Preparation and proportioning optimization method of solid waste non-sintered ceramsite

Abstract

The invention provides a preparation and proportioning optimization method of solid waste non-sintered ceramsite, and relates to the field of green building materials. Red mud, coal gangue and fly ash have a synergistic effect through a volcanic ash effect, so that the strength of the ceramsite is remarkably improved; the silica fume and the quick lime further enhance the activity of the raw materials and realize multi-component synergistic excitation; in the mix proportion design of the unfired ceramsite, a single-factor test is a basic link for determining the mixing amount range of each raw material, and a reasonable interval of key factors can be preliminarily identified by adjusting single variables one by one and testing performance indexes of the single variables; performing mix proportion optimization design in combination with a response surface method; the raw material mixing amount range is rapidly screened through factor tests, the multi-factor interaction effect is analyzed through a Box-Behnken response surface method, a quadratic regression model is established, and precise optimization of the ceramsite performance is achieved; by adopting an unfired process and a natural curing mode, high-temperature sintering is avoided, and energy consumption and carbon emission are remarkably reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of green building materials, and in particular to a preparation method and proportioning optimization method of solid waste unfired ceramsite. BACKGROUND

[0002] Resource comprehensive utilization and environmental protection are important components of China's sustainable development strategy. In recent years, the stockpiling of industrial solid waste has continued to increase, especially large-tonnage solid waste such as red mud, coal gangue, and steel slag, which has a low utilization rate, resulting in land occupation and environmental pollution problems; at the same time, the demand for green and low-carbon materials in the construction industry is increasing, and unfired ceramsite has become a research hotspot due to its low energy consumption and low carbon emissions.

[0003] With the acceleration of industrialization and urbanization, the stockpiling of industrial solid waste such as fly ash, red mud, iron tailings, and coal gasification slag continues to rise, not only occupying land resources, but also causing serious pollution to soil and water due to heavy metal ion leaching; traditional sintered ceramsite can partially absorb solid waste, but its high-temperature calcination process has high energy consumption and large waste gas emissions; in this context, unfired ceramsite technology has become a new direction for solid waste resource utilization due to its low energy consumption, low emissions, and simple process. Unfired ceramsite realizes the bonding and structure forming of solid waste particles through cementation, without the need for high-temperature sintering. The core lies in the proportioning design, which balances the strength, density, and water absorption of ceramsite by optimizing the proportioning of raw materials to meet the needs of different application scenarios.

[0004] Traditional ceramsite preparation process uses clay and shale as the main raw material and is prepared by high-temperature sintering; however, clay and shale are non-renewable resources, and their overexploitation has caused damage to the ecological environment. At the same time, the sintering process has high energy consumption and large carbon emissions, which is contrary to the national policy of energy saving and emission reduction. In contrast, unfired ceramsite uses industrial solid waste as the main raw material and realizes low-energy preparation through efficient resource utilization, which has broad application prospects; however, existing unfired ceramsite still faces the following problems: 1. Lack of scientificity in raw material selection and proportioning design, resulting in large fluctuations in product performance; 2. Insufficient process parameter optimization, affecting the key properties of ceramsite such as strength, density, and water absorption; 3. Insufficient combination of technology and actual engineering application needs.

[0005] Therefore, in view of the above problems, it is urgent to develop an unfired ceramsite based on industrial solid waste and its preparation method, which combines scientific proportioning optimization technology and process parameter design to significantly improve the performance of ceramsite and resource utilization efficiency. SUMMARY

[0006] This invention proposes a method for preparing and optimizing the proportion of solid waste-free ceramsite, in order to solve the problems of insufficient solid waste disposal and large fluctuations in product performance during the existing preparation process of non-fired ceramsite.

[0007] This invention provides a method for preparing and optimizing the proportion of non-combustible ceramsite for solid waste, specifically including the following steps: Step S1: Through single-factor experiments, multiple groups of ceramsite with different proportions were prepared and their performance was tested to determine the dosage range of various solid raw materials, including red mud, coal gangue, cement, fly ash, silica fume, quicklime and quick gypsum, on the performance of ceramsite. Step S2: Based on the dosage range determined in Step S1, select several key raw material factors that have a significant impact on the performance of ceramsite; Step S3: Using the Box-Behnken response surface methodology, design an experimental scheme for optimizing the mix proportion with the dosage of the key raw material factors as variables; Step S4: Prepare and cure ceramsite samples with multiple ratios according to the experimental scheme described in step S3; Step S5: Test the compressive strength of the ceramsite sample obtained in step S4; Step S6: Based on the cylinder compressive strength data measured in step S5, establish a regression model between the dosage of the key raw material factors and the cylinder compressive strength, and verify the effectiveness of the regression model through analysis of variance; Step S7: Using a validated model, determine the optimal raw material ratio for non-fired ceramsite from solid waste.

[0008] Further, the variable raw material factors in the single-factor experiment described in step S1 are red mud, coal gangue, fly ash, silica fume, and quicklime; five gradients are set for each variable, where the red mud content is 11%, 14%, 17%, 20%, and 23%; the coal gangue content is 14%, 17%, 20%, 23%, and 26%; the fly ash content is 17%, 20%, 23%, 26%, and 29%; the silica fume content is 6%, 9%, 12%, 15%, and 18%; and the quicklime content is 10%, 11%, 12%, 13%, and 14%. Twenty-five sets of experiments are conducted to study the influence of each raw material content on its performance. The specific methods for performance testing of expanded clay aggregate, including cylinder compressive strength determination, bulk density determination, 1-hour water absorption rate determination, and softening coefficient determination, include: (1) Cylinder compressive strength test; 5L of expanded clay aggregate was filled into the pressure-bearing cylinder. After filling, the pressure-bearing cylinder was vibrated. Then, the pressure-bearing cylinder was mounted on a stamping die, with the lower scale line of the stamping die aligned with the upper scale line of the guide cylinder. The assembled cylinder pressure tester was placed on a press, with the center of the pressure-bearing cylinder aligned with the center of the pressure plate. Pressure was applied at a test speed of 400N / s. When the depth of the stamping die reached 20mm, the load was stopped and the pressure data was recorded. The cylinder pressure strength test result was calculated using the following formula: Where R is the compressive strength of the expanded clay sample (MPa); p is the pressure value (N) when the indentation depth is 20 mm; and A is the bearing area, A=10. 4 mm 2 ; (2) Bulk density test; Take 15L of ceramsite sample, divide it into two portions, and dry them in an oven until constant weight. Pour the ceramsite sample evenly into a 5L standard metal measuring cylinder through a funnel, filling it to the point above the cylinder opening to form a cone. Use a steel ruler to scrape the surface level from the center to both sides along the edge of the measuring cylinder. Fill the surface depressions with smaller-diameter ceramsite and weigh it. Calculate the bulk density of the ceramsite sample according to the following formula, and take the arithmetic mean of the bulk density of the sample from three tests as the test result. Wherein, ρ is the bulk density of the ceramsite sample, kg / m³. 3 The results are accurate to 1 kg / m 3 m0 is the total mass of the ceramsite sample and the measuring cylinder, kg; m1 is the mass of the measuring cylinder, kg; V is the volume of the measuring cylinder, V=5L; (3) 1-hour water absorption rate test; According to the standard, take 4L of ceramsite sample, divide it into three portions, and dry them in an oven until constant weight. Weigh three 500g portions of sample and place them in graduated cylinders. Immerse them completely in water for 1 hour, then remove them and filter them through a sieve for 1-2 minutes. Place the sample on a wrung-out damp towel and roll it back and forth 8-10 times to remove surface moisture. Pour the sample into a dry graduated cylinder and weigh it. Calculate the water absorption rate of the ceramsite sample according to the following formula, and take the arithmetic mean of the water absorption rates of the three test samples as the test result. Where W is the water absorption rate of the ceramsite sample after 1 hour, %; the result is accurate to 0.1%; m1 is the mass of the ceramsite sample soaked in water, kg; m0 is the mass of the ceramsite sample dried, kg; (4) Softening coefficient; After absorbing water, the ceramsite was subjected to cylinder compressive strength test. The softening coefficient of the ceramsite sample was calculated according to the following formula, and the arithmetic mean of the softening coefficients of the three test samples was taken as the test result. Where K is the softening coefficient of the ceramsite sample after 1 hour, with the result accurate to 0.1%; R1 is the compressive strength of the ceramsite sample after soaking in water, in MPa; and R is the compressive strength of the ceramsite sample after drying, in MPa.

[0009] Furthermore, the key raw material factors mentioned in step S2 include the amount of red mud, coal gangue, fly ash, and silica fume.

[0010] Furthermore, in step S3, the dosage of each key raw material factor is taken at three levels, namely: The red mud content was 14%, 17%, and 20%, respectively. The coal gangue content is 20%, 23%, and 26%, respectively; The fly ash content is 20%, 23%, and 26%, respectively. The silica fume content is 9%, 12%, and 15%, respectively.

[0011] Furthermore, in step S4, the specific methods for preparing and curing multiple proportions of ceramsite samples according to the experimental scheme described in step S3 include: S4-1: Prepare the solid raw materials for the ceramsite sample, and dry and air-dry the raw materials and remove any foreign matter; S4-2: Weigh the components of red mud, coal gangue, cement, fly ash and other raw materials, mix them evenly, and obtain solid waste non-fired ceramsite dry material; S4-3: Pour the evenly mixed solid waste non-fired ceramsite dry material into a container with a rotation speed of 30-32 r / min and an inclination angle of 50°. o -55 o In a disc granulator, 25%-35% water of the solid raw material is sprayed in to granulate and obtain shaped ceramsite. S4-4: The shaped ceramsite was cured naturally for 3, 7, and 28 days to obtain ceramsite samples.

[0012] Furthermore, the specific method for establishing a regression model between the dosage of the key raw material factors and the cylinder compressive strength based on the cylinder compressive strength data measured in step S5, and verifying the effectiveness of the regression model through analysis of variance, as described in step S6, includes: The cylinder compressive strength data measured in step S5 and the key raw material factor admixture data are input into Design-Expert 13 software. The Design-Expert 13 software performs multinomial regression on the data to establish a quadratic multinomial regression model between the cylinder compressive strength of non-fired ceramsite and the admixtures of red mud, coal gangue, fly ash, and silica fume: Y=2.45-0.0125A+0.0292B+0.0517C-0.0283D-0.0050AB+0.0125AC-0.0050AD-0.0100BC+0.0075BD-0.0575CD-0.0425A 2 -0.0625B 2 -0.2312C 2 -0.1212D 2 Where A represents the amount of red mud added; B represents the amount of coal gangue added; C represents the amount of fly ash added; and D represents the amount of silica fume added. The regression model based on the Box-Behnken design was validated for effectiveness through analysis of variance and statistical analysis.

[0013] Further, the optimal solid raw material ratio of the solid waste non-fired ceramsite in step S7 is: 20% red mud, 9.5% cement, 23.5% coal gangue, 23.5% fly ash, 11.5% silica fume, 10% quicklime, and 2% gypsum.

[0014] Compared with the prior art, the present invention has the following advantages: This invention utilizes single-factor experiments combined with Box-Behnken response surface methodology to optimize the mix proportion of solid waste-free ceramsite, designing a method for preparing and optimizing the mix proportion of solid waste-free ceramsite. The invention first determines the raw materials and mix proportion design method for the solid waste-free ceramsite. Solid waste-free ceramsite is prepared using a disc granulator, and its performance is tested after natural curing. Based on the experimental results, the mix proportion is optimized and experimentally verified using Design-Expert 13. This invention not only utilizes large amounts of solid waste but also extends the industrial chain and increases product added value. The method of optimizing the mix proportion of solid waste-free ceramsite using single-factor experiments combined with Box-Behnken response surface methodology has many advantages. First, single-factor experiments can quickly assess the impact of each factor on the performance of ceramsite, while Box-Behnken response surface methodology can analyze the interactions between multiple factors, providing comprehensive optimization results. Second, this method is efficient, covering multiple dimensions with fewer experiments, reducing experimental costs and time. By constructing a regression model, the impact of different mix proportions on the performance of ceramsite can be accurately predicted, finding the optimal raw material mix proportion. Furthermore, this method is flexible, applicable to various raw materials and samples, and allows for adjustments to the experimental design based on specific circumstances. The Box-Behnken method also provides clear visual analysis, helping to identify relationships between influencing factors. This method has optimization potential; through multiple iterations, it can improve the strength of ceramsite and reduce water absorption, promoting the application and development of the material. In summary, combining these two methods not only allows for rapid identification of the influence of key factors but also enables systematic optimization of the ceramsite mix proportions, improving the performance of the final product.

[0015] Furthermore, this invention significantly improves the strength of ceramsite through the synergistic effect of red mud, coal gangue, and fly ash via the pozzolanic effect; silica fume and quicklime further enhance the activity of raw materials, achieving synergistic activation of multiple components; in the mix design of non-fired ceramsite, single-factor experiments are the basic step in determining the dosage range of each raw material. By adjusting individual variables one by one and testing their performance indicators, the reasonable range of key factors can be preliminarily identified; however, single-factor experiments cannot reveal the impact of multi-factor interactions on performance, so this paper combines response surface methodology for mix design optimization; factor experiments quickly screen the dosage range of raw materials, Box-Behnken response surface methodology analyzes multi-factor interactions, and a quadratic regression model is established to achieve precise optimization of ceramsite performance; the adoption of non-fired process and natural curing mode avoids high-temperature sintering, significantly reducing energy consumption and carbon emissions.

[0016] Based on the implementation methods provided in the above aspects, this application can be further combined to provide more implementation methods. Attached Figure Description

[0017] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are illustrated by way of example and not limitation, with the same or corresponding reference numerals denoteing the same or corresponding parts, wherein: Figure 1 A flowchart illustrating the preparation and proportioning optimization method of solid waste non-combustion ceramsite provided by this invention; Figure 2 The present invention provides a method for preparing and optimizing the proportion of solid waste non-fired ceramsite to obtain shaped non-fired ceramsite. Figure 3 shows the performance test results of the non-fired ceramsite provided by the present invention (taking red mud as an example in this experiment). Figure 4 The response surface and contour lines of fly ash and silica fume content to the compressive strength of non-fired ceramsite in the preparation and proportion optimization method of solid waste non-fired ceramsite provided by the present invention. Detailed Implementation

[0018] The exemplary embodiments disclosed in this application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art.

[0019] CombinationFigure 1 As shown, an optional embodiment of the present invention provides a method for preparing and optimizing the proportion of solid waste non-combustion ceramsite, specifically including the following steps: The first step is to determine the raw materials and the mix design method for solid waste non-fired ceramsite.

[0020] Literature review and X-ray fluorescence spectroscopy (XRF) and X-ray diffraction (XRD) analysis show that: The main components of solid wastes such as red mud and coal gangue are SiO2 and Al2O3, which are highly similar to the components of ceramsite raw materials; Fly ash has pozzolanic activity and a fast hydration rate, which can promote the pozzolanic activity of red mud and coal gangue, which have poor activity. Cement provides an alkaline environment for fly ash, activates the vitreous phase of fly ash, improves the activity of fly ash, and helps ceramsite to form pellets and improve strength; Silica fume increases density and drives "low activity" with "high activity", achieving the effect of combining and synergistically stimulating various solid wastes; Quicklime enhances the activity of raw materials; Gypsum promotes the action of quicklime.

[0021] The mix design method was determined to be a combination of single-factor experiments and Box-Behnken response surface methodology. Single-factor experiments determined the dosage of each raw material in the non-fired ceramsite, providing a reasonable numerical range for subsequent Box-Behnken response surface optimization experiments. Based on the single-factor experiments, the influence of each raw material dosage on the compressive strength of the non-fired ceramsite was determined. Factors with a significant impact on compressive strength were selected as independent variables for the response surface optimization experiment, using compressive strength as the response value. Based on the experimental results, a quadratic multinomial regression model was established between the compressive strength of the non-fired ceramsite and the dosage of the solid waste-based non-fired ceramsite raw materials. Analysis of variance was performed on the model to verify its effectiveness. Finally, the optimal mix proportion of the non-fired ceramsite was obtained. Experiments were conducted to verify the optimized mix proportion. A small relative error indicates that the model can reflect the influence of the interaction of different factors on the strength of the ceramsite and can predict the strength of the ceramsite.

[0022] The second step involves using red mud, coal gangue, cement, fly ash, silica fume, quicklime, gypsum, and tap water as raw materials for the non-fired solid waste ceramsite.

[0023] The cement is ordinary Portland cement PO42.5, the fly ash is Grade I fly ash, and the silica fume is 1000-mesh ultrafine powder with ultra-high activity.

[0024] The third step is to determine the mix design method as a combination of single-factor experiments and Box-Behnken response surface methodology.

[0025] The proportions of each raw material in non-fired ceramsite were determined through single-factor experiments, providing a reasonable numerical range for subsequent Box-Behnken response surface methodology (Box-Behnken Response Surface Methodology) optimization experiments. Optimizing the mix proportions of non-fired ceramsite using single-factor experiments combined with Box-Behnken Response Surface Methodology offers several advantages. First, single-factor experiments can quickly assess the impact of each factor on the performance of ceramsite, while Box-Behnken Response Surface Methodology analyzes the interactions between multiple factors, providing comprehensive optimization results. Second, this method is efficient, covering a multi-dimensional space with fewer experiments, reducing cost and time. By constructing a regression model, the impact of different proportions on ceramsite performance can be accurately predicted, finding the optimal raw material mix proportions. Furthermore, this method is flexible and applicable to various raw materials and samples, and the experimental design can be adjusted according to specific circumstances. Simultaneously, Box-Behnken Response Surface Methodology provides clear visual analysis, helping to identify the relationships between influencing factors. Finally, this method has optimization potential; through multiple iterations, the strength of ceramsite can be improved and the water absorption rate reduced, promoting the application and development of the material.

[0026] Single-factor experiment Single-factor experiments are a fundamental and efficient experimental design method. Their core lies in systematically exploring the influence of a single independent variable on a target performance index by adjusting it one by one while strictly fixing other parameters. In the mix design of non-fired ceramsite, this method first requires determining the reasonable range of values ​​for the target factor based on the characteristics of the raw materials and prior literature review. Then, with the dosage of other raw materials fixed, the levels of the target factor are adjusted in a gradient, and the performance test data of the non-fired ceramsite are recorded. A curve showing the relationship between the single variable and the performance test data is then plotted to determine the optimal level interval.

[0027] The main raw materials for non-fired ceramsite are red mud, coal gangue, cement, fly ash, silica fume, quicklime, gypsum, and tap water. This paper's single-factor experiment mainly studies the effects of the dosage of red mud, coal gangue, fly ash, silica fume, and quicklime on the compressive strength, bulk density, and 1-hour water absorption rate of non-fired ceramsite. Five gradients were set for each variable, with red mud dosages of 11%, 14%, 17%, 20%, and 23%; coal gangue dosages of 14%, 17%, 20%, 23%, and 26%; fly ash dosages of 17%, 20%, 23%, 26%, and 29%; silica fume dosages of 6%, 9%, 12%, 15%, and 18%; and quicklime dosages of 10%, 11%, 12%, 13%, and 14%. The specific mix design is shown in the table below, where "C", "T", "F", "G", and "Y" represent red mud, coal gangue, fly ash, silica fume, and quicklime, respectively, and the numbers are the specimen numbers.

[0028] Mix proportion table for different red mud admixtures Mix proportions for tests with different coal gangue content Test mix proportions for different fly ash contents Mix proportion table for different silica fume contents Mix proportion table for different quicklime dosage tests Based on the single-factor experimental design of non-fired ceramsite, 25 sets of experiments are required to study the influence of the dosage of each raw material on its performance.

[0029] The fourth step is to dry and air-dry the raw materials and remove any foreign matter. Weigh out the components of the raw materials, such as red mud, coal gangue, cement, and fly ash, and mix them evenly to obtain solid waste non-fired ceramsite dry material.

[0030] The proportions of each raw material in the dry solid waste non-fired ceramsite are shown in the table below (taking red mud in this experiment as an example).

[0031] Fifth step: Measure 30% of the mass of the solid waste non-fired ceramsite mixed powder with tap water and put it into a spray bottle for later use.

[0032] Step 6: Pour the evenly mixed solid waste non-fired ceramsite dry material into a container with a rotation speed of 30-32 r / min and an inclination angle of 50 degrees. o -55 o In a disc granulator, water is sprayed in at a rate of 30% of the solid raw material to produce shaped, non-fired ceramsite.

[0033] The resulting shaped, non-fired ceramsite is shown in the attached image. Figure 2 The prepared non-fired ceramic particles have a particle size of 5-15mm.

[0034] The seventh step involves naturally curing the formed ceramsite, with curing times of 3, 7, and 28 days.

[0035] The eighth step is to conduct performance tests on the non-fired ceramsite, including cylinder compressive strength measurement, bulk density measurement, 1-hour water absorption rate measurement, and softening coefficient measurement.

[0036] (1) Cylinder compressive strength test Referring to the standard "Test Methods for Lightweight Aggregates" GB2841-8, 5L of expanded clay aggregate was filled into the pressure cylinder, ensuring the aggregate level was higher than the cylinder opening edge. After filling, the pressure cylinder was vibrated. Then, the pressure cylinder was mounted on a stamping die, with the lower scale line of the die aligned with the upper scale line of the guide cylinder. The assembled cylinder compression tester was placed on a press, with the center of the pressure cylinder aligned with the center of the pressure plate. Pressure was applied at a test speed of 400N / s. When the stamping die reached a depth of 20mm, the load was stopped, and the pressure data was recorded. The cylinder compression strength test result was calculated according to the formula.

[0037] Where R is the compressive strength of the ceramsite sample (MPa); p is the pressure value (N) when the indentation depth is 20mm; A is the bearing area (A=10). 4 mm 2 ).

[0038] (2) Bulk density test According to the standard specifications, take 15L of ceramsite sample, divide it into two portions, and dry them in an oven until constant weight. Pour the ceramsite sample evenly into a 5L standard metal measuring cylinder through a funnel, filling it to the point above the cylinder opening to form a cone. Use a steel ruler to level the surface from the center outwards along the edge of the measuring cylinder, filling any depressions with smaller-diameter ceramsite particles, and then weigh the sample. Calculate the bulk density of the ceramsite sample according to the formula, and use the arithmetic mean of the bulk densities from three tests as the test result.

[0039] Where ρ is the bulk density of the ceramsite sample (kg / m³). 3 The results are accurate to 1 kg / m³. 3 m0 is the total mass (kg) of the ceramsite sample and the measuring cylinder; m1 is the mass (kg) of the measuring cylinder; V is the volume of the measuring cylinder (V=5L).

[0040] (3) 1-hour water absorption rate test According to the standard, take 4L of ceramsite sample, divide it into three portions, and dry them in an oven until constant weight. Weigh three 500g portions of sample and immerse them completely in water in a graduated cylinder for 1 hour. Remove them, pour the water through a sieve for 1-2 minutes, and then roll the specimens back and forth on a wrung-out damp towel 8-10 times to remove surface moisture. Pour the water into a dry graduated cylinder and weigh it. Calculate the water absorption rate of the ceramsite sample according to the formula, and take the arithmetic mean of the water absorption rates of the three tests as the test result.

[0041] Where W is the water absorption rate (%) of the ceramsite sample after 1 hour, accurate to 0.1%; m1 is the mass (kg) of the soaked ceramsite sample; and m0 is the mass (kg) of the dried ceramsite sample.

[0042] (4) Softening coefficient The expanded clay aggregates after absorbing water were subjected to cylinder compressive strength tests. The softening coefficient of the expanded clay aggregate samples was calculated according to the formula, and the arithmetic mean of the softening coefficients of the three test samples was taken as the test result.

[0043] Wherein, K is the softening coefficient of the ceramsite sample after 1 hour, with the result accurate to 0.1%; R1 is the compressive strength (MPa) of the ceramsite sample soaked in water; and R is the compressive strength (MPa) of the ceramsite sample dried in water.

[0044] The prepared non-fired ceramsite was subjected to performance tests following the above steps. The test results are attached. Figure 3a and 3b (Taking red mud as an example).

[0045] Step nine involves selecting the raw material dosages with the highest compressive strength, lowest bulk density, and lowest 1-hour water absorption rate as independent variables for the response surface methodology (RSM) optimization design. Single-factor experiments are conducted to determine the dosages of each raw material in the solid waste-fired ceramsite, providing a reasonable numerical range for subsequent Box-Behnken RSM optimization experiments. Based on the single-factor experiments, the influence of each raw material dosage on the compressive strength of the solid waste-fired ceramsite is determined. Factors with a significant impact on compressive strength are selected as independent variables for the RSM optimization experiment, and the RSM optimization experiment is designed using compressive strength as the response value.

[0046] Response Surface Methodology (RSM) is an optimization technique based on statistics and mathematical modeling. Its core objective is to efficiently determine optimal operating conditions by constructing approximate functions to describe the complex nonlinear relationship between the input and output of a multivariable system. This method aims to fit a response surface model to limited experimental data, replacing the real but difficult-to-analyze limit state function, thereby achieving parameter optimization and performance prediction. Box-Behnken Design (BBD), one of the most widely used design methods in RSM, avoids extreme level combinations by rationally selecting the distribution of experimental points, thus reducing experimental workload while ensuring model prediction accuracy.

[0047] In studies determining mix proportions, single-factor experiments combined with response surface methodology (RSM) demonstrate significant advantages over traditional methods. Traditional single-factor methods, by fixing other variables and changing only a single factor, can initially identify the influence trends of key parameters, but cannot reveal the interactions and nonlinear effects between multiple factors, making the optimization results susceptible to local extrema. In contrast, RSM, after determining the reasonable range of raw material dosages through single-factor experiments, further employs the BBD design method to construct a global response surface model, enabling systematic analysis of the optimal combination under the coupled effects of multiple factors. This method significantly reduces experimental costs by minimizing redundant experiments, and utilizes a quadratic polynomial model to reflect the curvature effect between parameters, avoiding the local optima limitations caused by fixed-level combinations in traditional orthogonal experiments. Furthermore, RSM quantifies the significance and interaction strength of each factor through analysis of variance, and combines model residual analysis and fit tests to ensure predictive reliability, ultimately outputting a statistically significant optimal mix proportion. Compared to manual experience methods or single-objective optimization strategies, this method transforms subjective experience into an objective data-driven decision-making process through mathematical modeling. It demonstrates high efficiency and practicality in fields such as materials science, chemical processes, and bioengineering, and provides a systematic solution for parameter tuning of complex systems.

[0048] Single-factor experiments were conducted to determine the influence trends of each raw material and thus the appropriate dosage range for each raw material in the non-fired ceramsite. This paper uses the 28-day curing test results as the standard and optimizes the design using Box-Behnken Design (RSM-BBD) in Design-Expert 13 software. The compressive strength (Y) of the non-fired ceramsite cylinder is used as the response value. Four factors significantly affecting Y are considered, each with three levels: red mud content (A), with levels of 14%, 17%, and 20%; coal gangue content (B), with levels of 20%, 23%, and 26%; fly ash content (C), with levels of 20%, 23%, and 26%; and silica fume content (D), with levels of 9%, 12%, and 15%. Based on this design, 27 sets of response surface optimization experiments were conducted for the non-fired ceramsite.

[0049] Step 10: Input the independent variable data into Design-Expert 13 to generate an experimental table for testing.

[0050] The experimental table is as follows: .

[0051] In terms of mix design, an optimization strategy combining single-factor experiments and response surface methodology was adopted. Single-factor experiments, by adjusting the proportions of five raw materials (red mud, coal gangue, fly ash, etc.) using gradients, preliminarily determined the influence of each component on compressive strength, bulk density, 1-hour water absorption rate, and softening coefficient. Based on the results of the single-factor experiments, a Box-Behnken experimental design was constructed using Design-Expert 13 software. Using compressive strength as the objective function, the nonlinear effects of the interaction of four factors—red mud, coal gangue, fly ash, and silica fume—on the performance of the non-fired ceramsite were systematically analyzed.

[0052] Step 11: Based on the experimental results, perform multiple regression fitting on the experimental data to establish a quadratic multiple regression model of the compressive strength of non-fired ceramsite cylinder and the amount of solid waste non-fired ceramsite raw material. Perform variance analysis on the model to verify the effectiveness of the compressive strength model of non-fired ceramsite cylinder.

[0053] This paper uses the 28-day curing test results as the standard and optimizes the experimental design using the Box-Behnken Design (RSM-BBD) in Design-Expert 13 software. The compressive strength (Y) of the unfired ceramsite cylinder is used as the response value. Four factors that significantly affect Y are considered, with three levels for each factor: red mud content (A), with three levels of 14%, 17%, and 20%; coal gangue content (B), with three levels of 20%, 23%, and 26%; fly ash content (C), with three levels of 20%, 23%, and 26%; and silica fume content (D), with three levels of 9%, 12%, and 15%. The cylinder compressive strength test was conducted according to the optimized experimental table based on the response surface, and the results are shown in the table below.

[0054] Response Surface Optimization Experiment Results Table Based on the experimental results, a quadratic multinomial regression model was established using Design-Expert 13 software to perform multinomial fitting regression on the data, relating the compressive strength of non-fired ceramsite to the content of red mud, coal gangue, fly ash, and silica fume. Y=2.45-0.0125A+0.0292B+0.0517C-0.0283D-0.0050AB+0.0125AC-0.0050AD-0.0100BC+0.0075BD-0.0575CD-0.0425A 2 -0.0625B 2 -0.2312C 2 -0.1212D 2 To verify the effectiveness of the compressive strength model for non-fired ceramsite, an analysis of variance was performed on the model, and the results are shown in the table below. In the response surface optimization experiment of the compressive strength of non-fired ceramsite, the regression model based on Box-Behnken design (BBD) was verified for its reliability and scientific validity through analysis of variance and statistical analysis.

[0055] Analysis of variance results table Statistical Analysis Table The analysis of variance (ANOVA) results showed that the regression model had a sum of squares of 0.3805, 14 degrees of freedom, a mean square of 0.0272, and an F-value of 49.98 (P < 0.0001). The F-value, calculated as the mean square of a factor divided by the mean square of the residuals, is the core test statistic in ANOVA. The F-value of 49.98 indicates that the model explains a variance far exceeding random error, suggesting that the model is highly significant overall. The significance level is marked as ** (extremely significant), indicating that the model effectively explains the variation in cylinder compressive strength and has an excellent fit. Looking at the main effects of the factors, the P-values ​​for coal gangue content (B), fly ash content (C), and silica fume content (D) are all less than 0.01, indicating high significance and a significant impact on cylinder compressive strength. The P-value represents the probability of observing a current or even larger F-value if a factor actually has no effect on the result. A smaller P-value indicates a lower degree of randomness in the factor's effect, meaning the factor's influence is more significant. The p-value for red mud admixture (A) was 0.0880, close to significant but not quite significant, indicating its relatively weak effect on cylinder compressive strength. Among the interactions, only the interaction between fly ash and silica fume admixture (CD) was significant; the other interactions were not significant, suggesting that the synergistic effect among factors mainly manifests in the CD combination. The p-values ​​for the quadratic terms A², B², C², and D² were all less than 0.01, indicating that the quadratic terms of each factor have a significant nonlinear effect on cylinder compressive strength, and the model needs to consider the curvature effect among factors. Residual analysis showed a residual sum of squares of 0.0065, 12 degrees of freedom, and a mean square of 0.0005. The p-value for the lack-of-fit term was not significant, indicating that the model had no significant lack of fit, and the error mainly stemmed from random factors, demonstrating high data reliability.

[0056] Statistical analysis further validated the model's excellent characteristics. The standard deviation (Std. Dev.) was only 0.0233, indicating low data dispersion and high measurement accuracy. The mean (Mean) was 2.25, reflecting the average level of cylinder pressure intensity. The coefficient of variation (CV%) was as low as 1.04%, indicating good data repeatability and strong model stability. The coefficient of determination (R²=0.9831) was close to 1, indicating that the model could explain 98.31% of the cylinder pressure intensity variation, demonstrating extremely high goodness of fit. The adjusted coefficient of determination (Adjusted R²=0.9635) was highly consistent with R², indicating that the model was not overfitting. The predicted coefficient of determination (Predicted R²=0.9047), although slightly lower, remained at a high level, indicating that the model's predictive ability for unknown data was reliable. The accuracy adequacy (AdeqPrecision=24.9321) was much greater than 4, validating the model's high signal-to-noise ratio, reasonable experimental design, and ability to accurately reflect the response relationship between various factors and cylinder pressure intensity.

[0057] In response surface methodology (RSM), contour plots and response surface plots serve as core visualization tools, revealing the complex relationships between independent and response variables from two-dimensional and three-dimensional perspectives, respectively. Contour plots, by fixing other variables, display the influence of two independent variables on the response variable in a two-dimensional plane. The shape of the contour lines visually reflects the strength of the interaction between factors. Circular contour lines indicate that the interaction effect between the two factors is insignificant, while elliptical contour lines suggest a significant synergistic effect. Dense areas of contour lines correspond to sensitive intervals where response values ​​change drastically. Response surface plots, through surface shape and color gradients, comprehensively present the nonlinear response characteristics between the three variables. The steepness of the surface reflects the significance of the main effect, changes in curvature reveal the moderating effect of the quadratic term on the response, and the direction of color transitions visually demonstrates the extreme trends of response values ​​as factor levels change, providing a theoretical basis for experimental optimization. The combination of both allows for a systematic analysis of the coupling effects of main effects, interaction effects, and curvature effects, providing multidimensional data support for determining the optimal combination of factor levels and optimizing process parameters.

[0058] As shown in the table above, the order of interaction strength is CD>AC>BC>BD>AD=AB, with CD being the most significant interaction between the two raw materials of the non-fired ceramsite. Therefore, response surface analysis is performed using the strongest interaction (CD) and the weakest interaction (AB). Figure 4 .

[0059] In the response surface analysis of non-fired ceramsite, the influence of the combination of two independent variables, AB and CD, on the cylinder compressive strength was analyzed. Through systematic comparison of contour line morphology, three-dimensional surface features and color gradient distribution, the significant differences in the interaction of different factors were clearly revealed.

[0060] Figure 4In a two-dimensional contour map, the contour lines gradually change from yellow to red, but the contour lines are nearly circular in shape, with blurred boundaries and low density. There are only faint red high-value areas in the central region, and the color transition in the outer region is smooth, indicating that the interaction between A and B is weak. Figure 4 (a) In the three-dimensional response surface plot, the overall tilt of the surface is gentle and the curvature change is not obvious. There is only a slight bulge in the central area, indicating that the AB synergistic effect is not significant.

[0061] Figure 4 (b) In the two-dimensional contour map, the contour lines gradually change from blue to green and then to red, and the contour lines are obviously elliptical in shape with clear boundaries and high density. The color gradient is steep, indicating that the interaction between C and D is significant. Figure 4 (b) In the three-dimensional response surface plot, the contour lines projected onto the bottom surface form closed loops with a significant elliptical shape and a distinct raised surface with drastic curvature changes and a steep color gradient, indicating that the CD synergistic effect is extremely significant.

[0062] Step 12: Model Optimization Results and Verification. In the performance optimization study of non-fired ceramsite, accurately determining its optimal proportion is a key step in improving the overall performance of the material. This study utilizes the Optimization function of Design-Expert 13 software to systematically analyze the interaction between the content of red mud (A), coal gangue (B), fly ash (C), and silica fume (D) on the compressive strength of the ceramsite cylinder by constructing a response surface model, and successfully found the optimal proportion (see table below). To verify the reliability and predictive ability of the model, three parallel experiments were conducted based on this optimal proportion, and the measured compressive strength of the non-fired ceramsite cylinder was 2.38 MPa. Comparative analysis showed that the relative error between the model prediction and the experimental actual value was only 0.83%. This result indicates that the constructed response surface model can effectively capture the nonlinear interaction between different influencing factors and achieve high-precision prediction of the compressive strength of the non-fired ceramsite cylinder.

[0063] Comparison table of model predictions and actual experimental values Therefore, the optimal mix proportions are: red mud 20%, cement 9.5%, coal gangue 23.5%, fly ash 23.5%, silica fume 11.5%, quicklime 10%, and gypsum 2%. This optimal mix proportion increases the amount of industrial solid waste used while reducing the amount of cement. Tests showed that the compressive strength of the optimal mix proportion of non-fired ceramsite is 2.38 MPa, and the bulk density is 552 kg / m³. 3The 1-hour water absorption rate was 15.2%, and the softening coefficient was 0.86. The above content systematically analyzes the synergistic regulation mechanism of the dosage of industrial solid wastes such as red mud, coal gangue, fly ash, and silica fume on the mechanical properties of non-fired ceramsite. Single-factor experiments determined the influence of gradient dosage changes of each raw material at the baseline ratio on the compressive strength, bulk density, 1-hour water absorption rate, and softening coefficient. The single-factor experiments provided a reasonable numerical range for subsequent response surface optimization experiments. Based on the Box-Behnken experimental design using the response surface methodology, a quadratic multinomial regression model with compressive strength as the objective was constructed. Key influencing factors were screened through variance analysis and significance testing, and the significance of the fly ash-silica fume interaction on the compressive strength of non-fired ceramsite was determined. The optimized model results were achieved with red mud (20%), coal gangue (23.51%), fly ash (23.47%), and silica fume (11.49%). The relative error between the model prediction and experimental values ​​was only 0.83%, verifying the reliability of the multi-factor coupling mechanism.

[0064] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for preparing and optimizing the proportion of solid waste non-fired ceramsite, characterized in that, Specifically, the steps include the following: Step S1: Through single-factor experiments, multiple groups of ceramsite with different proportions were prepared and their performance was tested to determine the dosage range of various solid raw materials, including red mud, coal gangue, cement, fly ash, silica fume, quicklime and quick gypsum, on the performance of ceramsite. Step S2: Based on the dosage range determined in Step S1, select several key raw material factors that have a significant impact on the performance of ceramsite; Step S3: Using the Box-Behnken response surface methodology, design an experimental scheme for optimizing the mix proportion with the dosage of the key raw material factors as variables; Step S4: Prepare and cure ceramsite samples with multiple ratios according to the experimental scheme described in step S3; Step S5: Test the compressive strength of the ceramsite sample obtained in step S4; Step S6: Based on the cylinder compressive strength data measured in step S5, establish a regression model between the dosage of the key raw material factors and the cylinder compressive strength, and verify the effectiveness of the regression model through analysis of variance; Step S7: Using a validated model, determine the optimal raw material ratio for non-fired ceramsite from solid waste.

2. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, The raw material variables in the single-factor experiment described in step S1 are red mud, coal gangue, fly ash, silica fume, and quicklime. Five gradients are set for each variable: red mud content is 11%, 14%, 17%, 20%, and 23%; coal gangue content is 14%, 17%, 20%, 23%, and 26%; fly ash content is 17%, 20%, 23%, 26%, and 29%; silica fume content is 6%, 9%, 12%, 15%, and 18%; and quicklime content is 10%, 11%, 12%, 13%, and 14%. Twenty-five sets of experiments are conducted to study the influence of each raw material content on its performance. The specific methods for performance testing of expanded clay aggregate, including cylinder compressive strength determination, bulk density determination, 1-hour water absorption rate determination, and softening coefficient determination, include: (1) Cylinder compressive strength test; 5L of expanded clay aggregate was filled into the pressure-bearing cylinder. After filling, the pressure-bearing cylinder was vibrated. Then, the pressure-bearing cylinder was mounted on a stamping die, with the lower scale line of the stamping die aligned with the upper scale line of the guide cylinder. The assembled cylinder pressure tester was placed on a press, with the center of the pressure-bearing cylinder aligned with the center of the pressure plate. Pressure was applied at a test speed of 400N / s. When the depth of the stamping die reached 20mm, the load was stopped and the pressure data was recorded. The cylinder pressure strength test result was calculated using the following formula: Where R is the compressive strength of the ceramsite sample, MPa; p is the pressure value when the indentation depth is 20 mm, N; and A is the bearing area, A=10. 4 mm 2 ; (2) Bulk density test; Take 15L of ceramsite sample, divide it into two portions, and dry them in an oven until constant weight. Pour the ceramsite sample evenly into a 5L standard metal measuring cylinder through a funnel, filling it to the point above the cylinder opening to form a cone. Use a steel ruler to scrape the surface level from the center to both sides along the edge of the measuring cylinder. Fill the surface depressions with smaller-diameter ceramsite and weigh it. Calculate the bulk density of the ceramsite sample according to the following formula, and take the arithmetic mean of the bulk density of the sample from three tests as the test result. Wherein, ρ is the bulk density of the ceramsite sample, kg / m³. 3 The results are accurate to 1 kg / m 3 m0 is the total mass of the ceramsite sample and the measuring cylinder, kg; m1 is the mass of the measuring cylinder, kg; V is the volume of the measuring cylinder, V=5L; (3) 1-hour water absorption rate test; According to the standard, take 4L of ceramsite sample, divide it into three portions, and dry them in an oven until constant weight. Weigh three 500g portions of sample and place them in graduated cylinders. Immerse them completely in water for 1 hour, then remove them and filter them through a sieve for 1-2 minutes. Place the sample on a wrung-out damp towel and roll it back and forth 8-10 times to remove surface moisture. Pour the sample into a dry graduated cylinder and weigh it. Calculate the water absorption rate of the ceramsite sample according to the following formula, and take the arithmetic mean of the water absorption rates of the three test samples as the test result. Where W is the water absorption rate of the ceramsite sample after 1 hour, %; the result is accurate to 0.1%; m1 is the mass of the ceramsite sample soaked in water, kg; m0 is the mass of the ceramsite sample dried, kg. (4) Softening coefficient; After absorbing water, the ceramsite was subjected to cylinder compressive strength test. The softening coefficient of the ceramsite sample was calculated according to the following formula, and the arithmetic mean of the softening coefficients of the three test samples was taken as the test result. Where K is the softening coefficient of the ceramsite sample after 1 hour, with the result accurate to 0.1%; R1 is the compressive strength of the ceramsite sample after soaking in water, in MPa; and R is the compressive strength of the ceramsite sample after drying, in MPa.

3. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, The key raw material factors mentioned in step S2 include the amount of red mud, coal gangue, fly ash, and silica fume.

4. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, In step S3, the dosage of each key raw material factor is taken at three levels, namely: The red mud content was 14%, 17%, and 20%, respectively. The coal gangue content is 20%, 23%, and 26%, respectively; The fly ash content is 20%, 23%, and 26%, respectively. The silica fume content is 9%, 12%, and 15%, respectively.

5. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, The specific methods for preparing and curing ceramsite samples with multiple proportions according to the experimental scheme described in step S3 in step S4 include: S4-1: Prepare the solid raw materials for the ceramsite sample, and dry and air-dry the raw materials and remove any foreign matter; S4-2: Weigh the components of red mud, coal gangue, cement, fly ash and other raw materials, mix them evenly, and obtain solid waste non-fired ceramsite dry material; S4-3: Pour the evenly mixed solid waste non-fired ceramsite dry material into a container with a rotation speed of 30-32 r / min and an inclination angle of 50°. o -55 o In a disc granulator, 25%-35% water of the solid raw material is sprayed in to granulate and obtain shaped ceramsite. S4-4: The shaped ceramsite was cured naturally for 3, 7, and 28 days to obtain ceramsite samples.

6. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, The specific method for establishing a regression model between the dosage of the key raw material factors and the cylinder compressive strength based on the cylinder compressive strength data measured in step S5, as described in step S6, and verifying the effectiveness of the regression model through analysis of variance, includes: The cylinder compressive strength data measured in step S5 and the key raw material factor admixture data are input into Design-Expert 13 software. The Design-Expert 13 software performs multinomial regression on the data to establish a quadratic multinomial regression model between the cylinder compressive strength of non-fired ceramsite and the admixtures of red mud, coal gangue, fly ash, and silica fume: Y=2.45-0.0125A+0.0292B+0.0517C-0.0283D-0.0050AB+0.0125AC-0.0050AD-0.0100BC+0.0075BD-0.0575CD-0.0425A 2 -0.0625B 2 -0.2312C 2 -0.1212D 2 Where A represents the amount of red mud added; B represents the amount of coal gangue added; C represents the amount of fly ash added; and D represents the amount of silica fume added. The regression model based on the Box-Behnken design was validated for effectiveness through analysis of variance and statistical analysis.

7. The method for preparing and optimizing the proportion of solid waste non-fired ceramsite according to claim 1, characterized in that, The optimal solid raw material ratio for the solid waste non-fired ceramsite mentioned in step S7 is: 20% red mud, 9.5% cement, 23.5% coal gangue, 23.5% fly ash, 11.5% silica fume, 10% quicklime, and 2% gypsum.

Images