A method for optimizing the proportions of slate-like materials in cold regions

By optimizing the proportions of similar materials for slate in cold regions through orthogonal experiments and analytic hierarchy process, the shortcomings of thermal conductivity and freeze-thaw damage evolution characteristics in freeze-thaw model tests of slate slopes in high-altitude and cold mines were solved. This achieved the applicability and reliability of similar materials under freeze-thaw cycle conditions and improved the accuracy and efficiency of model tests.

CN123111771APending Publication Date: 2026-11-17SOUTHWEST PETROLEUM UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611278545.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing studies on rock-similar materials have failed to fully consider thermal conductivity and freeze-thaw damage evolution characteristics in freeze-thaw model tests of slate slopes in high-altitude and cold mines. This makes it difficult to balance the synergistic similarity of the theoretical optimal mix ratio with the force-thermal parameters, and lacks systematic verification under freeze-thaw cycle conditions, which affects the reliability and repeatability of the model tests.

Method used

Orthogonal experimental design and analytic hierarchy process were adopted, combined with a multivariate regression prediction model, to determine the weights of each target parameter, construct a weighted comprehensive relative error function, optimize the mix proportions of similar materials to slate in cold regions, and verify the applicability of the materials in high-altitude and cold environments through freeze-thaw cycles.

Benefits of technology

This study achieved comprehensive performance matching between similar materials and prototype slate under freeze-thaw cycles, improving the reliability and repeatability of model tests, ensuring the synergistic similarity of mechanical and thermal parameters, and reducing experimental workload and costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111771A_ABST
    Figure CN123111771A_ABST
Patent Text Reader

Abstract

This invention discloses a method for optimizing the mix proportions of similar materials for slate in cold regions, relating to the field of similar materials for slate slope model tests in cold regions. This invention identifies multiple influencing factors and levels affecting the performance of similar slate materials in cold regions, and uses orthogonal experimental design to prepare multiple sets of samples with different mix proportions. The physical, mechanical, and thermal parameters of these samples are tested to obtain the measured values ​​of each target parameter. The weighting coefficients of each target parameter are determined using the analytic hierarchy process (AHP), and a weighted comprehensive relative error function is constructed. Constraint optimization is performed with the goal of minimizing the weighted comprehensive relative error to obtain the theoretically optimal mix proportion combination for each influencing factor. This invention enables the theoretically optimal mix proportion to simultaneously consider both mechanical and thermal similarity, ensuring the realism of the coupling between the temperature field and stress field in freeze-thaw model tests of slate slopes in high-altitude cold regions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of similar materials for model tests of slate slopes in cold regions, and specifically to a method for optimizing the proportions of similar materials for slate slopes in cold regions. Background Technology

[0002] Located in the transitional zone between the Qinghai-Tibet Plateau and the Sichuan Basin, the western Sichuan region is rich in mineral resources and is one of the important distribution areas of hard-rock lithium resources in my country. Many mines in this region are situated in high-altitude, cold-region permafrost areas with large diurnal and seasonal temperature differences, concentrated rainfall, and high annual evaporation. Open-pit mining exposes steep slate slopes to intense evaporation and significant temperature variations, creating conditions for wet-dry cycles and freeze-thaw cycles. Under the combined effects of mining disturbance and the complex environment, internal fissures in the slate gradually expand, pore structure and permeability characteristics continuously evolve, and strength and deformation parameters gradually deteriorate, thus affecting the long-term stability of high-altitude, cold-region mine slopes. Related studies show that freeze-thaw cycles reduce the longitudinal wave velocity, uniaxial compressive strength, and elastic modulus of slate, and the involvement of moisture exacerbates this deterioration. Wet-dry cycles, air-freeze cycles, and water-freeze cycles all increase slate porosity and permeability, with water-freeze cycles having the most significant effect. Therefore, in order to reveal the deformation and failure mechanism of slate slopes in high-altitude and cold regions under freeze-thaw conditions, it is of great significance to carry out corresponding physical model tests. The rational formulation of similar materials is the basis for the reliable conduct of model tests. It is necessary to develop similar materials suitable for freeze-thaw model tests of slate slopes in high-altitude and cold regions.

[0003] Rock mass physical model tests require that the model materials satisfy similarity relationships with the prototype rock mass in terms of main physical and mechanical properties. Existing studies on rock-similar materials mostly use parameters such as density, compressive strength, elastic modulus, cohesion, and internal friction angle as the main control indicators, and achieve material parameter control by adjusting the proportions of aggregates, cementing materials, and admixtures. Regarding mix design optimization methods, existing research, based on orthogonal design combined with range analysis, variance analysis, and intuitive analysis, has studied the influence of different components on the density, strength, deformation, and shear parameters of rock-similar materials. Furthermore, some studies have introduced regression analysis based on orthogonal experiments to establish a quantitative relationship between the mix design of similar materials and physical and mechanical parameters, gradually shifting the optimization of similar material mix design from factor influence analysis to parameter prediction and quantitative design. Other studies have developed similar materials for specific lithologies through orthogonal mix design experiments, sensitivity analysis, and multiple linear regression, and verified their applicability to model tests. As model tests of rock slopes and underground engineering have gradually shifted from single mechanical problems to complex environmental issues such as rainfall infiltration, reservoir water fluctuations, freeze-thaw cycles, and fluid-structure interaction, the control parameters for similar materials have also expanded from traditional mechanical parameters to include water absorption characteristics, softening effects, permeability characteristics, and thermal properties. The aforementioned research indicates that the control parameters for similar materials should be determined based on the specific similarity simulation objectives. For subsequent freeze-thaw model tests of slate slopes in high-altitude cold regions, in addition to meeting the conventional physical and mechanical parameter similarity requirements, the materials should also consider thermal conductivity matching, and their applicability should be verified through the attenuation patterns of compressive strength, elastic modulus, longitudinal wave velocity, and thermal conductivity after freeze-thaw cycles.

[0004] However, while existing research on rock-similar materials has yielded significant results in parameter control, orthogonal design, regression prediction, and model test applications, research on similar materials for freeze-thaw model tests of slate slopes in high-altitude and cold-climate mines remains relatively insufficient. Comprehensive consideration of the thermal conductivity and freeze-thaw damage evolution characteristics of slate is still inadequate. Specifically, existing mix optimization methods often employ single-index evaluation or simple weighting, failing to fully consider the varying importance of different target parameters to the needs of freeze-thaw model tests. This results in the theoretically optimal mix failing to balance the synergistic similarity of force and heat parameters. Furthermore, existing verification methods primarily focus on parameter testing under ambient temperature or single environmental conditions, lacking systematic verification methods for the similarity between similar materials and prototype slate damage evolution characteristics under freeze-thaw cycles. This makes it difficult to ensure the applicability of the developed similar materials in high-altitude and cold-climate freeze-thaw environments. Summary of the Invention

[0005] One objective of this invention is to provide a method for optimizing the mix proportions of slate-like materials in cold regions. This method employs orthogonal experiments to analyze the influence of various mix proportion factors on density, thermal conductivity, and mechanical parameters. Based on regression prediction, and considering the differences in the importance of different target parameters to the requirements of model experiments, the method of analytic hierarchy process (AHP) is used to determine the weights of the indicators, and a weighted comprehensive relative error function is constructed to determine the theoretically optimal mix proportions.

[0006] A method for optimizing the mix proportions of slate-like materials from cold regions includes the following steps:

[0007] S1, to determine multiple influencing factors and levels affecting the properties of similar materials to slate in cold regions, orthogonal experimental design was used to prepare multiple groups of samples with different proportions; the raw materials of the samples included magnetite powder, composite silicate cement, microsilica powder, mica powder, silicon carbide micro powder, polycarboxylate superplasticizer and defoamer, with magnetite powder as aggregate and composite silicate cement as binder; the orthogonal experiment adopted L16(4 5 An orthogonal array was designed with 5 influencing factors and 4 levels. The influencing factors include: moisture content, bone glue ratio, mass ratio of silica powder to binder, mass ratio of silicon carbide powder to aggregate, and mass ratio of mica powder to aggregate. Each influencing factor has 4 levels.

[0008] Each influencing factor does not affect only one specific performance attribute, but is determined comprehensively based on its main role in the material system, preliminary trial mixing results, and the target requirements for density, thermal conductivity, compressive strength, elastic modulus, cohesion, and internal friction angle from model tests. Specifically, moisture content primarily affects slurry formation, hydration process, and pore structure; the bone-bond ratio mainly regulates the degree of bonding, strength, and stiffness; microsilica powder mainly regulates particle filling and matrix density; silicon carbide micropowder mainly regulates thermal conductivity; and mica powder primarily utilizes its flaky particle characteristics to adjust the internal structure and deformation parameters of the material. Each factor is set at four levels. This approach allows for a more comprehensive reflection of the factor's changing trend within the experimental range and provides a data basis for regression interpolation. Furthermore, a five-factor, four-level full-factor experiment requires four... 5 =1024 groups of experiments, using L 16 (4 5 After using the orthogonal array, only 16 sets of experiments need to be completed, under the condition of mainly analyzing the main effects of each factor. The types of factors, the number of levels, and the level values ​​can be adjusted according to different target rock masses and material systems, but if the factors are changed or the number of levels is altered, the experiment needs to be redesigned and a regression model established; if the new levels exceed the original experimental range, the original regression model cannot be directly extrapolated in principle.

[0009] S2, conduct physical, mechanical and thermal parameter tests on the multiple groups of samples to obtain the measured values ​​of each target parameter; based on the measured values, use range analysis to calculate the mean and range of each influencing factor at each level, determine the order of influence of each influencing factor on each target parameter according to the magnitude of the range, and determine the influence law of each influencing factor on each target parameter according to the trend of the mean of the target parameter changing with the factor level;

[0010] S3. Based on the results of orthogonal experiments, a multiple regression prediction model is established between each influencing factor and each target parameter. This multiple regression prediction model is either a multiple linear regression model or a multiple regression model containing pure quadratic terms, used to establish the quantitative relationship between the influencing factors and each target parameter. The regression coefficients are not arbitrarily specified, but are estimated using the least squares method with 16 sets of orthogonal experimental results as samples, minimizing the sum of squared residuals between the predicted regression values ​​and the experimental values. Since 16 sets of experiments are insufficient to stably estimate a complete quadratic model containing all interaction terms, the model mainly retains linear terms and necessary pure quadratic terms, and comprehensively considers the degree of fit, model simplicity, residual degrees of freedom, and parameter variation patterns to determine the final equation.

[0011] S4. The weight coefficients of each objective parameter are determined using the analytic hierarchy process (AHP), and a weighted comprehensive relative error function is constructed. The weighted comprehensive relative error evaluation function is as follows:

[0012] In the formula, This is the weighted average relative error; Let be the predicted value of the i-th objective parameter of similar materials from the regression model; Let be the target value of the i-th target parameter; wi be the weight coefficient of the i-th target parameter, i=1,2,…,n, where n is the total number of target parameters, and satisfy . .

[0013] S5, based on the multiple regression prediction model in step S3, performs constrained optimization to minimize the weighted comprehensive relative error. The component ratio corresponding to the minimum is the theoretical optimal combination of each influencing factor.

[0014] S6. Prepare a verification sample according to the theoretically optimal ratio combination in step S5, and subject the verification sample to a preset number of freeze-thaw cycles; test the measured values ​​of each target parameter of the verification sample after freeze-thaw cycles; calculate the individual relative error between the measured value of each target parameter and the corresponding target value; based on the weighting coefficients of each target parameter determined in step S4, perform a weighted summation of each individual relative error to obtain a weighted comprehensive relative error; when the weighted comprehensive relative error is less than or equal to a preset similarity error threshold, determine that the theoretically optimal ratio combination meets the overall similarity requirement, and use it as the final ratio. The target parameters of the verification sample after freeze-thaw cycles include at least four of the following: density, thermal conductivity, uniaxial compressive strength, elastic modulus, cohesion, and internal friction angle; the preset similarity error threshold is 5%~15%.

[0015] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0016] This invention presents a method for optimizing the mix proportions of slate-like materials used in cold regions. Existing methods for optimizing the mix proportions of similar rock materials primarily use mechanical parameters such as density, strength, and elastic modulus as control indicators, rarely incorporating thermal conductivity into the simultaneous optimization system. This results in the developed similar materials failing to accurately simulate the temperature field distribution and heat conduction behavior of the prototype slate in freeze-thaw model tests. This invention uses thermal conductivity, along with density, uniaxial compressive strength, elastic modulus, cohesion, and internal friction angle, as target parameters. It determines the weights of each parameter using the analytic hierarchy process (AHP) and constructs a weighted comprehensive relative error function. This ensures that the theoretically optimal mix proportions simultaneously consider both mechanical and thermal similarity, guaranteeing the realistic coupling of temperature and stress fields in freeze-thaw model tests of slate slopes in high-altitude cold regions.

[0017] Existing technologies often use single-index range analysis or weighted averages of various indices to determine the optimal mix ratio, failing to consider the varying importance of different target parameters to the needs of freeze-thaw model tests. This can easily lead to imbalances where some parameters are highly similar while key parameters deviate significantly. This invention introduces the analytic hierarchy process (AHP), assigning differentiated weights to each target parameter based on the actual needs of freeze-thaw model tests. The optimization objective is to minimize the weighted overall relative error, ensuring that the final mix ratio most closely approximates the comprehensive performance of the prototype slate, thus avoiding subjective biases caused by human experience-based judgment.

[0018] Existing studies mostly verify the parameter matching of similar materials under ambient temperature conditions, lacking systematic verification of the damage evolution characteristics of materials under freeze-thaw cycles. This can lead to accelerated deterioration or failure of the developed materials in actual freeze-thaw tests. This invention, after determining the theoretically optimal mix ratio, adds a freeze-thaw cycle verification step. The weighted relative error between the measured values ​​and the target values ​​after freeze-thaw cycles is used as the final judgment criterion, and the error threshold is controlled within the range of 5% to 15%. This ensures that the similar materials not only meet the similarity requirements in the initial state but also maintain the same damage evolution pattern as the prototype slate after multiple freeze-thaw cycles, improving the reliability and repeatability of the model test results.

[0019] A five-factor, four-level full factorial experiment requires 1024 trials, which is time-consuming, labor-intensive, and costly. This invention uses L16(4 5 The orthogonal array only requires 16 sets of experiments to obtain the main effects and trends of each factor. The regression prediction model is established by the least squares method to achieve quantitative prediction of parameters. Then, the theoretical optimal ratio is obtained by combining the constraint optimization solution, which reduces the experimental workload. At the same time, the freeze-thaw verification step makes up for the limited sample size of the orthogonal experiment, and achieves an effective balance between experimental efficiency and optimization accuracy. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0021] Figure 1 This is a sensitivity analysis graph of density and thermal conductivity, where the horizontal axis represents the horizontal values ​​of each influencing factor, and the vertical axis represents the mean value of the target parameter. Figure 1 (a) is a diagram showing the degree of influence of each factor on the density of similar materials. Figure 1 (b) is a graph showing the degree of influence of each factor on the thermal conductivity of similar materials;

[0022] Figure 2 This is a sensitivity analysis diagram of compressive strength versus elastic modulus, where... Figure 2 (a) is a graph showing the influence of various factors on the compressive strength of similar materials. Figure 2 (b) is a diagram showing the degree of influence of each factor on the elastic modulus of similar materials;

[0023] Figure 3 This is a sensitivity analysis diagram of cohesion and internal friction angle, where... Figure 3 (a) is a graph showing the influence of various factors on the cohesiveness of similar materials. Figure 3 (b) is a diagram showing the influence of various factors on the internal friction angle of similar materials;

[0024] Figure 4 The graph shows the regression equations for each parameter and a comparison between the calculated regression values ​​and the experimental values. The horizontal axis represents the experimental group number, and the vertical axis represents the values ​​of each target parameter.

[0025] Figure 5 This is a comparison chart of the retention rates of force-thermal parameters after freeze-thaw cycles between the original rock and similar materials. The horizontal axis represents the number of freeze-thaw cycles, and the vertical axis represents the parameter retention rate. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments and accompanying drawings. However, this does not limit the invention to the scope of the embodiments described. Experimental methods in the following embodiments that do not specify specific conditions were performed according to conventional methods and conditions, or as selected according to the product instructions.

[0027] The endpoints and any values ​​of the ranges disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed herein.

[0028] Example 1

[0029] Step S1: Magnetite powder is selected as aggregate, composite silicate cement as binder, and supplemented with silica fume, mica powder, silicon carbide powder, polycarboxylate superplasticizer, and defoamer as raw materials. Five influencing factors are determined: moisture content (A), aggregate-to-cement ratio (B), mass ratio of silica fume to binder (C), mass ratio of silicon carbide powder to aggregate (D), and mass ratio of mica powder to aggregate (E). Each factor has four levels, as shown in the table below.

[0030] Table 1

[0031]

[0032] Using L 16 (4 5 An orthogonal array was used to arrange the experiment, and a total of 16 groups of similar material samples with different ratios were prepared. For each group of samples, the raw materials were weighed according to the corresponding ratio, polycarboxylate superplasticizer and defoamer were added, and the mixture was stirred evenly. The mixture was then poured into a standard mold to form the sample, and after curing to the specified age, the sample was tested.

[0033] Step S2.1: The measured values ​​of six target parameters (density, thermal conductivity, uniaxial compressive strength, elastic modulus, cohesion, and internal friction angle) were tested on 16 groups of samples. The measured values ​​are shown in Table 2.

[0034] Table 2

[0035]

[0036] Table 2 shows that the density of similar materials ranges from 2.57 to 2.76 g / cm³. 3 The thermal conductivity is 1.52–2.95 W·m. -1 ·K -1 The uniaxial compressive strength ranges from 3.10 to 6.50 MPa, the elastic modulus from 381.1 to 712.7 MPa, the cohesion from 0.27 to 3.63 MPa, and the internal friction angle from 13.72° to 53.30°. The density variation range is relatively small, while the thermal conductivity, compressive strength, elastic modulus, cohesion, and internal friction angle all have considerable adjustment potential. The selected raw material system can achieve synergistic control of thermal parameters, mechanical parameters, and water absorption characteristics while maintaining the basic self-weight characteristics of similar materials.

[0037] Step S2.2: Range analysis is used to study the influence of five factors (A to E) on the density, thermal conductivity, compressive strength, elastic modulus, cohesion, and internal friction angle of similar materials. Based on the orthogonal experimental results, the mean and range of the experimental indicators at different levels of each factor are calculated, and sensitivity analysis diagrams are plotted. The larger the range, the more significant the impact of changes in the factor level on the corresponding indicator.

[0038] Step S2.21, Analysis of factors affecting density and thermal conductivity;

[0039] Depend on Figure 1 (a) It can be seen that the influence of each factor on the density of similar materials is in the order of A > E > D > C > B, with factor A having the largest range, which is 0.094 g·cm³. -3 The results indicate that moisture content has the most significant impact on density. As the level of factor A increases, the mean density first increases and then decreases, reaching its maximum at level 2 (A=8%). Factor E is the next most significant factor, with its mean density generally decreasing as the level of factor E changes. The effects of factors D, C, and B weaken sequentially, and their regulatory effects on density are relatively limited.

[0040] Depend on Figure 1 (b) It can be seen that the influence of each factor on the thermal conductivity of similar materials is in the order of E > A > C > B > D, with factor E having the largest range, which is 0.634 W·m. -1 ·K -1 Factor E has the most significant impact on thermal conductivity. As the level of factor E increases, the mean thermal conductivity generally shows a fluctuating downward trend. Factors A and C also have a certain impact on thermal conductivity; as the level of these factors increases, both generally show a downward trend. The ranges of factors B and D are relatively small, and their effect on regulating thermal conductivity within the scope of this experiment is relatively limited.

[0041] Step S2.22, Analysis of factors affecting compressive strength and elastic modulus;

[0042] Depend on Figure 2 (a) It can be seen that the influence of each factor on the uniaxial compressive strength of similar materials is in the order of B > C > E > A > D. Among them, factor B has the largest range, which is 1.625 MPa, and plays a major regulatory role in compressive strength. As the level of factor B increases, the mean compressive strength generally decreases, reaching its maximum value when B=6. The influence of factors C and E is next, and the mean compressive strength corresponding to them shows a trend of first decreasing and then increasing, and fluctuating decreasing, respectively. The ranges of factors A and D are relatively small, and their regulatory role in compressive strength is relatively limited.

[0043] Depend on Figure 2 (b) It can be seen that the influence of each factor on the elastic modulus of similar materials is in the order of B > D > C > A > E. Among them, factor B has the largest range, which is 154.550 MPa, and is the main factor affecting the elastic modulus. As the level of factor B increases, the mean elastic modulus first increases and then decreases. The influence of factors D and C is next, with the mean value corresponding to factor D fluctuating more, while the elastic modulus increases significantly at higher levels of factor C. The ranges of factors A and E are small, and their regulatory effect on the elastic modulus is relatively limited.

[0044] Step S2.23, Analysis of factors affecting shear strength parameters;

[0045] Depend on Figure 3 (a) It can be seen that the influence of each factor on the cohesion of similar materials is in the order of C > B > D > E > A. Among them, factor C has the largest range, which is 1.580 MPa, and its influence on cohesion is the most significant. As the level of factor C increases, the mean cohesion shows a trend of first increasing, then decreasing, and then increasing again, reaching a maximum value of 2.353 MPa when C=15%. The influence of factor B is the second largest, and its mean cohesion shows a trend of first increasing and then decreasing; the influence of factors D and E is relatively weak, and the range of factor A is the smallest.

[0046] Depend on Figure 3 (b) It can be seen that the influence of each factor on the internal friction angle of similar materials is in the order of C > B > A > D > E. Among them, factor C has the largest range, which is 18.008°, and plays a major role in regulating the internal friction angle. As the level of factor C increases, the mean internal friction angle first increases and then decreases, reaching its maximum value when C=10%. The influence of factors B and A is next, while the ranges of factors D and E are relatively small, and their influence on the internal friction angle is relatively limited.

[0047] Step S3: Establish a multiple regression prediction model;

[0048] Based on the sensitivity analysis results of each parameter, in order to further establish the empirical quantitative relationship between similar material proportioning factors and test parameters, this embodiment uses moisture content (A), bone glue ratio (B), silica fume / binder (C), silicon carbide micropowder / aggregate (D), and mica powder / aggregate (E) as independent variables and each test parameter as the dependent variable for regression analysis. Considering that the parameters do not all show a linear relationship with the factor level, this embodiment uses a multiple linear model and a multiple regression model containing pure quadratic terms for fitting, and determines the regression equation for each parameter by combining the goodness of fit of the fitting curve, model simplicity, and parameter variation law.

[0049] Multiple linear models:

[0050] Yk=b0 + b1A + b2B + b3C + b4D + b5E;

[0051] Multiple regression model containing pure quadratic terms:

[0052] Yk=b0 + b1A + b2B + b3C + b4D + b5E+b6A 2 +b7B 2 +b8C 2 +b9D 2 +b10E 2 ;

[0053] In the formula: Yk represents the k-th test parameter, corresponding to density, thermal conductivity, compressive strength, elastic modulus, cohesion, and internal friction angle, respectively; b0 is a constant term, and bi is the regression coefficient corresponding to each regression term. Considering the limited number of test groups in this embodiment, no interaction term is introduced into the model to avoid excessive model parameters leading to reduced stability of the fitting results.

[0054] The regression equations for each parameter and the comparison between the calculated regression values ​​and the experimental values ​​are shown below. Figure 4 As shown, the fitted values ​​of density, compressive strength, cohesion, and internal friction angle show a relatively consistent trend with the experimental values, and can well reflect the overall pattern of their variation with the proportioning factors. However, there are certain deviations between the fitted values ​​of thermal conductivity and elastic modulus and the experimental values. Therefore, the regression equations for thermal conductivity and elastic modulus are mainly used for trend characterization and to assist in the optimization of overall errors.

[0055] S4, Weight determination and construction of the weighted comprehensive relative error function;

[0056] Using the theoretically optimal mix design similarity as the target layer, and density, compressive strength, elastic modulus, cohesion, internal friction angle, and thermal conductivity as the criterion layers, a judgment matrix was constructed based on the influence of each index on self-weight stress, overall strength, deformation response, shear strength, and heat conduction process in the freeze-thaw cycle slope model test. Calculated using the analytic hierarchy process (AHP), the weights of density, compressive strength, elastic modulus, cohesion, internal friction angle, and thermal conductivity were 0.075, 0.136, 0.232, 0.175, 0.122, and 0.26, respectively. The maximum eigenvalue of the judgment matrix was λmax = 6.0836, the consistency index CI = 0.0167, and the consistency ratio CR = 0.0135 < 0.10, indicating that the judgment matrix passed the consistency test and the obtained weights showed good consistency.

[0057] Construct a weighted comprehensive relative error evaluation function:

[0058] In the formula, This is the weighted average relative error; Let be the predicted value of the i-th objective parameter of similar materials from the regression model; Let be the target value of the i-th target parameter; wi be the weight coefficient of the i-th target parameter, and satisfy . .

[0059] S5, constrained optimization to find the theoretically optimal ratio;

[0060] Substituting the results of 16 orthogonal experiments into the comprehensive error function, the F-value of each group was calculated. The results showed that the F-value of the 9th group was the smallest, indicating that the 9th group had the highest comprehensive matching degree with the target parameters among the orthogonal experimental combinations. Based on the multiple regression fitting equations for each parameter, the combinations A, B, C, D, and E corresponding to the minimum comprehensive error were solved using the fmincon function in MATLAB, thus obtaining the theoretically optimal combination of proportions.

[0061] The theoretically optimal proportions were calculated as follows: A (moisture content) = 12%, B (bone glue ratio) = 6, C (microsilica powder / binder) = 7.6%, D (silicon carbide micropowder / aggregate) = 15%, and E (mica powder / aggregate) = 3%. This proportion was obtained through optimization using a regression model and a weighted comprehensive error function, and C = 7.6% falls within the orthogonal experimental range (5%~10%).

[0062] Example 2

[0063] Based on Example 1, the theoretically optimal mix ratio obtained in Example 1 is a theoretically optimal mix ratio derived from orthogonal experimental data, regression model, and comprehensive evaluation function. Since the physical and mechanical parameters of similar materials are affected by factors such as mixing uniformity, molding density, and curing conditions, it is necessary to verify their similarity effect through re-sample preparation and parameter testing.

[0064] To verify the similarity of the theoretically optimal mix ratio obtained from regression optimization, similar material samples were re-prepared according to the theoretically optimal mix ratio: A (moisture content) = 12%, B (bone glue ratio) = 6, C (silica powder / binder) = 7.6%, D (silicon carbide powder / aggregate) = 15%, and E (mica powder / aggregate) = 3%. Their density, compressive strength, elastic modulus, cohesion, internal friction angle, and thermal conductivity were then tested. Using existing techniques in the study of reef limestone similar materials, this embodiment employs a relative error method to evaluate the matching degree of parameters between the original rock and similar materials. Firstly, single-item relative error is used to evaluate the matching degree between each index of the theoretically optimal mix ratio sample and the target parameters. Simultaneously, combined with the index weights determined earlier, a weighted comprehensive error is used to evaluate the overall similarity of the theoretically optimal mix ratio sample. The relative error calculation formula is as follows:

[0065]

[0066] In the formula: δi is the relative error of the i-th index; Xi is the measured value of the theoretically optimal mix ratio sample verification; X0i is the target value determined by the similarity criterion.

[0067] Weighted composite error: ;

[0068] In the formula: δ is the weighted comprehensive error; ẋi is the weight of the i-th indicator; and satisfies: ;

[0069] The results of the verification of the initial mechanical-thermal parameters of the theoretical optimal mix proportion are shown in Table 3.

[0070] Table 3

[0071]

[0072] Table 3 shows that the relative errors of all measured parameters of the theoretically optimal mix sample are less than 25%. Specifically, the relative errors for density, elastic modulus, and thermal conductivity are 4.74%, 1.95%, and 5.65%, respectively, indicating good matching with the target values ​​in terms of initial physical, deformation, and thermal parameters. The relative errors for compressive strength, cohesion, and internal friction angle are 16.82%, 24.66%, and 13.07%, respectively, indicating some deviation in strength parameters. Further calculation based on the index weights shows that the weighted overall error of the sample is approximately 10.48%, with higher-weighted indicators such as thermal conductivity and elastic modulus showing smaller errors and contributing significantly to overall similarity. In summary, the theoretically optimal mix exhibits good overall similarity under multi-index constraints and can basically meet the similarity requirements for density, thermal conductivity, and main mechanical parameters of slate-like materials in the initial state.

[0073] Example 3

[0074] Based on the above embodiments, to further evaluate the performance evolution characteristics of the theoretically optimal mix design sample under freeze-thaw cycles, this embodiment tests the thermal conductivity, elastic modulus, compressive strength, and longitudinal wave velocity of the samples after 0, 5, 10, 15, 20, 25, 30, 35, and 40 freeze-thaw cycles. Thermal conductivity is used to evaluate the stability of the material's thermal conductivity after freeze-thaw cycles; elastic modulus reflects the degradation of the material's deformation stiffness; compressive strength characterizes the deterioration of the material's macroscopic load-bearing capacity; and longitudinal wave velocity reflects the integrity of the material's internal structure, the degree of pore and fracture development, and the damage state. Considering that this embodiment will subsequently conduct freeze-thaw cycle similarity verification, longitudinal wave velocity is used as an auxiliary verification indicator to evaluate the freeze-thaw damage evolution law of similar materials. Considering the differences in the absolute values ​​of various parameters between the original rock and similar materials, this embodiment uses parameter retention rate to compare and analyze the freeze-thaw deterioration trends of the two. The results are shown in […]. Figure 5 .

[0075] Figure 5 The comparison shows the retention rate of force-thermal parameters after freeze-thaw cycles between the original rock and similar materials. The solid line represents the original rock, and the dashed line represents the similar material.

[0076] Depend on Figure 5It can be seen that with the increase of the number of freeze-thaw cycles, the thermal conductivity, elastic modulus, compressive strength, and longitudinal wave velocity retention rate of both the original rock and similar materials generally show a downward trend, indicating that the mechanical properties and thermal conductivity of both materials deteriorate to varying degrees under freeze-thaw cycles. From the variation pattern, the retention rates of various indicators in the similar materials show a relatively consistent decline trend with those of the original rock, suggesting that the theoretically optimal mix design can better reflect the evolution characteristics of the mechanical and thermal properties of slate under freeze-thaw cycles.

[0077] Therefore, the theoretically optimal ratio not only meets the requirements for similarity in density, thermal conductivity and main mechanical parameters in the initial state, but also has good performance evolution similarity after freeze-thaw cycles.

[0078] In summary, based on the requirements for optimizing the proportion of similar materials for slate in cold regions and verifying freeze-thaw cycles, this invention conducts a 5-factor, 4-level orthogonal experiment to analyze the influence of each proportion factor on the density, thermal conductivity, and mechanical parameters of the similar materials, and verifies the theoretically optimal proportion sample through freeze-thaw cycles.

[0079] The similar materials of this invention have adjustable ranges for density, thermal conductivity, compressive strength, elastic modulus, cohesion, and internal friction angle, which can meet the formulation requirements of slate-like materials for density, thermal conductivity, and mechanical parameters. Range analysis shows that, within the set factor levels, the main sensitive factors of each target parameter differ. Density is relatively sensitive to moisture content and mica powder / aggregate ratio; thermal conductivity is mainly affected by mica powder / aggregate ratio and silica powder / binder ratio; compressive strength and elastic modulus respond significantly to the bone-to-cement ratio; and cohesion and internal friction angle are greatly affected by silica powder / binder ratio and bone-to-cement ratio.

[0080] Based on the regression prediction results and index weights of the six target parameters, the theoretically optimal mix ratio was determined to be: moisture content 12%, bone glue ratio 6, silica fume / cementing agent 7.6%, silicon carbide micropowder / aggregate 15%, and mica powder / aggregate 3%. This method can adjust the weights according to the main control similarity index of the model test, providing a reference for determining the mix ratio of rock-like materials under different similarity targets.

[0081] Furthermore, the weighted average error of the initial target parameters of the verification sample was 10.48%, which is generally in good agreement with the target parameters. After freeze-thaw cycles, the thermal conductivity, elastic modulus, compressive strength and longitudinal wave velocity attenuation trend of the verification sample were basically consistent with the original rock, which can provide a material basis for model tests of slate slopes in high-altitude and cold regions.

[0082] The above description is merely some specific embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the proportions of slate-like materials from cold regions, characterized in that, Includes the following steps: S1, identify multiple influencing factors and levels affecting the properties of similar materials of slate in cold regions, and prepare multiple groups of samples with different ratios using orthogonal experimental design; S2, Physical, mechanical and thermal parameters of the multiple groups of samples are tested to obtain the measured values ​​of each target parameter, including density, thermal conductivity and mechanical parameters; based on the measured values, the mean and range of each influencing factor at each level are calculated using range analysis; the order of influence of each influencing factor on each target parameter is determined according to the magnitude of the range; and the influence law of each influencing factor on each target parameter is determined according to the trend of the mean of the target parameter changing with the factor level. S3. Based on the results of orthogonal experiments, establish a multiple regression prediction model between each influencing factor and each target parameter; S4. The weight coefficients of each objective parameter are determined by the analytic hierarchy process (AHP) and a weighted comprehensive relative error function is constructed. S5. Based on the multiple regression prediction model in step S3, constrained optimization is performed with the goal of minimizing the weighted comprehensive relative error to obtain the theoretical optimal combination of each influencing factor.

2. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, The raw materials for the sample included magnetite powder, composite silicate cement, silica fume, mica powder, silicon carbide powder, polycarboxylate superplasticizer and defoamer, with magnetite powder as aggregate and composite silicate cement as binder.

3. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, In step S1, the orthogonal experiment uses L 16 (4 5 An orthogonal array was used to design an orthogonal experiment with 5 influencing factors and 4 levels.

4. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, In step S3, the multiple regression prediction model is a multiple linear regression model or a multiple regression model containing pure quadratic terms, used to establish the quantitative relationship between influencing factors and each target parameter.

5. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, In step S4, the weighted comprehensive relative error evaluation function is: In the formula, This is the weighted average relative error; Let be the predicted value of the i-th objective parameter of similar materials from the regression model; Let i be the target value of the i-th target parameter; wi is the weight coefficient of the i-th target parameter, and satisfies .

6. The method for optimizing the proportions of a cold-region slate-like material according to claim 5, characterized in that, In step S5, The component ratio corresponding to the minimum is the theoretical optimal combination of each influencing factor.

7. The method for optimizing the proportions of a cold-region slate-like material according to claim 2, characterized in that, In step S1, the multiple influencing factors include: moisture content, bone glue ratio, mass ratio of microsilica powder to binder, mass ratio of silicon carbide micro powder to aggregate, and mass ratio of mica powder to aggregate; each influencing factor is set with 4 levels.

8. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, The target parameters include density, thermal conductivity, uniaxial compressive strength, elastic modulus, cohesion, and internal friction angle.

9. The method for optimizing the proportions of a cold-region slate-like material according to claim 1, characterized in that, The verification sample is prepared according to the theoretical optimal ratio combination in step S5, and the verification sample is subjected to a preset number of freeze-thaw cycles. The measured values ​​of each target parameter of the test sample after freeze-thaw cycle treatment were verified. Calculate the individual relative error between the verified measured value and the corresponding target value for each target parameter; Based on the weight coefficients of each target parameter determined in step S4, the weighted summation of each individual relative error is performed to obtain the weighted comprehensive relative error. When the weighted overall relative error is less than or equal to the preset similarity error threshold, the theoretically optimal combination is determined to meet the overall similarity requirement and is used as the final combination.

10. The method for optimizing the proportions of a cold-region slate-like material according to claim 9, characterized in that, The target parameters of the verification sample after the freeze-thaw cycle treatment include at least four of the following: density, thermal conductivity, uniaxial compressive strength, elastic modulus, cohesion, and internal friction angle; the preset similarity error threshold is 5% to 15%.