Concrete mix proportion design method and device based on close packing theory

Abstract

The invention relates to the technical field of concrete mix proportion, in particular to a concrete mix proportion design method and device based on the close packing theory, and the method comprises the following steps: calculating the concrete mix proportion through a machine learning algorithm, and determining the total amount of coarse aggregate according to the concrete mix proportion; and according to the total use amount of the coarse aggregate and the particle size classification of the coarse aggregate in the concrete components, determining the optimal use amount of the coarse aggregate in the particle size classification of each coarse aggregate through a close packing equation. According to the method, on-site raw material grain composition distribution is fully considered, the close packing principle is introduced to determine the optimal use amount of coarse aggregate of each level, and the problem of landing application of the concrete multi-target mix proportion optimization design method based on machine learning in actual engineering is solved.

Description

Technical Field

[0001] The invention relates to the technical field of concrete mix ratio, in particular to a concrete mix ratio design method and device based on close packing theory. Background Art

[0002] At present, multi-objective concrete mix optimization design based on machine learning can realize the mix optimization design that comprehensively considers concrete performance, economy, and environmental protection, and solves the problems of traditional concrete mix design technology, such as high dependence on experience, low efficiency, high cost, and difficulty in ensuring the optimal mix ratio. However, in actual engineering applications, there are the following two shortcomings:

[0003] (1) In the process of establishing the concrete performance (mechanical properties, workability and durability) prediction model, the physical and chemical properties of concrete raw materials were not included in the input variables of the model. The established performance prediction model lacks consideration of the raw material properties (such as solid particle size distribution, coarse aggregate sphericity, roundness, etc.). When the properties of the actual raw materials on site are significantly different from those used in model training, it is difficult to directly use the existing multi-objective mix optimization design method to guide the actual mix design work on site.

[0004] (2) In the multi-objective mix optimization design of concrete, we usually use the total amount of various raw materials as the result output. However, in the actual construction process, coarse aggregate is often stored separately in particles of several different particle sizes. In order to achieve the particle gradation optimization of coarse aggregate, a common practice is to mix these aggregates of different particle sizes in a specific proportion to form a continuous gradation. However, the existing multi-objective mix optimization design often only provides the total amount of coarse aggregate, but does not provide specific guidance on the proportion of each grade of aggregate. Therefore, in order to obtain the optimal combination of coarse aggregate particle gradation, we need to further refine the design and provide a more specific mix design method to ensure the rational matching and use of coarse aggregate.

[0005] For example, the Chinese invention patent with announcement number CN117219211A discloses a method for designing a multi-objective optimization system for concrete mix ratio, including: S1: establishing a concrete database; S2: establishing a machine learning prediction model for concrete strength CCS, slump S and chloride ion diffusion coefficient CDC; S3: establishing a multi-objective joint optimization mathematical model for concrete, using the AEPSO algorithm to solve the mathematical model, obtaining the Pareto frontier, and combining the ideal point method to obtain the theoretical optimal mix ratio; S4: experimentally verifying and adjusting the theoretical optimal mix ratio to obtain the actual optimal mix ratio. Summary of the invention

[0006] The method of the present invention determines the optimal amount of coarse aggregates of each level by introducing the close packing principle, proposes a concrete mix design method and device based on the close packing theory, and provides a concrete multi-objective mix that combines concrete performance, economy and environmental protection.

[0007] In order to achieve the above objectives, the following technical solutions are proposed:

[0008] The concrete mix design method based on the close packing theory includes the following steps:

[0009] S1: Calculate the concrete mix ratio through machine learning algorithm and determine the total amount of coarse aggregate according to the concrete mix ratio;

[0010] S2: According to the total amount of the coarse aggregate and the particle size classification of the coarse aggregate in the concrete component, the optimal amount of the coarse aggregate in each particle size classification of the coarse aggregate is determined by a close packing equation.

[0011] Preferably, the method for determining the particle size classification of the coarse aggregate comprises the following steps:

[0012] Determine the particle size distribution of solid particles in concrete components at the construction site;

[0013] The particle sizes of the coarse aggregates in the solid particles are classified according to the particle size distribution of the solid particles.

[0014] Preferably, the concrete ingredients include: cement, fly ash, mineral powder, water, water reducing agent, coarse aggregate and fine aggregate.

[0015] Preferably, the particle size distribution is measured by a laser particle size analyzer or a square hole sieve, and the solid particles include powder particles, fine aggregate particles and coarse aggregate particles.

[0016] Preferably, the particle size of the coarse aggregate in the solid particles is classified according to the particle size distribution of the solid particles by a particle size classification method, and the particles are classified according to the particle size based on the physical properties of the particles.

[0017] Preferably, the particle size of the coarse aggregate in the solid particles is classified into three categories in order from small to large, and the range of each category is 4.75mm-9.5mm, 9.5mm-16mm, and 16mm-25mm respectively.

[0018] Preferably, determining the total amount of coarse aggregate according to the concrete mix ratio comprises the following steps:

[0019] S11: Establishing a concrete database: collecting data, including mix ratio parameters, performance parameters and target indicators of concrete raw materials, pre-processing the data through data cleaning and normalization, removing abnormal data, and eliminating dimensional effects between the data;

[0020] S12: Establish a machine learning prediction model for concrete strength CCS, slump S and chloride ion diffusion coefficient CDC, and input the mass of each component of concrete;

[0021] S13: Establish a concrete multi-objective joint optimization mathematical model, use the AEPSO algorithm to solve the mathematical model, obtain the Pareto frontier, and combine the ideal point method to obtain the theoretical optimal mix ratio;

[0022] S14: The theoretical optimal mix ratio is experimentally verified and adjusted to obtain the actual optimal mix ratio.

[0023] Preferably, the close packing equation is:

[0024]

[0025] Among them, D L is the maximum aggregate size, D S is the minimum aggregate particle size, D is the aggregate particle size, CPFT is the cumulative volume percentage of particles smaller than the particle size D, and q is the particle size distribution index.

[0026] Preferably, the close packing equation calculates the corresponding CPFT value to obtain the ratio of various coarse aggregates, so that the particles reach the optimal close packing state, and the corresponding ratio of various coarse aggregates is the optimal dosage.

[0027] Based on the same concept, a concrete mix design device based on the close packing theory is also proposed, which is characterized in that it includes a processor and a concrete mix equipment; the processor and the concrete mix equipment obtain the concrete mix ratio according to the concrete mix design method based on the close packing theory described in any one of the above claims.

[0028] Compared with the prior art, the invention has the following beneficial effects: the method of the invention combines the actual particle size distribution of raw materials on site, and further refines the optimal proportion design of each grade of coarse aggregate based on the close packing theory to ensure the reasonable matching and use of coarse aggregate. This improves the density of concrete particles under the existing construction conditions, thereby obtaining concrete performance with better performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of the method for concrete mix design based on close packing theory in Example 1;

[0030] Figure 2 is the particle size distribution diagram of the raw materials in Example 1;

[0031] Figure 3 It is a comparison diagram of the close packing of concrete particles in Example 1.

[0032] Markings in the figure:

[0033] (a) C30 third-level crushed stone is selected according to experience (1:6:3),

[0034] (b) C30 combined with improved A&A model for accurate calculation (1.4:4.1:4.5),

[0035] (c) C40 third-level crushed stone is selected according to experience (1:6:3),

[0036] (d) C40 combined with improved A&A model for accurate calculation (1.4:4.2:4.4),

[0037] (e) C50 third-level crushed stone is selected according to experience (1:6:3),

[0038] (f) C50 is accurately calculated by combining with the improved A&A model (1.5:4.2:4.3). DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0040] Example 1

[0041] Concrete mix design method based on close packing theory, the flow chart is as follows Figure 1 As shown, the following steps are included:

[0042] S1: Calculate the concrete mix ratio through machine learning algorithm and determine the total amount of coarse aggregate according to the concrete mix ratio;

[0043] S2: According to the total amount of the coarse aggregate and the particle size classification of the coarse aggregate in the concrete component, the optimal amount of the coarse aggregate in each particle size classification of the coarse aggregate is determined by a close packing equation.

[0044] S1 is a specific embodiment, using a machine learning method:

[0045] 1. Establish a concrete database: collect data, including mix parameters, performance parameters and target indicators of concrete raw materials, pre-process the data through data cleaning and normalization, remove abnormal data, and eliminate the dimensional influence between the data;

[0046] 2. Establish a machine learning prediction model for concrete strength CCS, slump S and chloride ion diffusion coefficient CDC. Input the mass of each component of concrete, including cement dosage x1, fly ash dosage x2, mineral powder dosage x3, crushed stone dosage x4, sand dosage x5, water reducer dosage x6, water dosage x7, and output chloride ion diffusion coefficient, concrete compressive strength and slump;

[0047] 3. Establish a mathematical model for multi-objective joint optimization of concrete, use the AEPSO algorithm to solve the mathematical model, obtain the Pareto frontier, and combine the ideal point method to obtain the theoretical optimal mix ratio;

[0048] 4. The theoretical optimal mix ratio is experimentally verified and adjusted to obtain the actual optimal mix ratio, and the optimal mix ratio is added to the concrete database.

[0049] As a specific embodiment, in step S1, a machine learning method is introduced to output a mix ratio that comprehensively considers concrete performance, economy, and environmental protection, including cement dosage, fly ash dosage, mineral powder dosage, water dosage, water reducer dosage, coarse aggregate dosage, and fine aggregate dosage. In step S2, a laser particle size analyzer is used to measure the particle size distribution of materials actually used at the construction site, including: the particle size distribution of solid particles such as cement, fly ash, mineral powder, coarse aggregate, and fine aggregate.

[0050] S3 is described as a specific embodiment. Combined with the storage conditions of coarse aggregate on site, the three-grade storage is taken as an example. The particle size grades are 4.75mm-9.5mm, 9.5mm-16mm, and 16mm-25mm respectively. The close packing equation (improved Andreasen and Andersen model, i.e. improved A&A) is introduced. As shown in formula (1), the calculation result of step S1 is used, the amount of solid particle material of each component is fixed, and the optimal proportion of the three grades of coarse aggregate is calculated by the close packing equation.

[0051]

[0052] Among them, D L is the maximum aggregate size, D S is the minimum aggregate particle size, D is the aggregate particle size, CPFT is the cumulative volume percentage smaller than particle size D (sieve), and q is the particle size distribution index, which is usually between 0.2 and 0.4.

[0053] The larger the q value is, the more coarse particles there are and the lower the workability of the mixture is.

[0054] The distribution coefficient q is usually an empirical coefficient, obtained according to previous studies:

[0055] For self-compacting concrete, q = 0.21-0.29 is recommended;

[0056] For conventional pumping and vibration compacted concrete, q = 0.3-0.34 is recommended;

[0057] For dry hard concrete, q=0.35-0.4 is recommended.

[0058] The multi-objective concrete mix ratio based on the close packing theory and taking into account the concrete performance, economy and environmental protection is obtained, that is, the raw material dosage of three grades of coarse aggregate. The corresponding CPFT value is calculated through the close packing equation to obtain the proportion of each type of coarse aggregate, so that the particles reach the optimal close packing state. At this time, the corresponding proportion of each type of coarse aggregate is the optimal dosage.

[0059] The reliability of the mix proportion designed by this method is verified through experiments.

[0060] The experimental steps of the present invention are as follows:

[0061] Step 1: The present invention adopts the machine learning method described in step S1 to calculate the mix ratios of C30, C40, and C50 that meet the concrete strength requirements and have the lowest cost and carbon emissions, and obtain the total amount of coarse aggregate in the total amount of concrete, as shown in Table 1.

[0062] Table 1 Concrete mix ratio

[0063]

[0064] Step 2: Conduct particle size tests on raw materials at the construction site, including cement (C), fly ash (FA), mineral powder (SB), three grades of coarse aggregate 4.75mm~9.5mm (stone1), 9.5mm~16mm (stone2), 16~25mm (stone3) and fine aggregate (sand). The results are as follows Figure 2 shown.

[0065] Step 3: According to the actual particle size distribution of raw materials, combined with the improved A&A model, the optimal proportion of the three grades of coarse aggregate in the current mix ratio is accurately calculated, so as to obtain the optimal amount of coarse aggregate in each of the three grades.

[0066] The ratios of C30, C40 and C50 crushed stones were calculated using empirical values ​​(1:6:3) and the improved A&A model. The particle size distribution index q was taken as 0.32, 0.3 and 0.28 respectively according to previous research and concrete fluidity requirements. The results are shown in Table 2. The corresponding concrete particles are tightly packed as shown in Figure 3 shown.

[0067] Table 2 Values ​​of three-level crushed stone

[0068]

[0069] (a) The graph shows the particle size distribution of three grades of crushed stone in C30 concrete according to the empirical value (1:6:3). The distribution range of particle size is wide and there may be some unevenness.

[0070] (b) shows the particle size distribution of C30 concrete with the improved A&A model (1.4:4.1:4.5). Compared with (a), the particle size distribution is more uniform and mainly concentrated in the smaller particle size range, which may help improve the performance of concrete.

[0071] Figures (c) and (d) show the particle size distribution of the three grades of crushed stone in C40 concrete according to the empirical value and the improved A&A model. Similar to the case of C30 concrete, C40 concrete also shows a more uniform particle size distribution under the improved A&A model.

[0072] Figures (e) and (f) show the particle size distribution of the three grades of crushed stone in C50 concrete according to the empirical value and the improved A&A model. When the improved A&A model is used for accurate calculation, C50 concrete may also obtain a more uniform particle size distribution.

[0073] By comparing the distribution of particle sizes under different conditions, the present invention can obtain a more uniform particle size distribution by accurately calculating the optimal proportion of the third grade coarse aggregate in the current mix ratio, thereby improving the performance of concrete.

[0074] The following is to verify the experimental results:

[0075] Step 4: Prepare 6 groups of concrete test blocks according to the test requirements of "Ordinary Concrete Mix Design Code" JGJ55-2011.

[0076] Step 5: Test the 28d strength of the concrete specimens. The results are shown in Table 3 below.

[0077] Table 3 Concrete 28d strength value

[0078] Serial number serial number Strength level Three levels of crushed stone ratio (large: medium: small) Strength value (MPa) 1 C30 C30 1:6:3 46.8 2 C40 C40 1:6:3 53.9 3 C50 C50 1:6:3 61.5 4 C30 - Improved A&A C30 1.4:4.1:4.5(q＝0.32) 49.8 5 C40 - Improved A&A C40 1.4:4.2:4.4(q＝0.3) 56.9 6 C50 - Improved A&A C50 1.4:4.2:4.4(q＝0.28) 66.2

[0079] Step 6: Result analysis shows that the 28d compressive strength value of each grade of coarse aggregate using the empirical value (1:6:3) can meet the concrete strength design requirements, and the 28d strength values ​​of the concrete mix design method integrating artificial intelligence technology and close packing theory proposed in the present invention are greater than those of the control group, and the strength growth rate is greater than 5%. This shows the feasibility and superiority of the concrete mix design method based on the close packing theory proposed in the present invention.

[0080] The present invention overcomes the problem that the current multi-objective concrete mix optimization design method based on machine learning cannot determine the proportional dosage of coarse aggregates at each level, and fails to consider the influence of raw material properties (such as solid particle size distribution, coarse aggregate sphericity, roundness, etc.) on the concrete performance prediction model, making it difficult to guide the actual mix design on site.

[0081] At the same time, the present invention provides a feasible solution for the optimization design of concrete mix ratio, and provides a multi-objective concrete mix ratio that comprehensively considers the performance, economy and environmental protection of concrete. On this basis, combined with the actual particle size distribution of raw materials on site, the optimal proportion design of each grade of coarse aggregate is further refined based on the close packing theory to ensure the reasonable matching and use of coarse aggregate. Then, the density of concrete particles under existing construction conditions is improved, so as to obtain concrete performance with better performance. Solve the problem of the implementation of the multi-objective concrete mix optimization design method based on machine learning in actual engineering.

[0082] Artificial intelligence has played a powerful role in predicting concrete performance and optimizing mix proportions, but how to implement the calculation results in the field of diversified and differentiated raw materials is one of the key issues to be solved in the new generation of concrete mix design systems. On the basis of meeting performance requirements, lowest cost and lowest carbon emissions, the present invention takes into account the actual particle size distribution of raw materials on site, and improves the density of concrete particles under existing construction conditions. It combines the close packing theory with artificial intelligence multi-objective mix design, solves the problems of low efficiency, high cost and not necessarily the optimal solution in the traditional mix design system, and also solves the problem that artificial intelligence calculation results are difficult to directly guide the actual mix design on site. The present invention provides technical support for the efficiency, accuracy and feasibility of the new generation of concrete mix design systems.

[0083] Finally, it should be noted that the embodiments described in detail above are only the best practices of the invention and cannot be used to limit the scope of rights of the invention. Equivalent replacement of the technical solutions recorded in the aforementioned embodiments does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the invention, and they should all be included in the scope of the claims and description of the invention.

Claims

1. A concrete mix design method based on close packing theory, characterized in that: The following steps are involved: S1: Calculate the concrete mix ratio through machine learning algorithm and determine the total amount of coarse aggregate according to the concrete mix ratio; S2: According to the total amount of the coarse aggregate and the particle size classification of the coarse aggregate in the concrete component, the optimal amount of the coarse aggregate in each particle size classification of the coarse aggregate is determined by a close packing equation.

2. The concrete mix design method based on the close packing theory as claimed in claim 1, characterized in that: The method for determining the particle size classification of the coarse aggregate comprises the following steps: Determine the particle size distribution of solid particles in concrete components at the construction site; The particle sizes of the coarse aggregates in the solid particles are classified according to the particle size distribution of the solid particles.

3. The concrete mix design method based on the close packing theory as claimed in claim 2, characterized in that: The concrete ingredients include cement, fly ash, mineral powder, water, water reducing agent, coarse aggregate and fine aggregate.

4. The concrete mix design method based on the close packing theory as claimed in claim 2, characterized in that: The particle size distribution is measured by a laser particle size analyzer or a square hole sieve, and the solid particles include powder particles, fine aggregate particles and coarse aggregate particles.

5. The concrete mix design method based on the close packing theory as claimed in claim 2, characterized in that: The particle size of the coarse aggregate in the solid particles is classified according to the particle size distribution of the solid particles. A particle size classification method is adopted to classify the particles according to the particle size based on the physical properties of the particles.

6. The method for concrete mix design based on close packing theory as claimed in claim 5, characterized in that: The particle size classification of the coarse aggregate in the solid particles is divided into three categories in order from small to large, and the range of each category is 4.75mm-9.5mm, 9.5mm-16mm, and 16mm-25mm respectively.

7. The concrete mix design method based on the close packing theory as claimed in claim 1, characterized in that: Determining the total amount of coarse aggregate according to the concrete mix ratio includes the following steps: S11: Establishing a concrete database: collecting data, including mix ratio parameters, performance parameters and target indicators of concrete raw materials, pre-processing the data through data cleaning and normalization, removing abnormal data, and eliminating dimensional effects between the data; S12: Establish a machine learning prediction model for concrete strength CCS, slump S and chloride ion diffusion coefficient CDC, and input the mass of each component of concrete; S13: Establish a concrete multi-objective joint optimization mathematical model, use the AEPSO algorithm to solve the mathematical model, obtain the Pareto frontier, and combine the ideal point method to obtain the theoretical optimal mix ratio; S14: The theoretical optimal mix ratio is experimentally verified and adjusted to obtain the actual optimal mix ratio.

8. The method for concrete mix design based on close packing theory as claimed in claim 1, characterized in that: The close packing equation is: Among them, D L is the maximum aggregate size, D S is the minimum aggregate particle size, D is the aggregate particle size, CPFT is the cumulative volume percentage of particles smaller than the particle size D, and q is the particle size distribution index.

9. The concrete mix design method based on close packing theory as claimed in claim 1, characterized in that: The close packing equation calculates the corresponding CPFT value to obtain the ratio of various coarse aggregates, so that the particles reach the optimal close packing state, and the corresponding ratio of various coarse aggregates is the optimal dosage.

10. A concrete mix design device based on close packing theory, characterized in that: It comprises a processor and a concrete proportioning device; the processor and the concrete proportioning device obtain the concrete proportion according to the concrete proportioning design method based on the close packing theory described in any one of claims 1-9.

Images