Optimization design method for mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder

By establishing a prediction model of ultra-low carbon high-strength concrete performance of stone powder, optimizing the amount of stone powder, solving the low utilization efficiency and environmental problems of stone powder, achieving the goal of maximizing stone powder and achieving low carbon, and suitable for stone powder concrete of different strength grades Match ratio.

CN118888064BActive Publication Date: 2024-12-06HUAQIAO UNIVERSITY +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411356963.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-27
Publication Date
2024-12-06
Estimated Expiration
2044-09-27

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively utilize stone powder, which leads to environmental and ecological problems in the stone processing industry. At the same time, there is a lack of a method that can target the general amount of stone powder for different concrete materials, making it difficult to design an efficient concrete mix ratio.

Method used

By establishing a test database of high-strength concrete mix ratio and raw material particle size distribution-density, combined with the close packing principle and machine learning algorithm, an ultra-low carbon high-strength concrete performance prediction model of calcium carbonate stone powder is constructed, and the amount of stone powder is optimized to maximize utilization Stone powder, comprehensive index for reducing carbon emissions and life cycle evaluation.

Benefits of technology

It has achieved the maximum amount of stone powder, the concrete strength meets requirements, and at the same time reduced carbon emissions and costs, and overcomes the problem that traditional design methods are difficult to meet multiple performances and multiple goals. It is suitable for stone powder concrete mix ratios of different strength grades.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118888064B_ABST
    Figure CN118888064B_ABST
Patent Text Reader

Abstract

The invention discloses a method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder, which comprises: establishing an experimental database of high-strength concrete mix ratio and raw material particle size distribution-density; establishing a calculation model for carbon emissions and costs of calcium carbonate stone powder solid waste transportation; based on the high-strength concrete mix ratio and the experimental database, obtaining a performance prediction model for ultra-low carbon high-strength concrete of calcium carbonate stone powder solid waste; inputting raw material dosage and concrete strength into the performance prediction model, outputting mix ratios of multiple groups of raw materials and the strength of each mix ratio, and calculating the carbon emissions and life cycle assessment comprehensive index ECM of each mix ratio; based on the mix ratio and the corresponding carbon emissions and ECM, setting the calcium carbonate stone powder dosage as a first optimization objective function, setting the carbon emissions as a second optimization objective function, correcting the ECM and maximizing it as a third optimization objective function, and obtaining the optimal mix ratio from the multiple mix ratios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of concrete mix design, and in particular to a mix optimization design method for ultra-low carbon high-strength concrete using stone powder to the maximum extent possible. Background Art

[0002] The amount of waste stone powder generated during the cutting, molding and polishing of building stones represented by marble is as high as 25% of the total processed stone. Some of the waste stone powder is reused in low value-added industries, while the commonly used treatment method is landfill, which has caused serious environmental and public health problems. The main components of marble powder and limestone powder (Limestone powder, LP) are calcite, the chemical composition is calcium carbonate crystals, and the average particle size is less than 75mm. Partially replacing cement or fine aggregate to make concrete materials is expected to break through the bottleneck faced by the current stone processing industry in its transformation to a green and sustainable industry, while reducing the large amount of greenhouse gas emissions caused by the cement production process and the ecological damage caused by excessive exploitation of sandy aggregates.

[0003] At present, a large number of experiments have been conducted to study the changes in the mechanical properties of concrete materials after the addition of stone powder. However, the test results are greatly affected by the raw materials, mix design methods, and substitution methods, and it is difficult to obtain a universal stone powder dosage for different concrete materials. Research on the microscopic aspects of concrete materials has shown that although calcium carbonate stone powder is an inert material, it can have filling effects, nucleation effects, and other effects, and can improve the performance of concrete within an appropriate dosage range. The reason is that under different raw materials and target strengths, the types and dosages of stone powder suitable for addition are different. Existing studies have not considered these factors, so the dosages obtained by each scholar are different. How to improve the traditional alternative design method, conduct efficient mix design and avoid a large number of experiments, and obtain the optimal mix ratio of concrete with the maximum amount of fossil powder for different strength grades needs to be proposed. Summary of the invention

[0004] The embodiment of the present invention proposes a mix ratio optimization design method for ultra-low carbon high-strength concrete that maximizes the utilization of stone powder to improve the above-mentioned problems.

[0005] The embodiment of the present invention provides a method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder, which comprises the following steps:

[0006] S1: Establish an experimental database of high-strength concrete mix proportion and raw material particle size distribution-density;

[0007] S2: Establish a calculation model for carbon emissions and costs of calcium carbonate stone powder solid waste production and transportation;

[0008] S3: Based on the establishment of an experimental database of high-strength concrete mix ratio and raw material particle size distribution-density, a performance prediction model for ultra-low carbon high-strength concrete made of calcium carbonate stone powder solid waste was built and trained in combination with the close packing principle and machine learning algorithm;

[0009] S4: inputting the raw material dosage and concrete strength into the performance prediction model, outputting the mix ratios of multiple groups of raw materials and the strength of each mix ratio, and calculating the carbon emissions and life cycle assessment comprehensive index ECM of each mix ratio according to the mix ratios of the multiple groups of raw materials and the calculation model;

[0010] S5: Based on the mix proportions of the multiple groups of raw materials and the corresponding carbon emissions and life cycle assessment comprehensive index ECM, the calcium carbonate stone powder dosage is set as the first optimization objective function, the carbon emissions are set as the second optimization objective function, the life cycle assessment comprehensive index ECM is corrected and its maximum is used as the third optimization objective function, and the optimal mix proportion is obtained from the multiple groups of mix proportions.

[0011] Preferably, in step S1, the raw materials include fine aggregate, coarse aggregate, water reducer and calcium carbonate stone powder; the fine aggregate includes one or more of river sand, mountain sand and machine-made sand, the coarse aggregate includes one or more of granite, basalt and bluestone, the calcium carbonate stone powder includes one or more of marble powder and limestone powder, the water reducer includes one or more of polycarboxylic acid-based high-efficiency water reducer and naphthalene-based water reducer, the raw material particle size distribution is the cumulative particle size throughput, and the raw material density is the apparent density.

[0012] Preferably, in step S2:

[0013] The calculation formula for the carbon emission factor of calcium carbonate stone powder solid waste transportation is:

[0014] C mp =dT

[0015] Among them, C mp represents the carbon emission factor of calcium carbonate stone powder solid waste in the transportation stage; d represents the average transportation distance of calcium carbonate stone powder solid waste; T represents the carbon emission factor of calcium carbonate stone powder solid waste per unit weight of transportation distance;

[0016] The calculation formula for the carbon emission cost of calcium carbonate stone powder solid waste transportation is:

[0017] S mp =dG

[0018] Among them, S mp It represents the unit cost of calcium carbonate stone powder solid waste in the transportation stage; G represents the cost of transporting calcium carbonate stone powder solid waste per unit weight.

[0019] Preferably, in step S3, the training method of the performance prediction model is as follows:

[0020] Collect training data samples, wherein the training data samples include a close packing degree k, a water-cement ratio b, a compressive strength m, and a colloid strength richness coefficient nf under different mix ratios;

[0021] Clean and organize the training data samples, and handle missing values ​​and outliers;

[0022] Input the training data samples into the machine learning algorithm to establish the relationship between m and k, b, and nf, and obtain the performance prediction model; the machine learning algorithm includes MLP, BP, SVR, and XGB; the collected high-strength concrete mix ratio and performance data are divided into a training data set and a verification data set in a ratio of 8:2, and the verification and training are synchronized for model optimization and debugging parameters. After the model is trained, it is tested;

[0023] Use cross-validation to evaluate model performance and ensure that the model has good predictive and generalization capabilities;

[0024] Through parameter adjustment and feature optimization, the accuracy and stability of the model can be further improved.

[0025] Preferably, in step S3, the raw material close packing degree index k is constructed according to the residual sum of squares RSS of the particle size distribution, wherein:

[0026]

[0027]

[0028] Where D is the particle size, P(D) is the proportion of particles smaller than D; D max is the maximum particle size; D min is the minimum particle size; q is the distribution modulus, is the maximum particle size under the distribution modulus q, is the minimum particle size under the distribution modulus q; n is the number of particles, e i is the residual of the ith particle, and The particle size classification of the mixture and the target curve respectively;

[0029] The cement dosage, auxiliary cementitious material dosage, fine aggregate dosage, coarse aggregate dosage, water dosage, water reducing agent dosage, stone powder dosage, and high-strength concrete compressive strength are used as the input of the test database. The data are normalized to eliminate the impact of the dimensions between the data to facilitate the convergence of the model. After the output, they are denormalized to obtain the actual mix ratio.

[0030] Preferably, the S5 includes:

[0031] S51: According to the target demand of maximizing the utilization of calcium carbonate stone powder, an optimization objective function of stone powder dosage is established, and on this basis, an optimization objective function of carbon emission and life cycle assessment comprehensive index ECM is constructed, and constraint conditions are set;

[0032] S52: Use the performance prediction model to predict the strength of calcium carbonate stone powder high-strength concrete, and set the optimal hyperparameters according to different algorithms to ensure a low error;

[0033] S53: Select the mix ratio that meets the strength grade requirements and has the largest amount of stone powder from the optimization results, and on this basis select the mix ratio with the smallest carbon emissions and the largest life cycle assessment comprehensive index ECM to obtain the optimal mix ratio.

[0034] Preferably, in S53, the stone powder dosage includes the dosage for satisfying close packing and the dosage for additionally replacing part of the fine aggregate while maintaining close packing and concrete strength.

[0035] Preferably, in S5, the optimization objective function includes three optimization objective functions in a sequential order, namely, a first optimization objective function, a second optimization objective function and a third optimization objective function, wherein the first optimization objective function is constructed based on the maximum amount of stone powder as a goal, the second optimization objective function is constructed based on the minimum carbon emission as a goal, and the third optimization objective function is constructed based on the maximum comprehensive evaluation index ECM of life cycle assessment as a goal;

[0036] The calculation formula of the first optimization objective function is:

[0037]

[0038] Among them, m mp1 Indicates the amount of stone powder replacing cement; m mp2 Indicates the amount of stone powder replacing fine aggregate; m c Indicates the amount of cement used; m fa Indicates the amount of fine aggregate.

[0039] In this embodiment, it should be noted that the mix ratio with the maximum stone powder content (for example, 30%) predicted by machine learning is often more than one, and there may be dozens or even hundreds of them. Therefore, further screening is required according to the second optimization objective function and the third optimization objective function.

[0040] The calculation formula of the second optimization objective function is:

[0041]

[0042] Among them, C c is the unit carbon emission factor of cement; C S is the unit carbon emission factor of auxiliary cementitious materials; C fa is the unit carbon emission factor of fine aggregate; C ca is the unit carbon emission factor of coarse aggregate; C w is the unit carbon emission factor of water; C sp is the unit carbon emission factor of water reducer;

[0043] The calculation formula of the third optimization objective function is:

[0044]

[0045] Among them, C S It represents the compressive strength of stone powder concrete; Cost represents the cost of each cubic meter of stone powder concrete; CO2 represents the carbon emission coefficient of each cubic meter of stone powder concrete.

[0046] The calculation formula of Cost in the second optimization objective function is:

[0047]

[0048] Among them, S c is the unit price of cement; S s is the unit price of auxiliary cementitious materials; S fa is the unit price of fine aggregate; S ca is the unit price of coarse aggregate; S w is the unit price of water; S sp is the unit price of water reducing agent; S mp is the unit price of stone powder;

[0049] The setting of preset constraints in the process of mix optimization includes key parameter constraints and volume constraints;

[0050] The formula for the key parameter constraint is:

[0051]

[0052]

[0053]

[0054]

[0055] Among them, w represents the water-cement ratio, s represents the sand ratio, r represents the water reducing agent dosage, p represents the stone powder dosage, and w min and w max Represents the lower and upper limits of the water-binder ratio, smin and max Represents the lower and upper limits of the sand ratio, r min and r max represents the lower and upper limits of water reducing agent dosage, p min and p max Represents the lower and upper limits of stone powder dosage;

[0056] The formula of the volume constraint is:

[0057]

[0058] Among them, m s The amount of auxiliary cementitious material; m w is the amount of water used; m ca is the amount of coarse aggregate; m sp is the dosage of water reducing agent; c is the density of cement; ρ s is the density of the auxiliary cementitious material; w is the density of water; ρ fa is the density of fine aggregate; ρ ca is the density of coarse aggregate; ρ mp is the density of stone powder; ρ sp is the density of the water reducer.

[0059] Preferably, the optimal solution of the optimization objective function must first satisfy the first optimization objective function, and then successively satisfy the second and third optimization objective functions to obtain the optimal solution for replacing cement, specifically:

[0060] The parameters of the mix ratio predicted by the performance prediction model include cement dosage, auxiliary cementitious material dosage, water dosage, fine aggregate dosage, coarse aggregate dosage, stone powder dosage, and water reducing agent dosage, and then classified according to strength grade and stone powder dosage;

[0061] The classified mix ratios are judged, and when it is judged that the predicted strength of the mix ratio differs from the set concrete strength by 5 MPa, the mix ratio is removed, and the remaining mix ratios are optimized and screened;

[0062] When the key parameter constraints and volume constraints are met, at the corresponding strength level, according to the first optimization objective function, it is judged whether a certain mix ratio is the optimal solution, wherein the mix ratio satisfies the maximum value of the first optimization objective function, that is, it satisfies the condition;

[0063] If yes, continue to optimize;

[0064] If not, stop the optimization and reselect other mix ratios for judgment;

[0065] When it is determined that the mix ratio satisfies the first optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the second optimization objective function, wherein the mix ratio satisfies the minimum value of the second optimization objective function, that is, satisfies the condition;

[0066] If so, the mix ratio is the optimal solution;

[0067] If not, reselect other mix ratios for judgment.

[0068] When it is determined that the mix ratio satisfies the second optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the third optimization objective function, wherein the mix ratio satisfies the maximum value of the third optimization objective function, that is, satisfies the condition;

[0069] If so, the mix ratio is the optimal solution;

[0070] If not, reselect other mix ratios for judgment.

[0071] Preferably, it also includes:

[0072] S6: The optimal mix ratio is experimentally verified and adjusted to obtain a final mix ratio.

[0073] Compared with the prior art, the advantages of the present invention are:

[0074] 1. Usually, the use of waste stone powder requires a large number of tests to design the mix ratio, so as to determine the proportion of stone powder to replace fine aggregate or cement. This is greatly affected by the raw materials and requires repeated tests. However, this method can ensure that the strength grade of concrete after stone powder replacement meets the requirements through machine learning, and can also maximize the amount of stone powder, as well as better environmental and economic benefits.

[0075] 2. The method provided by the present invention overcomes the problem that the traditional mix design method is difficult to satisfy the concrete material to achieve multiple performances and multiple objectives.

[0076] 3. The method provided by the present invention can achieve a high degree of prediction of the performance of stone powder concrete by establishing a database including raw material parameters and adopting multiple machine learning algorithms such as MLP, BP, SVR, and XGB.

[0077] 4. The method provided by the present invention is applicable to stone powder concrete mix proportions of different strength grades, which is more in line with actual engineering application requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0079] Figure 1 A schematic flow chart of a method for optimizing the mix ratio of ultra-low carbon high-strength concrete with maximum utilization of stone powder provided in an embodiment of the present invention.

[0080] Figure 2 A schematic diagram of the principle of a method for optimizing the mix ratio of ultra-low carbon high-strength concrete with maximum utilization of stone powder provided in an embodiment of the present invention.

[0081] Figure 3 This is the optimization result diagram of high-strength concrete mix ratio using marble powder instead of cement. DETAILED DESCRIPTION

[0082] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] See also Figure 1 and Figure 2 The first embodiment of the present invention provides a method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder, which comprises the following steps:

[0084] S1: Establish an experimental database of high-strength concrete mix proportions and raw material particle size distribution-density.

[0085] In this embodiment, the raw materials include fine aggregate, coarse aggregate, water reducer and calcium carbonate stone powder; the fine aggregate includes one or more of river sand, mountain sand and machine-made sand, the coarse aggregate includes one or more of granite, basalt and bluestone, the calcium carbonate stone powder includes one or more of marble powder and limestone powder, the water reducer includes one or more of polycarboxylic acid-based high-efficiency water reducer and naphthalene-based water reducer, the raw material particle size distribution is the cumulative particle size throughput, and the raw material density is the apparent density.

[0086] In this embodiment, 112 test data on the mix ratio and compressive strength of marble powder high-strength concrete were collected from different documents, and a test database was established, wherein the parameters of the test database include: cement dosage, auxiliary cementitious material dosage, fine aggregate dosage, coarse aggregate dosage, water dosage, water reducing agent dosage, marble powder dosage, raw material particle size, density, and compressive strength of marble powder high-strength concrete. Then, the data was subjected to standard normalization to eliminate the influence of the dimensions between the data, facilitate the convergence of the model, and then denormalized after output to obtain the actual concrete mix ratio result.

[0087] S2: Establish a calculation model for carbon emissions and costs of production and transportation of calcium carbonate stone powder solid waste.

[0088] Specifically, the calculation formula for the carbon emission factor of calcium carbonate stone powder solid waste transportation is:

[0089] C mp =dT

[0090] Among them, C mp represents the carbon emission factor of calcium carbonate stone powder solid waste in the transportation stage; d represents the average transportation distance of calcium carbonate stone powder solid waste; T represents the carbon emission factor of calcium carbonate stone powder solid waste per unit weight of transportation distance;

[0091] The calculation formula for the carbon emission cost of calcium carbonate stone powder solid waste transportation is:

[0092] S mp =dG

[0093] Among them, S mp It represents the unit cost of calcium carbonate stone powder solid waste in the transportation stage; G represents the cost of transporting calcium carbonate stone powder solid waste per unit weight.

[0094] S3: Based on the establishment of an experimental database of high-strength concrete mix proportions and raw material particle size distribution-density, combined with the principle of close stacking and machine learning algorithms, a performance prediction model for ultra-low carbon high-strength concrete made of calcium carbonate stone powder solid waste was built and trained.

[0095] Specifically, in this embodiment, the performance prediction model can be constructed based on a machine learning algorithm, and the machine learning algorithm includes MLP, BP, SVR, XGB, etc., which is not specifically limited in the present invention. In particular, this embodiment uses the XGB algorithm to establish a marble powder high-strength concrete compressive strength prediction model. The XGB algorithm has strong prediction performance, parallel processing capabilities, good generalization capabilities, and uses regularization to minimize overfitting. The specific parameters are set as follows: the maximum depth of the tree is set to 10, the number of trees is set to 100, the learning rate is set to 0.35, and the number of training rounds is set to 10.

[0096] After the performance prediction model is constructed, it needs to be trained. The training steps are as follows:

[0097] The training data samples are collected, wherein the training data samples include the compactness k, the water-cement ratio b, the compressive strength m and the colloid strength richness coefficient nf under different mix ratios.

[0098] The close packing index k reflects how closely the particles of each component in the mixture are arranged, affecting the density and strength of the concrete. The close packing index k is constructed based on the residual sum of squares RSS of the particle size distribution, where:

[0099]

[0100]

[0101] Where D is the particle size, P(D) is the proportion of particles smaller than D; D max is the maximum particle size; D min is the minimum particle size; q is the distribution modulus, is the maximum particle size under the distribution modulus q, is the minimum particle size under the distribution modulus q; n is the number of particles, e i is the residual of the ith particle, and Particle size classification for mixture and target curve respectively.

[0102] Clean and organize the training data samples, and handle missing values ​​and outliers;

[0103] The training data samples are input into the machine learning algorithm to establish the relationship between m and k, b, and n to obtain a performance prediction model.

[0104] Among them, by clarifying the relationship between these parameters, the optimal concrete mix can be designed to ensure that costs and resources are saved while meeting strength requirements.

[0105] Furthermore, this embodiment divides the collected training data samples into a training data set, a test data set, and a verification data set in a ratio of 8:2. Verification and training are synchronized for model optimization and debugging parameters. After the model is trained, it is tested.

[0106] S4: Input the raw material dosage and concrete strength into the performance prediction model, output the mix ratio of multiple groups of raw materials and the strength of each mix ratio, and calculate the carbon emissions and life cycle assessment comprehensive index ECM of each mix ratio based on the mix ratio of the multiple groups of raw materials and the calculation model.

[0107] S5: Based on the mix proportions of the multiple groups of raw materials and the corresponding carbon emissions and life cycle assessment comprehensive index ECM, the calcium carbonate stone powder dosage is set as the first optimization objective function, the carbon emissions are set as the second optimization objective function, the life cycle assessment comprehensive index ECM is corrected and its maximum is used as the third optimization objective function, and the optimal mix proportion is obtained from the multiple groups of mix proportions.

[0108] In this embodiment, S5 includes:

[0109] S51: According to the target demand of maximizing the utilization of calcium carbonate stone powder, an optimization objective function of stone powder dosage is established, and on this basis, an optimization objective function of carbon emission and life cycle assessment comprehensive index ECM is constructed, and constraint conditions are set;

[0110] S52: Use the performance prediction model to predict the strength of calcium carbonate stone powder high-strength concrete, and set the optimal hyperparameters according to different algorithms to ensure a low error;

[0111] S53: Select the mix ratio that meets the strength grade requirements and has the largest amount of stone powder from the optimization results, and on this basis select the mix ratio with the smallest carbon emissions and the largest life cycle assessment comprehensive index ECM to obtain the optimal mix ratio.

[0112] Among them, in S53, the amount of stone powder added includes the amount to meet the requirement of tight packing and the amount to replace part of the fine aggregate while maintaining tight packing and concrete strength.

[0113] Specifically, in S5, the optimization objective function includes three optimization objective functions in a sequential order, namely, a first optimization objective function, a second optimization objective function and a third optimization objective function, wherein the first optimization objective function is constructed based on the goal of maximizing the stone powder content, the second optimization objective function is constructed based on the goal of minimizing carbon emissions, and the third optimization objective function is constructed based on the goal of maximizing the life cycle assessment comprehensive index ECM.

[0114] The calculation formula of the first optimization objective function is:

[0115]

[0116] Among them, m mp1 Indicates the amount of stone powder replacing cement; m mp2 Indicates the amount of stone powder replacing fine aggregate; m c Indicates the amount of cement used; m fa Indicates the amount of fine aggregate.

[0117] The calculation formula of the second optimization objective function is:

[0118]

[0119] Where Cc is the unit carbon emission factor of cement; C S is the unit carbon emission factor of auxiliary cementitious materials; C fa is the unit carbon emission factor of fine aggregate; C ca is the unit carbon emission factor of coarse aggregate; C w is the unit carbon emission factor of water; C sp is the unit carbon emission factor of water reducer; C mp is the unit carbon emission factor of stone powder;

[0120] The calculation formula of the third optimization objective function is:

[0121]

[0122] Among them, C S It represents the compressive strength of stone powder concrete; Cost represents the cost of each cubic meter of stone powder concrete; CO2 represents the carbon emission coefficient of each cubic meter of stone powder concrete.

[0123] The calculation formula of Cost in the second optimization objective function is:

[0124]

[0125] Among them, S c is the unit price of cement; S s is the unit price of auxiliary cementitious materials; S fa is the unit price of fine aggregate; S ca is the unit price of coarse aggregate; S w is the unit price of water; S sp is the unit price of water reducing agent; S mp is the unit price of stone powder;

[0126] The setting of preset constraints in the process of mix optimization includes key parameter constraints and volume constraints;

[0127] The formula for the key parameter constraint is:

[0128]

[0129]

[0130]

[0131]

[0132] Among them, w represents the water-cement ratio, s represents the sand ratio, r represents the water reducing agent dosage, p represents the stone powder dosage, and wmin and w max Represents the lower and upper limits of the water-binder ratio, s min and max Represents the lower and upper limits of the sand ratio, r min and r max represents the lower and upper limits of water reducing agent dosage, p min and p max Represents the lower and upper limits of stone powder dosage;

[0133] The formula of the volume constraint is:

[0134]

[0135] Among them, m s The amount of auxiliary cementitious material; m w is the amount of water used; m ca is the amount of coarse aggregate; m sp is the dosage of water reducing agent; c is the density of cement; ρ s is the density of the auxiliary cementitious material; w is the density of water; ρ fa is the density of fine aggregate; ρ ca is the density of coarse aggregate; ρ mp is the density of stone powder; ρ sp is the density of the water reducer.

[0136] Preferably, the optimal solution of the optimization objective function must first satisfy the first optimization objective function, and then successively satisfy the second and third optimization objective functions to obtain the optimal solution for replacing cement, specifically:

[0137] The parameters of the mix ratio predicted by the performance prediction model include cement dosage, auxiliary cementitious material dosage, water dosage, fine aggregate dosage, coarse aggregate dosage, stone powder dosage, and water reducing agent dosage, and then classified according to strength grade and stone powder dosage;

[0138] The classified mix ratios are judged, and when it is judged that the predicted strength of the mix ratio differs from the set concrete strength by 5 MPa, the mix ratio is removed, and the remaining mix ratios are optimized and screened;

[0139] When the key parameter constraints and volume constraints are met, at the corresponding strength level, according to the first optimization objective function, it is judged whether a certain mix ratio is the optimal solution, wherein the mix ratio satisfies the maximum value of the first optimization objective function, that is, it satisfies the condition;

[0140] If yes, continue to optimize;

[0141] If not, stop the optimization and reselect other mix ratios for judgment;

[0142] When it is determined that the mix ratio satisfies the first optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the second optimization objective function, wherein the mix ratio satisfies the minimum value of the second optimization objective function, that is, satisfies the condition;

[0143] If so, the mix ratio is the optimal solution;

[0144] If not, reselect other mix ratios for judgment.

[0145] When it is determined that the mix ratio satisfies the second optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the third optimization objective function, wherein the mix ratio satisfies the maximum value of the third optimization objective function, that is, satisfies the condition;

[0146] If so, the mix ratio is the optimal solution;

[0147] If not, reselect other mix ratios for judgment.

[0148] In this embodiment, preferably, it also includes:

[0149] S6: The optimal mix ratio is experimentally verified and adjusted to obtain a final mix ratio.

[0150] The application of the present invention will be described below with a practical example:

[0151] First, the density, cost, and carbon emission factor of the raw materials. As shown in Table 1, the density, cost, and carbon emission factor of each raw material are summarized.

[0152] Table 1 Density, cost and carbon emission factor of each raw material

[0153]

[0154] Next are the constraints: key parameter constraints, as shown in Table 2.

[0155] Table 2 Key parameter constraints

[0156]

[0157] In addition, a volume constraint is defined, limiting the volume of the marble powder concrete to 1m 3 :

[0158]

[0159] In this embodiment, the optimal mix ratio is experimentally verified and adjusted to obtain the actual final mix ratio.

[0160] Based on the established prediction model, the optimization results of high-strength concrete mix ratio using marble powder instead of cement are as follows: Figure 3 As shown in the figure, the red line is the mix ratio with the maximum amount of stone powder under the corresponding substitution type. The maximum mix ratio of marble powder concrete with strength grade C120 is shown in Table 3:

[0161] Table 3 Maximum mix ratio of marble powder high-strength concrete with strength grade C120

[0162] Alternative Type Cement (kg / m3) Marble powder (kg / m3) Silica fume (kg / m3) Water (kg / m3) Water reducing agent (kg / m3) Coarse aggregate (kg / m3) Fine aggregate (kg / m3) Stone powder dosage cement 491.2 122.8 68.2 150.1 6.8 1046.4 622.3 20%

[0163] According to the above optimal mix ratio, actual concrete material preparation was carried out, and the standard value of cube compressive strength after 28 days of standard curing was 122.8 MPa.

[0164] In summary, compared with the prior art, the advantages of the present invention are:

[0165] 1. Usually, the use of waste stone powder requires a large number of tests to design the mix ratio in order to determine the proportion of stone powder to replace fine aggregate or cement. This is greatly affected by the raw materials and requires repeated tests. However, this method uses machine learning and the principle of close stacking to ensure that the strength grade of concrete after stone powder replacement meets the requirements, and can also maximize the amount of stone powder, as well as better environmental and economic benefits.

[0166] 2. The method provided by the present invention overcomes the problem that the traditional mix design method is difficult to meet the needs of concrete materials to achieve multiple performance and multiple goals.

[0167] 3. The method provided by the present invention can achieve a high degree of prediction of the performance of stone powder concrete by establishing a database including raw material parameters and adopting multiple machine learning algorithms such as MLP, BP, SVR, and XGB.

[0168] 4. The method provided by the present invention is applicable to stone powder concrete mix proportions of different strength grades, which is more in line with actual engineering application requirements.

[0169] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder, characterized in that: The following steps are involved: S1: Establish an experimental database of high-strength concrete mix proportion and raw material particle size distribution-density; S2: Establish a calculation model for carbon emissions and costs of calcium carbonate stone powder solid waste transportation; where: The calculation formula for the carbon emission factor of calcium carbonate stone powder solid waste transportation is: C mp =dT Among them, C mp represents the carbon emission factor of calcium carbonate stone powder solid waste in the transportation stage; d represents the average transportation distance of calcium carbonate stone powder solid waste; T represents the carbon emission factor of calcium carbonate stone powder solid waste per unit weight of transportation distance; The calculation formula for the carbon emission cost of calcium carbonate stone powder solid waste transportation is: S mp =dG Among them, S mp represents the unit cost of calcium carbonate stone powder solid waste in the transportation stage; G represents the cost of calcium carbonate stone powder solid waste per unit weight and transportation distance; S3: Based on the established experimental database of high-strength concrete mix ratio and raw material particle size distribution-density, combined with the close packing principle and machine learning algorithm, a performance prediction model for ultra-low carbon high-strength concrete made of calcium carbonate stone powder solid waste was built and trained; S4: inputting the raw material dosage and concrete strength into the performance prediction model, outputting the mix ratios of multiple groups of raw materials and the strength of each mix ratio, and calculating the carbon emission and life cycle assessment comprehensive index ECM of each mix ratio according to the mix ratios of the multiple groups of raw materials and the calculation model; S5: Based on the mix ratio of the multiple groups of raw materials and the corresponding carbon emissions and life cycle assessment comprehensive index ECM, the calcium carbonate stone powder content is set as the first optimization objective function, the carbon emissions are set as the second optimization objective function, the life cycle assessment comprehensive index ECM is corrected and its maximum is used as the third optimization objective function, and the optimal mix ratio is obtained from the mix ratio of the multiple groups of raw materials; wherein the optimization objective function includes three optimization objective functions in a sequential order, the first optimization objective function, the second optimization objective function and the third optimization objective function, wherein the first optimization objective function is constructed based on the maximum stone powder content as the goal, the second optimization objective function is constructed based on the minimum carbon emissions as the goal, and the third optimization objective function is constructed based on the maximum comprehensive evaluation index ECM of life cycle assessment as the goal; The calculation formula of the first optimization objective function is: Among them, m mp1 Indicates the amount of stone powder replacing cement; m mp2 Indicates the amount of stone powder replacing fine aggregate; m c Indicates the amount of cement used; m fa Indicates the amount of fine aggregate; The calculation formula of the second optimization objective function is: Among them, C c is the unit carbon emission factor of cement; m s is the amount of auxiliary cementitious material, C S is the unit carbon emission factor of auxiliary cementitious materials; C fa is the unit carbon emission factor of fine aggregate; m ca is the amount of coarse aggregate, C ca is the unit carbon emission factor of coarse aggregate; m w is the amount of water used, C w is the unit carbon emission factor of water; m sp is the dosage of water reducing agent, C sp is the unit carbon emission factor of water reducer; m mp is the amount of stone powder used; The calculation formula of the third optimization objective function is: Among them, C Sk It represents the compressive strength of stone powder concrete; Cost represents the cost of each cubic meter of stone powder concrete; CO2 represents the carbon emission coefficient of each cubic meter of stone powder concrete; The calculation formula of Cost in the third optimization objective function is: Among them, S c is the unit price of cement; S s is the unit price of auxiliary cementitious materials; S fa is the unit price of fine aggregate; S ca is the unit price of coarse aggregate; S w is the unit price of water; S sp is the unit price of water reducing agent; S mp is the unit price of stone powder; The setting of preset constraints in the process of mix optimization includes key parameter constraints and volume constraints; The formula for the key parameter constraint is: Among them, w represents the water-cement ratio, s represents the sand ratio, r represents the water reducing agent dosage, p represents the stone powder dosage, and w min and w max Represents the lower and upper limits of the water-binder ratio, s min and max Represents the lower and upper limits of the sand ratio, r min and r max represents the lower and upper limits of water reducing agent dosage, p min and p max Represents the lower and upper limits of stone powder dosage; The formula of the volume constraint is: Among them, ρ c is the density of cement; ρ s is the density of the auxiliary cementitious material; w is the density of water; ρ fa is the density of fine aggregate; ρ ca is the density of coarse aggregate; ρ mp is the density of stone powder; ρ sp is the density of the water reducer.

2. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 1, characterized in that: In step S1, the raw materials include cement, auxiliary cementitious materials, fine aggregate, coarse aggregate, water reducing agent and calcium carbonate stone powder; the cement includes one or more of silicate cement, sulphoaluminate cement and aluminate cement; the auxiliary cementitious materials include one or more of silica fume and fly ash; the fine aggregate includes one or more of river sand, mountain sand and machine-made sand, the coarse aggregate includes one or more of granite, basalt and bluestone, the calcium carbonate stone powder includes one or more of marble powder and limestone powder, the water reducing agent includes one or more of polycarboxylic acid-based high-efficiency water reducing agent and naphthalene-based water reducing agent, the particle size distribution of the raw material particles is the cumulative particle size throughput, and the raw material density is the apparent density.

3. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 1, characterized in that: In step S3, the training method of the performance prediction model is as follows: Collect training data samples, wherein the training data samples include a close packing degree k, a water-cement ratio b, a compressive strength m, and a colloid strength richness coefficient nf under different mix ratios; Clean and organize the training data samples, and handle missing values ​​and outliers; Input the training data samples into the machine learning algorithm to establish the relationship between m and k, b, and nf, and obtain the performance prediction model; the machine learning algorithm includes MLP, BP, SVR, and XGB; the collected high-strength concrete mix ratio and performance data are divided into a training data set and a verification data set in a ratio of 8:2, and the verification and training are synchronized for model optimization and debugging parameters. After the model is trained, it is tested; Use cross-validation to evaluate model performance and ensure that the model has good predictive and generalization capabilities; Through parameter adjustment and feature optimization, the accuracy and stability of the model can be further improved.

4. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 3, characterized in that: In step S3, the raw material close packing degree index k is constructed according to the residual sum of squares RSS of the particle size distribution, where: Where D is the particle size, P(D) is the proportion of particles smaller than D; D max is the maximum particle size; D min is the minimum particle size; q is the distribution modulus, is the maximum particle size under the distribution modulus q, is the minimum particle size under the distribution modulus q; n is the number of particles, e i is the residual of the ith particle, and The particle size classification of the mixture and the target curve respectively; The cement dosage, auxiliary cementitious material dosage, fine aggregate dosage, coarse aggregate dosage, water dosage, water reducing agent dosage, stone powder dosage, and high-strength concrete compressive strength are used as the input of the test database. The data are normalized to eliminate the impact of the dimensions between the data to facilitate the convergence of the model. After the output, they are denormalized to obtain the actual mix ratio.

5. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 3, characterized in that: The S5 includes: S51: According to the target demand of maximizing the utilization of calcium carbonate stone powder, an optimization objective function of stone powder dosage is established, and on this basis, an optimization objective function of carbon emission and life cycle assessment comprehensive index ECM is constructed, and constraint conditions are set; S52: Use the performance prediction model to predict the strength of calcium carbonate stone powder high-strength concrete, and set the optimal hyperparameters according to different algorithms to ensure a low error; S53: Select the mix ratio that meets the strength grade requirements and has the largest amount of stone powder from the optimization results, and on this basis select the mix ratio with the smallest carbon emissions and the largest life cycle assessment comprehensive index ECM to obtain the optimal mix ratio.

6. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 5, characterized in that: In S53, the stone powder dosage includes the dosage for satisfying close packing and the dosage for replacing part of the fine aggregate while maintaining close packing and concrete strength.

7. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 1, characterized in that: The optimal solution of the optimization objective function must first satisfy the first optimization objective function, and then successively satisfy the second and third optimization objective functions to obtain the optimal solution for cement replacement, specifically: The parameters of the mix ratio predicted by the performance prediction model include cement dosage, auxiliary cementitious material dosage, water dosage, fine aggregate dosage, coarse aggregate dosage, stone powder dosage, and water reducing agent dosage, and then classified according to strength grade and stone powder dosage; The classified mix ratios are judged, and when it is judged that the predicted strength of the mix ratio differs from the set concrete strength by 5 MPa, the mix ratio is removed, and the remaining mix ratios are optimized and screened; When the key parameter constraints and volume constraints are met, at the corresponding strength level, according to the first optimization objective function, it is judged whether a certain mix ratio is the optimal solution, wherein the mix ratio satisfies the maximum value of the first optimization objective function, that is, it satisfies the condition; If yes, continue to optimize; If not, stop the optimization and reselect other mix ratios for judgment; When it is determined that the mix ratio satisfies the first optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the second optimization objective function, wherein the mix ratio satisfies the minimum value of the second optimization objective function, that is, satisfies the condition; If so, the mix ratio is the optimal solution; If not, reselect other mix ratios for judgment; When it is determined that the mix ratio satisfies the second optimization objective function, continue to determine whether the mix ratio is an optimal solution according to the third optimization objective function, wherein the mix ratio satisfies the maximum value of the third optimization objective function, that is, satisfies the condition; If so, the mix ratio is the optimal solution; If not, reselect other mix ratios for judgment.

8. The method for optimizing the mix ratio of ultra-low carbon high-strength concrete by maximizing the utilization of stone powder as claimed in claim 1, characterized in that: Also includes: S6: The optimal mix ratio is experimentally verified and adjusted to obtain a final mix ratio.

Citation Information

Patent Citations

  • Construction method of shield tunnel segment concrete shrinkage prediction model

    CN113707241A

  • Concrete mix proportion multi-objective optimization system design method

    CN117219211A

  • Concrete full-life-cycle green quality control method based on big data of Internet of Things

    CN117764443A

  • High-performance concrete mix proportion design and performance prediction method based on deep learning

    CN117912594A

  • Sulfate corrosion-resistant concrete and method thereof for optimizing proportion and application

    US20230129027A1