Sodium ferric phosphate electrode material performance optimization method

Abstract

The invention provides a sodium ferric phosphate electrode material performance optimization method, which comprises the following steps: acquiring image data of an electrode material through imaging equipment, and determining particle size distribution characteristics by adopting an image processing technology; according to the particle size distribution characteristics, grouping the particles by using a classification method, and determining potential influence indexes of conduction efficiency and reaction activity; according to the surface property evaluation result, adjusting particle size distribution parameters, simulating reaction area change, and judging the reaction area increase degree; through the reaction area improvement degree, combining with a performance simulation model to input optimization parameters, running loop test simulation, and obtaining capacity fading rate prediction data; according to the capacity fading rate prediction data, constructing an objective function, adopting an optimization algorithm for iterative solution, and determining a balance configuration scheme of the particle size and the surface property; according to the balance configuration scheme, material synthesis control parameters are generated, the application effect is verified through numerical simulation, and a final performance optimization result is obtained.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for optimizing the performance of sodium iron phosphate electrode materials. Background Technology

[0002] In the fields of materials science and new energy, researching and optimizing battery material performance is considered one of the core driving forces for technological innovation. Especially cathode materials like sodium iron phosphate, whose performance directly affects battery energy density and cycle stability, have irreplaceable value for the development of electric vehicles and energy storage devices. However, how to improve overall performance through precise control of material properties remains a focus of industry attention.

[0003] Currently, although many studies and methods are dedicated to improving the performance of such materials, they often overlook the complex interactions between different material properties. Existing solutions mostly focus on improving single properties, failing to fully consider the synergistic effects between properties, resulting in limited optimization results. A deeper problem lies in the lack of a comprehensive understanding of the microscopic properties of materials, especially in practical applications where the performance of materials is often influenced by a combination of factors, the dynamic nature and complexity of which are underestimated.

[0004] Against this backdrop, particle size and surface properties have become key technical factors affecting material performance. Particle size not only determines the material's conductivity and reactivity in the battery but also further influences surface properties such as adsorption capacity and chemical stability. If the particle size is unsuitable, surface properties may not be fully utilized; for example, excessively large particles reduce the surface reaction area, thus limiting the material's electrochemical performance. In actual production, this problem often manifests during battery charging and discharging, resulting in uneven material reactions and even capacity decay, severely impacting product lifespan and safety.

[0005] Therefore, finding the optimal balance between particle size and surface properties, and achieving synergistic optimization through comprehensive regulation, has become a key issue in improving the performance of sodium iron phosphate materials. Solving this problem not only concerns the improvement of the material itself, but also directly impacts the overall progress of battery technology and the feasibility of its market applications. Summary of the Invention

[0006] This invention provides a method for optimizing the performance of sodium iron phosphate electrode materials, mainly including:

[0007] Image data of the electrode material is acquired using an imaging device, and image processing techniques are employed to determine the particle size distribution characteristics. Based on these characteristics, the particles are grouped using a classification method to identify potential influencing indicators for conductivity and reactivity. For these conductivity and reactivity indicators, response data from electrochemical tests is acquired, and quantitative values ​​of adsorption capacity and chemical stability are calculated through data analysis to obtain surface property evaluation results. Based on these surface property evaluation results, particle size distribution parameters are adjusted to simulate changes in reaction area and determine the degree of reaction area improvement. Using the degree of reaction area improvement, combined with optimization parameters input into a performance simulation model, cyclic testing simulations are run to obtain capacity decay rate prediction data. Based on the capacity decay rate prediction data, an objective function is constructed, and an optimization algorithm is used iteratively to determine a balanced configuration scheme for particle size and surface properties. For this balanced configuration scheme, material synthesis control parameters are generated, and the application effect is verified through numerical simulation to obtain the final performance optimization results. Furthermore, the step of acquiring image data of the electrode material through an imaging device and determining particle size distribution characteristics using image processing technology includes: acquiring raw image data of the electrode material through an imaging device; processing the raw image data using image segmentation technology to separate individual particle regions; applying a boundary detection method to determine the particle outline boundary for each individual particle region; calculating the diameter value of each particle based on the particle outline boundary to obtain particle size information; if abnormal data exists in the particle size information, filtering is performed through a preset threshold range to remove diameter values ​​that do not meet the conditions, obtaining corrected particle size information; determining the particle size distribution characteristics by statistical analysis of the corrected particle size information; and generating a distribution data table based on the particle size distribution characteristics to record the proportion of particles in different size ranges. Furthermore, the step of grouping particles according to the particle size distribution characteristics and determining potential influencing indicators of conductivity and reactivity includes: processing the particle size distribution characteristics using a pre-established classification model, grouping the particles by size using a support vector machine algorithm to obtain preliminary division results of small-size and large-size groups; obtaining the proportion information of the two groups of particles based on the preliminary division results to determine the proportion characteristics; obtaining the electrochemical response data of the two groups of particles under a specific environment based on the proportion characteristics; if the response data of the small-size group is higher than a preset threshold, it is marked as a high-activity group; if the response data of the large-size group is lower than a preset threshold, it is marked as a low-activity group; obtaining conductivity difference data through the identification information of the high-activity and low-activity groups, processing the difference data with statistical tools, judging the significance distinction results, and determining the potential correlation indicators between reactivity and particle size grouping.Furthermore, the step of acquiring response data from electrochemical tests for the conductivity and reactivity indicators, and calculating the quantitative values ​​of adsorption capacity and chemical stability through data analysis to obtain surface property evaluation results includes: acquiring material response data from electrochemical tests, recording multiple sets of response data to obtain an initial test dataset; cleaning and normalizing the initial test dataset using data processing methods to obtain a standard dataset; analyzing the response data using curve fitting methods based on the standard dataset to calculate the quantitative value of adsorption capacity and obtain adsorption performance evaluation data; using the adsorption performance evaluation data and the response data of chemical stability, performing comparative analysis using statistical tools to determine the quantitative value of chemical stability; if the quantitative value of chemical stability is lower than a preset threshold, performing secondary extraction on the response data to obtain a supplementary dataset; and comprehensively analyzing the supplementary dataset and the adsorption performance evaluation data to obtain the surface property evaluation results.

[0008] Furthermore, the step of adjusting particle size distribution parameters based on the surface property assessment results, simulating changes in reaction area, and determining the degree of reaction area improvement includes: obtaining adsorption capacity data based on the surface property assessment results; if the adsorption capacity data is lower than a preset threshold, adjusting the particle size distribution parameters using an optimization algorithm to obtain a preliminary distribution parameter scheme; performing simulation calculations for different size ratios based on the preliminary distribution parameter scheme to obtain corresponding data between the size ratio and the reaction area, and determining the trend of change; selecting parameter configurations with higher reaction area improvement based on the trend of change and the size ratio adjustment range, and obtaining an optimized size distribution scheme; obtaining distribution data for the optimized size distribution scheme, standardizing and organizing it using data processing tools, and determining the final parameter configuration result; and simulating reaction area performance data based on the final parameter configuration result to determine the stability of the improvement and obtain optimization effect data.

[0009] Furthermore, the step of obtaining capacity decay rate prediction data by combining the reaction area increase with the performance simulation model input optimization parameters and running cyclic test simulations includes: standardizing and organizing the reaction area increase using data processing tools to obtain structured improvement data records; initializing and configuring parameters based on the structured improvement data records and the performance simulation model to determine the basic parameter settings of the simulation environment; using the basic parameter settings, using a genetic algorithm to screen the optimized parameter combinations to obtain an adaptive parameter configuration scheme; executing cyclic simulation tests according to the adaptive parameter configuration scheme to obtain capacity decay data and determine the trend of change; using the trend of change to perform segmented statistics using data processing tools to obtain the decay rate distribution of different simulation stages; if the decay rate of a certain stage exceeds a preset threshold, the simulation data of that stage is marked to determine abnormal fluctuations, and the final capacity decay rate prediction result is obtained.

[0010] Furthermore, the step of constructing an objective function based on the capacity decay rate prediction data, and iteratively solving it using an optimization algorithm to determine a balanced configuration scheme for particle size and surface properties includes: obtaining an initial dataset using the capacity decay rate prediction data, cleaning and standardizing it using preprocessing tools to obtain a structured basic data set; constructing an objective function based on the structured basic data set, and using particle size and surface properties as variable inputs in conjunction with a collaborative optimization method to determine the mathematical expression of the objective function; iteratively solving the mathematical expression of the objective function using a genetic algorithm to obtain an approximate solution set for the function's minimum value; analyzing the correspondence between particle size and surface properties using the approximate solution set; if the deviation exceeds a preset threshold range, adjusting the variable weights and obtaining a new approximate solution set; generating a balanced configuration scheme based on the new approximate solution set, and determining the final configuration scheme set through multiple rounds of iterative verification.

[0011] Furthermore, the step of generating material synthesis control parameters for the balanced configuration scheme, verifying the application effect through numerical simulation, and obtaining the final performance optimization result includes: obtaining initial data through the balanced configuration scheme, constructing a preliminary material synthesis model, and determining the range of control parameters; simulating the synthesis process using numerical simulation tools based on the preliminary model to obtain synthesis results under different control parameters; if the synthesis results deviate from the actual environmental conditions, adjusting the control parameters, re-performing the numerical simulation, and judging the effect of the adjusted synthesis; determining the environmental adaptability assessment result by combining the adjusted synthesis effect data with key indicators of the actual environment; filtering the control parameters using the support vector machine algorithm based on the environmental adaptability assessment result to obtain an optimized parameter combination; generating the final synthesis process flow through the optimized parameter combination, determining the application effect, and generating an effect evaluation report.

[0012] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0013] This invention discloses a method for improving electrode material performance based on image processing and machine learning optimization. Addressing the unique business scenario of how particle size distribution affects conductivity, reactivity, and capacity decay, this method solves the challenge of material performance optimization through multi-dimensional technology integration. First, the invention acquires particle images using imaging equipment, employs image segmentation technology to accurately separate individual particles and calculate their diameters, obtaining size distribution characteristics. Then, machine learning clustering methods are used to classify particles into different size groups. Combined with electrochemical test data and curve fitting analysis, adsorption capacity and chemical stability are quantified. If performance does not meet standards, optimization algorithms are used to adjust size distribution parameters, simulating the effect of increased reaction area, and the capacity decay rate is predicted using a performance model. Finally, the objective function is iteratively solved to determine the balanced configuration of particle size and surface properties, generating synthesis control parameters and verifying the application effect. This invention, through the synergistic effect of image analysis, machine learning, and optimization algorithms, significantly improves the performance stability and lifespan of electrode materials, providing a scientific basis for material design. Attached Figure Description

[0014] Figure 1 This is a flowchart of a method for optimizing the performance of sodium iron phosphate electrode material according to the present invention.

[0015] Figure 2 This is a schematic diagram of a method for optimizing the performance of sodium iron phosphate electrode material according to the present invention.

[0016] Figure 3 This is another schematic diagram of a method for optimizing the performance of sodium iron phosphate electrode material according to the present invention.

[0017] Figure 4 This is another schematic diagram of a method for optimizing the performance of sodium iron phosphate electrode material according to the present invention.

[0018] Figure 5 This is another schematic diagram of a method for optimizing the performance of sodium iron phosphate electrode material according to the present invention. Detailed Embodiments

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1-5 The method for optimizing the performance of sodium iron phosphate electrode material in this embodiment may specifically include:

[0021] S101. Image data of the electrode material is acquired through an imaging device, individual particle regions are separated using image segmentation technology, and the diameter value of each particle is calculated to obtain the particle size distribution characteristics.

[0022] Images of the electrode material are acquired using an imaging device to obtain raw image data, ensuring that the acquired images clearly reflect the particle morphology. Preliminary denoising processing is performed on the acquired raw image data to generate a first processed image, reducing the impact of background interference on subsequent processing. For the first processed image, an image segmentation method is used to separate the particle regions from the background, resulting in multiple independent particle region images. For each separated particle region image, boundary extraction is performed to generate the contour boundary data for each particle. Based on the generated contour boundary data, the diameter value of each particle is calculated to obtain preliminary particle size data. The preliminary particle size data is then filtered using a preset threshold range to remove outliers, generating corrected particle size data. The corrected particle size data is then statistically processed to determine the proportion of particles within different size ranges, generating a particle size distribution characteristic data table. This data table visually reflects the particle size distribution characteristics of the electrode material.

[0023] For example, during the image acquisition process of electrode materials, scanning the material sample with an optical microscope as an imaging device can obtain high-resolution raw image data. This method helps to capture subtle morphological differences in particles, thus providing a reliable basis for subsequent processing.

[0024] Specifically, when using an optical microscope, the electrode material is first placed on a glass slide, and imaging is performed under uniform lighting conditions to ensure that the particle outlines are clearly visible in the image. This effectively avoids shadow interference caused by uneven lighting and improves data accuracy.

[0025] In one possible implementation, if the material particles are small, a scanning electron microscope can be used to obtain more detailed images of the surface. This device scans the sample surface with an electron beam to generate images that reflect the three-dimensional morphology of the particles, which is helpful for identifying micron-sized particles and thus improving the accuracy of the overall distribution characteristics.

[0026] For example, in the application of lithium battery electrode materials, this acquisition method can capture the aggregation state of active material particles, avoiding size deviations caused by subsequent segmentation errors.

[0027] In one possible implementation, the denoising of the original image data can be achieved by using median filtering to replace noise points in the image with the median of neighboring pixels, thereby generating a first processed image. This process helps to smooth background noise while preserving the sharpness of particle edges, which is beneficial for the accuracy of subsequent segmentation.

[0028] Specifically, median filtering effectively removes salt and pepper noise, such as random white or black dots that appear in an image, by selecting a fixed-size window, such as a 3x3 pixel area, sorting each pixel and replacing the original value with the median value. This makes granular areas easier to distinguish.

[0029] In one possible implementation, if the image is significantly affected by Gaussian noise, Gaussian filtering can be combined. First, a weighted average of the pixels within the window is calculated, where the weights are based on a Gaussian distribution function. This generates an image that reduces blur while maintaining the true size information of the particles, which is beneficial to the reliability of boundary extraction.

[0030] For example, when processing images of carbon-based electrode materials, this denoising can highlight the gaps between particles, preventing noise from being misjudged as part of particles, thus supporting a true reflection of the size distribution.

[0031] For example, for image segmentation of the first processed image, a threshold segmentation method can be used. A threshold is set according to the pixel gray value, and the area above the threshold is regarded as a particle, thereby separating the independent particle region image. This method is simple and efficient and is beneficial for quickly processing a large amount of image data.

[0032] Specifically, threshold segmentation first calculates the image histogram, finds gray-level peaks and valleys as threshold points, and then binarizes the pixels, for example, setting the particle region to white and the background to black. This results in a clear image of the region that isolates the particles, which is beneficial to the accuracy of subsequent boundary work.

[0033] In one possible implementation, for complex backgrounds, the Otsu method can be used to automatically calculate the optimal threshold. This method determines the threshold by maximizing the inter-class variance. For example, assuming the image has two classes, foreground and background, the peak variance at different thresholds is calculated to optimize the segmentation effect. This approach can adapt to the variability of images with different materials and is beneficial for generating accurate granular regions.

[0034] In one possible implementation, Canny edge detection can be used for boundary extraction. Gradient calculation and high / low threshold filtering are applied to the particle region image to generate contour boundary data. This detection first smooths the image, then finds the edge intensity by taking the derivative, and finally connects the edge points to form a closed boundary, which is beneficial for accurately capturing particle shape.

[0035] For example, during extraction, the horizontal and vertical gradients are first calculated using the Sobel operator, and then non-maximum suppression is used to remove false edges, thus generating reliable boundary data for diameter calculation.

[0036] In one possible implementation, if the particle shape is irregular, an active contour model can be combined to initialize a curve and iteratively converge to the real boundary. This can handle uneven particles and help avoid boundary breakage.

[0037] For example, when calculating the diameter value, the minimum enclosing circle is fitted to the contour boundary data to obtain the diameter of each particle as the diameter value of the circle, thereby forming preliminary particle size data. This fitting is achieved by solving the minimum enclosing circle of the boundary points, which is beneficial for quantifying the size of irregular particles.

[0038] Specifically, the minimum enclosing circle algorithm selects three points from the set of boundary points to form a circle, and iteratively checks whether all points are inside the circle. If not, it adjusts the circle. The diameter calculated in this way accurately reflects the particle scale.

[0039] In one possible implementation, for elliptical particles, the average of the major and minor axes can be calculated as the diameter, resulting in more comprehensive size data that is beneficial for targeted subsequent screening.

[0040] In one possible implementation, when screening the initial particle size data, a preset threshold range is set, such as the upper and lower limits based on the average size, to remove outliers that exceed the range and generate corrected data. This screening helps to eliminate erroneous measurements caused by imaging artifacts and is beneficial to data purity.

[0041] For example, if the average diameter is 10 micrometers, the threshold is set to 5 to 15 micrometers to eliminate values ​​that are too small or too large, thereby ensuring the representativeness of the distribution characteristics.

[0042] For example, during statistical processing, the corrected data is grouped and counted to determine the proportion of particles in size ranges such as 0-5 micrometers, 5-10 micrometers, etc., and a distribution characteristic data table is generated. This table is presented in the form of a bar chart or table, which is helpful for intuitive analysis of material uniformity.

[0043] Specifically, the statistics first sort the size values, and then calculate the frequency percentage by bin. This way, the data table can highlight the peak size range, which is helpful for evaluating the performance potential of electrode materials.

[0044] S102. Based on the particle size distribution characteristics, a machine learning classification method is used to cluster the distribution, dividing the particles into small-size groups and large-size groups, thereby determining the potential influencing indicators of conduction efficiency and reactivity.

[0045] Based on the particle size distribution characteristics, a pre-established classification rule is used to initially group the particle data, dividing the particles into an initial small-size group and an initial large-size group. By comparing the quantity distribution of the two groups, the proportion of the initial small-size and initial large-size groups in the total particles is obtained. Based on the proportion data of the initial small-size and initial large-size groups, the electrochemical response values ​​of the two groups under a specific environment are obtained. By comparing the response values ​​of the two groups with a preset threshold, it is determined whether the initial small-size group possesses high activity characteristics, and the group that meets the criteria is marked as the first active group. Based on the marking results of the first active group, the electrochemical response values ​​of the initial large-size group under the same environment are obtained and compared with a preset threshold. If the response value is lower than the threshold, it is marked as the second active group. The activity difference data of the two groups is compiled based on the marking results. Based on the activity difference data of the first and second active groups, the performance values ​​of the two groups of particles in terms of conductivity are obtained. Through data comparison processing, the potential correlation index between conductivity and particle size grouping is determined, completing the preliminary judgment of the factors affecting the reactivity.

[0046] For example, the generation steps are as follows:

[0047] 1. In one embodiment, the processing of particle size distribution characteristics first involves collecting particle sample data, such as size measurements obtained from scanning electron microscopy in nanomaterials research. These values ​​show a distribution ranging from 10 nanometers to 100 nanometers. Particles are grouped into an initial small-size group (e.g., particles less than 50 nanometers) and an initial large-size group (e.g., particles greater than 50 nanometers) using pre-established classification rules, such as a threshold division based on the median size. This grouping helps reveal the impact of size on material properties. Proportional data is calculated by comparing the quantity distribution of the two groups, for example, the small-size group accounts for 60% and the large-size group accounts for 40%. This proportional data provides a basis for subsequent activity assessment, ensuring the balance of the grouping to improve the accuracy of conductivity analysis.

[0048] 2. Specifically, based on the ratio data of the initial small-size group and the initial large-size group, the electrochemical response values ​​of the two groups in a specific environment, such as an electrolyte solution, are obtained. For example, the current density is measured using cyclic voltammetry. The small-size group may show a higher peak current, indicating a stronger response. By comparing it with a preset threshold, such as 0.5 mA per square centimeter, it is determined whether the initial small-size group has high activity characteristics. If it exceeds the threshold, it is marked as the first active group. This marking process helps to distinguish highly active particles, thereby optimizing the reaction rate of the material in battery applications and improving the overall energy conversion efficiency.

[0049] 3. In one embodiment, based on the labeling results of the first active group, the electrochemical response values ​​of the initial large-size group under the same electrolyte environment are further obtained. For example, a low current density such as 0.2 mA per square centimeter is measured and compared with a preset threshold. If it is lower than the threshold, it is labeled as the second active group. By sorting the labeling results of the two groups, the activity difference data such as an activity ratio of 3:1 is obtained. This difference data reveals the lower responsiveness of the large-size particles, which is beneficial for targeted improvement of the material synthesis process and enhancement of the stability of the conduction path.

[0050] 4. Specifically, based on the activity difference data of the first and second active groups, the performance values ​​of the two groups of particles in terms of conduction efficiency are obtained. For example, the conduction resistance value is obtained by impedance spectroscopy. The lower resistance of the smaller size group indicates efficient conduction. By comparing and processing the data, the potential correlation index between conduction efficiency and particle size grouping is determined. For example, the correlation coefficient shows that the smaller size group improves efficiency by 20%. This determination process completes the preliminary judgment of the factors affecting the reaction activity, which helps to guide particle optimization to achieve higher catalytic performance.

[0051] S103. For the aforementioned conductivity and reactivity indicators, obtain the material's response data in electrochemical testing, and calculate the quantitative values ​​of adsorption capacity and chemical stability through curve fitting analysis, thereby obtaining the surface property evaluation results.

[0052] For the aforementioned conductivity and reactivity indicators, response data of the material was obtained from electrochemical tests. Multiple sets of response data under different test conditions were recorded to form an initial response data set, completing preliminary data acquisition. For the initial response data set, data cleaning tools were used to denoise and impute missing values, and normalization was performed to organize the processed data into standardized response data sets, ensuring data consistency and usability. Based on the standardized response data sets, curve fitting methods were applied to process the data, calculating the quantitative value of the material's adsorption capacity and generating adsorption performance evaluation data, providing a basis for subsequent surface property evaluation. Based on the adsorption performance evaluation data, and combined with relevant chemical stability response data, statistical tools were used for comparative analysis to determine the quantitative value of chemical stability, ultimately forming the comprehensive surface property evaluation results.

[0053] In one possible implementation, when obtaining material response data from electrochemical tests, the relationship curve between current and voltage can be recorded by cyclic voltammetry. For example, when testing lithium-ion battery cathode materials in an acidic electrolyte environment, the change in peak current can be observed to reflect the conduction efficiency. This can accurately capture the electron transfer rate of the material at different potentials, which is beneficial to the accuracy of subsequent quantitative analysis.

[0054] It should be noted that the process of recording multiple sets of response data to form the initial response data set involves setting various test conditions, such as changing the scan rate from slow to fast. For example, when the scan rate is 10 mV / s, the curve obtained shows a higher peak current, indicating strong reactivity. This helps to verify the reliability of the data from multiple perspectives, avoids deviations under a single condition, and thus provides a comprehensive foundation for data cleaning.

[0055] For example, when testing graphene composite materials, collecting data at different pH values ​​can mutually support each other to reveal the material's sensitivity to the environment, ensuring the integrity of the initial data collection and improving the robustness of the overall assessment.

[0056] In one possible implementation, when using data cleaning tools to denoise the initial response dataset, median filtering can be used to remove noise points. For example, this method can be applied to current data sequences to replace abnormally high values ​​with the median of nearby points. This can smooth the curve while preserving key features, which is beneficial for improving the signal-to-noise ratio of the data.

[0057] It should be noted that missing value imputation can be achieved through linear interpolation. For example, if a voltage point lacks a current reading, interpolation is calculated based on the points before and after it to fill the missing value. This supports the continuity of the data from the perspective of accuracy. At the same time, normalization operations such as min-max scaling are performed to map the data to the 0-1 range, which helps to eliminate dimensional differences and facilitates subsequent curve fitting.

[0058] For example, when processing impedance spectral data of supercapacitor materials, this combination of cleaning and normalization mutually supports each other from multiple directions of noise interference and scale inconsistencies, ensuring the consistency of standardized response data sets and thus providing reliable input for analysis.

[0059] In one possible implementation, when applying curve fitting methods based on a standardized set of response data, the Langmuir adsorption isotherm can be used to fit the current-concentration curve. For example, the adsorption capacity can be calculated and quantified by fitting parameters such as the maximum adsorption capacity. For instance, the constant obtained after fitting reflects the affinity of the material surface for ions. This approach can reveal the adsorption mechanism from a quantitative perspective and is beneficial for generating accurate adsorption performance evaluation data.

[0060] It should be noted that the process of this method involves optimizing the fitting error using the least squares method.

[0061] Specifically, the model is first assumed to be of a certain form, and then the parameters are iteratively adjusted until the residual is minimized. This supports the accuracy of the results from the perspective of mathematical fitting and is beneficial to laying the foundation for chemical stability assessment.

[0062] For example, when evaluating metal-organic framework materials, fitting data at different temperatures can mutually support each other to demonstrate the temperature dependence of adsorption capacity, ensuring the comprehensiveness of the evaluation data.

[0063] In one possible implementation, when combining the assessment data of adsorption performance with the relevant response data of chemical stability, statistical tools such as t-tests are used for comparative analysis. For example, the difference in adsorption amount before and after cycling is compared to determine the quantification value. For instance, if the difference is significant, the quantification value is low, indicating poor stability. This approach can objectively assess the durability of the material and is beneficial for forming a comprehensive surface property assessment result.

[0064] It should be noted that the comparative analysis process includes calculating the mean and standard deviation and then applying statistical tests.

[0065] Specifically, grouping the data first and then testing the significance provides support for the reliability of the quantitative values ​​from the perspective of statistical significance, which helps to avoid subjective bias.

[0066] For example, when testing polymer electrolytes, the mutual support from multiple directions of cycling stability and adsorption decay can generate comprehensive surface property assessment results, thereby achieving a complete evaluation of conductivity and reactivity indicators.

[0067] S104. If the surface property evaluation results show that the adsorption capacity is lower than the threshold, the optimization algorithm is used to adjust the parameter scheme of particle size distribution, simulate the reaction area change under different size ratios, and thus determine the degree of improvement of the optimized reaction area.

[0068] If the surface property assessment results show that the adsorption capacity is lower than a preset threshold, specific adsorption capacity data is obtained from the assessment results, and a parameter adjustment process is initiated based on this data to determine the initial adjustment direction of the particle size distribution, resulting in an initial size distribution parameter scheme. For the initial size distribution parameter scheme, the distribution of particle size at different ratios is simulated, and the reaction area data corresponding to each ratio is calculated to obtain the trend of reaction area change. Based on the trend of reaction area change, combined with the adjustment range of the size ratio, parameter configurations with higher reaction area improvement are selected using data processing tools to form an optimized size distribution scheme. For the optimized size distribution scheme, the reaction area performance data under the adjusted particle size distribution is simulated, specific area change values ​​are obtained, and the degree of improvement in reaction area after optimization is determined.

[0069] For example, when the surface property assessment results show that the adsorption capacity is lower than a preset threshold, specific adsorption capacity data is obtained from the assessment results. This data usually includes adsorption rate indicators of the particle surface, such as the number of molecules adsorbed per unit area measured experimentally. If this number is insufficient to meet the reaction requirements, a parameter adjustment process is initiated. This process involves analyzing the deviation of the current particle size distribution, such as particles that are too large, resulting in a reduction of adsorption sites, thereby determining the initial adjustment direction, such as increasing the proportion of small particles to increase the total surface area, and obtaining an initial size distribution parameter scheme. Doing so can effectively identify the root cause of the problem and initially plan the optimization path, which is beneficial to the accuracy of subsequent simulations.

[0070] In one possible implementation, for an initial size distribution parameter scheme, the distribution of particle size at different proportions is simulated. For example, assuming that the proportion of small particles in the initial scheme is 30%, the simulation is performed by adjusting it to 40% or 50%, and the reaction area data corresponding to each proportion is calculated. This calculation process is based on the surface area formula, where the reaction area is inversely proportional to the particle size. Small particles provide more exposed surface, and the trend of reaction area change is obtained. For example, the trend shows that the area increases linearly when the proportion of small particles increases. Doing so can reveal the influence of size ratio on area, which is beneficial for quantifying optimization potential and providing a data basis for screening.

[0071] For example, based on the changing trend of the reaction area and combined with the adjustment range of the size ratio, such as limiting the range to 20% to 60% to avoid aggregation problems, the parameter configuration with a higher increase in reaction area is screened out by data processing tools. This tool can use a sorting algorithm to arrange the simulation data and select the configuration that increases the initial area by more than 15%, forming an optimized size distribution scheme. This can extract the best option from multiple sets of simulations, which is beneficial to ensuring the feasibility of the scheme and its stability in practical applications.

[0072] In one possible implementation, for the optimized size distribution scheme, the reaction area performance data under the adjusted particle size distribution is simulated. For example, after the reduction of large particles in the optimization scheme, the reaction process is reproduced in a virtual environment to obtain specific area change values. For example, the change shows that the total area has increased from the initial value to a higher level. The degree of improvement of the optimized reaction area is judged. This judgment is based on numerical comparison. For example, if the increase is stable within a preset range, this can verify the effectiveness of the optimization and help confirm the overall improvement of adsorption capacity by particle size adjustment, thereby achieving the technical goal.

[0073] S105. Based on the degree of increase in reaction area, combined with the optimization parameters input into the performance simulation model, a cyclic test simulation is run to obtain the predicted data of capacity decay rate.

[0074] By analyzing the improvement data of the reaction area and combining it with a pre-established performance simulation tool, the improvement data is organized and input to complete the parameter initialization configuration in the simulation environment and determine the basic operating conditions. Based on the basic operating conditions, the range of optimized parameters is filtered to obtain a set of suitable parameter configuration schemes for subsequent simulation testing preparation. For the suitable parameter configuration scheme, cyclic simulation tests are run to obtain capacity decay data after each simulation and record the trend of decay data changes. Based on the trend of decay data changes, data processing tools are used to perform segmented statistical analysis on the data to obtain the decay rate distribution at different simulation stages, providing a basis for the final capacity decay rate prediction data.

[0075] For example, when processing data on the degree of increase in reaction area, initial area values ​​are first obtained from experimental measurements, such as baseline data obtained by observing the particle surface through scanning electron microscopy. Then, combined with a pre-established performance simulation tool, which is a software environment based on the finite element method to simulate the behavior of materials under cyclic loading, these data are organized into a standardized input format to complete the parameter initialization configuration. This helps to ensure that the simulation accurately reflects the actual conditions and is beneficial to the reliability of subsequent predictions.

[0076] In one possible implementation, the range of optimization parameters is filtered based on the basic operating conditions. For example, upper and lower limits are set for temperature and pressure variables, and a suitable parameter configuration scheme is obtained through a grid search method. This scheme can optimize resource allocation, avoid invalid calculations, and help improve simulation efficiency and reduce errors.

[0077] For example, running cyclic simulation tests for the adapted parameter configuration scheme, such as simulating 100 battery charge-discharge cycles, recording capacity decay data and plotting the trend curve each time, can reveal the decay mode, which is helpful in identifying potential problems and guiding material improvement.

[0078] In one possible implementation, data processing tools are used to perform segmented statistics based on the changing trend of the decay data. For example, the simulation is divided into initial, intermediate and late stages, and the average decay rate distribution of each stage is calculated. This provides a basis for capacity decay rate prediction data, which is beneficial for quantifying risks and supporting decision optimization.

[0079] For example, the specific process of a performance simulation tool involves mapping input data onto a virtual mesh, such as simulating ion transport by iteratively solving diffusion equations, without the need for complex formulas, only by adjusting the mesh resolution to match the actual scenario. This ensures consistency from area enhancement to attenuation prediction, which is beneficial to the robustness of the overall model.

[0080] In one possible implementation, when selecting parameter configuration schemes, one can examine them from multiple perspectives, such as verifying the effectiveness of the scheme by combining historical datasets and testing its stability under different load conditions. This mutually supports the comprehensiveness of the scheme and is beneficial for adapting to changing environments.

[0081] For example, the example of cyclic simulation testing can be extended to multivariate scenarios, such as simultaneously changing the humidity factor to observe the change in the decay trend. This indirectly supports the accuracy of trend recording and is beneficial for capturing nonlinear effects.

[0082] In one possible implementation, the segmented statistics process involves using statistical software to group the data, for example, summarizing the decay rate by time intervals. This is directly linked to the aforementioned trend, which helps to generate accurate distribution and ultimately support capacity decay rate prediction.

[0083] For example, the entire process, from area data processing to distribution information acquisition, forms a closed loop. For instance, the initial input affects the final output, which ensures rigorous logic and is beneficial for improving prediction accuracy in practical applications.

[0084] S106. Based on the predicted capacity decay rate data, obtain the collaborative optimization objective function, and use an optimization algorithm to iteratively solve for the minimum value of the function, thereby determining a balanced configuration scheme for particle size and surface properties.

[0085] Based on the predicted capacity decay rate data, an initial predicted data set is obtained from the data storage source. This data set is then cleaned and formatted, outliers are removed, and standardization is performed to obtain a cleaned basic data set, which serves as the input for constructing the objective function. For this cleaned basic data set, particle size and surface property status are used as variables to construct an objective function expression for collaborative optimization. The minimum value of this expression is solved iteratively to obtain a preliminary function solution set, which is used to analyze the relationships between variables. Based on the preliminary function solution set, the correspondence between particle size and surface property status is analyzed. If the deviation between the two exceeds a preset threshold, the weight ratio of the variables is adjusted, and an updated function solution set is recalculated for further configuration optimization. For the updated function solution set, a balanced configuration scheme for particle size and surface property status is generated. This scheme is verified to meet the requirements of the predicted capacity decay rate data, resulting in the final configuration result used to determine the balanced configuration scheme for particle size and surface properties.

[0086] For example, the generation steps are as follows:

[0087] For example, when processing the predicted data of capacity decay rate, an initial set of predicted data is obtained from the data storage source. This set of data usually contains numerical records from multiple sources. By cleaning and formatting to remove outliers and performing standardization, noise interference can be effectively reduced, thus obtaining a cleaned basic data set. This process helps to ensure that the input basis for the subsequent construction of the objective function is more accurate and reliable, because outliers may amplify the bias, while standardization can unify the units of measurement and promote fair comparison between variables. Doing so can bring about the technical effect of improving the optimization accuracy.

[0088] In one possible implementation, for the processed basic data set, particle size and surface property state are used as variables to construct an objective function expression for collaborative optimization. This expression often integrates multi-dimensional parameters, and a preliminary function solution set is obtained by iteratively calculating and solving for the minimum value. This can reveal the intrinsic relationship between variables. For example, in the scenario of battery material optimization, an excessively large particle size may increase the stability of the surface property state, but it will affect the overall capacity decay rate. Through this construction and solution, the benefits of more accurate analysis of variable relationships can be obtained, avoiding the inefficiency caused by blind adjustments.

[0089] For example, the correspondence between particle size values ​​and surface property states is analyzed based on the preliminary function solution set. If the deviation exceeds the preset threshold range, the weight ratio of the variables is adjusted and the updated function solution set is obtained by recalculation. This adjustment can dynamically balance the influence of variables. For example, when the particle size value is too small, the surface property state may be unstable. The deviation can be corrected by optimizing the weight ratio, thereby further optimizing the configuration scheme. This helps to achieve a more stable balanced configuration because it can support the deliberation of variable relationships from multiple perspectives, such as avoiding the technical effect of rapid capacity decay caused by excessive deviation in actual material design.

[0090] In one possible implementation, a balanced configuration scheme for particle size and surface property states is generated for the updated function solution set, and its compliance with the predicted capacity decay rate is verified to obtain the final configuration result. This generation and verification process can support each other from different directions. For example, combining electrochemical response data to verify the high activity of the small-size group can ensure the comprehensiveness of the scheme. This approach can bring practical results in determining the balanced configuration scheme for particle size and surface properties because it progressively connects the logical chain from data processing to final optimization, avoiding inconsistencies caused by isolated steps.

[0091] For example, in an extended example, the objective function for the overall process of collaborative optimization is constructed by using the predicted data as the basic integrated variable to form a mathematical expression to solve for the minimum value. This can explain in principle why iterative calculation is needed, because it can gradually approach the optimal solution.

[0092] In one embodiment, such as when optimizing lithium battery particles, adjusting the surface properties can support a balance in particle size values, thereby achieving the technical effect of reducing capacity decay rate. This multi-faceted approach ensures the integrity and diversity of the solution.

[0093] In one possible implementation, the threshold range for deviation analysis is set based on historical data statistics. The process can be clearly explained, such as first collecting multiple sets of prediction data to calculate the average deviation, and then defining the threshold to determine whether to adjust the weight. This can bring the benefits of more refined optimization. For example, after marking low-activity groups in large-size groups, the configuration generation of high-activity groups can be supported through correspondence analysis, avoiding business silos and forming a consistent thought chain.

[0094] For example, the process of verifying that the configuration scheme meets the requirements of the predicted data is explained in detail, such as comparing the value range of the generated balanced configuration scheme with the initial predicted data. If they match, the final result is confirmed. This can be explained from the perspective of business cause and effect. For example, the initial cleaning ensures the reliability of the data, and the subsequent verification prevents invalid scheme output, thereby bringing about the overall technical effect of improving material performance and supporting the realization of balanced configuration of particle size and surface properties in multiple directions.

[0095] S107. For the balanced configuration scheme, generate material synthesis control parameters, and verify the application effect of the scheme in the actual environment through numerical simulation, thereby obtaining the final performance optimization result.

[0096] Initial data is obtained from the balanced configuration scheme to construct a preliminary simulation model for material synthesis, determine the initial range of control parameters, and simulate the material synthesis process using numerical simulation tools to obtain preliminary synthesis data under different control parameters. For the preliminary synthesis data, its matching degree with actual environmental conditions is assessed. If deviations exist, the control parameter range is adjusted, and the simulation is repeated using numerical simulation tools to obtain adjusted synthesis data. For the adjusted synthesis data, combined with key indicators in the actual environment, an environmental adaptability assessment result is constructed to determine the adaptability of the control parameters to environmental conditions, generating intermediate performance data after parameter adjustment. For the intermediate performance data, parameter combinations are screened using data processing tools to determine the optimized control parameter combinations, generate the final synthesis process flow, verify its application performance in the actual environment, and obtain performance optimization results.

[0097] In one possible implementation, obtaining initial data from the balanced configuration scheme involves extracting the pre-defined material composition ratios and environmental variables in the scheme, such as using the temperature and pressure values ​​described in the scheme as a starting point. This ensures that the initial simulation model is built on a reliable foundation, which is beneficial to the accuracy of subsequent simulations.

[0098] Specifically, constructing a preliminary simulation model for material synthesis refers to using the finite element method to simulate the molecular-level synthesis reaction process. The finite element method divides the material synthesis region into multiple small units, calculates the stress and diffusion rate for each unit, and thus predicts the overall synthesis behavior. This method can improve the technical effect of improving model accuracy because it allows for fine-grained simulation of complex interactions.

[0099] For example, determining the initial range of control parameters involves analyzing initial data to set the temperature limit from room temperature to high temperature. For instance, for polymer synthesis, the pH range is set to the neutral to acidic range. This limits the simulation variables to avoid invalid calculations and facilitates the efficient acquisition of initial synthesis data under different control parameters.

[0100] In one possible implementation, simulating the material synthesis process using numerical simulation tools refers to running iterative calculations with finite element software. The tool generates a concentration distribution map of the synthesized products based on the input parameters. This simulation can provide the technical effect of verifying the feasibility of the scheme because it reveals the impact of parameter changes on the results.

[0101] In one possible implementation, determining the degree of matching between preliminary synthetic data and actual environmental conditions involves comparing the deviation between simulated concentrations and field measurements. For example, if the simulation shows high yields but actual environmental humidity interference leads to low yields, the deviation is identified. This allows for timely detection of problems and facilitates adjustments to the control parameter range to improve the matching degree.

[0102] Specifically, re-simulating using numerical simulation tools means rerunning the finite element calculation after updating the parameters to obtain the adjusted synthetic data. This iteration can bring about the technical effect of optimizing the simulation's realism because it gradually approaches the actual conditions.

[0103] For example, constructing environmental adaptability assessment results by combining adjusted synthetic data with key indicators in the actual environment refers to integrating data such as durability and stability indicators to form an assessment score. For instance, matching simulated data with environmental temperature fluctuations to calculate the adaptation percentage can quantify the degree of adaptability of control parameters to environmental conditions. This is beneficial for generating intermediate performance data after parameter adjustment to guide further optimization.

[0104] In one possible implementation, filtering parameter combinations based on intermediate performance data using data processing tools refers to using a sorting algorithm to arrange the efficiency of combinations, such as filtering out the temperature-pH combination with the highest yield. This can determine the optimized combination of control parameters, which is beneficial for generating the final synthesis process flow.

[0105] Specifically, verifying the application performance in a real-world environment involves inputting the optimized combination into a field test model to obtain performance optimization results. This verification can bring about the technical effect of confirming the effectiveness of the overall solution because it bridges simulation and real-world application.

[0106] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for optimizing the performance of sodium iron phosphate electrode materials, characterized in that, include: Image data of the electrode material is acquired using imaging equipment, and particle size distribution characteristics are determined using image processing techniques. Based on the particle size distribution characteristics, the particles are grouped using a classification method to determine potential influencing indicators of conduction efficiency and reactivity. For the aforementioned conductivity and reactivity indicators, response data from electrochemical tests are obtained, and quantitative values ​​of adsorption capacity and chemical stability are calculated through data analysis to obtain surface property evaluation results. Based on the surface property evaluation results, the particle size distribution parameters are adjusted to simulate changes in the reaction area and determine the degree of increase in the reaction area. By measuring the increase in reaction area and combining the optimized parameters with the performance simulation model input, a cyclic test simulation is run to obtain the predicted capacity decay rate data. Based on the predicted capacity decay rate data, an objective function is constructed, and an optimization algorithm is used to iteratively solve the problem to determine a balanced configuration scheme for particle size and surface properties. For the aforementioned balanced configuration scheme, material synthesis control parameters are generated, and the application effect is verified through numerical simulation to obtain the final performance optimization results.

2. The electrode material performance optimization method as described in claim 1, characterized in that, The process of acquiring image data of the electrode material through an imaging device and determining particle size distribution characteristics using image processing technology includes: Raw image data of the electrode material is acquired using imaging equipment; The original image data is processed using image segmentation technology to separate individual particle regions; For the individual particle region, a boundary detection method is applied to determine the particle outline boundary; Based on the particle outline boundary, the diameter value of each particle is calculated to obtain particle size information; If there is abnormal data in the particle size information, the diameter value that does not meet the conditions is filtered out by a preset threshold range to obtain the corrected particle size information. The particle size distribution characteristics are determined by statistical analysis of the corrected particle size information. Based on the particle size distribution characteristics, a distribution data table is generated to record the proportion of particles in different size ranges.

3. The method for optimizing electrode material performance as described in claim 1, characterized in that, The step of grouping particles according to their size distribution characteristics and determining potential indices affecting conductivity and reactivity includes: Based on the particle size distribution characteristics, a pre-established classification model is used for processing. The particle size is grouped by the support vector machine algorithm to obtain preliminary division results of small size group and large size group; Based on the preliminary division results, obtain the particle quantity ratio information of the two groups and determine the proportional characteristics; Based on the aforementioned proportional characteristics, electrochemical response data of two sets of particles under specific environments were obtained; If the response data of the small-size group is higher than a preset threshold, it is marked as a high-activity group; If the response data of the large-size group is lower than a preset threshold, it is marked as a low-activity group; By using the identification information of the high-activity group and the low-activity group, the difference data of conduction efficiency is obtained. The difference data is processed using statistical tools to determine the significance of the distinction results and to identify the potential correlation index between reactivity and particle size grouping.

4. The method for optimizing electrode material performance as described in claim 1, characterized in that, The process involves acquiring response data from electrochemical tests for the aforementioned conductivity and reactivity indicators, calculating quantitative values ​​of adsorption capacity and chemical stability through data analysis, and obtaining surface property evaluation results, including: The material's response data is obtained from electrochemical tests, and multiple sets of response data are recorded to obtain an initial test dataset; For the initial test dataset, data processing methods are used to clean and normalize it to obtain a standard dataset; Based on the standard dataset, the response data is analyzed using curve fitting methods to calculate the quantitative value of adsorption capacity and obtain adsorption performance evaluation data. By combining the adsorption performance evaluation data with the chemical stability response data, statistical tools are used for comparative analysis to determine the quantitative value of chemical stability. If the quantification value of the chemical stability is lower than the preset threshold, the response data is extracted a second time to obtain a supplementary dataset; The surface property evaluation results were obtained through comprehensive analysis based on the supplementary dataset and adsorption performance evaluation data.

5. The method for optimizing electrode material performance as described in claim 1, characterized in that, The step of adjusting the particle size distribution parameters based on the surface property evaluation results, simulating changes in the reaction area, and determining the degree of increase in the reaction area includes: Adsorption capacity data are obtained based on the surface property evaluation results. If the adsorption capacity data is lower than a preset threshold, an optimization algorithm is used to adjust the particle size distribution parameters to obtain a preliminary distribution parameter scheme. Based on the preliminary distribution parameter scheme, simulation calculations are performed for different size ratios to obtain the corresponding data of size ratio and reaction area, and to determine the trend of change. Based on the aforementioned trend and the size ratio adjustment range, parameter configurations with a higher increase in reaction area are selected to obtain an optimized size distribution scheme. For the optimized size distribution scheme, the distribution data is obtained, standardized and organized using data processing tools, and the final parameter configuration result is determined. Based on the final parameter configuration results, simulated reaction area performance data are used to determine the stability of the improvement and obtain optimization effect data.

6. The method for optimizing electrode material performance as described in claim 1, characterized in that, The process involves increasing the reaction area, combining optimized parameters with a performance simulation model, and running cyclic test simulations to obtain capacity decay rate prediction data, including: Based on the degree of increase in reaction area, data processing tools are used to standardize and organize the data to obtain structured improvement data records. Based on the structured improvement data records, the parameters are initialized and configured in conjunction with the performance simulation model to determine the basic parameter settings of the simulation environment. By setting the basic parameters, a genetic algorithm is used to screen the optimal parameter combinations to obtain a suitable parameter configuration scheme. Based on the aforementioned adaptation parameter configuration scheme, perform cyclic simulation tests to obtain capacity decay data and determine the trend of change. Based on the aforementioned trend, data processing tools were used to perform segmented statistics to obtain the attenuation rate distribution at different simulation stages. If the decay rate in a certain stage exceeds the preset threshold, the simulation data for that stage is marked to identify abnormal fluctuations and obtain the final capacity decay rate prediction result.

7. The method for optimizing electrode material performance as described in claim 1, characterized in that, The step of constructing an objective function based on the predicted capacity decay rate data, and iteratively solving it using an optimization algorithm to determine a balanced configuration scheme for particle size and surface properties includes: Using the capacity decay rate prediction data, an initial dataset is obtained, and preprocessing tools are used for cleaning and standardization to obtain a structured basic data set. Based on the structured basic data set, an objective function is constructed, and the particle size and surface properties are used as variable inputs in combination with the collaborative optimization method to determine the mathematical expression of the objective function; For the mathematical expression of the objective function, a genetic algorithm is used for iterative solution to obtain an approximate solution set for the minimum value of the function; The relationship between particle size and surface properties is analyzed using the approximate solution set. If the deviation exceeds the preset threshold range, the variable weights are adjusted and a new approximate solution set is obtained. Based on the new approximate solution set, a balanced configuration scheme is generated, and the final configuration scheme set is determined through multiple rounds of iterative verification.

8. The method for optimizing electrode material performance as described in claim 1, characterized in that, The process of generating material synthesis control parameters for the balanced configuration scheme, verifying the application effect through numerical simulation, and obtaining the final performance optimization results includes: Using the aforementioned balanced configuration scheme, initial data is obtained, a preliminary model for material synthesis is constructed, and the range of control parameters is determined. Based on the preliminary model, the synthesis process was simulated using numerical simulation tools to obtain synthesis results under different control parameters; If the synthesis result deviates from the actual environmental conditions, the control parameters are adjusted, the numerical simulation is repeated, and the effect of the adjusted synthesis is judged. The environmental adaptability assessment results are determined by combining the adjusted synthetic effect data with key indicators of the actual environment. Based on the environmental adaptability assessment results, the support vector machine algorithm is used to screen the control parameters to obtain an optimal parameter combination; By using the optimized parameter combination, the final synthesis process flow is generated, the application effect is determined, and an effect evaluation report is generated.

Images