A new material data intelligent management system based on feature analysis

Through the new material data intelligent management system based on feature analysis, the size control of nanoplatinum catalysts is monitored and optimized in real time, and the problem of performance of nanoplatinum catalysts is solved, which improves catalytic efficiency and stability and reduces the risk of environmental pollution.

CN118737325BActive Publication Date: 2024-11-05LONGYAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411236266.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-11-05
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

In the production process of nanoplatinum catalysts, changes in key characteristics of dimensional control may lead to catalytic performance not meeting standards, affecting the efficiency of industrial emission treatment and environmental pollution risks, and it is difficult to early warning and optimize the existing technology.

Method used

Adopting a new material data intelligent management system based on feature analysis, the nanoparticle size control situation is monitored and analyzed in real time through synthetic data acquisition, key feature extraction, stress risk marking, chemical change monitoring, aggregation behavior analysis and feature change prediction modules, and early warning And optimize the synthesis process.

Benefits of technology

Real-time monitoring and optimization of nanoplatinum catalyst size control is achieved, which reduces trial and error costs in production, improves catalytic efficiency and stability, reduces environmental pollution risks and waste generation, and improves the efficiency of raw material use.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118737325B_ABST
    Figure CN118737325B_ABST
Patent Text Reader

Abstract

The invention discloses a new material data intelligent management system based on feature analysis, which specifically relates to the technical field of production data monitoring, and is used to solve the existing problem that it is impossible to give early warning of the situation in which the probability of the key feature with the greatest impact on size control is high; by real-time data collection, key parameters and performance indicators in the production process are continuously monitored, in-depth data analysis is performed to understand the key factors affecting size control and catalytic performance, and high-risk stress areas and low-risk stress areas of nanomaterials are marked; by monitoring the change of the surface plasma resonance frequency of nanoparticles in the low-risk stress areas and the comprehensive analysis of dynamic monitoring and evaluation, the probability of the key feature with the greatest impact on size control being changed is predicted; by combining the early warning signal with the risk assessment mechanism of the redesigned judgment module, not only the ability to respond to production anomalies is enhanced, but also the environmental burden is significantly reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of production data monitoring, and more specifically, to a new material data intelligent management system based on feature analysis. Background Art

[0002] In industrial emission treatment, the use of nanometal catalysts of specific sizes (such as nanoplatinum) to catalytically convert harmful gases such as carbon monoxide (CO) and nitrogen oxides (NOx) is a common environmental protection measure. These nanocatalysts provide a large number of active sites due to their extremely high specific surface area, significantly improving the efficiency and speed of chemical reactions.

[0003] The key features that have the greatest impact on size control usually include the average particle size, particle size distribution, shape, surface properties, etc. of the nanoparticles, which directly determine the activity and selectivity of the nanoparticles in the catalytic reaction. In the production process of nanocatalysts, the stability of these size-related features is the key to ensuring efficient catalytic performance. If there is no early warning of the high probability of changes in the key features that have the greatest impact on size control, the performance of the nanocatalyst may not meet the standards, thereby affecting the catalytic efficiency and stability, which will result in the failure to effectively convert harmful gases in industrial emission treatment and increase the risk of environmental pollution.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a new material data intelligent management system based on feature analysis to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A new material data intelligent management system based on feature analysis, including a synthetic data acquisition module, a key feature extraction module, a stress risk marking module, a chemical change monitoring module, an aggregation behavior analysis module, a feature change prediction module and a redesign judgment module;

[0008] Synthesis data acquisition module: collects synthesis data during the synthesis of nanoplatinum, including key parameters and performance data;

[0009] Key feature extraction module: extracts key features related to particle size control and catalytic performance from synthetic data, and screens out the key features that have the greatest impact on size control;

[0010] Stress risk marking module: Through multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of nanomaterials are marked respectively;

[0011] Chemical change monitoring module: For low-risk stress areas, the probability of nanoparticle size control failure due to chemical changes on the nanoparticle surface is evaluated by monitoring the changes in the nanoparticle surface plasma resonance frequency;

[0012] Aggregation behavior analysis module: For low-risk stress areas, the dynamic and nonlinear characteristics of nanoparticle aggregation behavior are analyzed to assess the risk probability of uneven size distribution caused by aggregation behavior;

[0013] Feature change prediction module: Based on a comprehensive analysis of the probability of failure of nanoparticle size control due to chemical changes on the nanoparticle surface and the risk probability of uneven size distribution due to aggregation behavior, the module predicts the probability of changes in the key features that have the greatest impact on size control, and issues a feature change warning signal when the probability is greater than a preset threshold;

[0014] Redesign judgment module: Mark the data in the time interval corresponding to the warning signal of feature change as risk synthesis data, and judge whether it is necessary to redesign the nanoplatinum synthesis process based on the frequency of occurrence of recent risk synthesis data.

[0015] In a preferred embodiment, the synthesis data during the synthesis of nano-platinum is collected, specifically:

[0016] Key parameters include temperature, pressure, reaction time, precursor concentration, and stirring rate;

[0017] Performance data include particle size distribution, surface morphology, crystal structure, catalytic activity, and surface chemistry.

[0018] In a preferred embodiment, key features related to particle size control and catalytic performance are extracted from the synthetic data, and the key features with the greatest impact on size control are screened out, specifically including:

[0019] Principal component analysis was used to extract the main components in the synthetic data and preliminarily screen out the features related to particle size and catalytic performance;

[0020] Pearson correlation coefficient was applied to calculate the correlation between the characteristics and particle size control and catalytic performance;

[0021] The importance score of each feature is calculated through the random forest algorithm to identify the key features that have the greatest impact on size control;

[0022] A recursive feature elimination method is used to gradually remove low-importance features and optimize the final feature set.

[0023] In a preferred embodiment, by multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of the nanomaterials are marked respectively, specifically including:

[0024] Construct a multi-scale mechanical model including micro and macro scales to simulate the stress distribution of nanomaterials at different scales;

[0025] Parameterize the physical properties of nanomaterials and characterize the mechanical behavior of nanomaterials at various scales;

[0026] Use finite element analysis to configure multi-scale models and divide the mesh into fine grids;

[0027] Perform time-step dynamic simulations to calculate the stress distribution of the nanomaterial at each moment during the synthesis process and capture areas of stress concentration;

[0028] The simulated stress distribution is compared with the preset stress threshold to evaluate whether the stress at each point exceeds the preset stress threshold, and the high-risk stress areas and low-risk stress areas of the nanomaterial are marked respectively.

[0029] In a preferred embodiment, the probability of failure of nanoparticle size control due to chemical changes on the surface of the nanoparticles is evaluated by monitoring the changes in the surface plasmon resonance frequency of the nanoparticles, specifically comprising:

[0030] Continuously collect resonant frequency data on the nanoparticle surface;

[0031] Use time series analysis techniques to analyze the trend of surface resonance frequency data and identify changes in surface properties caused by chemical adsorption or molecular rearrangement;

[0032] Predicting the probability of size control failure by mathematically modeling the relationship between surface chemistry changes and size control failure of nanoparticles: A logistic regression model was constructed to link surface chemistry changes and the probability of size control failure.

[0033] In a preferred embodiment, the risk probability of uneven size distribution caused by the aggregation behavior is evaluated by analyzing the dynamic and nonlinear characteristics of the nanoparticle aggregation behavior, specifically:

[0034] Nanoparticles are considered as nodes in the network, and the physical or chemical interactions between particles are considered as edges in the network; node connections in the network are defined based on the spatial position, interaction energy or aggregation state of the nanoparticles;

[0035] Through particle tracking technology, the position and interaction force data of nanoparticles are collected in real time to construct a dynamic network structure;

[0036] The modularity of the network is calculated as: ;in: For modularity, Nodes in the network and The actual connection weight between and The nodes are and The degree, is the total number of edges in the network;

[0037] :When the node and 1 if they belong to the same module, otherwise 0;

[0038] Update the structure and connection weights of the nanoparticle network in real time and calculate modularity on the fly;

[0039] A correlation model is established between the modularity value and the nanoparticle size distribution to predict the risk probability of uneven size distribution under a specific aggregation state. The expression is: ;in, is the risk probability of uneven size distribution caused by aggregation behavior, is a scaling factor calibrated according to experimental data, It is a functional relationship fitted by statistical analysis or machine learning models.

[0040] In a preferred embodiment, based on a comprehensive analysis of the probability of failure of nanoparticle size control due to chemical changes on the nanoparticle surface and the risk probability of uneven size distribution due to aggregation behavior, the probability of changes in the key features that have the greatest impact on size control is predicted, specifically:

[0041] The probability of failure of nanoparticle size control and the risk probability of uneven size distribution caused by aggregation behavior are dimensionally processed;

[0042] The probability of failure of nanoparticle size control after dimensionless processing and the risk probability of uneven size distribution caused by aggregation behavior are weighted summed to calculate the probability of change of the key feature that has the greatest impact on size control;

[0043] The probability of a change in the key feature with the greatest impact on dimensional control is compared to a preset threshold:

[0044] When the probability of a change in the key feature with the greatest impact on size control is greater than a preset threshold, a feature change warning signal is issued; when the probability of a change in the key feature with the greatest impact on size control is less than or equal to the preset threshold, a feature normal signal is issued.

[0045] In a preferred embodiment, the data in the time interval corresponding to the warning signal of the characteristic change is marked as risk synthesis data, and whether the nano-platinum synthesis process needs to be redesigned is determined according to the frequency of occurrence of recent risk synthesis data, specifically:

[0046] After issuing a warning signal of feature change, the synthetic data in the corresponding time interval is marked as risky synthetic data;

[0047] Set a preset interval; mark the ratio of risk synthetic data within the preset interval to total synthetic data as the risk synthetic data frequency;

[0048] Set the risk synthetic data frequency threshold and compare the risk synthetic data frequency with the risk synthetic data frequency threshold:

[0049] When the risk synthesis data frequency is greater than the risk synthesis data frequency threshold, it is determined that the nano-platinum synthesis process needs to be redesigned;

[0050] When the risk synthesis data frequency is less than or equal to the risk synthesis data frequency threshold, it is determined that there is no need to redesign the nano-platinum synthesis process.

[0051] Technical effects and advantages of a new material data intelligent management system based on feature analysis of the present invention:

[0052] 1. Through real-time data acquisition of the synthetic data acquisition module, the system can continuously monitor key parameters and performance indicators in the production process to ensure that key production variables are in the best state. In-depth data analysis of the key feature extraction module further enables the production team to understand the key factors affecting size control and catalytic performance, thereby optimizing the characteristics of nanoparticles. This real-time monitoring and rapid response mechanism reduces the cost of trial and error in production and improves the consistency and predictability of production, thereby ensuring the high quality and high performance of the final product.

[0053] 2. The stress risk marking module provides early warnings for potential risk areas in the production process, allowing timely adjustments and interventions. For low-risk stress areas, the feature change prediction module predicts the probability of changes in key features that have the greatest impact on size control by monitoring changes in the surface plasma resonance frequency of nanoparticles and analyzing the dynamic and nonlinear characteristics of nanoparticle aggregation behavior. The risk assessment mechanism of the early warning signal and the redesign judgment module is combined to not only enhance the ability to respond to production anomalies, but also significantly reduce the environmental burden by reducing waste generation and improving the efficiency of raw material use. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a structural schematic diagram of a new material data intelligent management system based on feature analysis in the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example

[0057] Figure 1 A structural schematic diagram of a new material data intelligent management system based on feature analysis of the present invention is given, which includes a synthetic data acquisition module, a key feature extraction module, a stress risk marking module, a chemical change monitoring module, an aggregation behavior analysis module, a feature change prediction module and a redesign judgment module.

[0058] Synthesis data acquisition module: collects synthesis data during the synthesis process of nano-platinum, including key parameters and performance data.

[0059] Key feature extraction module: extracts key features related to particle size control and catalytic performance from synthetic data, and screens out the key features that have the greatest impact on size control.

[0060] Stress risk marking module: Through multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of nanomaterials are marked respectively.

[0061] Chemical Change Monitoring Module: For low stress risk areas, the probability of nanoparticle size control failure due to chemical changes on the nanoparticle surface is evaluated by monitoring the changes in the nanoparticle surface plasmon resonance frequency.

[0062] Aggregation behavior analysis module: For low-risk stress areas, the dynamic and nonlinear characteristics of nanoparticle aggregation behavior are analyzed to evaluate the risk probability of uneven size distribution caused by aggregation behavior.

[0063] Feature change prediction module: Based on a comprehensive analysis of the probability of failure of nanoparticle size control due to chemical changes on the nanoparticle surface and the risk probability of uneven size distribution due to aggregation behavior, the probability of change in the key features that have the greatest impact on size control is predicted, and a feature change warning signal is issued when the probability is greater than a preset threshold.

[0064] Redesign judgment module: Mark the data in the time interval corresponding to the warning signal of feature change as risk synthesis data, and judge whether it is necessary to redesign the nanoplatinum synthesis process based on the frequency of occurrence of recent risk synthesis data.

[0065] Collect synthesis data during the synthesis of nanoplatinum, specifically:

[0066] During the synthesis process of nanoplatinum, key parameter data such as temperature, pressure, reaction time, precursor concentration and stirring rate are collected in real time through sensors and online monitoring equipment.

[0067] Adaptive data acquisition algorithms can be used to dynamically adjust the data acquisition frequency to ensure high-precision data at critical stages of the reaction while reducing the storage pressure of redundant data.

[0068] Performance data such as the crystal structure, particle size distribution, surface morphology and catalytic activity of nano-platinum are collected through physical testing and chemical analysis techniques such as X-ray diffraction (XRD), transmission electron microscopy (TEM), and spectral analysis.

[0069] After data collection, all data are standardized to eliminate biases between different data sources, and data cleaning techniques are used to remove noise and outliers.

[0070] Synthesis data include key parameters and performance data, where key parameters refer to the experimental conditions and operating variables that directly affect the formation and characteristics of nanomaterials during the synthesis of nanoplatinum, usually including:

[0071] Temperature: The temperature inside the reactor or reactor is a key parameter that controls the rate of chemical reactions and the formation of crystal structures.

[0072] Pressure: The pressure inside the reactor affects the crystallization process and particle size of nanomaterials.

[0073] Reaction time: The time from the start to the end of the reaction, the total time and the reaction time of each stage affect the particle size and distribution of the final product.

[0074] Precursor concentration: The concentration of the precursor substances (such as platinum compounds) involved in the reaction directly affects the generation rate and morphology of nano-platinum.

[0075] Stirring rate: The speed of the stirrer during the reaction, which affects the uniform mixing of the reactants and the nucleation and growth of nanoparticles.

[0076] Performance data refers to the data obtained by characterizing the physical and chemical properties of the final synthesized nano-platinum material. Performance data directly reflects the quality and application potential of the material, and performance data includes:

[0077] Particle size distribution: The size range and distribution of nanoparticles are key factors affecting catalytic performance and stability. Particle size distribution data is measured and recorded by dynamic light scattering (DLS) or transmission electron microscopy (TEM).

[0078] Surface morphology: The surface structure and morphology of nanoparticles affect their activity and selectivity in chemical reactions. Use scanning electron microscopy (SEM) or atomic force microscopy (AFM) to analyze surface morphology.

[0079] Crystal structure: The internal atomic arrangement and crystal type of nanoplatinum determine its physical properties such as mechanical strength and conductivity. The crystal structure is analyzed by X-ray diffraction (XRD) to record the composition and lattice parameters of the crystal phase.

[0080] Catalytic activity: The efficiency of nano-platinum as a catalyst, such as conversion rate and selectivity in a specific chemical reaction. Through catalytic reaction experiments, the reaction rate, product selectivity, and relevant data are measured.

[0081] Surface chemical properties: The types and distribution of surface functional groups and chemical bonds affect the interaction between nanoplatinum and reactants. Use X-ray photoelectron spectroscopy (XPS) or Fourier transform infrared spectroscopy (FTIR) to analyze surface chemical properties.

[0082] Key features related to particle size control and catalytic performance were extracted from the synthetic data, and the key features with the greatest impact on size control were screened out, including:

[0083] Principal component analysis was used to extract the main components in the synthetic data and preliminarily screen out the features related to particle size and catalytic performance:

[0084] Principal component analysis is a commonly used data dimensionality reduction technique that converts the original high-dimensional data into a set of uncorrelated principal components through linear transformation. Its goal is to reduce the data dimension and simplify the data set by retaining the principal component with the largest variance in the data.

[0085] Select the first several principal components so that the cumulative variance explanation rate reaches more than 95% to ensure that sufficient data information is retained. The specific calculation formula is: ;in, is the data matrix after dimensionality reduction (representing the projection of the original data in the direction of the principal component), is the original data matrix (including all key parameters and performance data), is the eigenvector matrix (obtained by eigenvalue decomposition of the covariance matrix). Ensure that the data after dimensionality reduction retains more than 95% of the original information to verify the effectiveness of principal component analysis.

[0086] The Pearson correlation coefficient was used to calculate the correlation between the characteristics and the particle size control and catalytic performance:

[0087] The Pearson correlation coefficient is used to measure the linear correlation between two variables. Its value range is [-1, 1], where 1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no correlation.

[0088] Calculated variables and The Pearson correlation coefficient between them is expressed as: ;in, For variables and The Pearson correlation coefficient between is the data point of the particle size control related characteristics, are data points of characteristics related to catalytic performance, and Variables and The mean of .

[0089] A collection of values ​​that represents a characteristic, usually one related to particle size control. For example, It can be a collection of data points representing the particle size distribution measured under different experimental conditions.

[0090] A collection of values ​​that represents another characteristic, usually something related to catalytic performance. For example, It can be a collection of data points representing the catalytic activity or selectivity measured under different experimental conditions.

[0091] is the index, indicating the samples or data points. It is used to traverse each element in the data set.

[0092] The Pearson correlation coefficient between all features and the target variable was calculated, and the features with an absolute value greater than 0.7 were selected for the next step of analysis to ensure the significance of the correlation.

[0093] The importance score of each feature is calculated by the random forest algorithm to identify the key features that have the greatest impact on size control:

[0094] Random forest is an ensemble learning method that improves the accuracy and robustness of the model by building multiple decision trees and taking their average. It can effectively evaluate the contribution of each feature to the target variable.

[0095] The number of trees is set to 100 to balance computational cost and model stability. Calculation formula: ;in, It refers to the importance of features to the model prediction results. Indicates The first tree The importance of features, is the total number of decision trees.

[0096] The performance of the random forest model was evaluated by cross validation, and the top 20% features were selected as key features based on the importance score.

[0097] A recursive feature elimination method is used to gradually remove low-importance features and optimize the final feature set:

[0098] Recursive feature elimination is a feature selection technique that recursively trains a model and removes the least important features, gradually reducing the feature set to optimize the final model performance.

[0099] The step size is set to 1, and the least important features are removed one at a time to ensure fine-grained feature selection.

[0100] Train the model and calculate the importance of all features.

[0101] Remove the least important features and retrain the model.

[0102] The above process is repeated until the predetermined number of features is reached or the model performance no longer improves.

[0103] The model performance of different feature sets is evaluated through cross-validation, and the feature set that optimizes the model performance is selected.

[0104] Through multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of nanomaterials are marked respectively, including:

[0105] Construct a multi-scale mechanical model including microscopic (nanoscale) and macroscopic (material level) to simulate the stress distribution of nanomaterials at different scales:

[0106] A comprehensive multi-scale mechanical model is constructed to combine the nanoscale microstructure with the material-level macrostructure. At the micro level, molecular dynamics (MD) is used to simulate the stress distribution of a single nanoparticle, and at the macro level, a continuum mechanics model is used to describe the stress behavior of the entire material. This multi-scale modeling method innovatively combines molecular dynamics and continuum mechanics, and can capture both microscopic and macroscopic stress changes, providing more accurate stress field predictions.

[0107] The specific multi-scale mechanical model can be realized by the following formula: ;in, is the stress tensor, which represents the internal stress distribution of the material under stress; It is the macroscopic energy density function, which describes the energy distribution of the entire material on a macroscopic scale; is the strain tensor, which indicates the degree of deformation of the material after being subjected to force; It is the sum of different energy terms at the microscopic scale, representing the microscopic contribution in the multi-scale combination; It is A microscopic energy term, representing the energy state of a specific structure at the microscopic scale; It is with The strain tensor corresponding to each microscopic energy term represents the degree of deformation of the microstructure; is the number of the microscopic energy term.

[0108] Multi-scale modeling provides a basis for subsequent material property parameterization, allowing the model to seamlessly connect between different scales and achieve accurate simulation of the stress field.

[0109] Parameterize the physical properties of nanomaterials and characterize the mechanical behavior of nanomaterials at various scales:

[0110] In order to accurately describe the mechanical behavior of nanomaterials in the model, it is necessary to parameterize the physical properties of nanomaterials, including Young's modulus, Poisson's ratio, shear modulus, density, etc., and the mechanical response characteristics of nanomaterials at different scales must be covered.

[0111] Young's modulus and Poisson's ratio at the microscopic scale were obtained through nanoindentation experiments and calculated using the indentation depth and load curve.

[0112] The material properties obtained from the experiment are input into the mechanical model, corresponding to different scales and regions (such as grain boundaries and inside grains), and each part is given parameters that adapt to its physical properties.

[0113] Finite element analysis is used to configure a multi-scale model and refine the mesh to ensure that the stress changes in the microstructure are accurately captured:

[0114] Finite element analysis (FEA) is an important tool for solving multi-scale models. Through reasonable mesh division and boundary condition setting, it can accurately capture the details of stress concentration areas and stress changes.

[0115] An adaptive meshing strategy is adopted, using fine meshes in areas with large stress gradients and coarser meshes in areas with small and uniform stresses to balance computational efficiency and accuracy.

[0116] According to the operating conditions of the actual synthesis process, the boundary conditions of the model (such as fixed support, symmetric boundary, etc.) are set to ensure that the model can truly reflect the stress state of the material.

[0117] Through adaptive meshing technology, the rigidity of traditional meshing is overcome, making it possible for finite element analysis to more effectively handle stress changes in nanoscale structures and achieve high-precision simulation.

[0118] Perform a time-stepped dynamic simulation to calculate the stress distribution of the nanomaterial at each moment during the synthesis process and capture areas of stress concentration:

[0119] Depending on the reaction rate during the synthesis process and the mechanical response characteristics of the material, an appropriate time step (usually in the nanosecond to microsecond range) is selected to ensure that dynamic changes can be accurately captured.

[0120] Computer simulation tools (such as ABAQUS and ANSYS) are used to calculate the stress distribution of the material in each time step and store the stress state data at each moment.

[0121] Compare the simulated stress distribution with the experimental data or the preset stress threshold in the literature to evaluate whether the stress at each point exceeds the preset stress threshold, and determine whether the nanomaterial has reached the stress threshold that may cause material defects, thereby marking the high-risk stress area and low-risk stress area of ​​the nanomaterial respectively:

[0122] Based on material mechanics experiments, the stress threshold that may cause material failure (such as cracks, damage) is determined, usually obtained through nanoindentation experiments or fatigue tests.

[0123] The stress data in the dynamic simulation is compared point by point with the stress threshold, and the areas exceeding the stress threshold are identified and marked as high-risk stress areas. That is, the nanomaterials corresponding to the high-risk stress areas have reached the stress threshold that may cause material defects; the areas that do not exceed the stress threshold are marked as low-risk stress areas.

[0124] The following measures can be taken for high stress risk areas:

[0125] Perform stress relief treatments on high-risk stress areas, such as heat treatment or mechanical loosening, to reduce local stress concentration. For heat treatment, temperature adjustment techniques such as annealing or normalizing can be used to rearrange the atomic structure inside the material and reduce internal stress. Mechanical loosening can change the stress state on the surface or inside the material through local grinding or vibration treatment.

[0126] Add more monitoring points in high-risk stress areas, use real-time data monitoring technology to track stress changes in real time, and adjust process parameters or take measures in a timely manner. Use embedded stress sensors or surface acoustic wave sensors to monitor stress status in real time, and feed data back to the control system to adjust process parameters. Combined with intelligent manufacturing systems, process conditions such as temperature, pressure or reaction rate can be automatically adjusted to respond to stress exceeding the standard detected in real time.

[0127] Local reinforcement of areas identified as high risk, such as using high-strength material coatings or fiber reinforcement. Carbon nanotubes, fiber-reinforced composites or other high-performance coatings are used to reinforce high-risk areas to improve the local strength and durability of the material. Surface engineering treatments, such as plating or surface hardening, are performed to enhance the mechanical properties of the surface layer.

[0128] By monitoring the changes in the surface plasmon resonance frequency of the nanoparticles, the probability of failure of nanoparticle size control due to changes in the surface chemistry of the nanoparticles is evaluated, including:

[0129] Surface plasmon resonance (SPR) sensors continuously collect resonant frequency data on the surface of nanoparticles to monitor chemical state changes in real time:

[0130] An SPR sensor is installed in the nanoparticle synthesis device to collect surface resonance frequency data in real time. The SPR sensor can detect changes in the surface plasmon resonance frequency, which reflects changes in the chemical state of the nanoparticle surface.

[0131] The collected data is first subjected to noise filtering and signal enhancement, using digital filtering techniques (such as low-pass filters) to clean up environmental noise and retain chemical change signals.

[0132] Analyze the trend of surface resonance frequency data to identify changes in surface properties caused by chemical adsorption or molecular rearrangement:

[0133] Use time series analysis technology, such as the autoregressive moving average (ARMA) model, to analyze the collected surface resonance frequency data and identify changes in surface characteristics caused by chemical adsorption or molecular rearrangement. The existing technology for analyzing data using the autoregressive moving average (ARMA) model is relatively mature, and we will not go into details here on how to analyze the collected surface resonance frequency data using the autoregressive moving average (ARMA) model.

[0134] By establishing a mathematical model between surface chemical changes and nanoparticle size control failure, the probability of size control failure is predicted:

[0135] A logistic regression model was constructed to link the probability of surface chemistry changes and dimensional control failure, which can be achieved through the following formula: ;in:

[0136] :Given chemical change index The probability of nanoparticle size control failure when .

[0137] : Predictor variables derived from the surface resonance frequency data through the ARMA model.

[0138] and are the parameters of the logistic regression model, determined by the maximum likelihood estimation method.

[0139] : Binary outcome variable for failure of dimensional control, where Indicates that the size control fails. Indicates that dimensional control has not failed.

[0140] The larger it is, the greater the probability of failure of nanoparticle size control due to chemical changes on the nanoparticle surface.

[0141] By real-time monitoring of changes in the resonance frequency of the nanoparticle surface plasmon, the probability of size control failure due to changes in the surface chemical state can be effectively evaluated, thus providing a scientific basis for the optimization of the nanoparticle synthesis process. By using a surface plasmon resonance (SPR) sensor, the tiny signals of chemical changes on the surface of nanoparticles can be captured in real time to ensure that the quality and functionality of the nanoparticles during the synthesis process meet the design requirements. In addition, the combination of time series analysis and the application of logistic regression models not only enhances the accuracy of data processing, but also improves the reliability of predicting the probability of size control failure. This method not only improves the synthesis efficiency of nanomaterials, but also reduces the scrap rate of materials due to size control failure, thereby achieving dual optimization of cost savings and production efficiency.

[0142] By analyzing the dynamic and nonlinear characteristics of nanoparticle aggregation behavior, the risk probability of uneven size distribution caused by aggregation behavior is evaluated, specifically:

[0143] Nanoparticles are considered as nodes in the network, and the physical or chemical interactions between particles are considered as edges in the network. The node connections in the network are defined based on the spatial position, interaction energy or aggregation state of the nanoparticles.

[0144] Each nanoparticle is regarded as a node, and the edge weights between nodes are determined according to the force, distance or surface energy between the two particles.

[0145] Through particle tracking technology or molecular dynamics simulation, the position and interaction force data of nanoparticles are collected in real time to construct a dynamic network structure.

[0146] The modularity of the network was calculated, which evaluates the clustering of groups in the network structure and their distribution in the entire network. Modularity reflects the tightness of nanoparticle aggregation and the significance of clustering.

[0147] The expression of modularity is: ;in:

[0148] : Modularity, used to evaluate the clustering of the network.

[0149] : Node in the network and The actual connection weight between them.

[0150] and The nodes are and The degree of the nodes (i.e. the number of edges connected to them).

[0151] : The total number of edges in the network.

[0152] :When the node and 1 if they belong to the same module, 0 otherwise.

[0153] During the nanoparticle synthesis process, as time passes and the particle aggregation behavior evolves, the modularity of the network is updated in real time to capture the dynamic changes in the particle aggregation behavior.

[0154] The structure and connection weights of the nanoparticle network are updated in real time, and modularity is calculated on the fly.

[0155] By dynamically updating the network modularity, it is possible to reflect changes in the aggregation state in real time, capture key nodes and turning points in nonlinear aggregation behavior, and innovatively adapt to the real-time needs of dynamic systems.

[0156] A correlation model is established between the modularity value and the nanoparticle size distribution to predict the risk probability of uneven size distribution under a specific aggregation state. The expression is: ;in, is the risk probability of uneven size distribution caused by aggregation behavior, is a proportionality factor calibrated according to experimental data (reflecting the particle distribution characteristics under different aggregation states), It is a functional relationship fitted by statistical analysis or machine learning models.

[0157] Among them, the relationship between modularity and uneven size distribution is verified through experiments, and the function parameters are calibrated to ensure the accuracy and reliability of the prediction.

[0158] in, ; are the model coefficients determined through the fitting process and are used to express the effect of modularity on the risk probability of uneven size distribution.

[0159] The larger it is, the greater the risk probability of uneven size distribution due to aggregation behavior.

[0160] The modularity analysis of complex networks is used to evaluate in real time the risk probability of uneven size distribution caused by the aggregation behavior of nanoparticles, which significantly improves the ability to accurately control the synthesis process of nanomaterials. By dynamically updating the modularity value, this method can capture the nonlinear aggregation behavior of nanoparticles in the synthesis process in real time, accurately identify high-risk aggregation states, and effectively reduce the probability of uneven size. Compared with traditional methods, it not only enhances the adaptability to dynamic complex systems, but also improves the automation and intelligence level of the synthesis process, reduces the scrap rate of materials and optimizes production efficiency, providing strong support for the stable production of high-quality nanomaterials.

[0161] Based on a comprehensive analysis of the probability of failure of nanoparticle size control due to changes in the surface chemistry of nanoparticles and the risk probability of uneven size distribution due to aggregation behavior, the probability of changes in the key features that have the greatest impact on size control is predicted, specifically:

[0162] The probability of failure of nanoparticle size control and the risk probability of uneven size distribution caused by aggregation behavior are dimensionlessly processed to eliminate the influence of different dimensions, so that the probability of failure of nanoparticle size control and the risk probability of uneven size distribution caused by aggregation behavior can be comprehensively analyzed on the same scale.

[0163] Based on historical data and experimental results, determine the coefficients used for the weighted summation. The coefficients should reflect the relative importance of each factor in affecting dimensional control. Two weighting coefficients will be selected and To perform weighted summation.

[0164] The probability of failure of nanoparticle size control after dimensionless processing and the risk probability of uneven size distribution caused by aggregation behavior are weighted summed to calculate the probability of change of the key feature with the greatest impact on size control, and its expression is: ;in, is the probability of a change in the key feature that has the greatest impact on dimensional control, and They are respectively the probability of failure of nanoparticle size control after dimensionless processing and the risk probability of uneven size distribution caused by aggregation behavior.

[0165] and It is a weighting coefficient determined through experiments or data analysis, which reflects the impact of various factors on the final probability.

[0166] , the sum of the weighted coefficients is 1 to ensure the correctness of the synthesis of the comprehensive probability value.

[0167] The probability of a change in the key feature with the greatest impact on dimensional control is compared to a preset threshold:

[0168] When the probability of a change in the key feature that has the greatest impact on size control is greater than a preset threshold, a warning signal for feature change is issued, which means that the key process parameters in the nanoparticle synthesis process may have deviated from the ideal state, resulting in potential failure of size control. This situation indicates that the synthesis environment or conditions of the nanoparticles are undergoing significant changes that may affect the quality and consistency of the final product.

[0169] After issuing a warning signal of characteristic change, the measures that can be taken include:

[0170] Automatic adjustment of process parameters: The system can automatically adjust key process parameters (such as reaction temperature, pressure, stirring rate, etc.) immediately to restore the synthesis process to the expected control range.

[0171] Pause production for inspection: Stop the current production process and perform equipment inspection and maintenance to ensure there are no hardware or sensor failures.

[0172] Trigger an alarm to notify the operator: Sound or visual alarms are issued to notify the operator to manually intervene and adjust the synthesis process.

[0173] Data recording and analysis: All relevant data of the abnormal situation will be recorded for subsequent analysis and process optimization.

[0174] When the probability of the key feature with the greatest impact on size control changing is less than or equal to the preset threshold, a normal feature signal is issued, which means that the size control of the nanoparticles is still within an acceptable range and the production process is basically in a stable state. This indicates that the current synthesis conditions and process parameters have not deviated significantly from expectations, and product quality and consistency can be guaranteed.

[0175] The preset threshold is determined by analyzing a large amount of historical production data to determine the probability distribution range of changes in key characteristics under normal production conditions, and extract a statistically significant value as the threshold. For example, a 99% confidence interval can be used to determine the threshold to ensure that under normal circumstances, only a 1% probability will trigger an early warning. Simulation experiments are conducted in the laboratory to observe the changes in key characteristics by changing process conditions and determine under what conditions product quality begins to be significantly affected. The threshold is calibrated and optimized through these experimental data.

[0176] The data in the time interval corresponding to the warning signal of feature change is marked as risk synthesis data. According to the frequency of recent risk synthesis data, it is judged whether the nano-platinum synthesis process needs to be redesigned, specifically:

[0177] After the feature change warning signal is issued, the synthetic data in the corresponding time interval is marked as "risk synthetic data" and stored and managed separately from the regular synthetic data for further analysis:

[0178] Set up a marking system, and when the feature change warning signal is triggered, automatically mark the data within the time period (including key parameters and performance data) as "risk synthetic data". The time period is set according to the actual situation.

[0179] The "risk synthesis data" is stored in a dedicated database and classified with normal production data for subsequent independent analysis.

[0180] Regularly collect statistics and analyze the frequency of occurrence of "risk synthetic data" to identify whether there are abnormal synthetic conditions or equipment status that cause frequent triggering of warnings.

[0181] Set a preset interval. The preset interval is a real-time interval, that is, the end point of the preset interval is always a real-time time point.

[0182] The ratio of risk synthesis data within a preset interval to total synthesis data is marked as the risk synthesis data frequency. The larger the risk synthesis data frequency is, the greater the frequency of occurrence of recent risk synthesis data is, and the more it is necessary to redesign the nanoplatinum synthesis process.

[0183] Set the risk synthetic data frequency threshold and compare the risk synthetic data frequency with the risk synthetic data frequency threshold:

[0184] When the risk synthesis data frequency is greater than the risk synthesis data frequency threshold, it is determined that the nano-platinum synthesis process needs to be redesigned; specifically:

[0185] According to the results of risk data analysis, re-evaluate and adjust key process parameters such as reaction temperature, pressure, catalyst concentration, reaction time, stirring speed, etc. These parameters may deviate from the optimal values ​​in long-term operation and need to be readjusted to ensure the stability of the synthesis process and product quality.

[0186] Use optimization algorithms (such as response surface method and genetic algorithm) to systematically optimize existing process parameters and find new optimal parameter combinations. Test new parameter combinations and verify their effectiveness and stability through small batch trial production.

[0187] Check and maintain production equipment to ensure its normal operation. Equipment aging or sensor failure may lead to unstable process control. Upgrade or replace equipment when necessary.

[0188] Carry out comprehensive overhaul of key equipment (such as reactors, agitators, and temperature control systems), replace aging components, and upgrade to higher-precision sensors and control systems. Introduce intelligent monitoring equipment to improve the level of process automation.

[0189] Review and strengthen quality control measures for raw materials to ensure that raw materials entering the production process meet standards. Batch differences in raw materials may be one of the reasons for instability in the synthesis process.

[0190] Introduce stricter raw material testing standards and processes, re-evaluate raw material suppliers, and establish a raw material traceability system to ensure that the quality of raw materials is controllable and stable.

[0191] Redesign or optimize existing process flows, simplify process steps, reduce uncertainties in intermediate links, and improve overall production efficiency and product quality consistency.

[0192] Through value stream analysis (VSM) of the process, waste and inefficiency in the process are identified and eliminated, and a more lean production method is introduced.

[0193] These measures, combined with problems encountered in actual production, can significantly improve the stability of the nanoplatinum synthesis process and product quality through systematic redesign and optimization, ensuring that the production process meets the expected process standards and quality requirements.

[0194] When the risk synthesis data frequency is less than or equal to the risk synthesis data frequency threshold, it is determined that there is no need to redesign the nano-platinum synthesis process.

[0195] Among them, the setting of the frequency threshold of risk synthetic data is based on the statistical analysis and experimental verification of historical production data. First, by retrospectively analyzing a large amount of historical synthetic data, the normal frequency range of risk synthetic data under different production conditions is determined. Then, combined with laboratory simulation tests, by adjusting process parameters, observe which frequency levels will significantly affect product quality. Finally, statistical methods, such as setting a 95% confidence interval, or based on expert experience, are used to determine a threshold that can balance stability and sensitivity in actual production. This threshold ensures that when a significant deviation from the normal production state is detected, a redesign recommendation can be triggered in a timely manner to avoid the accumulation and amplification of quality problems.

[0196] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0197] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0198] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0199] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0200] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0201] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0202] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0203] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0204] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0205] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A new material data intelligent management system based on feature analysis, characterized in that: It includes synthetic data acquisition module, key feature extraction module, stress risk marking module, chemical change monitoring module, aggregation behavior analysis module, feature change prediction module and redesign judgment module; Synthesis data acquisition module: collects synthesis data during the synthesis of nanoplatinum, including key parameters and performance data; Key feature extraction module: extracts key features related to particle size control and catalytic performance from synthetic data, and screens out the key features that have the greatest impact on size control; Stress risk marking module: Through multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of nanomaterials are marked respectively; Chemical change monitoring module: For low-risk stress areas, the probability of nanoparticle size control failure due to chemical changes on the nanoparticle surface is evaluated by monitoring the changes in the nanoparticle surface plasma resonance frequency; Aggregation behavior analysis module: For low-risk stress areas, the module analyzes the dynamic and nonlinear characteristics of nanoparticle aggregation behavior to assess the risk probability of uneven size distribution caused by aggregation behavior. Feature change prediction module: Based on a comprehensive analysis of the probability of failure of nanoparticle size control due to chemical changes on the nanoparticle surface and the risk probability of uneven size distribution due to aggregation behavior, the module predicts the probability of changes in the key features that have the greatest impact on size control, and issues a feature change warning signal when the probability is greater than a preset threshold; Redesign judgment module: Mark the data in the time interval corresponding to the warning signal of feature change as risk synthesis data, and judge whether the nanoplatinum synthesis process needs to be redesigned according to the frequency of occurrence of recent risk synthesis data; Key features related to particle size control and catalytic performance were extracted from the synthetic data, and the key features with the greatest impact on size control were screened out, including: Principal component analysis was used to extract the main components in the synthetic data and preliminarily screen out the features related to particle size and catalytic performance; Pearson correlation coefficient was applied to calculate the correlation between the characteristics and particle size control and catalytic performance; The importance score of each feature is calculated through the random forest algorithm to identify the key features that have the greatest impact on size control; A recursive feature elimination method is used to gradually remove low-importance features and optimize the final feature set; Through multi-scale analysis of the stress field during the synthesis of nanomaterials, high-risk stress areas and low-risk stress areas of nanomaterials are marked respectively, including: Construct a multi-scale mechanical model including micro and macro scales to simulate the stress distribution of nanomaterials at different scales; Parameterize the physical properties of nanomaterials and characterize the mechanical behavior of nanomaterials at various scales; Use finite element analysis to configure multi-scale models and divide the mesh into fine grids; Perform time-step dynamic simulations to calculate the stress distribution of the nanomaterial at each moment during the synthesis process and capture areas of stress concentration; Compare the simulated stress distribution with the preset stress threshold, evaluate whether the stress at each point exceeds the preset stress threshold, and mark the high-risk stress area and low-risk stress area of ​​the nanomaterial respectively; By monitoring the changes in the surface plasmon resonance frequency of the nanoparticles, the probability of failure of nanoparticle size control due to changes in the surface chemistry of the nanoparticles is evaluated, including: Continuously collect resonant frequency data on the nanoparticle surface; Use time series analysis techniques to analyze the trend of surface resonance frequency data and identify changes in surface properties caused by chemical adsorption or molecular rearrangement; Predict the probability of size control failure by establishing a mathematical model between surface chemical changes and nanoparticle size control failure: Construct a logistic regression model to link surface chemical changes and the probability of size control failure; this can be achieved through the following formula: ; in, is a given chemical change indicator The probability of nanoparticle size control failure is is the predictor variable derived from the surface resonance frequency data through the ARMA model, and are the parameters of the logistic regression model, is the binary outcome variable of dimensional control failure; By analyzing the dynamic and nonlinear characteristics of nanoparticle aggregation behavior, the risk probability of uneven size distribution caused by aggregation behavior is evaluated, specifically: Nanoparticles are considered as nodes in the network, and the physical or chemical interactions between particles are considered as edges in the network; node connections in the network are defined based on the spatial position, interaction energy or aggregation state of the nanoparticles; Through particle tracking technology, the position and interaction force data of nanoparticles are collected in real time to construct a dynamic network structure; The modularity of the network is calculated as: ;in: For modularity, Nodes in the network and The actual connection weight between and The nodes are and The degree, is the total number of edges in the network; :When the node and 1 if they belong to the same module, otherwise 0; Update the structure and connection weights of the nanoparticle network in real time and calculate modularity on the fly; A correlation model is established between the modularity value and the nanoparticle size distribution to predict the risk probability of uneven size distribution under a specific aggregation state. The expression is: ;in, is the risk probability of uneven size distribution caused by aggregation behavior, is a scaling factor calibrated according to experimental data, It is a functional relationship fitted by statistical analysis or machine learning models; in, ; are the model coefficients determined through the fitting process; Based on a comprehensive analysis of the probability of failure of nanoparticle size control due to changes in the surface chemistry of nanoparticles and the risk probability of uneven size distribution due to aggregation behavior, the probability of changes in the key features that have the greatest impact on size control is predicted, specifically: The probability of failure of nanoparticle size control and the risk probability of uneven size distribution caused by aggregation behavior are dimensionally processed; The probability of failure of nanoparticle size control after dimensionless processing and the risk probability of uneven size distribution caused by aggregation behavior are weighted summed to calculate the probability of change of the key feature that has the greatest impact on size control; The probability of a change in the key feature with the greatest impact on dimensional control is compared to a preset threshold: When the probability of a change in the key feature with the greatest impact on size control is greater than a preset threshold, a feature change warning signal is issued; when the probability of a change in the key feature with the greatest impact on size control is less than or equal to the preset threshold, a feature normal signal is issued.

2. The new material data intelligent management system based on feature analysis according to claim 1 is characterized in that: Collect synthesis data during the synthesis of nanoplatinum, specifically: Key parameters include temperature, pressure, reaction time, precursor concentration, and stirring rate; Performance data include particle size distribution, surface morphology, crystal structure, catalytic activity, and surface chemistry.

3. The new material data intelligent management system based on feature analysis according to claim 2 is characterized in that: The data in the time interval corresponding to the warning signal of feature change is marked as risk synthesis data. According to the frequency of recent risk synthesis data, it is judged whether the nano-platinum synthesis process needs to be redesigned, specifically: After issuing a warning signal of feature change, the synthetic data in the corresponding time interval is marked as risky synthetic data; Set preset intervals; The ratio of risk synthetic data within a preset interval to total synthetic data is marked as the frequency of risk synthetic data; Set the risk synthetic data frequency threshold and compare the risk synthetic data frequency with the risk synthetic data frequency threshold: When the risk synthesis data frequency is greater than the risk synthesis data frequency threshold, it is determined that the nano-platinum synthesis process needs to be redesigned; When the risk synthesis data frequency is less than or equal to the risk synthesis data frequency threshold, it is determined that there is no need to redesign the nano-platinum synthesis process.