Synthetic grading evaluation method and system

Abstract

The invention relates to the technical field of building material science, in particular to a synthetic gradation evaluation method and system.The method comprises the steps that aggregate with different particle sizes is separated by adopting a method of reducing the size of screen holes step by step, and screening parameters are adjusted in real time through an optimization algorithm; calculating the ratio of the aggregates with various particle sizes, and performing particle size distribution analysis on the uniformly mixed aggregates to obtain high-precision particle size distribution data; drawing a gradation curve according to the high-precision particle size distribution data, and marking particle size section information influencing the material performance; using an image processing algorithm to extract compactness characteristics of the sample image, and using a convolutional neural network model to perform compactness evaluation to obtain a compactness evaluation result; integrating particle size distribution and compactness evaluation results, and analyzing the relationship between the particle size distribution and the compactness by using a data mining algorithm and a regression analysis model; and establishing an optimization model according to an analysis result between the particle size and the compactness, setting a standardized evaluation index, and carrying out comprehensive evaluation on gradation. By means of the method, the defects of an existing method in the aspects of particle size distribution accuracy, grading curve reliability and compactness evaluation scientificity are effectively overcome, and the precision and credibility of synthesis grading evaluation are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of building material science, and in particular to a synthetic gradation evaluation method and system. Background Art

[0002] In the field of building materials and civil engineering, synthetic grading is a key technology that mixes aggregates of different particle sizes in a certain proportion to obtain a specific particle size distribution. The main goal of this process is to optimize the density, strength and durability of the material to meet engineering requirements.

[0003] However, the existing synthetic grading evaluation methods have some obvious shortcomings. First, the accuracy of particle size distribution is insufficient. The existing methods are not accurate enough in the distribution of aggregates of different particle sizes, resulting in uneven particle size distribution of the mixed material, which affects the overall performance of the material. Second, the reliability of the grading curve is poor. The traditional method often ignores the influence of certain key particle sizes when drawing the grading curve, and cannot fully reflect the true situation of the material. In addition, the density evaluation lacks scientificity. The existing methods lack a systematic scientific basis when evaluating the density of materials and are easily affected by subjective factors, resulting in unreliable evaluation results. Summary of the invention

[0004] The present invention provides a synthetic grading evaluation method and system, thereby effectively solving the problems pointed out in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A method for evaluating synthetic gradation, the method comprising: Aggregates of different particle sizes are separated by gradually reducing the size of the sieve holes, and the screening parameters are adjusted in real time through the optimization algorithm; Calculate the proportion of aggregates of different particle sizes, and analyze the particle size distribution of the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; Draw a gradation curve based on the high-precision particle size distribution data, and mark the particle size segment information that affects material properties; The density features of the sample image are extracted using an image processing algorithm, and the density is evaluated using a convolutional neural network model to obtain a density evaluation result; Integrate the particle size distribution and the density evaluation results, and apply data mining algorithms and regression analysis models to analyze the relationship between the particle size distribution and the density; An optimization model is established based on the analysis results between particle size and density, and standardized evaluation indicators are set to conduct a comprehensive evaluation of grading.

[0006] Furthermore, the aggregates of different particle sizes are separated by gradually reducing the size of the sieve holes, including: Select the appropriate initial sieve size according to the maximum particle size of the target aggregate, and start primary screening to separate the large-particle aggregate; The sieve hole size is reduced step by step, and the undersize material is screened in multiple stages. After each screening stage, the weight and particle size distribution of the aggregates on and under the sieve are recorded.

[0007] Furthermore, the proportion of aggregates of different particle sizes is calculated, including: From the step-by-step screening process, collect the aggregate quantity and particle size data after each screen layer; According to the preset target gradation requirements, the required proportion of aggregates of each particle size is calculated, and the weight of each particle size aggregate is measured by a weighing sensor; Using computer algorithms, the actual measured data is compared with the preset target ratio, and the distribution amount of each size aggregate is adjusted in real time based on the difference; Record the results of each calculated and adjusted mix ratio, conduct test mixing, verify whether the calculated mix ratio meets the expected performance standards, and make corrections based on the test results.

[0008] Furthermore, a gradation curve is drawn based on the high-precision particle size distribution data, and particle size segment information that affects material properties is annotated, including: Collect and organize high-precision particle size distribution data obtained from dynamic light scattering analysis and calculate cumulative distribution percentage; Import the collated data using the drawing tool to generate a cumulative distribution curve for particle size distribution; Determine the critical particle size segments that affect material properties, mark these key points on the grading curve, and add explanations.

[0009] Furthermore, the density features of the sample image are extracted using an image processing algorithm, including: De-noising, gray-scaling and binarization are performed on the image to obtain the pre-processed image; The Canny edge detection algorithm is used to extract particle edges and pore boundaries, analyze the morphological characteristics of the samples, and obtain morphological characteristic results; Applying morphological operations to remove noise and enhance structural features, the morphological operations including but not limited to dilation, erosion, opening and closing operations; Using a connected region analysis method and based on the morphological feature results, marking particles and pore regions in the image, and calculating the area and shape features of the particles and pore regions; The porosity and morphological characteristics of the particles are statistically analyzed to extract parameters reflecting the density of the sample.

[0010] Furthermore, the connected region analysis method adopts the 8-connectivity standard, scans the binary image row by row and column by column from the upper left corner, marks the connected regions of the foreground pixels, and processes the label merging.

[0011] Furthermore, the particle size distribution and the density evaluation results are integrated, and the relationship between the particle size distribution and the density is analyzed by applying a data mining algorithm and a regression analysis model, including: collecting, cleaning and fusing the particle size distribution data and the compactness evaluation results to obtain a comprehensive data set; Select key features and apply data mining techniques to identify potential patterns between particle size distribution and compactness; Constructing and training a regression analysis model, analyzing the relationship between the particle size distribution and density, and obtaining a relationship analysis result; According to the relationship analysis results, the specific influence and mechanism of particle size distribution on density are obtained.

[0012] Furthermore, an optimization model is established based on the analysis results between particle size and density, and standardized evaluation indicators are set to conduct a comprehensive evaluation of gradation, including: Select genetic algorithm as the optimization method, define the fitness function, perform selection, crossover and mutation operations, and adjust parameters to optimize particle size distribution and density; According to the optimization model results, select optimization indicators, set standardized evaluation standards and grades, and build a comprehensive evaluation system based on the weighted algorithm; According to the set evaluation standards and grades, the loan matching is comprehensively evaluated to obtain the specific evaluation results of each indicator, and the grading is comprehensively scored using the comprehensive evaluation system to obtain the final evaluation result.

[0013] A synthetic grading evaluation system, the system comprising: The screening parameter adjustment module uses the method of gradually reducing the size of the sieve holes to separate aggregates of different particle sizes, and adjusts the screening parameters in real time through the optimization algorithm; The particle size data acquisition module calculates the proportion of aggregates of different particle sizes and performs particle size distribution analysis on the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; A gradation curve drawing module draws a gradation curve according to the high-precision particle size distribution data and marks particle size segment information that affects material properties; The density result evaluation module uses an image processing algorithm to extract the density features of the sample image, and uses a convolutional neural network model to perform density evaluation to obtain a density evaluation result; A correlation analysis module integrates the particle size distribution and the density evaluation results, and applies a data mining algorithm and a regression analysis model to analyze the relationship between the particle size distribution and the density; The gradation comprehensive evaluation module establishes an optimization model based on the analysis results between particle size and density, and sets standardized evaluation indicators to conduct a comprehensive evaluation of the gradation.

[0014] Furthermore, the grading comprehensive evaluation module includes: The parameter adjustment optimization unit selects the genetic algorithm as the optimization method, defines the fitness function, performs selection, crossover and mutation operations, and adjusts the parameters to optimize the particle size distribution and density; The evaluation system construction module selects optimization indicators according to the optimization model results, sets standardized evaluation standards and grades, and builds a comprehensive evaluation system based on a weighted algorithm; The evaluation result output module comprehensively evaluates the gradation according to the set evaluation standards and grades, obtains the specific evaluation results of each indicator, and uses the comprehensive evaluation system to comprehensively score the gradation to obtain the final evaluation result.

[0015] The technical solution of the present invention can achieve the following technical effects: The present invention effectively solves the shortcomings of existing methods in terms of accuracy of particle size distribution, reliability of grading curves and scientificity of density evaluation, and significantly improves the accuracy and credibility of synthetic grading evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of a flow chart of a synthetic gradation evaluation method; Figure 2 A schematic diagram of the process of drawing grading curves and marking particle size segment information; Figure 3 A schematic diagram of the process for analyzing the relationship between particle size distribution and density; Figure 4 Schematic diagram of the process for comprehensive evaluation of grading. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0020] Embodiment 1 like Figure 1 As shown, the present invention provides a method for evaluating synthetic gradation, the method comprising: S1: The method of reducing the sieve size step by step is used to separate aggregates of different particle sizes, and the screening parameters are adjusted in real time through the optimization algorithm; Specifically, the precise separation of aggregates of different particle sizes is achieved by gradually reducing the size of the sieve holes, and the screening parameters are adjusted in real time through the optimization algorithm to ensure that the screening process is always in the best state at different stages. This method can improve the efficiency and accuracy of screening, obtain more accurate particle size distribution data, ensure the uniformity and consistency of the mixed material, and thus improve the overall performance of the material, including density, strength and durability; in addition, accurate particle size distribution data provides reliable basic data for subsequent dynamic light scattering analysis, grading curve drawing and density evaluation, ensuring the scientificity and reliability of the entire evaluation process; by optimizing the screening parameters in real time, it is also possible to maximize the use of screening equipment and material resources, reduce waste, improve production efficiency and economic benefits, and thus significantly improve the accuracy and reliability of synthetic grading evaluation.

[0021] S2: Calculate the proportion of aggregates of different particle sizes, and analyze the particle size distribution of the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; Specifically, in the data collection stage, the dispersed samples are injected into the dynamic light scattering instrument measurement cell, and multiple measurements are performed to collect light scattering data in real time. The collected data is processed using DLS software to eliminate abnormal data and noise, and the particle size distribution of the sample is calculated through an algorithm to generate a high-precision particle size distribution graph and statistical data. Finally, multiple measurements are performed on the same sample to ensure data consistency and reliability, and the accuracy of the particle size distribution data is confirmed by comparing multiple measurement results to ensure that high-precision particle size distribution information is obtained.

[0022] S3: Draw the gradation curve based on the high-precision particle size distribution data, and mark the particle size information that affects the material performance; Specifically, by visualizing the particle size distribution data, the grading of aggregates can be intuitively displayed, which facilitates the analysis and understanding of the particle size distribution characteristics of the material, ensures a comprehensive evaluation of the material performance, and provides a scientific basis to guide formula adjustments and material optimization.

[0023] S4: Use image processing algorithm to extract density features of sample images, and use convolutional neural network model to evaluate density to obtain density evaluation results; Specifically, by building and training a convolutional neural network model (CNN), the extracted image features are input into the CNN model to perform density assessment and obtain density assessment results. Finally, the assessment results are verified and compared with the actual density data to evaluate the accuracy and reliability of the model. Advanced image processing algorithms and deep learning technologies are used to accurately extract and evaluate the density features of samples, providing scientific and systematic density assessment results. Automated and intelligent image analysis eliminates the interference of subjective factors in traditional methods and improves the accuracy and consistency of assessment results. At the same time, the use of CNN models for density assessment can quickly and efficiently process large amounts of sample data and provide real-time assessment results.

[0024] S5: Integrate the particle size distribution and density evaluation results, apply data mining algorithm and regression analysis model to analyze the relationship between particle size distribution and density; Specifically, by quantifying the relationship between particle size distribution and density, revealing the specific impact of particle size distribution on density, and conducting in-depth analysis and establishing mathematical models, we can not only accurately understand and predict the density of materials, but also provide data support for optimizing grading design and improving material performance and engineering quality.

[0025] S6: Establish an optimization model based on the analysis results between particle size and density, set standardized evaluation indicators, and conduct a comprehensive evaluation of grading.

[0026] Specifically, by establishing an optimization model and setting standardized evaluation indicators, a scientific and systematic comprehensive evaluation of the grading is carried out to ensure that the performance of the material meets the expected requirements; its significance lies in that through the application of the optimization model and the establishment of a standardized evaluation system, not only can the grading design of the material be scientifically evaluated and optimized, but the density, strength and durability of the material can also be improved.

[0027] The present invention effectively solves the shortcomings of existing methods in terms of accuracy of particle size distribution, reliability of grading curves and scientificity of density evaluation, and significantly improves the accuracy and credibility of synthetic grading evaluation.

[0028] As a preferred embodiment of the above, a method of gradually reducing the size of the sieve holes is used to separate aggregates of different particle sizes, including: A10: Select the appropriate initial sieve size according to the maximum particle size of the target aggregate and start primary screening to separate the large aggregate particles; A20: Gradually reduce the size of the sieve holes and perform multi-stage screening on the material under the sieve. After each stage of screening, record the weight and particle size distribution of the aggregates on and under the sieve.

[0029] Specifically, first, according to the maximum particle size of the target aggregate, a suitable initial sieve hole size is selected, and primary screening is started. During the primary screening process, large aggregate particles are separated so that they can pass through the initial sieve hole, thereby achieving preliminary particle size separation; then, the sieve hole size is gradually reduced, and the undersize material after the primary screening is subjected to multi-stage screening. During each screening process, a smaller sieve hole size is used to further screen the undersize material to achieve a finer particle size separation effect. After each screening operation, the weight and particle size distribution data of the oversize and undersize aggregates are recorded. These recorded data include the mass ratio of the oversize material and the undersize material under each sieve hole size, as well as the respective particle size distribution characteristics. By reducing the sieve hole size step by step, the accurate separation of aggregates of different particle sizes can be achieved, ensuring that the aggregates of each particle size segment can be effectively separated and their detailed distribution data can be recorded. This step-by-step screening and recording method not only improves the accuracy and efficiency of screening, but also provides reliable basic data for subsequent high-precision particle size distribution analysis and grading curve drawing.

[0030] As a preferred embodiment of the above embodiment, the proportion of aggregates of different particle sizes is calculated, including: B10: Collect the aggregate quantity and particle size data after screening at each screen layer from the step-by-step screening process; B20: Calculate the required proportion of aggregates of each particle size according to the preset target gradation requirements, and measure the weight of each particle size aggregate through a weighing sensor; B30: Use computer algorithms to compare the actual measured data with the preset target ratio, and adjust the distribution amount of aggregates of different particle sizes in real time based on the differences; B40: Record the results of each calculated and adjusted mix, conduct test mixing, verify whether the calculated mix meets the expected performance standards, and make corrections based on the test results.

[0031] Specifically, first, from the step-by-step screening process, the aggregate quantity and particle size data after screening each screen layer are collected. These data include the aggregate weight and particle size distribution characteristics of each screen layer, ensuring that the aggregate quantity and distribution information of all particle size segments are accurately recorded; then, according to the preset target grading requirements, the required proportion of aggregates of each particle size is calculated. According to these proportions, the weight of each particle size aggregate is measured using a weighing sensor to ensure that each particle size aggregate can be mixed according to the preset proportion. The use of weighing sensors ensures that the weight of aggregates in each particle size segment can be accurately measured, providing data support for subsequent precise proportioning; then, a computer algorithm is used to compare the actual measured data with the preset target proportion. The computer algorithm analyzes the difference between the actual weight and the target weight of each size of aggregate and adjusts the distribution of each size of aggregate in real time based on this difference. Through computer control, the input amount of each size of aggregate can be automatically adjusted to ensure that the final mixture meets the preset grading requirements. In this process, the results of each calculation and adjustment are recorded in detail. These records include the actual ratio after each adjustment, the target ratio, and the specific steps and parameters of the adjustment. After that, a test mix is ​​carried out to mix the aggregates of each size according to the adjusted ratio to prepare the test samples; finally, the performance test is carried out on the samples after the test mix to verify whether the calculated ratio meets the expected performance standards. These performance tests may include evaluations of density, strength, and durability. According to the test results, if the actual performance is found to be different from the expected, the computer algorithm and the ratio parameters are further corrected until the performance of the mixture meets the expected standards.

[0032] As a preferred embodiment of the above, Figure 2 As shown, step S3, drawing a gradation curve based on high-precision particle size distribution data, and marking the particle size segment information that affects material properties, including: S31: Collect and organize high-precision particle size distribution data obtained from dynamic light scattering analysis and calculate the cumulative distribution percentage; S32: Use a drawing tool to import the sorted data and generate a cumulative distribution curve of particle size distribution; S33: Determine the critical particle size segments that affect material properties, mark these key points on the grading curve, and add explanations.

[0033] Specifically, first, collect and organize the high-precision particle size distribution data obtained from the dynamic light scattering technology analysis. These data include the distribution of aggregates of different particle sizes in the sample. Then, calculate the cumulative distribution percentage based on these data, that is, the percentage of aggregates in each particle size segment in the total aggregate. Next, use appropriate drawing tools (such as Excel, Origin, MATLAB, etc.) to import the organized data, import the cumulative distribution percentage data of each particle size segment into the drawing tool, and use these data to generate a cumulative distribution curve of the particle size distribution. This curve shows the distribution from the smallest particle size to the largest particle size. The cumulative percentage of aggregates intuitively reflects the particle size distribution of the entire sample; then, determine the key particle size segments that have a significant impact on the material properties. These key particle size segments may include the maximum particle size, the minimum particle size, the median particle size, etc. According to the specific performance requirements of the material, identify which particle size segments have an important impact on the material's density, strength, durability and other properties. On the grading curve, mark the position of these key particle size segments, and indicate their specific values ​​and cumulative distribution percentages for clear display; finally, add text descriptions to the grading curve to describe in detail the impact of these key particle size segments on material properties. For example, it can explain how the aggregate of a certain particle size segment contributes to the density and strength of the material, and how to optimize the material performance by adjusting the proportion of this particle size segment. These descriptions help users better understand the impact of grading curves and particle size distribution on material properties, and provide a scientific basis for further optimizing material proportions and process parameters.

[0034] As a preferred embodiment of the above embodiment, using an image processing algorithm to extract density features of the sample image includes: C10: De-noise, grayscale and binarize the image to obtain the pre-processed image; C20: Use the Canny edge detection algorithm to extract particle edges and pore boundaries, analyze the morphological characteristics of the sample, and obtain the morphological characteristics results; C30: Apply morphological operations to remove noise and enhance structural features. Morphological operations include but are not limited to dilation, erosion, opening and closing operations; C40: Using the connected region analysis method and based on the morphological feature results, the particles and pore regions in the image are marked, and the area and shape characteristics of the particles and pore regions are calculated; C50: Statistical analysis of porosity and particle morphology to extract parameters reflecting sample density.

[0035] Specifically, first, the collected sample images are preprocessed. This step includes denoising, grayscale and binarization. Denoising is performed by using filters (such as Gaussian filtering or median filtering) to reduce noise interference in the image; then, the color image is converted into a grayscale image to simplify subsequent processing steps; finally, a threshold segmentation algorithm (such as the Otsu algorithm) is applied to convert the grayscale image into a binary image, thereby highlighting the structural features of particles and pores and obtaining a preprocessed image. Next, the Canny edge detection algorithm is used to extract the particle edges and pore boundaries in the preprocessed image. The Canny algorithm can accurately detect the edges in the image and analyze the morphological characteristics of the sample, such as the outline of the particles and the shape of the pores, to obtain the morphological characteristic results; then, the image is further processed by applying morphological operations to remove noise and enhance structural features. Morphological operations include but are not limited to dilation, erosion, opening and closing operations. The dilation operation can fill small holes in the image, the erosion operation can remove small noise points in the image, the opening operation (first erosion and then dilation) can remove small structural elements, and the closing operation (first dilation and then erosion) can fill small holes in the structure. Through these morphological operations, the structural characteristics of particles and pores can be enhanced. The image is further purified; on this basis, the connected region analysis method is used, and according to the morphological feature results, the particles and pore areas in the image are marked. The connected region analysis method can identify and mark the interconnected pixel areas in the image, namely the particles and pore areas, and then calculate the area and shape characteristics of these particles and pore areas, including the area, perimeter, shape factor, etc. of each area; finally, the porosity and morphological characteristics of the particles in the image are statistically analyzed to extract parameters reflecting the density of the sample. The porosity is calculated by calculating the ratio of the total area of ​​the pore area to the total area of ​​the sample image. The morphological characteristics of the particles include the average size and shape factor of the particles. These parameters comprehensively reflect the density characteristics of the sample and provide a reliable data basis for subsequent density evaluation.

[0036] As a preferred embodiment of the above embodiment, the connected region analysis method adopts the 8-connectivity standard, scans the binary image row by row and column by column from the upper left corner, marks the connected regions of the foreground pixels, and processes label merging.

[0037] Specifically, first, starting from the upper left corner of the image, scan each pixel in the binary image row by row and column by column. The 8-connectivity criterion considers the adjacent pixels of each pixel in the horizontal, vertical, and diagonal directions. Therefore, each pixel has 8 adjacent pixels to check. When a foreground pixel (i.e., a pixel with a value of 1) is scanned, check the labeling of its 8 adjacent pixels: Unlabeled neighboring pixels: If all neighboring pixels are unlabeled, assign a new label to the current pixel and record this label as a new connected region; Single labeled neighboring pixel: If there is a labeled neighboring pixel, the current pixel is assigned the same label, indicating that they belong to the same connected region; Multiple labeled adjacent pixels: If multiple adjacent pixels have been labeled, select a label from an adjacent pixel and assign it to the current pixel. At the same time, record the equivalence relationship between these different labels for subsequent merging processing.

[0038] After completing the initial scan and assigning labels to all foreground pixels, the label merging is processed. For the recorded label equivalence relationship, different labels belonging to the same connected region are merged to ensure that each connected region corresponds to only one unique label. This step can be completed by constructing an equivalence class set and updating the corresponding pixel labels in the image. Through the above steps, the connected region analysis method can accurately mark and identify all connected regions (i.e. particles and pores) in the image, and calculate the area and shape characteristics of each region. This method can not only accurately mark and identify each connected region, but also effectively process complex image structures to ensure the accuracy and completeness of the analysis results.

[0039] As a preferred embodiment of the above, Figure 3 As shown, step S5 integrates the particle size distribution and density evaluation results, applies data mining algorithm and regression analysis model, and analyzes the relationship between particle size distribution and density, including: S51: Collect, clean and fuse particle size distribution data and compactness evaluation results to obtain a comprehensive data set; S52: Select key features and apply data mining techniques to identify potential patterns between particle size distribution and compactness; S53: construct and train a regression analysis model, analyze the relationship between particle size distribution and density, and obtain relationship analysis results; S54: Based on the relationship analysis results, the specific influence and mechanism of particle size distribution on density are obtained.

[0040] Specifically, first, the particle size distribution data and density evaluation results are collected, cleaned and fused to obtain a comprehensive data set. The collected data include the distribution data of each particle size segment and the corresponding density evaluation results. In the data cleaning process, outliers and incomplete data are removed to ensure the quality and consistency of the data. The cleaned data are fused to form a comprehensive data set containing particle size distribution and density evaluation results, which provides a basis for subsequent analysis; then, through data mining algorithms (such as cluster analysis, association rules, etc.), the key features that have a significant impact on particle size distribution and density are screened out from the comprehensive data set. Data mining technology helps to identify the particle size distribution and density. The potential relationships and patterns between them lay the foundation for further analysis; then, an appropriate regression analysis model (such as linear regression, multivariate regression, etc.) is selected, and the fused comprehensive data set is used for model training. By adjusting the model parameters, the fitting accuracy and prediction ability of the model are improved. The trained regression model is used to quantify the specific impact of particle size distribution on density. By analyzing the model output, detailed relationship analysis results are obtained; finally, through the analysis results of the regression model, it is clear which particle size characteristics have a significant impact on density, as well as the specific manifestations and mechanisms of these impacts. Such analysis helps to optimize the particle size distribution design and improve the density, strength and durability of the material.

[0041] As a preferred embodiment of the above, Figure 4 As shown, step S6, based on the analysis results between particle size and density, an optimization model is established, and standardized evaluation indicators are set to conduct a comprehensive evaluation of the gradation, including: S61: Select genetic algorithm as the optimization method, define the fitness function, perform selection, crossover and mutation operations, and adjust parameters to optimize particle size distribution and density; S62: According to the optimization model results, select optimization indicators, set standardized evaluation standards and grades, and build a comprehensive evaluation system based on the weighted algorithm; S63: According to the set evaluation standards and grades, a comprehensive evaluation is conducted on the matching to obtain the specific evaluation results of each indicator, and the matching is comprehensively scored using the comprehensive evaluation system to obtain the final evaluation results.

[0042] Specifically, first, the genetic algorithm is selected as the optimization method. The genetic algorithm is a search algorithm based on natural selection and genetic mechanism, which has the advantages of strong global optimization ability and wide adaptability. Through the genetic algorithm, the optimal combination of particle size distribution and density can be found in a large search space. In the optimization process, the fitness function is first defined. This is the standard for evaluating the quality of individuals in the genetic algorithm. A fitness function that comprehensively considers the particle size distribution and density is defined. The quality of each individual is measured by the weighted sum of the particle size and density. The formula is as follows: ,in, and are the weights of particle size and density, and are the function values ​​of particle size and density, respectively. The selection, crossover and mutation operations in the genetic algorithm are carried out in sequence. The selection operation adopts the roulette selection method to select excellent individuals as parents according to the fitness value of the individuals. The crossover operation adopts the single-point crossover method or the multi-point crossover method to exchange genes for the selected parent individuals to generate new offspring individuals. The mutation operation adopts the random mutation method to randomly change some genes of the offspring individuals to increase the diversity of the population and avoid local optimality. According to the evaluation results of the fitness function, the parameters in the genetic algorithm, such as population size, crossover probability and mutation probability, are continuously adjusted to obtain a better combination of particle size distribution and density. Finally, according to the results of the optimization model, the optimization index is selected, and the standardized evaluation criteria and grades are set. The evaluation criteria include the uniformity of particle size distribution, the maximum value of density, etc. The evaluation grades are divided into several grades according to the standardized indicators, such as excellent, good, medium, poor, etc. A comprehensive evaluation system is constructed based on the weighted algorithm. The comprehensive evaluation system obtains the comprehensive score of each individual by weighted summation of each evaluation index. The formula is as follows: ,in, For the comprehensive rating, is the weight of the 𝑖th indicator, is the evaluation value of the 𝑖th indicator; according to the set evaluation criteria and grades, the particle size and density are comprehensively evaluated to obtain the specific evaluation results of each indicator, and the comprehensive evaluation system is used to comprehensively score them to obtain the final evaluation results.

[0043] Embodiment 2 Based on the same inventive concept as the evaluation method for synthetic gradation in the aforementioned embodiment, the present invention further provides an evaluation system for synthetic gradation, the system comprising: The screening parameter adjustment module uses the method of gradually reducing the size of the sieve holes to separate aggregates of different particle sizes, and adjusts the screening parameters in real time through the optimization algorithm; The particle size data acquisition module calculates the proportion of aggregates of different particle sizes and performs particle size distribution analysis on the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; The gradation curve drawing module draws the gradation curve based on the high-precision particle size distribution data and marks the particle size segment information that affects the material performance; The density result evaluation module uses an image processing algorithm to extract the density features of the sample image, and uses a convolutional neural network model to perform density evaluation to obtain a density evaluation result; The correlation analysis module integrates the particle size distribution and density evaluation results, and applies data mining algorithms and regression analysis models to analyze the relationship between particle size distribution and density; The gradation comprehensive evaluation module establishes an optimization model based on the analysis results between particle size and density, and sets standardized evaluation indicators to conduct a comprehensive evaluation of the gradation.

[0044] The above-mentioned evaluation system in the present invention can effectively implement the evaluation method of synthetic grading, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0045] As a preferred embodiment of the above, the grading comprehensive evaluation module includes: The parameter adjustment optimization unit selects the genetic algorithm as the optimization method, defines the fitness function, performs selection, crossover and mutation operations, and adjusts the parameters to optimize the particle size distribution and density; The evaluation system construction module selects optimization indicators according to the optimization model results, sets standardized evaluation standards and grades, and builds a comprehensive evaluation system based on a weighted algorithm; The evaluation result output module conducts a comprehensive evaluation of the gradation according to the set evaluation standards and grades, obtains the specific evaluation results of each indicator, and uses the comprehensive evaluation system to comprehensively score the gradation to obtain the final evaluation result.

[0046] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0047] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for evaluating synthetic grading, characterized in that: include: Aggregates of different particle sizes are separated by gradually reducing the size of the sieve holes, and the screening parameters are adjusted in real time through the optimization algorithm; Calculate the proportion of aggregates of different particle sizes, and analyze the particle size distribution of the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; Draw a gradation curve based on the high-precision particle size distribution data, and mark the particle size segment information that affects material properties; The density features of the sample image are extracted using an image processing algorithm, and the density is evaluated using a convolutional neural network model to obtain a density evaluation result; Integrate the particle size distribution and the density evaluation results, and apply data mining algorithms and regression analysis models to analyze the relationship between the particle size distribution and the density; An optimization model is established based on the analysis results between particle size and density, and standardized evaluation indicators are set to conduct a comprehensive evaluation of grading.

2. The evaluation method for synthetic grading according to claim 1, characterized in that: Aggregates of different particle sizes are separated by gradually reducing the size of the sieve holes, including: Select the appropriate initial sieve size according to the maximum particle size of the target aggregate, and start primary screening to separate the large-particle aggregate; The sieve hole size is reduced step by step, and the undersize material is screened in multiple stages. After each screening stage, the weight and particle size distribution of the aggregates on and under the sieve are recorded.

3. The evaluation method for synthetic gradation according to claim 1, characterized in that: Calculate the proportion of aggregates of different particle sizes, including: From the step-by-step screening process, collect the aggregate quantity and particle size data after each screen layer; According to the preset target gradation requirements, the required proportion of aggregates of each particle size is calculated, and the weight of each particle size aggregate is measured by a weighing sensor; Using computer algorithms, the actual measured data is compared with the preset target ratio, and the distribution amount of each size aggregate is adjusted in real time based on the difference; Record the results of each calculated and adjusted mix ratio, conduct test mixing, verify whether the calculated mix ratio meets the expected performance standards, and make corrections based on the test results.

4. The evaluation method for synthetic grading according to claim 1, characterized in that: Draw a gradation curve based on the high-precision particle size distribution data, and mark the particle size segment information that affects the material properties, including: Collect and organize high-precision particle size distribution data obtained from dynamic light scattering analysis and calculate cumulative distribution percentage; Import the collated data using the drawing tool to generate a cumulative distribution curve for particle size distribution; Determine the critical particle size segments that affect material properties, mark these key points on the grading curve, and add explanations.

5. The method for evaluating synthetic gradation according to claim 1, characterized in that: Use image processing algorithms to extract density features of sample images, including: De-noising, gray-scaling and binarization are performed on the image to obtain the pre-processed image; The Canny edge detection algorithm is used to extract particle edges and pore boundaries, analyze the morphological characteristics of the samples, and obtain morphological characteristic results; Applying morphological operations to remove noise and enhance structural features, the morphological operations including but not limited to dilation, erosion, opening and closing operations; Using a connected region analysis method and based on the morphological feature results, marking particles and pore regions in the image, and calculating the area and shape features of the particles and pore regions; The porosity and morphological characteristics of the particles are statistically analyzed to extract parameters reflecting the density of the sample.

6. The method for evaluating synthetic gradation according to claim 5, characterized in that: The connected region analysis method adopts the 8-connectivity standard, scans the binary image row by row and column by column from the upper left corner, marks the connected regions of foreground pixels, and processes label merging.

7. The method for evaluating synthetic gradation according to claim 1, characterized in that: The particle size distribution and the density evaluation results are integrated, and the relationship between the particle size distribution and the density is analyzed by applying data mining algorithms and regression analysis models, including: collecting, cleaning and fusing the particle size distribution data and the compactness evaluation results to obtain a comprehensive data set; Select key features and apply data mining techniques to identify potential patterns between particle size distribution and compactness; Constructing and training a regression analysis model, analyzing the relationship between the particle size distribution and density, and obtaining a relationship analysis result; According to the relationship analysis results, the specific influence and mechanism of particle size distribution on density are obtained.

8. The method for evaluating synthetic gradation according to claim 1, characterized in that: According to the analysis results between particle size and density, an optimization model is established, and standardized evaluation indicators are set to conduct a comprehensive evaluation of gradation, including: Select genetic algorithm as the optimization method, define the fitness function, perform selection, crossover and mutation operations, and adjust parameters to optimize particle size distribution and density; According to the optimization model results, select optimization indicators, set standardized evaluation standards and grades, and build a comprehensive evaluation system based on the weighted algorithm; According to the set evaluation standards and grades, the loan matching is comprehensively evaluated to obtain the specific evaluation results of each indicator, and the grading is comprehensively scored using the comprehensive evaluation system to obtain the final evaluation result.

9. A synthetic grading evaluation system, characterized in that: The system comprises: The screening parameter adjustment module uses the method of gradually reducing the size of the sieve holes to separate aggregates of different particle sizes, and adjusts the screening parameters in real time through the optimization algorithm; The particle size data acquisition module calculates the proportion of aggregates of different particle sizes and performs particle size distribution analysis on the mixed aggregates based on dynamic light scattering technology to obtain high-precision particle size distribution data; A gradation curve drawing module draws a gradation curve according to the high-precision particle size distribution data and marks the particle size segment information that affects the material performance; The density result evaluation module uses an image processing algorithm to extract the density features of the sample image, and uses a convolutional neural network model to perform density evaluation to obtain a density evaluation result; A correlation analysis module integrates the particle size distribution and the density evaluation results, and applies a data mining algorithm and a regression analysis model to analyze the relationship between the particle size distribution and the density; The gradation comprehensive evaluation module establishes an optimization model based on the analysis results between particle size and density, and sets standardized evaluation indicators to conduct a comprehensive evaluation of the gradation.

10. The synthetic grading evaluation system according to claim 9, characterized in that: The grading comprehensive evaluation module comprises: The parameter adjustment optimization unit selects the genetic algorithm as the optimization method, defines the fitness function, performs selection, crossover and mutation operations, and adjusts the parameters to optimize the particle size distribution and density; The evaluation system construction module selects optimization indicators according to the optimization model results, sets standardized evaluation standards and grades, and builds a comprehensive evaluation system based on a weighted algorithm; The evaluation result output module comprehensively evaluates the gradation according to the set evaluation standards and grades, obtains the specific evaluation results of each indicator, and uses the comprehensive evaluation system to comprehensively score the gradation to obtain the final evaluation result.

Images