Method for optimizing anti-seismic performance of UHPC pier body structure

Abstract

The invention discloses a UHPC pier body structure anti-seismic performance optimization method, and relates to the technical field of pier body structure seismic resistance. According to the method, the anti-seismic property of the UHPC pier body structure is greatly improved by accurately regulating and controlling the density, the particle size distribution and the grading curve of the aggregate and monitoring the physical property of the concrete in real time; through dual screening and optimization of density and particle size, an optimal aggregate ratio is calculated in combination with an advanced algorithm, so that the mechanical property and durability of the concrete are remarkably enhanced; manual errors are reduced through the automatic detection and screening process, and the construction efficiency and quality are improved; in addition, the material waste is effectively reduced by optimizing the aggregate proportion, the construction cost is further reduced, and the economic benefit is improved. According to the method, structural safety is guaranteed, and meanwhile double optimization of construction efficiency and cost effectiveness is promoted.

Description

Technical Field

[0001] This application belongs to the technical field of seismic resistance of pier structures, and particularly relates to a method for optimizing the seismic performance of UHPC pier structures. Background Art

[0002] With the continuous development of modern construction technologies, UHPC (Ultra-High Performance Concrete), as a new type of building material with high strength, high durability, and excellent seismic performance, has been widely used in structural engineering such as bridges, high-rise buildings, and tunnels. As an important part of them, the seismic performance of UHPC pier structures directly affects the safety and stability of the entire structure.

[0003] However, in the actual design and construction processes of UHPC pier structures, how to effectively optimize their seismic performance has always been a technical problem. Traditional optimization methods often rely on empirical formulas and experimental data, lacking systematicness and precision, and it is difficult to meet the high requirements of modern engineering for the seismic performance of UHPC pier structures. In actual engineering, the density and particle size distribution of aggregates are often interrelated. Aggregates with high density often have a more uniform particle size distribution, while aggregates with low density may be accompanied by discreteness in particle size distribution. Therefore, when optimizing the seismic performance of concrete, it is necessary to comprehensively consider these two factors of aggregate density and particle size distribution.

[0004] In the preparation process of UHPC, the screening and matching of aggregate density are a key technical issue. First of all, the density and gradation of aggregates directly affect the compactness and uniformity of concrete. In actual operation, it is necessary to screen out particles with densities meeting the requirements from a large number of aggregates. This process not only requires high-precision equipment but also needs to consider the balance between screening efficiency and aggregate loss. In the actual production process, due to the diversity of aggregate sources, there may be significant differences in the density and particle size distribution of different batches of aggregates. This requires that in the screening and matching process, dynamic adjustments must be made according to the actual situation of each batch of aggregates. For example, when the supply of aggregates of a certain particle size is insufficient, how to make up for this defect by adjusting the proportion of other particle sizes of aggregates while ensuring the continuity of the overall gradation is a complex technical problem. Summary of the Invention

[0005] The purpose of this application is to provide a method for optimizing the seismic performance of UHPC pier structures, which optimizes the seismic performance of UHPC pier structures by optimizing the aggregate density and particle size.

[0006] To achieve the above objective, an embodiment of this application provides a method for optimizing the seismic performance of UHPC pier structures, including:

[0007] Obtain an initial aggregate sample from the aggregate storage bin, and obtain the density value of the initial aggregate sample through a density detection device to obtain a density data set; according to the density data set, set a preset density threshold, and perform screening according to the density threshold to obtain an aggregate set;

[0008] Obtain the particle size value of each aggregate particle in the aggregate set through a particle size analysis device to obtain a particle size data set, and according to the particle size data set, classify the aggregate particles according to a preset particle size grading standard to obtain a classified aggregate set;

[0009] Extract the aggregate distribution ratio of each particle size grade from the classified aggregate set, calculate the current grading curve to obtain grading curve data; if the deviation between the grading curve data and the preset grading curve exceeds the first preset threshold, then adjust the aggregate distribution ratio of each particle size grade and recalculate the grading curve until the deviation between the recalculated grading curve and the preset grading curve does not exceed the first preset threshold to obtain the final grading curve;

[0010] According to the final grading curve, use the aggregate proportioning algorithm based on the minimum void ratio to calculate the optimal proportion of each particle size grade of aggregate to obtain an aggregate proportioning plan; input the aggregate proportioning plan into the concrete mixing equipment, and monitor the slump and spread of UHPC to determine whether the slump is lower than the second preset threshold or the spread is less than the third preset threshold; if the slump is lower than the second preset threshold or the spread is less than the third preset threshold, then readjust the aggregate proportioning plan until the slump is not lower than the second preset threshold or the spread is not less than the third preset threshold.

[0011] According to the above method of the embodiments of the present application, the following additional technical features may also be provided:

[0012] Further, according to the density data set, calculate the mean and standard deviation of the density values to determine the density distribution characteristics; if the density distribution characteristics meet the fourth preset threshold, then use a clustering algorithm to classify the density data set to obtain a first classification result; according to the first classification result, extract the aggregate samples with abnormal density values to generate a first abnormal sample set;

[0013] Use a regression algorithm to analyze the first abnormal sample set to obtain a first correlation analysis result of the density value and the aggregate characteristics, and according to the first correlation analysis result, generate an aggregate density optimization model, and screen the aggregates in the aggregate storage bin through the aggregate density optimization model to obtain an optimized aggregate set.

[0014] Further, according to the particle size data set, calculate the mean and standard deviation of the particle size values to determine the particle size distribution characteristics; if the particle size distribution characteristics meet the fifth preset threshold, use a clustering algorithm to classify the particle size data set to obtain a second classification result; according to the second classification result, extract the aggregate samples with abnormal particle size values to generate a second abnormal sample set;

[0015] Use a regression algorithm to analyze the second abnormal sample set to obtain a second correlation analysis result between the particle size value and the aggregate characteristics. According to the second correlation analysis result, generate an aggregate particle size optimization model, and through the aggregate particle size optimization model, screen the aggregates in the aggregate set to obtain an optimized particle size data set.

[0016] Further, according to the classified aggregate set, use a clustering algorithm to group the aggregates of each particle size grade to obtain a particle size grouping set; if there are outliers in the particle size grouping set, use a regression algorithm to analyze the correlation between the outliers and the aggregate characteristics to obtain an outlier analysis result;

[0017] According to the outlier analysis result, use an optimization algorithm to correct the outliers to obtain a corrected aggregate set; use statistical analysis to calculate the proportion of the aggregates of each particle size grade in the corrected aggregate set to generate a proportion distribution characteristic;

[0018] According to the proportion distribution characteristic, use a classification algorithm to optimize the classification of the corrected aggregate set to obtain an optimized classification result; use a screening algorithm to extract the aggregate particles that meet the preset particle size threshold in the optimized classification result to form a target aggregate set; according to the target aggregate set, use a verification algorithm to evaluate the preset particle size classification standard to generate a classification standard evaluation report.

[0019] Further, use an extraction algorithm to obtain the distribution ratio of the aggregates of each particle size grade from the classified aggregate set to generate distribution ratio data; according to the distribution ratio data, use a fitting algorithm to calculate the parameters of the current grading curve to generate fitting parameter data;

[0020] If there are outliers in the fitting parameter data, use a regression algorithm to analyze the correlation between the outliers and the aggregate distribution to generate an outlier analysis result; according to the outlier analysis result, use an optimization algorithm to correct the fitting parameter data to generate corrected parameter data;

[0021] For the corrected parameter data, use a statistical analysis method to calculate the characteristic values of the current grading curve to generate characteristic value data; according to the characteristic value data, use a classification algorithm to classify the current grading curve to generate classification result data; for the classification result data, use a screening algorithm to extract the aggregate set that meets the preset grading standard to obtain a grading standard aggregate set.

[0022] Further, determine whether the deviation between the gradation curve data and the preset gradation curve exceeds the first preset threshold. If the deviation exceeds the first preset threshold, adjust the aggregate ratio of the particle size grade according to the standard aggregate set for gradation.

[0023] Adopting the UHPC pier structure seismic performance optimization method based on BIM and the finite element method provided by the embodiments of the present application, compared with the prior art, has the following beneficial technical effects:

[0024] By precisely controlling the density, particle size distribution, and gradation curve of the aggregate, and monitoring the physical properties of the concrete, the embodiments of the present application significantly improve the seismic performance of the UHPC pier structure, enabling it to better resist the damage of natural disasters such as earthquakes; by screening and optimizing the aggregate, the optimal aggregate mixing ratio plan is obtained, improving the mechanical properties and durability of the concrete; through the automated detection and screening process, the cumbersome manual operations and errors are reduced, improving the construction efficiency and quality; by optimizing the aggregate mixing ratio and reducing waste, the construction cost of the UHPC pier structure is reduced, improving economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The flowchart shows a method for optimizing the seismic performance of a UHPC pier structure according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To make the above objects, features, and advantages of the present application more obvious and understandable, the following will describe in detail the specific embodiments of the present application with reference to the drawings. It can be understood that the specific embodiments described herein are only for explaining the present application, rather than limiting the present application. Additionally, it should be noted that for the sake of description, only parts related to the present application are shown in the drawings, not all structures. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0027] The terms "including" and "having" and any variations thereof in the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products, or devices.

[0028] References to "embodiments" in this application mean that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of this application. The phrase may not necessarily refer to the same embodiment each time it appears in the specification, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described in this application can be combined with other embodiments.

[0029] As Figure 1 shown, an embodiment of this application provides a method for optimizing the seismic performance of a UHPC pier structure, including the following steps:

[0030] Step 101: Obtain an initial aggregate sample from the aggregate library and obtain the density value of the initial aggregate sample through a density detection device to obtain a density data set; set a preset density threshold according to the density data set, and perform screening according to the density threshold to obtain an aggregate set.

[0031] Step 101 involves the acquisition and screening of the initial aggregate sample. In this step, a certain number of aggregates are randomly or selected according to certain rules from a pre-prepared aggregate library as the initial sample. The aggregate library usually contains various types, sizes, and densities of aggregates to meet different engineering requirements. A professional density detection device (such as a densitometer, pycnometer, etc.) is used to measure the density of each initial aggregate sample. These devices can accurately measure the volume and mass of the aggregates, thereby calculating their density. The density values of all initial aggregate samples are summarized to form a density data set. This set contains the density information of all samples and provides a basis for subsequent analysis and screening.

[0032] Set one or more density thresholds according to engineering requirements and material properties. These thresholds are used to screen aggregate samples that meet specific density requirements to obtain a more uniform and stable aggregate set. Perform statistical analysis on the density data set, and calculate the mean and standard deviation of the densities of all samples. These statistics can reflect the overall level and dispersion degree of the aggregate density, and help to understand the density distribution characteristics of the aggregates.

[0033] If the density distribution characteristics (such as the mean, standard deviation, etc.) meet a certain preset threshold (the fourth preset threshold), then use a clustering algorithm (such as K-means, hierarchical clustering, etc.) to classify the density data set. The clustering algorithm can group similar samples into the same class, thereby identifying aggregate sets with different density characteristics.

[0034] In the classification results, identify the aggregate samples whose density values deviate significantly from those of other samples. These samples are considered to have abnormal densities. Extract these abnormal samples to form a first set of abnormal samples. Analyze the first set of abnormal samples using regression algorithms (such as linear regression, non - linear regression, etc.). The regression algorithm can establish a mathematical model between the density value and aggregate characteristics (such as composition, shape, size, etc.), thereby revealing the correlation between them.

[0035] According to the analysis results of the regression algorithm, generate an aggregate density optimization model. This model can predict the density values of different aggregate characteristics and is used to guide the screening and optimization of aggregates. Use the aggregate density optimization model to screen the aggregates in the aggregate library. By comparing the predicted density with the actual density, select the aggregates that meet specific density requirements to obtain an optimized set of aggregates.

[0036] Specifically, the aggregate density optimization model of the embodiment of the present application includes:

[0037] The first data pre - processing module: This module is responsible for processing the density data set of the initial aggregate samples obtained from the density detection device, which may include steps such as data cleaning, missing value processing, and outlier detection.

[0038] The first statistical analysis module: Calculate the mean and standard deviation of the density values to determine the density distribution characteristics. This step helps to identify the overall trend and variability in the data.

[0039] The first classification and anomaly detection module: If the density distribution characteristics meet specific preset thresholds, use clustering algorithms (such as K - means, DBSCAN, etc.) to classify the density data set to identify the aggregate samples with abnormal density values. These abnormal samples will be used for further analysis.

[0040] The first regression analysis module: Use regression algorithms (such as linear regression, decision tree regression, random forest regression, etc.) to analyze the set of abnormal samples to explore the correlation between the density value and aggregate characteristics. This step aims to establish a prediction model for predicting and optimizing the density of aggregates.

[0041] The first optimization and screening module: Based on the results of the regression analysis, generate an aggregate density optimization model. This model will be used to screen the aggregates in the aggregate library to eliminate the aggregates that do not meet the optimization criteria, thereby obtaining an optimized set of aggregates.

[0042] Specifically, in the embodiments of the present application, initial aggregate samples are randomly selected from the aggregate storage bin. The number of samples is 100, and the weight of each sample is controlled between 500 grams and 1000 grams to ensure the representativeness of the samples. A high-precision density detection device is used to measure the density of each aggregate sample. The precision of the device is 0.1 grams per cubic centimeter. During the measurement process, it is necessary to ensure that the ambient temperature remains at 20 degrees Celsius to avoid the influence of temperature on the density measurement results. After the measurement is completed, 100 density data are obtained, forming a density data set.

[0043] Through statistical analysis of the density data set, parameters such as its average value and standard deviation are calculated. For example, the average density is 65 grams per cubic centimeter, and the standard deviation is 0.8 grams per cubic centimeter. Further, a normal distribution test method, such as the Shapiro-Wilk test, is used to determine whether the density data conforms to the normal distribution. The test result shows that the p-value is greater than 0.5, indicating that the data follows the normal distribution. Based on the distribution characteristics of the density data, a clustering analysis method, such as the K-means algorithm, can be used to classify the aggregate samples into three categories: high density, medium density, and low density. The clustering centers are 75 grams per cubic centimeter, 65 grams per cubic centimeter, and 55 grams per cubic centimeter respectively. Through these analyses, a scientific basis can be provided for the subsequent quality control and optimization of the aggregates.

[0044] Based on the density data set, a preset density threshold of 70 grams per cubic centimeter is set, and an automated screening algorithm is used to efficiently screen the aggregate samples. Aggregate particles with a density value greater than or equal to 70 grams per cubic centimeter are screened out to form a high-density aggregate set. During the screening process, programming scripts are used to compare the density data item by item to ensure the accuracy and efficiency of the screening. For example, 25 aggregate samples that meet the high-density requirements are screened out from 100 density data, and their density values are distributed between 70 grams per cubic centimeter and 80 grams per cubic centimeter.

[0045] To further verify the reliability of the screening results, statistical analysis methods are used to calculate the average density and standard deviation of the high-density aggregate set, which are 72 grams per cubic centimeter and 2 grams per cubic centimeter respectively, indicating that the screened high-density aggregates have high density consistency. By analyzing the distribution characteristics of the high-density aggregate set, it is found that its density values are mainly concentrated in the range of 71 grams per cubic centimeter to 73 grams per cubic centimeter, meeting the preset high-density requirements.

[0046] Based on this, the high-density aggregate set can be further used for subsequent quality optimization. For example, high-density aggregates are preferentially used in the concrete mix ratio to improve the strength and durability of the concrete, providing a better material basis for engineering applications.

[0047] In summary, step 101 describes in detail the process of aggregate density screening and optimization in the seismic performance optimization method of the UHPC pier structure. Through the application of precise density measurement, statistical analysis, clustering algorithm and regression algorithm, the aggregate density can be effectively controlled and optimized, thereby improving the seismic performance of the UHPC pier structure.

[0048] Step 102, obtaining the particle size value of each aggregate particle in the aggregate set by a particle size analysis device to obtain a particle size data set, and classifying the aggregate particles into particle size grades according to the particle size data set and a preset particle size classification standard to obtain a graded aggregate set.

[0049] Step 102 involves particle size analysis and classification. The particle size value of each aggregate particle in the aggregate set is accurately measured by a particle size analysis device; these particle size values ​​are aggregated into a particle size data set; according to a preset particle size classification standard (such as being divided into different grades according to particle size), the aggregate particles are classified into particle size grades; and a graded aggregate set is obtained, that is, an aggregate set classified by particle size.

[0050] Furthermore, based on the particle size data set, the mean and standard deviation of the particle size values ​​are calculated, and the particle size distribution characteristics are determined through these statistics; if the particle size distribution characteristics meet the fifth preset threshold (such as the mean and standard deviation are within a specific range), the next step is to classify the particle size data set using a clustering algorithm (such as K-means, DBSCAN, etc.); a second classification result is obtained, that is, the aggregate category divided by the clustering algorithm; based on the second classification result, aggregate samples with abnormal particle size values ​​(such as samples that deviate far from the cluster center) are extracted to generate a second abnormal sample set.

[0051] The second abnormal sample set is analyzed by using a regression algorithm (such as linear regression, nonlinear regression, etc.); the second correlation analysis result of the particle size value and the aggregate characteristics (such as the relationship between particle size and strength and durability) is obtained; according to the second correlation analysis result, an aggregate particle size optimization model is generated. Aggregates in the aggregate set are screened by the aggregate particle size optimization model to obtain an optimized particle size data set.

[0052] Specifically, the aggregate particle size optimization model of the embodiment of the present application includes:

[0053] The second data preprocessing module: Similar to the aggregate density optimization model, this module is responsible for processing the particle size data set obtained from the particle size analysis device.

[0054] Second statistical analysis module: Calculate the mean and standard deviation of the particle size values ​​to determine the particle size distribution characteristics. This step also helps to identify the overall trend and variability in the data.

[0055] Second Classification and Anomaly Detection Module: If the particle size distribution characteristics meet specific preset thresholds, a clustering algorithm is used to classify the particle size data set to identify aggregate samples with abnormal particle size values.

[0056] Second Regression Analysis Module: A regression algorithm is used to analyze the abnormal particle size samples to explore the correlation between the particle size values and the aggregate characteristics. This step aims to establish a prediction model for predicting and optimizing the particle size of the aggregates.

[0057] Second Optimization and Screening Module: Based on the results of the regression analysis, an aggregate particle size optimization model is generated. This model will be used to screen the aggregates in the aggregate set to obtain an optimized particle size data set.

[0058] According to the classified aggregate set, a clustering algorithm is used to group the aggregates of each particle size grade; a particle size grouping set is obtained. If there are outliers in the particle size grouping set (such as points far from the clustering center), then proceed to the next step, use a regression algorithm to analyze the correlation between the outliers and the aggregate characteristics, and obtain the outlier analysis results.

[0059] According to the outlier analysis results, an optimization algorithm (such as gradient descent, genetic algorithm, etc.) is used to correct the outliers; a corrected aggregate set is obtained; statistical analysis is used to calculate the proportion of the aggregates of each particle size grade in the corrected aggregate set, and a proportion distribution characteristic is generated.

[0060] According to the proportion distribution characteristic, a classification algorithm (such as decision tree, random forest, etc.) is used to optimize the classification of the corrected aggregate set, and an optimized classification result is obtained. A screening algorithm is used to extract the aggregate particles that meet the preset particle size threshold from the optimized classification result to form a target aggregate set.

[0061] According to the target aggregate set, a verification algorithm (such as cross-validation, A / B test, etc.) is used to evaluate the preset particle size grading standard; a grading standard evaluation report is generated to evaluate the accuracy and applicability of the grading standard.

[0062] Specifically, in the embodiment of the present application, a particle size analyzer is used to measure the particle size of each aggregate particle to obtain a particle size data set. Through an automated data acquisition system, the particle size analyzer is connected to a computer to record the particle size value of each aggregate particle in real time. For example, the particle size measurements of 25 high-density aggregate samples are obtained, and the particle size values are distributed between 5 mm and 20 mm.

[0063] The particle size data is sorted and classified using data processing algorithms to generate a particle size distribution histogram, showing that the particle size is mainly concentrated in the range of 8 mm to 12 mm, with a proportion of 60%. To further analyze the particle size distribution characteristics, a clustering algorithm is used to divide the particle size data into three categories: small particle size (5 mm to 8 mm), medium particle size (8 mm to 12 mm), and large particle size (12 mm to 20 mm), and the proportions of each category are calculated as 20%, 60%, and 20% respectively.

[0064] Through statistical analysis, the average particle size and standard deviation of the particle size data set are calculated as 10 mm and 3 mm respectively, indicating that the particle size distribution of high-density aggregates is relatively concentrated.

[0065] Combining the particle size and density data, a multi-dimensional analysis method is used to screen out aggregate particles with a particle size of 8 mm to 12 mm and a density of 71 g / cm³ to 73 g / cm³ to form an optimized high-density aggregate subset, providing a more accurate material selection basis for subsequent concrete mix design. According to the preset particle size grading standard, the high-density aggregates are divided into multiple particle size grades to form a graded aggregate set.

[0066] First, the particle size grading range is set from 5 mm to 20 mm and is subdivided into four grades: 5 mm to 8 mm, 8 mm to 10 mm, 10 mm to 12 mm, and 12 mm to 20 mm.

[0067] Through an automated data processing system, the particle size data collected by the particle size analyzer is classified according to the above grading standard to generate a graded aggregate set. For example, when grading 30 high-density aggregate samples, the results show that the proportion of aggregates with a particle size of 5 mm to 8 mm is 15%, the proportion of those with a particle size of 8 mm to 10 mm is 30%, the proportion of those with a particle size of 10 mm to 12 mm is 35%, and the proportion of those with a particle size of 12 mm to 20 mm is 20%.

[0068] To further optimize the grading results, a weighted average algorithm is used to calculate the average particle size of each particle size grade, which are 5 mm, 2 mm, 11 mm, and 18 mm respectively. Through variance analysis, the standard deviation of each particle size grade is calculated as 8 mm, 7 mm, 6 mm, and 2 mm respectively, indicating a relatively high internal consistency of the particle size grading results.

[0069] Combining the graded particle size data and density data, a multi-dimensional clustering algorithm is used to further divide the aggregate particles into three categories: high-density small particle size, high-density medium particle size, and high-density large particle size, and the average density of each category is calculated as 73 g / cm³, 78 g / cm³, and 71 g / cm³ respectively, providing a more accurate graded material basis for subsequent concrete mix design.

[0070] In summary, step 102 constitutes a complete process for aggregate particle size processing and optimization, aiming to improve the seismic performance of the UHPC pier structure through means such as precise measurement, statistical analysis, clustering classification, regression analysis, optimization and correction, and verification and evaluation.

[0071] Step 103: Extract the aggregate distribution ratio of each particle size grade from the classified aggregate set, calculate the current grading curve, and obtain grading curve data; if the deviation between the grading curve data and the preset grading curve exceeds the first preset threshold, adjust the aggregate distribution ratio of each particle size grade, recalculate the grading curve until the deviation between the recalculated grading curve and the preset grading curve does not exceed the first preset threshold, and obtain the final grading curve.

[0072] Step 103 involves the extraction of the aggregate distribution ratio and the calculation of the grading curve. In this step, first, it is necessary to extract the aggregate distribution ratio of each particle size grade from the classified aggregate set, and this step is completed through statistical analysis of the classified aggregates, aiming to understand the proportion of aggregates with different particle sizes in the overall.

[0073] Next, according to these distribution ratio data, a fitting algorithm is used to calculate the parameters of the current grading curve. The fitting algorithm is a mathematical method used to estimate the parameters of a function (here, the grading curve) based on known data points (here, the aggregate distribution ratios of different particle sizes). Through this process, a grading curve describing the current aggregate grading situation can be obtained.

[0074] After obtaining the fitting parameter data, it is necessary to perform outlier detection on it. Outliers refer to those data points that are significantly different from most data points, and they may be caused by measurement errors, data entry errors, or other reasons. If there are outliers in the fitting parameter data, then these outliers may have a negative impact on the accuracy of the grading curve.

[0075] Therefore, a regression algorithm is used to analyze the correlation between the outliers and the aggregate distribution. The regression algorithm is a statistical method used to study the relationship between one or more independent variables and a dependent variable. Here, the independent variable is the aggregate particle size distribution, and the dependent variable is the outliers in the fitting parameter data. By analyzing this relationship, an anomaly analysis result can be generated, so as to understand the reasons for the generation of outliers.

[0076] According to the anomaly analysis result, an optimization algorithm is used to correct the fitting parameter data. The optimization algorithm is a mathematical method used to find the solution that makes the objective function reach the optimal (maximum or minimum) under given constraints. Here, the objective function is the accuracy of the fitting parameter data, and the constraint is to keep the overall grading of the aggregates unchanged. Through this process, the corrected parameter data can be obtained, so as to more accurately describe the aggregate grading situation.

[0077] For the corrected parameter data, statistical analysis methods are used to calculate the characteristic values of the current gradation curve. Characteristic values are a set of numerical measures that can describe the characteristics of a data set. Here, the characteristic values can include parameters such as the slope, intercept, and curvature of the gradation curve, which can reflect information on the uniformity and continuity of the aggregate gradation.

[0078] Based on these characteristic value data, a classification algorithm is used to classify the current gradation curve. The classification algorithm is a machine learning technique used to divide input data into one of the predefined classes. Here, the classes can be "excellent gradation", "good gradation", "average gradation", etc., which reflect different quality levels of the aggregate gradation. Through this process, classification result data can be obtained, thereby understanding the quality class to which the current gradation curve belongs.

[0079] Finally, for the classification result data, a screening algorithm is used to extract the aggregate set that meets the preset gradation standard. The screening algorithm is a technique for selecting data items that meet the conditions from a data set according to specific conditions. Here, the specific condition is the preset gradation standard, which stipulates the quality requirements that the aggregate gradation should meet. Through this process, an aggregate set meeting the gradation standard can be obtained, and these aggregates will be used in the subsequent concrete preparation process.

[0080] After obtaining the final gradation curve, it is also necessary to determine whether the deviation from the preset gradation curve exceeds the first preset threshold. If the deviation exceeds the threshold, it indicates that there is a significant difference between the current gradation curve and the ideal state, and it is necessary to adjust the aggregate ratio of the particle size grade according to the aggregate set meeting the gradation standard to improve the aggregate gradation. This process is an iterative process and needs to be repeated until the preset gradation requirements are met.

[0081] Specifically, in the embodiment of the present application, the aggregate distribution ratio of each particle size grade is extracted from the classified aggregate set. First, the data acquisition module is used to obtain the particle size distribution data of the aggregate sample. For example, among 30 samples, the aggregate with a particle size of 5 mm to 8 mm accounts for 15%, the aggregate with a particle size of 8 mm to 10 mm accounts for 30%, the aggregate with a particle size of 10 mm to 12 mm accounts for 35%, and the aggregate with a particle size of 12 mm to 20 mm accounts for 20%.

[0082] Using these ratio data, the cumulative distribution function of the aggregate is constructed, and the gradation curve is generated through the interpolation algorithm. For example, with the particle size as the abscissa and the cumulative percentage as the ordinate, the cubic spline interpolation method is used when plotting the curve to ensure the smoothness and accuracy of the curve. Further, the gradation curve is fitted by the least squares method, and the fitted equation is y = 85x^3 - 2x^2 + 35x + 15, where x is the particle size and y is the cumulative percentage.

[0083] By calculating the slope change of the curve, the uniformity of aggregate distribution is analyzed. For example, a large slope in the range of 8 mm to 10 mm indicates that the aggregate distribution in this range is relatively concentrated.

[0084] Combined with the gradation curve data, the Monte Carlo simulation method is used to predict the distribution effect of aggregates with different particle sizes in concrete. For example, the simulation results show that the distribution density of aggregates with a size of 12 mm to 20 mm in concrete is 45, indicating good filling effect.

[0085] Finally, through the K-means clustering algorithm, the gradation curve is divided into different regions, such as the low particle size region, the medium particle size region, and the high particle size region, and the center points of each region are calculated to be 6 mm, 11 mm, and 16 mm respectively, providing a more accurate optimization basis for subsequent concrete mix design. When analyzing the gradation curve, if the deviation from the preset ideal gradation curve exceeds 5%, the aggregate ratio of each particle size grade needs to be adjusted according to engineering experience or design specifications. For example, assume that the cumulative percentage in the range of 8 mm to 10 mm of the current gradation curve is 30%, while the corresponding range of the ideal curve should be 35%, and the deviation is 5%.

[0086] At this time, the adaptive optimization algorithm is used to adjust the aggregate ratio, increasing the proportion in the range of 8 mm to 10 mm to 33% and reducing the proportion in the range of 10 mm to 12 mm by 2% to maintain the overall ratio balance.

[0087] The gradient descent method is used to iteratively optimize the adjusted data to ensure that the deviation of the adjusted gradation curve is controlled within 1%. Further, the adjustment effect is verified through nonlinear regression analysis. For example, the fitting equation is y = 88x^3 - 3x^2 + 32x + 18, where x is the particle size and y is the cumulative percentage, and the goodness of fit R 2 reaches 98, indicating a high degree of coincidence between the adjusted gradation curve and the ideal curve.

[0088] Combined with the adjusted curve data, the finite element analysis method is used to simulate the distribution effect of aggregates in concrete. For example, the simulation results show that the distribution density of aggregates with a size of 8 mm to 10 mm increases from 30 to 33, and the filling effect is significantly improved.

[0089] Finally, the gradation curve is dimensionally reduced through principal component analysis to extract key features. For example, the contribution rate of the first principal component is 85% and the contribution rate of the second principal component is 10%, providing a more accurate scientific basis for subsequent concrete mix design.

[0090] In summary, step 103 constitutes the core part of the aggregate gradation optimization in the seismic performance optimization method of the UHPC pier structure. Through the implementation of these steps, it can be ensured that the finally obtained aggregate gradation meets the design requirements, thereby improving the seismic performance of the UHPC pier.

[0091] In step 104, according to the final gradation curve, the optimal aggregate ratio for each particle size grade is calculated using an aggregate ratio algorithm based on the minimum void ratio to obtain an aggregate ratio plan. The aggregate ratio plan is input into the concrete mixing equipment, and the slump and spread of the UHPC are monitored to determine whether the slump is lower than the second preset threshold or the spread is less than the third preset threshold. If the slump is lower than the second preset threshold or the spread is less than the third preset threshold, the aggregate ratio plan is readjusted until the slump is not lower than the second preset threshold or the spread is not less than the third preset threshold.

[0092] Step 104 involves determining the optimal aggregate ratio based on the final gradation curve and ensuring that the prepared UHPC meets specific physical property requirements.

[0093] First, the optimal aggregate ratio needs to be calculated according to the final gradation curve. The final gradation curve is obtained through a series of steps (such as aggregate screening, particle size grading, gradation curve adjustment, etc.) and reflects the proportion distribution of aggregates with different particle sizes in the mixture. The aggregate ratio algorithm based on the minimum void ratio aims to minimize the void ratio inside the concrete by optimizing the aggregate ratio, thereby improving the density and mechanical properties of the concrete. The algorithm takes into account factors such as the shape, size, and density of the aggregates, as well as their interactions, to determine the optimal aggregate ratio.

[0094] Then, the calculated optimal aggregate ratio plan is input into the concrete mixing equipment to prepare UHPC, and the slump and spread of the UHPC are monitored. Among them, the slump is an important indicator to measure the fluidity of concrete and reflects the deformation ability of concrete under its own weight; the spread is the area where the concrete spreads on the horizontal plane and is also an indicator to measure the fluidity of concrete. During the preparation process, it is necessary to monitor the slump and spread of the UHPC in real time to ensure that they meet specific performance requirements.

[0095] If the monitoring results show that the slump is lower than the second preset threshold or the spread is less than the third preset threshold, it indicates that there may be problems with the current aggregate ratio plan and it needs to be adjusted. The basis for adjustment can be the deviation between the actual measured values of the slump and spread and the preset thresholds, as well as the possible impact of these deviations on the concrete performance. The adjusted aggregate ratio plan needs to be input into the concrete mixing equipment again for preparation, and the monitoring steps are repeated until the performance requirements are met.

[0096] By precisely controlling the aggregate ratio in step 104, the density and mechanical properties of the concrete can be significantly improved, thereby enhancing the seismic resistance of the pier structure. In addition, real-time monitoring of the slump and spread can ensure that the prepared UHPC meets specific physical property requirements and avoid quality problems during subsequent construction.

[0097] Specifically, according to the adjusted grading curve, the embodiments of the present application adopt an aggregate proportioning algorithm based on the minimum void ratio, aiming to optimize the aggregate packing density, and calculate the optimal proportion of aggregates of each particle size grade. First, based on the aggregate particle size distribution data, the initial void ratio is calculated to be 42% using the void ratio calculation formula V = 1 - (ρ_b / ρ_a), where ρ_b is the aggregate packing density and ρ_a is the true density of the aggregate.

[0098] The aggregate proportion is optimized through a genetic algorithm. The population size is set to 100, the number of iterations is 50, the crossover probability is 8, and the mutation probability is 1. The proportion of aggregates in each particle size interval is gradually adjusted. For example, the proportion in the interval of 15 mm to 3 mm is adjusted from 12% to 15%, and at the same time, the proportion in the interval of 36 mm to 75 mm is reduced from 18% to 16%.

[0099] After optimization, the calculated minimum void ratio is 38%, which is 4% lower than the initial value.

[0100] Furthermore, the Monte Carlo simulation method is adopted to randomly generate 1000 groups of aggregate proportioning schemes to verify the stability of the optimization results. The simulation results show that the void ratio fluctuation range of the optimized proportioning scheme is from 35% to 35%, indicating that the optimization results have high robustness.

[0101] Combined with the optimized aggregate proportion, the discrete element method is used to simulate the packing state of aggregates in concrete. The simulation results show that the aggregate packing density is increased from 58% to 62%, and the number of contact points between aggregates is increased by 15%, effectively improving the workability of concrete.

[0102] Finally, an aggregate proportioning scheme is generated based on the optimization results. For example, the proportion in the interval of 15 mm to 3 mm is 15%, the proportion in the interval of 3 mm to 6 mm is 20%, the proportion in the interval of 6 mm to 18 mm is 18%, the proportion in the interval of 18 mm to 36 mm is 17%, the proportion in the interval of 36 mm to 75 mm is 16%, and the proportion in the interval of 75 mm to 5 mm is 14%.

[0103] After the optimized aggregate proportioning scheme is input into the concrete mixing equipment, the system automatically activates the real-time monitoring module, and uses laser sensors and image recognition technology to dynamically detect the slump and spread of concrete.

[0104] The slump height of the concrete is obtained through a slump tester, and the spread diameter of the concrete is collected in combination with a spread meter. The data is transmitted to the control center in real time for comparison and analysis.

[0105] If the slump is detected to be less than 160 mm or the spread is less than 450 mm, the mixing ratio adjustment algorithm will be automatically triggered. Based on the fuzzy logic control theory, the aggregate mixing ratio will be dynamically adjusted according to the current concrete performance deviation value. For example, when the slump is 150 mm, the calculated deviation value is 10 mm, and based on this deviation value, the proportion of aggregates in the range of 15 mm to 3 mm is increased from 15% to 16%, while the proportion in the range of 3 mm to 6 mm is decreased from 20% to 19%. After the adjustment, the concrete is remixed and its performance is detected again until the slump reaches 165 mm and the spread reaches 460 mm, meeting the preset construction performance standards.

[0106] To verify the stability of the adjusted mixing ratio, the embodiment of this application also uses a grey prediction model to predict the slump and spread of the subsequent 10 mixings. The prediction results show that the slump fluctuates in the range of 162 mm to 168 mm, and the spread fluctuates in the range of 455 mm to 465 mm, indicating that the adjusted mixing ratio scheme has high stability and reliability. Finally, the optimized aggregate mixing ratio scheme is stored in the database to provide a reference for subsequent concrete production.

[0107] In summary, step 104 involves multiple links such as aggregate mixing ratio calculation, concrete preparation, performance monitoring and adjustment. By precisely controlling these links, the density and mechanical properties of the concrete can be significantly improved, thereby enhancing the seismic resistance of the pier structure.

[0108] It should be noted that in this application, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of this application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, the features described with reference to certain examples may be combined in other examples.

[0109] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

Claims

1. A method for optimizing the seismic performance of a UHPC pier structure, characterized in that: The method comprises: Obtaining an initial aggregate sample through an aggregate library, and obtaining a density value of the initial aggregate sample through a density detection device to obtain a density data set; setting a preset density threshold according to the density data set, and screening according to the density threshold to obtain an aggregate set; The particle size value of each aggregate particle in the aggregate set is obtained by a particle size analysis device to obtain a particle size data set, and the aggregate particles are classified into particle size grades according to the particle size data set and a preset particle size classification standard to obtain a graded aggregate set; Extracting the aggregate distribution ratio of each particle size grade from the graded aggregate set, calculating the current grading curve, and obtaining grading curve data; if the deviation between the grading curve data and the preset grading curve exceeds a first preset threshold, adjusting the aggregate distribution ratio of each particle size grade, recalculating the grading curve, until the deviation between the recalculated grading curve and the preset grading curve does not exceed the first preset threshold, and obtaining a final grading curve; According to the final grading curve, an aggregate proportioning algorithm based on minimum void ratio is used to calculate the optimal aggregate proportion of each particle size grade to obtain an aggregate proportioning scheme; the aggregate proportioning scheme is input into a concrete mixing device, and the slump and expansion of the UHPC are monitored to determine whether the slump is lower than a second preset threshold or whether the expansion is less than a third preset threshold; if the slump is lower than the second preset threshold or the expansion is less than the third preset threshold, the aggregate proportioning scheme is readjusted until the slump is not lower than the second preset threshold or the expansion is not less than the third preset threshold.

2. The method for optimizing the seismic performance of a UHPC pier structure according to claim 1, characterized in that: The method comprises: According to the density data set, the mean and standard deviation of the density value are calculated to determine the density distribution characteristics; if the density distribution characteristics meet the fourth preset threshold, the density data set is classified by using a clustering algorithm to obtain a first classification result; according to the first classification result, aggregate samples with abnormal density values ​​are extracted to generate a first abnormal sample set; A regression algorithm is used to analyze the first abnormal sample set to obtain a first correlation analysis result between the density value and the aggregate property. An aggregate density optimization model is generated based on the first correlation analysis result. Aggregates in the aggregate library are screened using the aggregate density optimization model to obtain an optimized aggregate set.

3. The method for optimizing the seismic performance of a UHPC pier structure according to claim 1, characterized in that: The method comprises: According to the particle size data set, the mean and standard deviation of the particle size values ​​are calculated to determine the particle size distribution characteristics; if the particle size distribution characteristics meet the fifth preset threshold, the particle size data set is classified by using a clustering algorithm to obtain a second classification result; according to the second classification result, aggregate samples with abnormal particle size values ​​are extracted to generate a second abnormal sample set; The second abnormal sample set is analyzed by using a regression algorithm to obtain a second correlation analysis result between the particle size value and the aggregate property. An aggregate particle size optimization model is generated based on the second correlation analysis result. Aggregates in the aggregate set are screened by using the aggregate particle size optimization model to obtain an optimized particle size data set.

4. The method for optimizing the seismic performance of a UHPC pier structure according to claim 1, characterized in that: The method comprises: According to the graded aggregate set, a clustering algorithm is used to group the aggregates of each particle size grade to obtain a particle size grouping set; if there are outliers in the particle size grouping set, a regression algorithm is used to analyze the correlation between the outliers and aggregate characteristics to obtain an outlier analysis result; According to the outlier analysis result, the outliers are corrected by using an optimization algorithm to obtain a corrected aggregate set; the proportion of aggregates of each particle size grade in the corrected aggregate set is calculated by using statistical analysis to generate a proportion distribution feature; According to the proportion distribution characteristics, a classification algorithm is used to optimize the classification of the corrected aggregate set to obtain an optimized classification result; a screening algorithm is used to extract aggregate particles that meet a preset particle size threshold in the optimized classification result to form a target aggregate set; based on the target aggregate set, a verification algorithm is used to evaluate the preset particle size grading standard to generate a grading standard evaluation report.

5. The method for optimizing the seismic performance of a UHPC pier structure according to claim 1, characterized in that: The method comprises: An extraction algorithm is used to obtain the aggregate distribution ratio of each particle size grade from the graded aggregate set to generate distribution ratio data; a fitting algorithm is used to calculate the parameters of the current grading curve according to the distribution ratio data to generate fitting parameter data; If there are abnormal values ​​in the fitting parameter data, a regression algorithm is used to analyze the correlation between the abnormal value and aggregate distribution to generate an abnormal analysis result; based on the abnormal analysis result, an optimization algorithm is used to correct the fitting parameter data to generate corrected parameter data; For the correction parameter data, a statistical analysis method is used to calculate the characteristic value of the current grading curve to generate characteristic value data; based on the characteristic value data, a classification algorithm is used to classify the current grading curve to generate classification result data; for the classification result data, a screening algorithm is used to extract a set of aggregates that meet the preset grading standards to obtain a set of graded standard aggregates.

6. The method for optimizing the seismic performance of a UHPC pier structure according to claim 5, characterized in that: The method comprises: It is determined whether the deviation between the grading curve data and the preset grading curve exceeds a first preset threshold value. If the deviation exceeds the first preset threshold value, the aggregate proportion of the particle size grade is adjusted according to the grading standard aggregate set.

Images