters are input into the probabilistic regression model to obtain the macrorheological expectation value and the prediction uncertainty.

7. The method as described in claim 1, characterized in that, The comprehensive anomaly index is generated based on the dimensionless deviation of the measured macroscopic rheological characteristic parameters relative to the expected macroscopic rheological value and the current molding process benchmark, specifically including: The standardized deviation between the measured macroscopic rheological characteristic parameters and the expected macroscopic rheological values ​​is calculated to obtain the cross-scale consistency deviation. The structural response deviation is obtained by calculating the deviations of the structural recovery parameters and the ratio of the loss modulus to the energy storage modulus relative to the current molding process reference. The process matching deviation is obtained by calculating the distance by which the apparent viscosity at the process shear rate exceeds the allowable range of the current molding process. The cross-scale consistency deviation, the structural response deviation, and the process matching deviation are converted into dimensionless values, where a larger value indicates a larger deviation. The values ​​are then weighted using weights determined and fixed by historical paired samples of the current molding process to obtain the comprehensive anomaly index.

8. The method as described in claim 1, characterized in that, When the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process, the rheological properties of the sanitary ceramic slurry are determined to be abnormal. Specifically, determining the anomaly judgment upper limit and determining the normal state of the rheological properties of the sanitary ceramic slurry include: Select the corresponding sample from the independent verification samples according to the current molding process type, and determine the anomaly judgment upper limit based on the comprehensive anomaly index distribution of normal samples and the preset false alarm upper limit; When the comprehensive anomaly index does not exceed the upper limit of the anomaly judgment, the rheological properties of the sanitary ceramic slurry are determined to be normal.

9. The method as described in claim 4, characterized in that, The measuring fixture is a coaxial cylindrical structure or a conical plate structure.

10. A system for testing the rheological properties of sanitary ceramic slurry, characterized in that, include: The extraction module is used to acquire microscopic morphology images of the same batch of sanitary ceramic slurry through the microscopic imaging data acquisition unit in the intelligent sensing acquisition component, and extract microscopic structural feature parameters that characterize the dispersion state of solid particles and the texture of particle boundary neighborhoods. The processing module is used to divide the same batch of sanitary ceramic slurry into independent equal samples, perform the same shear test at different target temperatures, acquire measured macroscopic rheological response data through the rheological response data acquisition unit in the intelligent sensing acquisition component, and determine the structural recovery parameters, storage modulus, loss modulus and apparent viscosity at the process shear rate to form measured macroscopic rheological characteristic parameters. The processing module is further configured to establish a probability mapping from microstructural feature parameters to macroscopic rheological feature parameters based on paired samples consisting of microstructural feature parameters of historical batches of mud, measured macroscopic rheological feature parameters, and molding quality labels, to obtain the expected macroscopic rheological value and prediction uncertainty corresponding to the current microstructural feature parameter; and to generate a comprehensive anomaly index that increases with the increase of the dimensionless deviation based on the dimensionless deviation of the measured macroscopic rheological feature parameter relative to the expected macroscopic rheological value and the current molding process benchmark. The execution module is used to determine that the rheological properties of the sanitary ceramic slurry are abnormal when the comprehensive anomaly index exceeds the anomaly judgment upper limit determined by independent verification samples of the current molding process.