A method for physicochemical coupling strengthening and multi-objective optimization of slag recycled aggregate

Abstract

The application discloses a kind of slag regeneration aggregate physical and chemical coupling reinforcement and optimization method, first to slag is pretreated and magnetic separation impurity;Subsequently, mechanical activation is carried out using high-pressure roller mill combined with microwave irradiation, optimize particle size distribution and stimulate surface activity;Again, the gradient coating process of silane coupling agent, water-based epoxy resin and nano SiO2 Composite solution reduces water absorption;Finally, the mapping relationship between mechanical-chemical parameters and performance indicators is established based on experiment and machine learning model, combined with genetic algorithm for multi-objective optimization, output Pareto optimal matching scheme.The application improves the performance of slag regeneration aggregate through the synergistic effect of mechanical activation and chemical coating, and predicts the crushing value and water absorption of the regeneration aggregate through the machine learning model, optimizes the matching and process parameters, improves the quality and production efficiency of the regeneration aggregate, realizes the high-value utilization of construction slag, and provides an efficient technical path for the high-value utilization of construction solid waste.

Description

Technical Field

[0001] This invention belongs to the field of building material recycling technology and relates to a method for preparing recycled aggregates. Specifically, it relates to a method for enhancing the performance of recycled aggregates from slag and soil through mechanical activation and chemical coating coupling, and optimizing the proportioning parameters based on machine learning. Background Technology

[0002] With the rapid advancement of urbanization, the annual production of construction waste has exceeded 3 billion tons, making its resource utilization crucial for addressing environmental pollution and resource shortages. However, untreated recycled aggregates from construction waste suffer from defects such as high internal porosity, low surface activity, and weak interfacial bonding with the cement matrix, making it difficult to meet the mechanical strength and durability requirements of high-performance concrete.

[0003] Existing technologies mainly enhance recycled aggregates through mechanical or chemical means, but there are still significant problems such as limited mechanical activation effect, poor uniformity and durability of chemical modification, and low efficiency of proportion optimization methods. Summary of the Invention

[0004] This invention provides a method for physical-chemical coupling enhancement and multi-objective optimization of recycled construction waste aggregate. This method improves the performance of recycled construction waste aggregate through the synergistic effect of mechanical activation and chemical coating, and uses machine learning models to predict the crushing value and water absorption rate of recycled aggregate, optimize the mix ratio and process parameters, improve the quality and production efficiency of recycled aggregate, and realize the high-value utilization of construction waste.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A method for physical-chemical coupling enhancement and multi-objective optimization of recycled waste aggregate includes the following steps:

[0007] Step 1, Pre-treatment of excavated soil:

[0008] Step 1.1: Screen the original slag to remove oversized lumps and impurities;

[0009] Step 1.2: After screening, the slag particles are magnetically separated to remove metal impurities;

[0010] Step 1.3: Place the pretreated slag in a constant temperature drying oven and dry it to constant weight;

[0011] Step 2, Mechanical activation:

[0012] Step 2.1: Crush the slag into continuously graded particles of 0.15~4.75mm;

[0013] Step 2.2: Perform microwave irradiation treatment to expand the microcracks inside the particles by utilizing the thermal stress effect, stimulate the surface hydroxyl active sites, and increase the specific surface area.

[0014] Step 3, Chemical Coating:

[0015] Step 3.1: Use a high-pressure airless sprayer to precisely coat the composite modifier. The spraying process is carried out in two stages. After the first spray, let it stand for 10-15 minutes, and then perform a second spray to ensure that the coating thickness reaches 50-200μm. The composite modifier is prepared from silane coupling agent, nano SiO2 sol, waterborne epoxy resin and deionized water. The amount of silane coupling agent added is 2-5% of the total mass of the composite modifier, the amount of nano SiO2 sol added is 1-3% of the total mass of the composite modifier, and the amount of waterborne epoxy resin added is 3-8% of the total mass of the composite modifier.

[0016] Step 3.2: The coated aggregate is cured by gradient temperature to optimize the film performance: First, it is treated in a hot air circulating drying oven at 80~100℃ for 20~60 minutes, then heated to 120~150℃ and kept at that temperature for 0.5~1.5 hours, and finally cooled naturally to room temperature;

[0017] Step 4, Optimize the proportions:

[0018] Step 4.1: Construct a five-level experimental matrix with four factors (mechanical pressure, irradiation time, composite modifier ratio, and coating thickness) based on the central composite design (CCD) to test the crushing value and water absorption rate of each sample;

[0019] Step 4.2: Construct a BP neural network model. The input layer consists of four process parameters: mechanical pressure, irradiation time, composite modifier ratio, and coating film thickness. The output layer consists of crushing value and water absorption rate. A random forest algorithm is used for comparative analysis, and the robustness of the model is verified by ten-fold cross-validation.

[0020] Step 4.3: With the goal of minimizing crushing value and water absorption rate, use the NSGA-II algorithm to solve for the Pareto optimal solution set, and combine it with the engineering constraints (crushing value ≤ 12%, water absorption rate ≤ 6%) to select the optimal parameter combination.

[0021] This invention proposes a synergistic enhancement technology for the treatment of recycled slag and soil aggregates through mechanical activation and chemical coating, and establishes a data-driven multi-objective optimization method. Compared with existing technologies, it has the following advantages:

[0022] 1. The use of engineering waste to prepare recycled aggregates improves resource utilization. At the same time, the use of selective heating and short-time treatment with microwave irradiation reduces energy consumption and CO2 emissions, which meets the needs of green building materials development.

[0023] 2. By adopting a coupled approach of mechanical activation (high-pressure crushing + microwave irradiation) and chemical coating (vacuum impregnation + spray drying), the problems of weak interfacial bonding and high energy consumption in traditional step-by-step processes are solved.

[0024] 3. Based on machine learning (BP neural network, random forest) and NSGA-II algorithm, multi-objective optimization has a short development cycle, high parameter optimization accuracy, supports dynamic formula adjustment, and is suitable for real-time quality control in large-scale production.

[0025] 4. Breaking through the limitations of single strengthening methods, it significantly improves the performance and durability of aggregates, providing an efficient technical path for the high-value utilization of construction solid waste. Attached Figure Description

[0026] Figure 1 This is a flowchart of a method for strengthening and optimizing the physical and chemical coupling of recycled aggregates from slag and soil. Detailed Implementation

[0027] The technical solution of the present invention will be further described below with reference to the accompanying drawings, but it is not limited thereto. Any modifications or equivalent substitutions to the technical solution of the present invention that do not depart from the spirit and scope of the technical solution of the present invention should be covered within the protection scope of the present invention.

[0028] This invention provides a method for the physical-chemical coupling strengthening and optimization of recycled slag aggregate. First, the slag is pretreated and magnetically separated to remove impurities. Then, mechanical activation is performed using a high-pressure roller mill combined with microwave irradiation to optimize particle size distribution and stimulate surface activity. Next, a gradient coating process using a silane coupling agent, water-based epoxy resin, and nano-SiO2 composite solution reduces water absorption. Finally, a mapping relationship between mechanical-chemical parameters and performance indicators is established based on experiments and machine learning models. Multi-objective optimization is then performed using a genetic algorithm to output the Pareto optimal ratio scheme, such as... Figure 1 As shown, the specific steps include the following:

[0029] Step 1, Pre-treatment of excavated soil:

[0030] Step 1.1: Use a vibrating screen (screen mesh size 30mm) to classify the raw slag and remove oversized lumps (>30mm) and debris.

[0031] Step 1.2: After screening, the slag particles are passed through a permanent magnet drum magnetic separator (magnetic field strength 0.8~1.2T) to remove metal impurities and ensure the safety of subsequent process equipment.

[0032] Step 1.3: Place the pretreated slag in a constant temperature drying oven (105±5℃) and dry it to constant weight, controlling the moisture content to ≤1%.

[0033] Step 2, Mechanical activation:

[0034] Step 2.1: Use a double-roll high-pressure crusher (roller spacing adjustable from 0.1 to 5 mm) to crush the slag and soil in multiple stages under a pressure of 5 to 15 MPa, forming continuously graded particles of 0.15 to 4.75 mm. After crushing, the particle distribution is monitored in real time using a laser particle size analyzer to ensure that the gradation meets the "Standard for Recycled Aggregates for Construction" (GB / T 25177).

[0035] Step 2.2: In a closed microwave reaction chamber, the crushed aggregate is evenly spread and irradiated with 2.45GHz microwave (800~1200W) for 3~8 minutes to promote selective heat absorption of internal moisture and defect areas of the particles, generate microcrack network and increase surface activity, and stimulate surface hydroxyl active sites to provide a highly reactive interface for subsequent chemical coating.

[0036] Step 3, Chemical Coating:

[0037] Step 3.1: The composite modifier is prepared by mixing silane coupling agent, nano-SiO2 sol, waterborne epoxy resin, and deionized water in a specific ratio. Specifically, the amount of silane coupling agent (KH550 type) added is 2-5% of the total mass of the composite modifier, used to enhance the interfacial bonding between the aggregate surface and the organic phase; the amount of nano-SiO2 sol (particle size 10-20nm, solid content 30%) added is 1-3% of the total mass of the composite modifier, whose nanoparticles can fill surface micropores and improve the hardness of the coating layer; the amount of waterborne epoxy resin (epoxy equivalent 450-500) added is 3-8% of the total mass of the composite modifier, serving as a film-forming matrix to provide a continuous and dense coating structure; the remainder is supplemented by deionized water to adjust the viscosity. During preparation, the silane coupling agent and nano-SiO2 sol are premixed and treated with an ultrasonic disperser for 20 minutes to ensure uniform dispersion of the nanoparticles. Then, waterborne epoxy resin is added, and the mixture is continuously mixed for 30 minutes under mechanical stirring at 800 rpm to form a stable composite modifier suspension. After resin mixing is complete, deionized water is gradually added in three portions, each 5 minutes apart, with the stirring speed reduced to 500 rpm to prevent bubble formation. The amount of water added is dynamically adjusted according to the target viscosity (monitored using an NDJ-5S viscometer, with the final viscosity controlled between 200 and 400 mPa·s). After mixing, the mixture is allowed to stand for 30 minutes to allow water to fully penetrate between the modifier molecular chains, forming a stable quaternary composite system of "silane-nanoparticle-epoxy resin-water".

[0038] Step 3.2: Use a high-pressure airless sprayer to precisely coat the composite modifier. The operating pressure of the equipment should be controlled at 20~25MPa, and the nozzle diameter should be 0.3~0.5mm to ensure uniform atomized particle size distribution. The spraying process is carried out in two stages: after the first spray, allow it to stand for 10~15 minutes to allow the modifier to initially penetrate into the microcracks on the surface of the aggregate; then perform a second spray to ensure that the coating thickness reaches 50~200μm.

[0039] Step 3.3: The coated aggregate needs to be cured under gradient temperature to optimize the film performance. First, it is treated in an 80℃ hot air circulating drying oven for 30 minutes to promote the evaporation of moisture in the modifier and complete the initial condensation, forming a preliminary cross-linked network; then the temperature is raised to 120℃ and held for 1 hour. During this stage, the hydrolysis and condensation reaction of the silane coupling agent and the curing reaction of the epoxy resin proceed synergistically to form a three-dimensional interpenetrating network structure, which significantly improves the heat resistance and mechanical strength of the coating layer; finally, it is naturally cooled to room temperature to avoid thermal stress cracking caused by sudden cooling.

[0040] Step 4, Optimize the proportions:

[0041] Step 4.1: Construct a five-level experimental matrix with four factors (mechanical pressure, irradiation time, composite modifier ratio, and coating thickness) based on a central composite design (CCD), with a total of 25 samples. Test the crushing value and 24-hour water absorption rate of each sample.

[0042] Step 4.2: Construct a BP neural network model based on experimental data. The input layer includes four process parameters: mechanical pressure, irradiation time, composite modifier ratio, and coating film thickness. The hidden layer has six neurons and uses the ReLU activation function to capture nonlinear relationships. The output layer corresponds to two performance indicators: crushing value and 24-hour water absorption rate. The model is trained using the Adam optimizer with a learning rate of 0.001, a batch size of 16, and a training set to test set ratio of 8:2. To verify the model's generalization ability, a random forest algorithm (100 trees, maximum depth 5, and mean squared error feature selection) is used for comparative analysis. The robustness of the model is verified by ten-fold cross-validation.

[0043] Step 4.3: With the goal of minimizing crushing value and water absorption rate, the NSGA-II algorithm (population size = 50, crossover probability = 0.8, mutation probability = 0.1) is used to solve for the Pareto optimal solution set. Candidate solutions that meet the engineering constraints (crushing value ≤ 12%, water absorption rate ≤ 6%) are screened through non-dominated sorting and crowding calculation, and the optimal parameter combination is finally determined.

Claims

1. A method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates, characterized in that... The method includes the following steps: Step 1, Pre-treatment of excavated soil: Step 1.1: Screen the original slag to remove oversized lumps and impurities; Step 1.2: After screening, the slag particles are magnetically separated to remove metal impurities; Step 1.3: Place the pretreated slag in a constant temperature drying oven and dry it to constant weight; Step 2, Mechanical activation: Step 2.1: Crush the slag into continuously graded particles of 0.15~4.75mm; Step 2.2: Perform microwave irradiation treatment to expand the microcracks inside the particles by utilizing the thermal stress effect, stimulate the surface hydroxyl active sites, and increase the specific surface area. Step 3, Chemical Coating: Step 3.1: Use a high-pressure airless sprayer to precisely coat the composite modifier. The spraying process is carried out in two stages. After the first spraying, let it stand for 10 to 15 minutes, and then carry out the second spraying to ensure that the coating thickness reaches 50 to 200 μm. Step 3.2: The coated aggregate is cured at a gradient temperature to optimize the film performance; Step 4, Optimize the proportions: Step 4.1: Construct a four-factor, five-level experimental matrix based on the central composite design to test the crushing value and water absorption rate of each sample. The four factors include mechanical pressure, irradiation time, composite modifier ratio, and coating film thickness. Step 4.2: Construct a BP neural network model. The input layer consists of four process parameters: mechanical pressure, irradiation time, composite modifier ratio, and coating film thickness. The output layer consists of crushing value and water absorption rate. A random forest algorithm is used for comparative analysis, and the robustness of the model is verified by ten-fold cross-validation. Step 4.3: With the goal of minimizing crushing value and water absorption rate, use the NSGA-II algorithm to solve for the Pareto optimal solution set, and combine it with engineering constraints to select the optimal parameter combination.

2. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 1.1, a vibrating screen is used to classify the original slag and soil. The screen mesh size is 30mm, and the oversized particles are >30mm.

3. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 1.2, the magnetic field strength is 0.8~1.2T.

4. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 1.3, the drying temperature is 105±5℃ and the moisture content is ≤1%.

5. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 2.1, a double-roll high-pressure crusher is used to crush the slag in multiple stages under a pressure of 5~15MPa, and the roller spacing is adjustable from 0.1 to 5mm.

6. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 2.2, the microwave irradiation time is 3 to 8 minutes and the power is 800 to 1200W.

7. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 3.1, the composite modifier is prepared from silane coupling agent, nano SiO2 sol, waterborne epoxy resin and deionized water. The amount of silane coupling agent added is 2-5% of the total mass of the composite modifier, the amount of nano SiO2 sol added is 1-3% of the total mass of the composite modifier, and the amount of waterborne epoxy resin added is 3-8% of the total mass of the composite modifier.

8. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 3.1, the working pressure of the high-pressure airless sprayer is controlled at 20~25MPa, and the nozzle diameter is selected as 0.3~0.5mm.

9. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... The specific steps of step 3.2 are as follows: First, treat in a hot air circulating drying oven at 80~100℃ for 20~60 minutes, then raise the temperature to 120~150℃ and keep it at that temperature for 0.5~1.5 hours, and finally let it cool naturally to room temperature.

10. The method for physical-chemical coupling enhancement and multi-objective optimization of recycled aggregates according to claim 1, characterized in that... In step 4.3, the crushing value of the engineering constraints is ≤12% and the water absorption rate is ≤6%.

Images