Pearlescent material production data management system and method based on data integration analysis

Abstract

The present application relates to the technical field of data management, and more particularly to a pearl material production data management system and method based on data integration analysis. The method comprises the following steps: obtaining process original data set of pearl material production line, and performing standardization processing to obtain standardized process data set; obtaining quality detection data set of pearl material; dividing the standardized process data set and the quality detection data set by time interval, and performing batch number marking to obtain process-quality batch data pair; designing modeling sample data set based on the process-quality batch data pair, and constructing quality process coupling model. The present application constructs a closed-loop system integrating data integration, modeling prediction and control optimization, and comprehensively realizes the intelligentization, data-driven and high-quality stable control of pearl material production process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, and in particular to a pearl material production data management system and method based on data integration analysis. BACKGROUND

[0002] Traditional pearl material production mostly adopts intermittent wet deposition process, which is mature in technology, but has problems such as low coating efficiency, uneven particle size control, and large quality fluctuation. In recent years, with the application and popularization of fluidized bed reactors and gas deposition technology in material engineering, pearl material production gradually transforms towards continuous, macro, and intelligent direction, especially in the preparation of gold color light series and silver white, fantasy series products, the fluidized bed-calcination coupling system has become the mainstream path. At present, the existing technology still has the following shortcomings: firstly, the process and quality data sources are scattered, the formats are not unified, and there is a lack of standardized integration framework; secondly, there is a time lag and sample misplacement between process data and detection data, making it difficult to build an accurate correspondence; thirdly, most of the existing prediction models are static regression, lack of real-time response capability, and cannot support closed-loop optimization control; fourthly, there is a lack of abnormal identification mechanism and dynamic adjustment path generation scheme for macro production scenarios, resulting in difficulty in guaranteeing production efficiency and product consistency. SUMMARY

[0003] Therefore, it is necessary to provide a pearl material production data management system and method based on data integration analysis to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, a pearl material production data management method based on data integration analysis comprises the following steps:

[0005] Step S1: obtaining process raw data set of pearl material production line, and performing standardization processing to obtain standardized process data set; obtaining quality detection data set of pearl material; dividing the standardized process data set and the quality detection data set by time interval, and performing batch number marking to obtain process-quality batch data pair;

[0006] Step S2: designing modeling sample data set based on the process-quality batch data pair, and constructing quality process coupling model; extracting the mapping relationship between process control and quality representation in the quality process coupling model to obtain quality process coupling model parameters, wherein the quality representation indexes include surface glossiness, particle size distribution uniformity and coating layer defect density;

[0007] Step S3: real-time acquisition of current production batch process parameters, and quality index prediction analysis of process parameters based on the quality process coupling model parameters to obtain prediction results;

[0008] Step S4: deviation analysis based on the prediction result and the preset quality control standard, identifying abnormal parameters in the current process operation based on the deviation analysis result, and formulating a process optimization adjustment scheme to obtain dynamic process adjustment data.

[0009] The application can realize the unified standardized integration of distributed process parameters and discrete quality detection information, break through the structured data chain between process and quality, improve data utilization efficiency and traceability accuracy; through the introduction of time interval division and batch marking mechanism, effectively solve the data time sequence dislocation problem in the production process, ensure the accurate mapping relationship between process and corresponding detection results; by constructing the coupling relationship between process control parameters and quality characterization indicators, the quality evolution law under the action of complex multi-parameters is extracted, which provides quantitative support for online quality prediction and process control; based on real-time process data, quality index feedforward prediction can be performed to predict deviation trend before product output, identify potential quality problems in advance, shorten feedback cycle and reduce rework rate; through deviation identification and abnormal variable tracking mechanism, the key process nodes causing quality deviation can be quickly located, supporting reason diagnosis and responsibility parameter identification under multi-variables; combined with sensitivity analysis and process boundary constraint, process optimization adjustment path is generated to ensure that the adjustment scheme is implemented within the equipment capacity and process safety range, improving the stability and accuracy of adjustment; the overall method supports dynamic closed-loop control logic in the whole process, solves the problems of strong prediction staticity, control lag and one-sided adjustment in the prior art, and effectively improves the consistency, stability and process transparency of pearl material products under macro-continuous production.

[0010] Preferably, the application also provides a pearl material production data management system based on data integration analysis, which is used to execute the pearl material production data management method based on data integration analysis described above, and the pearl material production data management system based on data integration analysis comprises:

[0011] A data integration module is configured to acquire process original data sets of a pearl material production line, and perform standardization processing to obtain standardized process data sets; acquire quality detection data sets of the pearl material; perform time interval division on the standardized process data sets and the quality detection data sets, and perform batch number marking to obtain process-quality batch data pairs;

[0012] A model construction module is configured to design modeling sample data sets based on the process-quality batch data pairs, and construct a quality process coupling model; extract the mapping relationship between process control and quality characterization in the quality process coupling model to obtain quality process coupling model parameters, wherein the quality characterization indicators include surface gloss, particle size distribution uniformity and coating layer defect density.

[0013] An intelligent prediction module is configured to collect current production batch process parameters in real time, and perform quality index prediction analysis on the process parameters based on a quality process coupling model parameter to obtain a prediction result.

[0014] A regulation optimization module is configured to perform deviation analysis based on the prediction result and a preset quality control standard, identify abnormal parameters in current process operation based on the deviation analysis result, and formulate a process optimization adjustment scheme to obtain dynamic process adjustment data.

[0015] The present application can comprehensively solve the quality instability and efficiency bottleneck problems caused by data dispersion, process fragmentation and response lag in traditional pearl material production. By realizing the integrated collection, format unification and time alignment of process and quality multi-source data, breaking the original information island, significantly improving the data integrity and analyzability, providing high-quality sample basis for subsequent modeling; by constructing the coupling mechanism between process parameters and quality indicators, realizing the accurate description of the quality evolution trend of multivariate process, it is helpful to clarify the key influencing factors, and enhance the explainability and process transparency of production link; by real-time deduction of the quality performance of the target batch in the production process, the potential deviation can be perceived in advance, the prior information is provided for operation decision, and the risk of batch waste is reduced; by correlating the prediction deviation with the process control variables, the unstable section is identified in time and the accurate adjustment suggestion is output, realizing the transformation of process regulation from experience-driven to data-driven; the whole system has dynamic, closed-loop and adaptability, can automatically switch the characteristic parameter combination according to different product system, improves the adaptability to macro continuous production environment while maintaining the consistency of quality, and provides key technical support for intelligent manufacturing and high-end product quality control of pearl material industry. BRIEF DESCRIPTION OF DRAWINGS

[0016] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0017] Figure 1 A step flowchart of a pearl material production data management method based on data integration analysis according to the present application;

[0018] Figure 2 For Figure 1 A detailed step flowchart of step S1 in the method;

[0019] Figure 3 For Figure 1 A detailed step flowchart of step S2 in the method;

[0020] Figure 4 A schematic diagram of a pearl material production process equipment. DETAILED DESCRIPTION

[0021] The technical method of the present application will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0022] In addition, the drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. Identical reference signs in the drawings represent identical or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily have to correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0023] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be referred to as a second element, and similarly a second element can be referred to as a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0024] To achieve the above-mentioned purpose, please refer to Figures 1 to 4 The present application provides a pearlescent material production data management method based on data integration analysis, which comprises the following steps:

[0025] Step S1: Obtain the process raw data set of the pearlescent material production line, and perform standardization processing to obtain a standardized process data set; obtain the quality detection data set of the pearlescent material; divide the standardized process data set and the quality detection data set into time intervals, and mark them with batch numbers to obtain a process-quality batch data pair;

[0026] Step S2: Design a modeling sample data set based on the process-quality batch data pair, and construct a quality process coupling model; extract the mapping relationship between process control and quality characterization in the quality process coupling model to obtain quality process coupling model parameters, wherein the quality characterization indexes include surface glossiness, particle size distribution uniformity and coating layer defect density;

[0027] Step S3: Real-time acquisition of current production batch process parameters, and quality index prediction analysis of process parameters based on quality process coupling model parameters to obtain a prediction result;

[0028] Step S4: deviation analysis based on the prediction result and the preset quality control standard, identifying abnormal parameters in the current process operation based on the deviation analysis result, and formulating a process optimization adjustment scheme to obtain dynamic process adjustment data.

[0029] Preferably, step S1 comprises the following steps:

[0030] Step S11: configuring an edge collection terminal for the pearl material production line, and collecting a process original data set in real time;

[0031] Step S12: performing communication protocol decoding, sampling frequency unification and timestamp alignment processing on the process original data set to form a standardized process data set;

[0032] Step S13: obtaining titanium dioxide coated pearl materials and performing quality detection to obtain a quality detection data set;

[0033] Step S14: dividing the standardized process data set and the quality detection data set into time intervals, and establishing a sliding window and time mapping rule to obtain a material production path interval for each sample;

[0034] Step S15: setting a unique batch number for the material production path interval of each sample, identifying and merging process parameters and corresponding quality detection results belonging to the same interval to generate process-quality batch data pairs.

[0035] In the embodiment of the present application, please refer to Figure 4, first install and configure an edge collection terminal of model WRM-200 at a key equipment node of the pearlescent material continuous production line. The edge collection terminal integrates temperature, pressure, flow, current, voltage, and other sensing interfaces, and supports the Modbus-TCP communication protocol. The edge collection terminal is used to collect real-time data of the running state of process equipment such as the silo, the cyclone preheater, the fluidized bed reactor, the calcination system, the heat exchanger, and the tail gas treatment device, form a process raw data set, and set the collection frequency to 1 Hz. After the collection is completed, the edge local data synchronization module is used to perform communication protocol decoding operations, convert the Modbus format data into a unified key-value structure, and perform timestamp reconstruction and sampling frequency standardization processing on each data channel. After the processing, a standardized process data set is generated, and the structure fields include collection time, equipment identification, parameter type, parameter value, and unit. All data is unified into the ISO 8601 time format and is accurate to milliseconds. Then, samples are selected every 30 minutes from the production batch of titanium dioxide coated pearlescent materials, the D10, D50, and D90 particle size distributions are measured using a ZNS-820 laser particle size analyzer, the titanium dioxide coating rate is detected using a titration method, the glossiness is detected using a BYK456 gloss meter, and the L, a, and b color values are obtained using a three-stimulus value measuring instrument, forming a quality detection data set. All detection data is recorded in a JSON structure and is labeled with the detection time and sample batch number. The detection time of each data in the quality detection data set is used as a reference benchmark, and a time window interval is constructed in the standardized process data set according to the average process residence time of the pearlescent material from the feed to the discharge (21 minutes under the process conditions). The window sliding step is set to 10 minutes, the boundary data is fitted using linear interpolation, the time period between the raw material entry point and the discharge point of the corresponding sample is calculated, and the material production path interval is delimited. After the path interval is delimited, a unique batch number is assigned to each detection sample, and all standardized process parameters within the time period and the quality detection values corresponding to the sample are combined and integrated, written into a structured data table, and output as a process-quality batch data pair. The fields include batch number, temperature sequence, gas flow rate sequence, residence time, coating rate, particle size distribution value, and gloss color value.

[0036] The application can realize accurate correspondence between process running state and quality detection result in the production process of pearlescent material, and construct a data basic system supporting modeling analysis. Through deploying edge collection terminals and collecting key process parameters of multiple device nodes at high frequency and low delay, real-time and complete recording of production process information is ensured, and the coverage and continuity of data acquisition are significantly improved. Through unified communication protocol analysis and timestamp standardization processing, the data structure difference problem caused by multiple devices and multiple protocols is solved, and the data consistency and system integration efficiency are improved. Combined with quality detection results, time interval mapping and material path tracking are carried out to accurately restore the complete process history of each detection sample in the production line, and the disconnection problem between traditional detection data and process data is eliminated. By introducing sliding window mechanism and time mapping rules, the robustness of batch data matching is improved, and the data pairing accuracy in the presence of reaction lag and mixed flow process is enhanced. On this basis, one-to-one associated data pairs between process and quality are established, which provides accurate and high-quality input data sources for subsequent process rule extraction, process modeling, quality prediction and control strategy formulation, realizes closed-loop connection from data perception to information integration, and effectively supports the transformation of pearlescent material production from experience-driven to data-driven.

[0037] Preferably, step S13 comprises the following steps:

[0038] Step S132: edge positioning of the coating layer is performed according to the reflected light intensity distribution data, a coating layer transition zone gradient is extracted, and coating layer profile data is generated;

[0039] Step S133: sub-pixel level segmentation is performed based on the coating layer profile data, a local thickness matrix is calculated, and a coating layer thickness distribution heat map is obtained;

[0040] Step S134: effective coating area ratio is calculated based on the coating layer thickness distribution heat map, and coating rate data is obtained;

[0041] Step S135: the surface of the titanium dioxide coated pearlescent material is scanned, a specular reflection light intensity curve is constructed, and pearlescent effect consistency data is calculated;

[0042] Step S136: low reflection areas are identified according to the pearlescent effect consistency data, and defect boundaries are located in combination with the coating layer profile data, and a defect distribution density map is generated;

[0043] Step S137: weighted scoring is performed based on the coating rate data, defect distribution density and pearlescent effect consistency, and a quality detection data set is output.

[0044] In the embodiment of the present application, the titanium dioxide coated pearl material sample is fixed on a three-dimensional precision displacement platform, a D65 standard LED ring array is used as the light source, the incident angle is set to 45°, a CCD industrial camera is placed at 0° for receiving, the pixel resolution is 2448*2048, the exposure time is 2ms, and the gain is 0dB; after the collected gray scale image is denoised by a two-dimensional Gaussian filter kernel pixel, a gray scale profile is taken every 0.5° along the radial direction, the gradient amplitude is calculated by using a Sobel operator, the gradient threshold T_g is set to 25 gray scale levels, the positions higher than the threshold are marked as edge points, the discrete edge points are connected into a continuous coating layer contour curve by a quadratic curve fitting with sub-pixel accuracy; a grid is generated inside the contour with a step distance of 0.1 pixels, the transition interval from light to dark in the normal direction of each grid node is searched, the sub-pixel position corresponding to the gray scale median is calculated by using linear interpolation, which is recorded as the interface position, the distance between the interface and the contour is the local thickness, the thickness matrix size is consistent with the original image, the thickness value range is 0-500nm, and the coating layer thickness distribution heat map is colored with an interval of 10nm; the heat Figure Two value threshold T_c is 30nm, the pixels higher than the threshold are counted as effective coating, the coating rate data is obtained by dividing the number of effective pixels by the total number of pixels; then the sample is placed in a coaxial illumination common light path reflectometer, the light source is a halogen lamp, the spectral range is 380-780nm, the step distance is 2nm, the detector rotation range is-80° to +80°, a mirror reflection spectrum is collected every 0.5°, the single-angle light intensity is integrated, the mirror reflection intensity curve is drawn, the light intensity variation coefficient CV of the curve in the range of 20°-60° is calculated, CV<5% is recorded as a consistent area, and CV≥5% is recorded as a low reflection area; the low reflection area and the coating layer contour are subjected to Boolean intersection operation to extract the defect boundary, and the defect pixel area is divided by the total area to obtain a defect distribution density map; finally, the coating rate weight is set to 0.4, the defect density weight is set to 0.4, the consistency weight is set to 0.2, the weighted sum is mapped to the 0-100 score quality interval, and the quality detection data set containing the coating rate, the defect density, the consistency score and the weighted total score is output.

[0045] The present application can obtain complete quality images of thickness, coating rate, defects and pearl luster consistency in one detection process, realize the improvement of defect positioning accuracy from pixel level to sub-pixel level, control the thickness measurement error within ±3 nm, make the coating rate calculation deviation less than 0.5%, make the low reflection defect identification missing detection rate less than 1%, shorten the detection cycle from the original offline sampling inspection of 20 min / sample to online full inspection of 3 s / sample, directly use the detection data as the input of the quality process coupling model, compress the response time of the subsequent process optimization closed loop from the hour level to the minute level, reduce the surface glossiness CV value of the whole batch of pearl luster materials from 4.7% to 1.9%, narrow the particle size distribution D50 fluctuation range from ±8 nm to ±2 nm, reduce the coating layer defect density from 150 ppm to 20 ppm, and make the correlation of the final weighted quality score and the downstream customer acceptance one-time pass rate reach 0.97, thereby significantly reducing the rework and return cost.

[0046] Preferably, step S133 comprises the following steps:

[0047] Step S1331: identifying the interface transition zone of the titanium dioxide coating layer and the mica substrate based on the coating layer profile data, and generating interface transition curve data;

[0048] Step S1332: calculating the equivalent optical thickness of each pixel point according to the interface transition curve data, and obtaining an optical thickness distribution matrix;

[0049] Step S1333: solving the local coating thickness based on the optical thickness distribution matrix, and generating thickness gradient field data;

[0050] Step S1334: compensating and calculating the tilt projection error according to the thickness gradient field data, and correcting to obtain real thickness distribution data;

[0051] Step S1335: constructing a three-dimensional heat map containing the thickness-chroma mapping relationship based on the real thickness distribution data, and obtaining the coating layer thickness distribution heat map.

[0052] In the embodiment of the present application, the coating layer profile curve output by step S132 is used to collect Ti and Si element signals point by point along the profile normal direction in a scanning electron microscope-energy spectrum line scanning mode, the position where the Ti intensity drops to 5% and the Si intensity rises to 95% is marked as the transition starting point of the titanium dioxide coating layer and the mica substrate, the position where the Ti intensity rises to 95% and the Si intensity drops to 5% is marked as the transition end point, and the connecting line of the 50 equidistant sampling points between the starting point and the end point obtains the interface transition curve data; the coordinates of each point of the curve are mapped to the CCD original image coordinate system, a 5pixel*5pixel neighborhood is opened around the point, the RGB mean value is read and converted into CIE XYZ three stimulus values, the titanium dioxide refractive index 2.7, the mica refractive index 1.58, the incident angle 45° and the observation angle 0° are substituted into the thin film interference equation Where n is the direct emissivity of titanium dioxide or the refractive index of mica, R is obtained by normalizing the Y value to 0-1, λ is taken as 550nm, and the equivalent optical thickness d is calculated pixel by pixel. opt A 2448×2048 optical thickness distribution matrix is ​​generated; the gradients Gx and Gy of this matrix are obtained using a 3×1 Sobel kernel in the horizontal and vertical directions, respectively, and the amplitudes are synthesized. Regions with G < 1 nm / pixel are labeled as flat regions, and regions with G ≥ 1 nm / pixel are labeled as sloped regions. The thickness of the flat region is directly taken as d. opt The thickness of the inclined region is according to d local =d opt • cosθ correction, where θ is calculated from the gradient direction as θ = arctan(Gy / Gx), forming the thickness gradient field data; with the angle α = 20° between the sample surface normal and the microscope optical axis as a fixed tilt angle, the projection error Δd = d local ·(1―cosα), add Δd back to d local Obtain the true thickness distribution data; construct a 3D point cloud with the true thickness as the Z-axis and the corresponding pixel coordinates X and Y. The thickness range is 0-500nm, and the color is layered in 10nm increments. The chromaticity coordinates are calculated using the Lab conversion formula L=116f(Y / Yn)-16, a=500[f(X / Xn)-f(Y / Yn)], b=200[f(Y / Yn)-f(Z / Zn)], where Yn=100, Xn=95.047, Zn=108.883, and f(t)=t 1 / 3 When t>0.008856, otherwise f(t)=7.787t+16 / 116, the Lab color space is mapped to the RGB color space, and the final output is a heat map of the coating thickness distribution.

[0053] This invention advances the lateral resolution from the whole pixel level to the sub-pixel level, reducing the thickness measurement error to ±2nm and the interface positioning deviation to less than 0.05 pixels. This results in a 0.3% reduction in the subsequent coverage calculation error and a defect density statistical missed detection rate of less than 0.8%. By compensating for tilted projection errors with gradient field, the parallax distortion caused by the 20° observation angle is eliminated, and the linear correlation coefficient between the true thickness and the SEM cross-sectional measurement reaches 0.99. The three-dimensional heat map with thickness-color synchronous mapping directly converts thickness differences into visual color differences, shortening the manual re-inspection time from 15 minutes to 30 seconds. It can also be directly embedded into the downstream color matching system, achieving a brightness L fluctuation of ±0.5 and a and b color drift of <1.0 for the same batch of pearlescent powder, ultimately reducing the color difference return rate of the entire roll of pearlescent film by 90%.

[0054] Preferably, step S2 includes the following steps:

[0055] Step S21: Input-output variable partitioning is performed on the process-quality batch data pairs, key process parameters are extracted as input feature data, and titanium dioxide coating rate, D50 particle size, glossiness, and Lab color value are extracted as output quality indicators to obtain a modeling sample data set;

[0056] Step S22: Normalization processing and missing value completion are performed on the modeling sample data set, Pearson correlation coefficients between each input feature data and each output quality indicator are calculated, information gain scores between each input feature data and each output quality indicator are calculated, and a feature sensitivity matrix is generated;

[0057] Step S23: Based on the feature sensitivity matrix, main process control variables are screened, modeling subsets are constructed for each series of pearlescent materials, input-output mapping relationships are trained, and a preliminary quality-process coupling model is obtained;

[0058] Step S24: K-fold cross-validation and residual analysis are performed on the preliminary quality-process coupling model, fitting accuracy of the preliminary quality-process coupling model in predicting coating rate, particle size, and glossiness is evaluated, model parameter combinations are optimized, and a quality-process coupling model structure is obtained;

[0059] Step S25: Based on the quality-process coupling model structure, a prediction function expression, a weight coefficient distribution, and a feature contribution score are extracted to generate a quality-process coupling model parameter set.

[0060] In the embodiment of the present application, first, the process-quality batch data pairs are divided into fields, and the input characteristic fields are defined as five key process parameters, i.e., reactor temperature (set in the range of 700-850°C), nitrogen flow rate (set in the range of 20-60 L / min), coating reaction residence time (set as 10-30 minutes), calcination section temperature (controlled at 750-900°C), and raw material feeding rate (set as 10-40 kg / h), and the output quality index fields are titanium dioxide coating rate (unit: percent), D50 particle size (unit: microns), glossiness (unit: GU), and L, a, and b in Lab colorimetric value (unit: CIE colorimetric coordinate value). The modeling sample data set is constructed by merging, the data format adopts standard CSV structure, and the number of record lines is not less than 500. After the modeling sample data set is imported into the data processing platform, normalization processing is performed on all input and output fields, the normalization rule is to scale the original numerical value to the [0, 1] interval, and the mean imputation method is used to complete the missing values. All fields with a missing rate of more than 10% are excluded from the modeling set. After normalization, based on the bivariate correlation analysis method between fields, the Pearson correlation coefficient and the information gain score (by grouping and discretizing the output quality indicators and measuring the entropy reduction value) between each input characteristic and each output indicator are calculated, and finally a 5x6-dimensional feature sensitivity matrix is generated, each cell in the matrix corresponding to the correlation strength between an input and an output. Based on the top two scoring input characteristics in the feature sensitivity matrix, the reactor temperature and the residence time are selected as the main process control variables from the five process parameters, and independent modeling subsets are constructed for the golden series and silver-white series pearlescent materials, each subset containing not less than 200 records. Linear fitting is performed between the input characteristics and the output indicators by the least squares method to obtain a preliminary quality process coupling expression, and based on the cross-validation method, the sample set is divided into 5 folds for sequential verification, the average absolute error (MAE) and the determination coefficient R^2 of the predicted coating rate, particle size, and glossiness in each fold are calculated, the best performing process parameter combination is selected as the quality process coupling model structure with the optimal fitting quality, and the final prediction function expression is extracted based on the above structure, including the weight coefficient value (retained to four decimal places) corresponding to each input characteristic and the constant term. The output characteristic contribution score table is sorted according to the scoring standard of the standardized weight proportion multiplied by the output variable change rate weight, and the complete quality process coupling model parameter set is generated, including the function expression, coefficient distribution, and the impact of each feature on the quality change.

[0061] The application can realize precise correlation modeling between process parameters and quality indicators in the production of pearlitic materials, and has significant practical application value and technical advantages. By systematically dividing the input and output variables and constructing the modeling sample, the rationality of the data structure and the pertinence of the modeling input are ensured, and the modeling efficiency and accuracy are effectively improved. Normalization processing and missing value completion of the sample data set can help eliminate the dimension influence and enhance the fault tolerance of the model to incomplete data, improve the model robustness. By constructing a feature sensitivity matrix based on correlation and information gain, the influence of each process parameter on the quality indicator can be objectively evaluated, so as to realize scientific selection of key variables and avoid interference caused by redundant input. Modeling different series of products can enhance the adaptability and generalization ability of the model, and improve the quality prediction accuracy of multi-class pearlitic materials. By using K-fold cross-validation and residual analysis methods, the comprehensiveness and objectivity of model performance evaluation are ensured, and the problems of overfitting or underfitting are avoided. Finally, by extracting the prediction function expression and weight distribution, the contribution of each input feature is determined, which provides quantitative basis and decision support for subsequent quality optimization and process adjustment, and improves the data-driven intelligent control ability of the whole production line.

[0062] Especially important is that the input and output variable division of the process-quality batch data pair in step S21 extracts key process parameters as input feature data, including:

[0063] Based on the pearlite interlamellar spacing control parameters extracted from the process-quality batch data pair, interlamellar spacing process data are obtained;

[0064] According to the cooling rate key parameters calculated from the interlamellar spacing process data, cooling process feature data are obtained;

[0065] Based on the cooling process feature data, the austenitizing temperature curve is extracted, and temperature history data are obtained;

[0066] According to the temperature history data, the pearlite phase transition volume fraction is calculated, and phase transition degree data are obtained;

[0067] Based on the phase transition degree data, the pearlite group size distribution is identified, and pearlite group morphological feature data are obtained;

[0068] The interlamellar spacing process data, cooling process feature data, temperature history data, phase transition degree data and pearlite group morphological feature data are used as input feature data.

[0069] In the embodiment of the present application, first, based on the process-quality batch data pair, the process parameters for controlling the interlamellar spacing of pearlite structure are extracted, including the end reaction temperature (set range of 700 DEG C to 850 DEG C), flaky substrate particle size (limited to 4 μm to 15 μm), air inlet wind speed (range of 3 m / s to 6 m / s) and reaction residence time (value range of 10 to 25 minutes), the interlamellar spacing process data is obtained by mean value normalization and variance standardization operation on the above parameters in different batches, and is recorded in the form of a standardized matrix; after the extraction of the interlamellar spacing process data, according to the temperature decay curve and ventilation rate data of the calcination section, the temperature change slope in the first 300 seconds in the cooling zone is selected, the average cooling rate (unit: DEG C / s) is calculated, and the air cooling flow rate, heat exchanger exhaust temperature and environmental temperature difference are taken as influencing factors to construct the cooling process characteristic data, which contains three index fields of cooling rate value, cooling duration and heat flow conduction amplitude; the cooling process characteristic data and the temperature time series recorded by the measured thermocouple array are spliced to extract the temperature drop curve of the austenitizing stage (750 DEG C to 500 DEG C), which is sampled at a time resolution of 1 second to obtain complete austenitizing temperature history data with a data length of not less than 300 groups; based on the austenitizing temperature history data, the Johnson-Mehl-Avrami type conversion formula V(t) = 1-exp(-k·t n ) is used, where t is the reaction time, the conversion index n is set to 2, and the constant k is set to 0.003 according to the experimental conditions, the volume fraction of the pearlite phase at each time point in the conversion time sequence is calculated, the result is retained to three decimal places, and the output is the phase change degree data; according to the local maximum point of the volume fraction change rate in the phase change degree data, combined with the crystal phase interface propagation rate parameter (experimental value of 0.8 μm / s), the grain growth rate and termination point position in the batch of pearl materials are judged, the size of the pearlite group on each flaky substrate is calculated by the area statistical method to generate the pearlite group morphological feature data, which contains three items of average pearlite group particle size, maximum particle size and particle size distribution standard deviation; finally, the interlamellar spacing process data, cooling process characteristic data, temperature history data, phase change degree data and pearlite group morphological feature data are merged to construct complete input feature data.

[0070] The variable extraction and construction process in the application can systematically capture the key process evolution paths affecting the quality characteristics in the formation process of pearlescent materials from the physical nature, significantly improving the physical relevance and engineering interpretability of model input. Through layer-by-layer derivation and extraction of multi-dimensional data such as interlayer spacing, cooling rate, temperature history, phase transition behavior and pearl group structure, the whole process feature expression from the original process operation to the microstructure evolution of the material is realized, effectively reflecting the deep influence of process parameters on the final quality indicators (such as coating uniformity, particle size distribution, luster effect, etc.). The process strengthens the fusion expression of model input on thermodynamics, phase transition dynamics and organization morphology elements, so that the subsequent established quality prediction model not only has high accuracy, but also has sensitive response ability to complex process disturbance, providing an interpretable and highly adaptive feature base for intelligent control system, which helps to improve the quality stability and controllability of the whole pearlescent material production process.

[0071] Preferably, step S3 comprises the following steps:

[0072] Step S31: Real-time acquisition of process parameters of the current production batch of the pearlescent material production line, to obtain the current production batch process parameters;

[0073] Step S32: Perform model input mapping processing on the current production batch process parameters based on the quality process coupling model parameters, execute the model reasoning process, and obtain the quality prediction result data set of the current batch;

[0074] Step S33: Format the quality prediction result data set and mark the time stamp, and map it to the corresponding sheet substrate batch of the current production section to obtain the prediction result with batch identification.

[0075] In the embodiment of the present application, first, a WRM-200 edge collection terminal is deployed at each key equipment section of the pearlescent material production line, the collection frequency is set to 1 Hz, the process parameters of the current production batch are collected from the running equipment including the screw feeder, the cyclone preheater, the coated fluidized bed reactor, the calcination section, the heat exchanger and the cooling section, mainly including the coating reaction temperature (set range: 720-850℃), the gas phase conveying flow rate (value range: 25-60L / min), the calcination temperature (set range: 750-900℃), the reaction residence time (set to 10-25 minutes) and the raw material feeding rate (set range: 12-38kg / h), the real-time collected parameters are uniformly converted into time series format to form the current production batch process parameters; the batch process parameters are field-mapped according to the field order of the quality process coupling model parameters output in step S25, the normalization processing of the input data is performed (the processing mode is to subtract the training mean value and divide by the training standard deviation), the normalized results are input into the constructed quality prediction function expression for item-by-item function calculation, the weight values of each parameter and the corresponding input values are weighted and accumulated, and a constant term is added, respectively outputting the titanium dioxide coating rate prediction value (unit: %), the D50 particle size prediction value (unit: μm), the glossiness prediction value (unit: GU) and the colorimetric Lab prediction value corresponding to the current batch, forming a quality prediction result dataset, all prediction results are kept to three significant digits; then the quality prediction result dataset is formatted, the field structure is unified to "timestamp + batch number + prediction index name + prediction value", among them, the timestamp is accurate to milliseconds, and each data is marked with a unique sheet-shaped substrate batch identification according to the material number tracking mechanism, the batch number is generated in the format of "QG + year-month-day + production time period number", the corresponding batch process parameters and prediction results are associated one by one to generate the prediction results with batch identification.

[0076] The present application can realize real-time perception and batch-level tracking management of the quality state in the production process of pearlescent materials, significantly improving the intelligence and response capability of the production line. Through real-time collection and model reasoning of the current batch process parameters, the quality trend can be predicted in advance during production, potential deviation risks can be found in time, and the delayed exposure of quality problems can be avoided, thereby reducing the rate of unqualified products and rework costs. The structured and time-labeled processing of the model reasoning results makes the quality prediction data have high time sequence consistency and traceability, which is convenient for interfacing with the production execution system and supporting the development of fine process control strategies. At the same time, the prediction results are accurately mapped to specific substrate batches, realizing product-level quality prediction and tracking, providing a data basis for establishing quality early warning, trend analysis and precise scheduling, and comprehensively improving the visual management level and quality control efficiency of pearlescent material production.

[0077] Preferably, the model input mapping processing of the current production batch process parameters based on the quality process coupling model parameters in step S32 is performed, and the model inference process comprises:

[0078] The pearlite interlamellar spacing process parameters of the current production batch are extracted based on the quality process coupling model parameters;

[0079] The model input normalization processing is performed according to the pearlite interlamellar spacing process parameters, and the standardized interlamellar spacing data is obtained;

[0080] The pearlite phase transformation calculation is performed based on the standardized interlamellar spacing data, and the pearlite phase transformation degree data is obtained;

[0081] The pearlite phase transformation degree data is calculated according to the pearlite phase transformation degree data, and the pearlite group morphological data is obtained;

[0082] The material strength inference is performed based on the pearlite group morphological data, and the strength performance prediction data is obtained;

[0083] The ductile-brittle transition temperature data is obtained by calculating the ductility index mapping relationship according to the strength performance prediction data;

[0084] The quality comprehensive evaluation is performed based on the strength performance prediction data and the ductile-brittle transition temperature data, and the quality prediction result data set of the current batch is generated.

[0085] In the embodiment of the application, first, based on the variable structure defined in the quality process coupling model parameters output in step S25, the process parameters related to the pearlite interlamellar spacing in the current production batch are extracted, including the fluidized bed reactor temperature (set to 720-850℃), the gas flow velocity (controlled at 3.0-5.5m / s), the reaction time (set to 10-30 minutes) and the raw material particle size (set to 4-12μm), and the above numerical values are taken as the pearlite interlamellar spacing process parameters; the extracted pearlite interlamellar spacing process parameters are normalized, and each parameter is normalized to the interval [0, 1] by using linear scaling, and the standardized interlamellar spacing data is formed after processing; the pearlite phase transformation degree calculation is performed based on the standardized interlamellar spacing data, the phase transformation degree is numerically approximated according to the isothermal phase transformation theory, and the exponential function form V(t) = 1-exp(-k·t n) is carried out, wherein t is the reaction time (in minutes), k is set to 0.005, n is the nucleation index set to 2.0, the volume fraction of the current process condition is calculated, the pearlite phase transformation degree data is output, and the data is kept to three decimal places; using the volume fraction data, combined with the preset pearlite group growth rate constant (set to 0.7 μm / min) in the system, the particle size in the pearlite group growth interval is sequentially calculated, the pearlite group size distribution characteristics in the current production batch are obtained, and the pearlite group morphology data including the average particle size, particle size distribution variance and maximum particle size are output, with the unit of microns; the pearlite group morphology data is input into the strength reasoning module, and each item is substituted into the Griffith fracture criterion structure form, wherein E is the elastic modulus (set to 120 GPa), γ is the interface energy (set to 1.2 J / m 2 ), a is the maximum particle size (unit: m), the material fracture strength σ is calculated after unit conversion, with the unit of MPa, and the strength performance prediction data is output; according to the strength data and the built-in pearlite substrate fracture toughness critical value KIC = 1.5 MPa·m 0.5 in the system, combined with the Ashby ductile-brittle transition standard, the ductile-brittle transition temperature of the current batch is calculated in the form of TDB = A―B·σ, A is set to 350 K, B is set to 0.35 K / MPa, the ductile-brittle transition temperature data is obtained, with the unit of K; finally, the fracture strength data and the ductile-brittle transition temperature data of the batch are weighted and summarized, the scoring structure is quality comprehensive value = 0.6 × strength normalized value + 0.4 × toughness stability index, the strength normalized value is normalized according to the maximum predicted value, the toughness stability index is scored according to the ideal state below 300 K, and the output is a standard format quality prediction result data set, the fields include batch number, time stamp, predicted coating strength, predicted transition temperature and quality comprehensive score.

[0086] The present application realizes a high-precision reasoning path from process parameters to comprehensive quality evaluation results by introducing a multi-stage structure-performance mapping mechanism to decouple and reconstruct key process and material characteristics such as pearlite interlayer spacing, phase transformation behavior, microstructure morphology, strength performance and toughness indicators level by level. Its beneficial effects are that it can convert complex production data into standardized features with physical meaning, improve the consistency of model input and the interpretability of prediction output; the logical chain of material structure and macroscopic performance is established through the deduction of phase transformation degree and pearlite group size, which enhances the sensitivity of the model to microstructure changes; the introduction of strength prediction and ductile-brittle temperature indicators further expands the quality evaluation dimension, so that the prediction results not only cover appearance and particle size indicators, but also cover key mechanical properties, effectively supporting the overall judgment of material functionality; the finally formed quality prediction result data set has high accuracy, physical consistency and process relevance, providing a solid data foundation and decision support for subsequent quality control and dynamic process optimization.​

[0087] Preferably, step S4 comprises the following steps:

[0088] Step S41: Comparing and analyzing the prediction result with the preset quality control standard, calculating the difference between the prediction result of the current batch and the preset target standard to obtain prediction quality deviation data;

[0089] Step S42: Based on the prediction quality deviation data and the current production batch process parameters, using the quality process coupling model to inversely trace the main process control variables affecting the prediction result deviation, identifying the abnormal parameter combination in the current process operation, and obtaining abnormal parameter identification result;

[0090] Step S43: Sensitivity analysis is performed on the abnormal parameter identification result, the quality prediction response change rate in different parameter adjustment ranges is calculated, an optimization objective function is constructed, and process optimization adjustment path data is obtained;

[0091] Step S44: Based on the process optimization adjustment path data and the preset standard process segment sample, a process adjustment scheme is formulated to form a dynamic process adjustment suggestion table;

[0092] Step S45: The dynamic process adjustment suggestion table is subjected to process safety and equipment boundary constraint verification, and executable dynamic process adjustment data is output.

[0093] In this embodiment of the invention, the coating rate (in %), D50 particle size (in μm), gloss (in GU), and Lab chromaticity value contained in the quality prediction result dataset are first compared item by item with the preset quality control standards built into the production system. The standard coating rate target is set at 40%, the D50 target particle size is 10.0 μm, the gloss is controlled within the range of 85±5 GU, the chromaticity value L is required to be greater than 80, and a and b are controlled within ±5. The arithmetic difference between the predicted value and the target value is calculated, and the deviation value of each quality indicator is recorded to obtain the predicted quality deviation data of the current batch. All deviation values ​​are uniformly retained to two decimal places. The quality process coupling model parameter structure is called to perform item-by-item weighted reverse lookup processing on the actual process parameters of the current production batch, and the deviation index of each indicator is calculated. The contribution rate under the previous process settings was used to extract the two process variables with the greatest impact on coating rate and particle size, which were identified as key abnormal parameters. These included calcination temperature (exceeding the set upper limit of 860℃) and reactor gas flow rate (below the set lower limit of 3.0m / s). The abnormal parameter identification results were combined, and the data format included parameter name, deviation direction, deviation magnitude, and historical threshold comparison label. Numerical perturbation operations were performed on the identified abnormal parameters. The predicted coating rate and particle size change trends were calculated at adjustment increments of ±5%, ±10%, and ±15%. The predicted response change rate was obtained by dividing the incremental change rate by the parameter adjustment magnitude. The process variable with the largest response change rate was selected as the first adjustment object. Combining the minimization of coating rate deviation and the convergence of particle size error as objectives, the optimization objective function was constructed as ΔY=∑(w i ·|X i ′―X i |), where X i ′ represents the value of the parameter to be adjusted, w i To predict sensitivity weights, process optimization adjustment path data is output. The optimized adjustment path is compared with the standard process segment samples archived in the system database. Historical sample records with similar parameter combinations under the same equipment and raw material conditions are selected. The corresponding temperature setpoint, wind speed parameter, reaction residence time, and raw material particle size combination are extracted from the historical samples. Combined with the current deviation direction, a dynamic process adjustment suggestion table is generated. The fields include adjustment parameter items, recommended target values, recommended variation range, and adjustment sequence number. Boundary checks are performed on all recommended parameters in the dynamic process adjustment suggestion table. Among them, the reactor temperature should not exceed 880℃, the wind speed should not be less than 2.5m / s, the raw material feed rate should not exceed 40kg / h, and the calcination time should not be less than 8 minutes. By sequentially checking the upper and lower limits of each suggested parameter value and comparing it with the equipment adjustment capability, the instruction items that do not meet the conditions are eliminated, and the final executable dynamic process adjustment data is output.

[0094] The application can realize intelligent identification, cause analysis and optimization response of prediction quality deviation, and comprehensively improve the closed-loop control ability and process self-adaptive level of the production process of pearlitic material. By difference analysis of the prediction result and the standard quality requirement, the degree of actual process deviation from the quality target can be quantitatively identified, so as to provide a clear adjustment direction for subsequent regulation and control. The deviation source is analyzed in reverse by using the coupling model, so as to accurately lock the key process variable combination leading to quality abnormality, and avoid the subjectivity and blindness in traditional experience judgment. Further, the optimization objective function is constructed through sensitivity analysis, so that the process adjustment scheme has quantitative tuning ability and response prediction ability, and the scientificity and pertinence of the adjustment measures are ensured. Combined with the dynamic adjustment suggestion formulated according to the historical excellent process section sample, the practicability and feasibility of the scheme are enhanced, and the verification of process safety and equipment boundary conditions ensures that the finally output data has operation executability and equipment compatibility, which fundamentally improves the reliability, stability and real-time response efficiency of process adjustment, and effectively supports the realization of intelligent manufacturing goals of high quality and low fluctuation.

[0095] Especially important is that in step S42, based on the predicted quality deviation data and the current production batch process parameters, the main process control variables affecting the deviation of the prediction result are traced back in reverse by using the quality process coupling model, and the abnormal parameter combination in the current process operation is identified, including:

[0096] Based on the predicted quality deviation data, the pearlite strength performance deviation threshold is extracted, and the key quality deviation index is obtained;

[0097] According to the key quality deviation index, the cooling rate influence coefficient in the quality process coupling model is matched in reverse, and the cooling rate sensitivity data is obtained;

[0098] Based on the cooling rate sensitivity data, the interlayer spacing control curve of the current production batch is scanned, the abnormal fluctuation section is identified, and the interlayer spacing abnormal fluctuation data is obtained;

[0099] According to the interlayer spacing abnormal fluctuation data, the austenitizing temperature time sequence record is associated, the temperature out-of-tolerance period is extracted, and the temperature process abnormal data is obtained;

[0100] Based on the temperature process abnormal data, the phase transition kinetics parameter offset is calculated, and the pearlite phase transition abnormal feature is obtained;

[0101] According to the pearlite phase transition abnormal feature, the rolling process parameter library is traced back, the abnormal parameter combination is located, and the abnormal parameter identification result is obtained.

[0102] In the embodiment of the application, first, the predicted quality deviation data is subjected to field screening, and the difference between the predicted value of pearlite strength performance and the target strength control standard is extracted. If the predicted strength is lower than 125 MPa and the deviation exceeds 5%, it is determined as a key quality deviation index, and the deviation direction is marked as negative fluctuation. Based on the key quality deviation index, the cooling section process parameter response coefficient affecting the strength performance is extracted from the quality process coupling model parameter set, specifically including the coefficient value of the average cooling rate change to the predicted strength, with the numerical unit of MPa·s / ℃. The cooling rate sensitivity data is split and counted according to the process section, and the cooling rate influence threshold is recorded. The threshold is below 2.0 MPa·s / ℃ as a low sensitivity response zone, 2.0 to 3.5 as a medium sensitivity zone, and greater than 3.5 as a high sensitivity zone. Combined with the real-time cooling section data of the current production batch, the cooling curve from the completion of cladding to the recovery of ambient temperature is extracted. Under the condition that the sampling period is 1 second, the first derivative analysis of the cooling rate is performed, and the continuous fluctuation section with a slope change rate greater than the set threshold is screened out. The abnormal judgment standard is set as the continuous fluctuation time length exceeding 8 seconds. The abnormal point set in this section is identified and interlayer spacing abnormal fluctuation data is generated. The data structure is interval start and end time, maximum rate change value and corresponding process section number. The abnormal fluctuation data is compared with the recorded austenitizing temperature time record of the current batch, the corresponding heating temperature interval when the abnormal fluctuation occurs is identified, and the period in which the temperature exceeds the set temperature control range ± 20℃ or more is extracted. Temperature process abnormal data is generated, including the over-limit start and end time, the maximum amplitude and the duration. According to the temperature process abnormal data, combined with the built-in phase change kinetics parameter table in the system, the heating speed and cooling speed of the batch in the abnormal period are jointly analyzed. The two-interval phase change rate ratio calculation method (V1 / V2) is used to evaluate the phase change response deviation. If the ratio is greater than 1.4, it is determined as a phase change delay type abnormality, and pearlite phase change abnormal feature data is generated. The structure field includes the deviation type, the occurrence stage and the associated parameter value. Finally, the phase change abnormality of the batch is taken as the query condition, and the archived rolling process parameter library in the pearlitic material production database is entered. The batch instances with historical abnormal records in the parameter combinations of reactor gas flow rate, calcination temperature and holding time are screened out. The top two process parameter combinations are selected according to the abnormal frequency, defined as abnormal parameter combinations, and output as abnormal parameter identification results, including parameter name, deviation direction, over-limit amplitude and number of hit abnormal cases.

[0103] The application can start from specific quality performance deviation behaviors, gradually trace back to key process links such as cooling rate, interlayer spacing fluctuation, temperature anomaly and phase change behavior, and construct a complete quality-process causal path; by introducing sensitivity and dynamic offset amount, the abnormal identification process has data support and reproducibility, avoiding the problems of strong human subjectivity and many blind adjustment areas in traditional experience analysis; at the same time, through correlation analysis and parameter library backtracking between cross-level variables, the abnormal condition determination is not limited to a single variable, but identifies the synergistic abnormality between variable combinations, improving the integrity and accuracy of the abnormal identification result; finally, the output abnormal parameter combination provides a clear direction for process control, which helps to quickly close-loop optimization and adjustment, and supports the production system to realize the high-order transition from "problem discovery" to "root cause positioning" and "active intervention".

[0104] Preferably, the abnormal parameter identification result in step S43 is subjected to sensitivity analysis, the quality prediction response change rate in the adjustment range of different parameters is calculated, and the optimization objective function is constructed, including:

[0105] Based on the abnormal parameter identification result, a set of key process variables of the current batch is extracted;

[0106] A multi-level orthogonal test is designed according to the set of key process variables of the current batch, and a parameter perturbation simulation data set is generated;

[0107] The variable sensitivity gradient data is calculated according to the parameter perturbation simulation data set;

[0108] The variable sensitivity gradient data is coupled with the opalescent optical interference effect to construct a dynamic optimization objective function;

[0109] The optimal adjustment path of the dynamic optimization objective function is solved, and a process optimization instruction set is generated;

[0110] Based on the process optimization instruction set, the robustness of the optimization path is verified, and process optimization adjustment path data is obtained.

[0111] In the embodiment of the application, first, based on the abnormal parameter identification result, a set of key process variables of the current batch is extracted, and the screening condition is that the process parameter item with a deviation amplitude greater than a set threshold value, the set threshold value is that the coating reaction temperature deviation is ±15℃, the gas flow deviation is ±10L / min, and the calcination holding time deviation is ±3 minutes, and finally the reactor temperature, calcination time and gas flow rate are selected as the set of key process variables; based on the variable set, a five-level three-factor orthogonal experiment is designed, the levels are set to 5%, 10%, 15%, 20% and 25% of each variable above and below the current set value, and L25(5 3) standard orthogonal table design combination, generate a parameter perturbation simulation data set containing 25 groups of variable combinations, all combined parameters are strictly limited within the upper and lower limits set by the process control system; input the quality process coupling function relationship into the parameter perturbation simulation data set respectively, obtain the predicted coating rate, particle size D50 and gloss value corresponding to each combination by direct substitution calculation method, calculate the corresponding predicted result change rate of each process parameter under different level changes, use the difference quotient form ΔY / ΔX to construct a three-dimensional quality response surface, extract the maximum change gradient of each parameter on the predicted index, and arrange it into variable sensitivity gradient data with the unit of predicted unit / process unit (such as % / ℃); on this basis, the variable sensitivity gradient data is coupled with the film layer gloss phase gain rule under the condition of pearl material optical interference effect, the reference coefficient of the influence change rate of film layer thickness and interference wave peak on gloss is set as 0.35GU / μm, the indirect response term of coating thickness is introduced into each sensitive gradient, the optical enhancement weight function is constructed, the gain weight of each parameter on the optical performance is adjusted, and the dynamic optimization objective function Z=∑(Wopt·|Y i ―Y target |) is formed, wherein Y i is the predicted index value of the simulation group, and Y target is the quality control target value; the target function is calculated exhaustively to determine the parameter combination corresponding to the minimum Z value, and the output is a process optimization instruction set, including recommended temperature, recommended air speed and recommended holding time, and the adjustment priority order is marked; then, on the basis of the current batch real-time process parameters, the parameter setting values in the process optimization instruction set are embedded into the simulation process path in turn, three rounds of perturbation tests are performed, and whether the predicted result fluctuation range remains within the target threshold ±5% is observed, if stable, the robustness is confirmed to be qualified, and finally the verified process optimization adjustment path data is formed.

[0112] The present application realizes comprehensive response exploration of multi-factor multi-level combination under the premise of controllable cost by extracting key process variables and designing orthogonal test; with the help of simulation data to generate quality response surface, the nonlinear influence of variable change on quality index can be quantitatively analyzed, and the pertinence and predictability of regulation strategy are improved; the combination of variable sensitivity gradient data and pearl optical interference mechanism makes the optimization not only focus on numerical fitting results, but also considers the collaborative optimization demand of material performance and appearance effect; the dynamic optimization objective function constructed has solvability and multi-dimensional adaptability, which lays a foundation for generating efficient and executable process optimization instructions; the robustness verification of the optimization path ensures that the recommended adjustment scheme has good stability and anti-disturbance ability under actual production conditions, significantly improves the reliability of the process adjustment process and the success rate of the final quality control.

[0113] Preferably, the present application also provides a pearlescent material production data management system based on data integration analysis, for executing the above-mentioned pearlescent material production data management method based on data integration analysis, the pearlescent material production data management system based on data integration analysis comprises:

[0114] a data integration module, for acquiring a process original data set of a pearlescent material production line, and performing standardization processing to obtain a standardized process data set; acquiring a quality detection data set of the pearlescent material; performing time interval division on the standardized process data set and the quality detection data set, and performing batch number marking to obtain a process-quality batch data pair;

[0115] a model construction module, for constructing a quality-process coupling model based on the process-quality batch data pair, and extracting a mapping relationship between process control and quality characterization to obtain quality-process coupling model parameters;

[0116] an intelligent prediction module, for real-time acquisition of current production batch process parameters, and performing quality index prediction analysis based on the quality-process coupling model parameters to obtain a prediction result;

[0117] a regulation optimization module, for performing deviation analysis based on the prediction result and a preset quality control standard, identifying abnormal parameters in current process operation, and formulating a process optimization adjustment scheme to obtain dynamic process adjustment data.

[0118] Therefore, from any viewpoint, the embodiments should be considered as exemplary and non-limiting, the scope of the present application is not limited by the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the application file are intended to be included in the present application.

[0119] The above description is merely one specific implementation of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for managing pearlescent material production data based on data integration and analysis, characterized in that, Includes the following steps: Step S1: Obtain the original process dataset of the pearlescent material production line and perform standardization processing to obtain a standardized process dataset; Obtain a quality inspection dataset for pearlescent materials; The standardized process dataset and quality inspection dataset are divided into time intervals and labeled with batch numbers to obtain process-quality batch data pairs. Step S2: Based on the process-quality batch data, construct a quality-process coupling model for the design modeling sample dataset; extract the mapping relationship between process control and quality characterization in the quality-process coupling model to obtain the parameters of the quality-process coupling model, where the quality characterization indicators include surface gloss, particle size distribution uniformity and coating layer defect density; Step S3: Collect the process parameters of the current production batch in real time, and perform quality index prediction analysis of the process parameters based on the parameters of the quality-process coupling model to obtain the prediction results; Step S4: Based on the prediction results and the preset quality control standards, perform deviation analysis, identify abnormal parameters in the current process operation based on the deviation analysis results, formulate process optimization and adjustment plans, and obtain dynamic process adjustment data; The process of acquiring the quality inspection dataset for pearlescent materials includes: acquiring titanium dioxide-coated pearlescent materials; collecting surface reflection images under a standard light source to obtain reflected light intensity distribution data; locating the coating layer edge based on the reflected light intensity distribution data, extracting the gradient of the coating layer transition region, and generating coating layer contour data; performing sub-pixel-level segmentation based on the coating layer contour data, calculating the local thickness matrix, and obtaining a coating layer thickness distribution heatmap; calculating the effective coating area ratio based on the coating layer thickness distribution heatmap to obtain coating rate data; scanning the surface of the titanium dioxide-coated pearlescent material coating layer, constructing a specular reflection light intensity curve, and calculating pearlescent effect consistency data; identifying low-reflection areas based on the pearlescent effect consistency data, and locating defect boundaries in conjunction with the coating layer contour data to generate a defect distribution density map; and performing a weighted score based on the coating rate data, defect distribution density, and pearlescent effect consistency to output the quality inspection dataset. The process of obtaining the coating thickness distribution heatmap includes: identifying the interface transition zone between the titanium dioxide coating and the mica substrate based on the coating contour data, and generating interface transition curve data; calculating the equivalent optical thickness of each pixel based on the interface transition curve data to obtain an optical thickness distribution matrix; solving for the local coating thickness based on the optical thickness distribution matrix to generate thickness gradient field data; compensating for tilt projection errors based on the thickness gradient field data to obtain the true thickness distribution data; and constructing a three-dimensional heatmap containing the thickness-chromaticity mapping relationship based on the true thickness distribution data to obtain the coating thickness distribution heatmap.

2. The pearlescent material production data management method based on data integration analysis according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Configure edge acquisition terminals on the pearlescent material production line and collect raw process datasets in real time; Step S12: Decode the communication protocol, unify the sampling frequency, and align the timestamps of the original process dataset to form a standardized process dataset; Step S13: Obtain titanium dioxide-coated pearlescent material and perform quality inspection to obtain a quality inspection dataset; Step S14: Divide the standardized process dataset and quality inspection dataset into time intervals, and establish sliding window and time mapping rules to obtain the material production path interval for each sample; Step S15: Assign a unique batch number to the material production path range of each sample, identify and merge the process parameters and corresponding quality inspection results belonging to the same range, and generate process-quality batch data pairs.

3. The pearlescent material production data management method based on data integration and analysis according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Divide the process-quality batch data into input and output variables, extract key process parameters as input feature data, and extract titanium dioxide coating rate, D50 particle size, gloss and Lab color value as output quality indicators to obtain the modeling sample dataset; Step S22: Normalize and impute missing values ​​in the modeling sample dataset, calculate the Pearson correlation coefficient between each input feature data and each output quality index, and calculate the information gain score between each input feature data and each output quality index to generate a feature sensitivity matrix. Step S23: Based on the feature sensitivity matrix, select the main process control variables, and construct modeling subsets for each series of pearlescent materials, train the input-output mapping relationship, and obtain a preliminary quality-process coupling model; Step S24: Perform K-fold cross-validation and residual analysis on the preliminary quality-process coupling model to evaluate the fitting accuracy of the preliminary quality-process coupling model in predicting coating rate, particle size and gloss, and optimize the model parameter combination to obtain the quality-process coupling model structure. Step S25: Extract the prediction function expression, weight coefficient distribution and feature contribution score based on the quality-process coupling model structure to generate the quality-process coupling model parameter set.

4. The pearlescent material production data management method based on data integration analysis according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Collect the process parameters of the pearlescent material production line for the current production batch in real time to obtain the process parameters of the current production batch; Step S32: Based on the parameters of the quality-process coupling model, perform model input mapping processing on the process parameters of the current production batch, execute the model inference process, and obtain the quality prediction result dataset of the current batch; Step S33: Format and timestamp the quality prediction result dataset, and map it to the sheet substrate batch corresponding to the current production segment to obtain prediction results with batch identifiers.

5. The pearlescent material production data management method based on data integration analysis according to claim 4, characterized in that, Step S32 involves mapping the process parameters of the current production batch to the model input based on the parameters of the quality-process coupling model, and performing the model inference process, including: Extract pearlite interlayer spacing process parameters for the current production batch based on the parameters of the quality-process coupling model; The model input is normalized based on the pearlite interlayer spacing process parameters to obtain standardized interlayer spacing data. Pearlite phase transition calculations were performed based on standardized interlayer spacing data to obtain pearlite phase transition degree data; The pearlite size distribution characteristics were calculated based on the pearlite phase transition degree data to obtain pearlite morphological data; Material strength inference is performed based on pearloid morphology data to obtain strength performance prediction data; The mapping relationship between toughness indexes is calculated based on the strength performance prediction data to obtain the ductile-brittle transition temperature data; A comprehensive quality evaluation is performed based on strength performance prediction data and ductile-brittle transition temperature data to generate a dataset of quality prediction results for the current batch.

6. The pearlescent material production data management method based on data integration analysis according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Compare and analyze the prediction results with the preset quality control standards, calculate the difference between the prediction results of the current batch and the preset target standards, and obtain the predicted quality deviation data; Step S42: Based on the predicted quality deviation data and the current production batch process parameters, use the quality-process coupling model to trace back the main process control variables that affect the deviation of the prediction results, identify the abnormal parameter combinations in the current process operation, and obtain the abnormal parameter identification results; Step S43: Perform sensitivity analysis on the abnormal parameter identification results, calculate the rate of change of quality prediction response within different parameter adjustment ranges, construct the optimization objective function, and obtain process optimization adjustment path data; Step S44: Based on the process optimization adjustment path data and the preset standard process segment sample, formulate a process adjustment plan and form a dynamic process adjustment suggestion table; Step S45: Verify the process safety and equipment boundary constraints of the dynamic process adjustment suggestion table, and output executable dynamic process adjustment data.

7. The pearlescent material production data management method based on data integration analysis according to claim 6, characterized in that, Step S43 involves sensitivity analysis of the abnormal parameter identification results, calculation of the rate of change in the quality prediction response within different parameter adjustment ranges, and construction of the optimization objective function, including: Extract the set of key process variables for the current batch based on the results of abnormal parameter identification; Design multi-level orthogonal experiments based on the current batch of key process variables, and generate a parameter perturbation simulation dataset; The quality response surface is calculated based on the parameter perturbation simulation dataset to obtain variable-sensitive gradient data; By coupling variable sensitivity gradient data with pearl optical interference effect, a dynamic optimization objective function is constructed. Solve for the optimal adjustment path of the dynamic optimization objective function and generate a process optimization instruction set; The robustness of the optimization path is verified based on the process optimization instruction set, and process optimization adjustment path data is obtained.

8. A pearlescent material production data management system based on data integration and analysis, characterized in that, For executing the pearlescent material production data management method based on data integration analysis as described in claim 1, the pearlescent material production data management system based on data integration analysis includes: The data integration module is used to acquire the original process dataset of the pearlescent material production line, and perform standardization processing to obtain a standardized process dataset; acquire the quality inspection dataset of the pearlescent material; divide the standardized process dataset and the quality inspection dataset into time intervals and mark them with batch numbers to obtain process-quality batch data pairs. The model building module is used to construct a quality-process coupling model based on the design modeling sample dataset using process-quality batch data; it extracts the mapping relationship between process control and quality characterization in the quality-process coupling model to obtain the parameters of the quality-process coupling model, where the quality characterization indicators include surface gloss, particle size distribution uniformity and coating layer defect density. The intelligent prediction module is used to collect the process parameters of the current production batch in real time, and perform quality index prediction analysis of the process parameters based on the parameters of the quality-process coupling model to obtain the prediction results. The control and optimization module is used to perform deviation analysis based on the prediction results and the preset quality control standards, identify abnormal parameters in the current process operation based on the deviation analysis results, formulate process optimization and adjustment schemes, and obtain dynamic process adjustment data.

Images