A multi-scale evaluation method and system for the mechanical strength of construction waste mixtures for roads

Abstract

The present invention discloses a multi-scale evaluation method and system for the mechanical strength of road-use construction waste mixtures, relating to the field of foundation construction technology. The method comprises: testing the mechanical properties of prepared test specimens; organizing and analyzing the mechanical property data obtained from the tests; establishing a mechanical strength evaluation model for the road-use construction waste mixture based on the test data and the results of the organization and analysis, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture; and performing a multi-scale evaluation of the mechanical strength of the road-use construction waste mixture based on the mechanical strength evaluation model. The present invention utilizes an autoregressive integral moving average model and a random forest algorithm, each taking into account multiple independent variables, to comprehensively predict mechanical strength. Furthermore, the method considers maintenance conditions and performs scenario-based mechanical strength prediction based on these conditions, thereby improving the accuracy of the final multi-scale evaluation of mechanical strength.

Description

Technical Field

[0001] The present invention relates to the field of foundation construction technology, and more particularly to a multi-scale evaluation method and system for the mechanical strength of road construction waste mixtures. Background Art

[0002] As an important renewable resource, construction waste plays a key role in urbanization. Typically, it includes abandoned soil, materials, slag, sludge, and other waste generated during the construction, renovation, or expansion of various buildings and structures. With the rapid development of urban construction, the amount of construction waste continues to increase, posing a serious challenge to the environment.

[0003] According to relevant data, the amount of construction waste generated globally each year has reached an alarming level, with most of it being landfilled or dumped in the open air, causing serious environmental pollution and waste of resources. In this context, finding efficient and sustainable methods for the reuse of construction waste has become particularly urgent.

[0004] The use of construction waste in road engineering is gaining increasing attention. However, methods for evaluating the mechanical strength of construction waste mixtures have not been widely studied or applied. The complex composition and heterogeneous mechanical properties of construction waste lead to uncertainties when using construction waste mixtures in road engineering.

[0005] Therefore, how to propose a multi-scale evaluation method and system for the mechanical strength of construction waste mixtures for roads and effectively evaluate the mechanical properties of construction waste mixtures in road engineering is an urgent problem that technical personnel in this field need to solve. Summary of the Invention

[0006] In view of this, the present invention provides a multi-scale evaluation method and system for the mechanical strength of construction waste mixtures for road use, which effectively evaluates the mechanical properties of construction waste mixtures in road engineering. To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] A multi-scale evaluation method for the mechanical strength of construction waste mixture for road use, comprising:

[0008] Aggregates are screened from construction waste and evenly mixed with inorganic binders according to a preset mix ratio to prepare standard mechanical test specimens;

[0009] Test the mechanical properties of the prepared specimens;

[0010] Organize and analyze the mechanical properties data obtained from the test;

[0011] Based on the test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, a mechanical strength evaluation model for road construction waste mixture was established;

[0012] A multi-scale evaluation of the mechanical strength of road construction waste mixture is performed based on the mechanical strength evaluation model.

[0013] Optionally, the aggregate includes: granular materials processed from concrete, mortar, stone, bricks and tiles in construction waste, which are crushed and sieved to obtain particles of different sizes;

[0014] The inorganic binder includes cement, lime and fly ash. The inorganic binder reacts chemically or physically with aggregate to form a road structure layer.

[0015] Optionally, the testing of mechanical properties of the prepared specimen includes: performing compressive strength, flexural strength and tensile strength mechanical property tests on the prepared specimen.

[0016] Optionally, arranging and analyzing the mechanical property data obtained from the test includes: arranging and analyzing the mechanical property data obtained from the mechanical property test, and using statistical methods to calculate the mean value, standard deviation and variance of each mechanical property.

[0017] Optionally, the mechanical strength evaluation model of the road construction waste mixture is established based on the test data and the collated analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, including:

[0018] Pre-process the test data and analysis results, as well as the mix ratio, raw material properties, and curing condition data of the construction waste mixture;

[0019] The autoregressive integrated moving average model is used to predict mechanical strength;

[0020] The random forest algorithm is used to predict the mechanical strength by considering multiple independent variables, and the final mechanical strength prediction result is formed by combining the autoregressive integral moving average model;

[0021] Considering the maintenance conditions, the maintenance conditions are selected for scenario mechanical strength prediction.

[0022] Optionally, the random forest algorithm is used to predict mechanical strength including:

[0023] Mechanical strength was used as the dependent variable, and test data and analysis results, as well as the mix ratio and raw material properties of the construction waste mixture were used as independent variables and input into the random forest model.

[0024] Set the number of random forest trees K, the training ratio, use MSE as the node splitting evaluation criterion, set sampling with replacement, set the maximum depth of the tree to a, set the maximum number of leaf nodes to b, and the node partition impurity threshold to c;

[0025] After the predicted random forest model is run, the parameters are adjusted and the random forest model is retrained based on whether the observation evaluation criteria are reasonable;

[0026] The independent variable data to be predicted is input into the random forest model to obtain the prediction results.

[0027] Optionally, it also includes: estimation accuracy verification of the random forest model, including actual sample point verification, the actual sample point verification data comes from monitoring data at different time points, and the accuracy verification indicators include the determination coefficient R 2 , root mean square error RMSE and mean relative error MRE, R 2 The higher it is, the smaller the RMSE and MRE are, which means the better the performance of the random forest model.

[0028] Optionally, the consideration of curing conditions and the selection of curing conditions for scenario mechanical strength prediction include:

[0029] Considering the maintenance conditions, the scenario mechanical strength prediction is carried out according to the preset quarter, and the historical high temperature weather times, monthly maximum temperature average, non-rainy days times, and rainy days times in the preset quarter are collected according to the preset quarter;

[0030] Normalize the preset seasonal climate characteristics and give a comprehensive score;

[0031] Determine the climate characteristics of the scenario, and use the determined climate characteristics as search conditions to find similar days to the day to be predicted, and weight the mechanical strength prediction results of the similar days with the final mechanical strength prediction results as the predicted mechanical strength of the prediction day in the preset quarter.

[0032] Optionally, weighting the mechanical strength prediction results of similar days and the final mechanical strength prediction results also includes calculating the mechanical strength prediction results of similar days and the final mechanical strength prediction results through Euclidean distance, and performing weighted calculation when the calculation threshold is met.

[0033] Optionally, a multi-scale evaluation system for the mechanical strength of road construction waste mixtures includes:

[0034] Proportioning module: used to select aggregates from construction waste and mix them evenly with inorganic binders according to the preset proportions to prepare standard mechanical test specimens;

[0035] Testing module: used to test the mechanical properties of prepared specimens;

[0036] Arrangement module: used to organize and analyze the mechanical properties data obtained from the test;

[0037] Model building module: used to establish a mechanical strength evaluation model for road construction waste mixtures based on test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture;

[0038] Evaluation module: used for performing multi-scale evaluation of the mechanical strength of road construction waste mixture based on the mechanical strength evaluation model.

[0039] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a multi-scale evaluation method and system for the mechanical strength of road construction waste mixtures, which has the following beneficial effects:

[0040] The present invention proposes a multi-scale evaluation method for the mechanical strength of a road construction waste mixture, comprising: screening aggregates from construction waste, and uniformly mixing them with an inorganic binder according to a preset mix ratio to prepare standard mechanical test specimens; testing the mechanical properties of the prepared specimens; collating and analyzing the mechanical property data obtained from the tests; establishing a mechanical strength evaluation model for the road construction waste mixture based on the test data and the collation and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture; and performing a multi-scale evaluation of the mechanical strength of the road construction waste mixture based on the mechanical strength evaluation model. The present invention discloses the use of an autoregressive integral sliding average model to predict mechanical strength; the use of a random forest algorithm to predict mechanical strength considering multiple independent variables, and the combination of the autoregressive integral sliding average model to form a final mechanical strength prediction result; the maintenance conditions are considered and the maintenance conditions are selected to predict the scene mechanical strength. The present invention uses an autoregressive integral sliding average model to predict mechanical strength, and combines it with the random forest algorithm to improve the accuracy of the final mechanical strength prediction result. By considering climate conditions and conducting multi-scenario mechanical strength prediction analysis based on seasons, it effectively responds to the impact of maintenance conditions on roads and achieves effective evaluation of the mechanical properties of construction waste mixtures in road engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0042] Figure 1This is a schematic flow chart of a multi-scale evaluation method for the mechanical strength of road construction waste mixtures provided by the present invention.

[0043] Figure 2 This is a structural framework diagram of a multi-scale evaluation system for the mechanical strength of road construction waste mixtures provided by the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] The embodiment of the present invention discloses a multi-scale evaluation method for the mechanical strength of road construction waste mixture, such as Figure 1 Shown, including:

[0046] Aggregates are screened from construction waste and evenly mixed with inorganic binders according to a preset mix ratio to prepare standard mechanical test specimens;

[0047] Test the mechanical properties of the prepared specimens;

[0048] Organize and analyze the mechanical properties data obtained from the test;

[0049] Based on the test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, a mechanical strength evaluation model for road construction waste mixture was established;

[0050] A multi-scale evaluation of the mechanical strength of road construction waste mixture is performed based on the mechanical strength evaluation model.

[0051] Furthermore, the aggregate includes: granular materials processed from concrete, mortar, stone, bricks and tiles in construction waste, which are crushed and sieved to obtain particles of different sizes;

[0052] The inorganic binder includes cement, lime, and fly ash. The inorganic binder reacts chemically or physically with aggregate to form a road structure layer with certain strength and stability.

[0053] Furthermore, the mechanical property testing of the prepared test specimens includes: performing compressive strength, flexural strength and tensile strength mechanical property testing on the prepared mechanical test specimens.

[0054] In a specific embodiment, the mechanical property test includes:

[0055] S1: Take representative aggregate and determine its air-dried moisture content;

[0056] S2: Weigh the mass of the test tube (m1), fix the test tube on the bottom plate, put the pad into the tube, put a piece of filter paper on the pad, and install the ring;

[0057] S3: Pour the prepared sample into the cylinder three times and compact it layer by layer;

[0058] S4: After every 3 compacted specimens, take a representative sample for moisture content test;

[0059] S5: Remove the collar, use a straight scraper to level the compacted specimen along the top of the test tube, repair any uneven surfaces with fine material, remove the pad, and weigh the mass of the test tube and specimen (m2);

[0060] S6: When CBR specimens are made by static pressing, the required sample volume is calculated based on the determined compaction degree and static pressing is performed once;

[0061] S7: Soak in water to measure expansion;

[0062] S8: Penetration test.

[0063] Furthermore, in S1, three specimens are prepared according to the optimal moisture content. After the aggregates are fully mixed with water, they are placed in a sealed container or plastic bag for soaking. The soaking time of clay soil shall not be less than 24 hours, the soaking time of silt soil can be shortened to 12 hours, the soaking time of sandy soil can be shortened to 6 hours, and the soaking time of natural gravel can be shortened to about 2 hours. When necessary, three dry density specimens can be prepared to control the dry density of the specimens between 90% and 100% of the maximum dry density. If three specimens of each dry density are prepared, a total of 9 specimens are prepared, and a total of about 55 kg of specimens are required for the 9 specimens. When compaction is used to form specimens, the number of blows per layer is generally 30, 50 and 98 respectively. When static pressing is used to form specimens, the required sample volume is calculated based on the determined compaction degree, and static pressing is performed once.

[0064] Furthermore, in S3, each layer requires about 1500 to 1750 g of sample, and the amount should be such that the compacted sample is 1 to 2 mm higher than 1 / 3 of the cylinder height; after compacting the large test cylinder, the sample should not be higher than 10 mm of the cylinder height.

[0065] Furthermore, the specific steps of S7 are as follows:

[0066] S71. After the specimen is prepared, remove the damaged filter paper from the top surface of the specimen, replace it with a new piece of filter paper, and install a porous plate with an adjustment rod on top. Place four load plates on top of the porous plate.

[0067] S72. Place the test tube and the porous plate into the tank (without adding water), tighten the mold with the tie rod, install the dial indicator, and take the initial reading.

[0068] S73. Fill the water tank with water until it covers the top of the test tube. During the soaking period, the water level in the tank should be maintained approximately 25 mm above the top of the test tube. The test piece should usually be soaked for 4 days and nights.

[0069] S74. At the end of soaking, read the final reading of the dial indicator on the specimen and calculate the expansion rate using the formula:

[0070]

[0071] Where: δ e -Expansion rate of the specimen after soaking in water, calculated to 0.1%;

[0072] H1-the height of the specimen after immersion in water (mm);

[0073] H0-initial height of specimen (mm);

[0074] S75. Remove the specimen from the water tank, pour out the water on the top surface of the specimen, let it stand for 15 minutes to allow it to drain, then remove the additional load and the porous plate, bottom plate and filter paper, and weigh it (m3) to calculate the change in humidity and density of the specimen.

[0075] Furthermore, the specific steps of S8 are as follows:

[0076] S81. A force ring of appropriate tonnage should be selected, and the force ring reading at the end of penetration should be at least 1 / 3 of its range;

[0077] S82. Level the surface of the specimen and ensure that the penetration rod is in full contact with the top surface of the specimen. Place four load plates around the penetration rod.

[0078] S83. First, apply a small load to the penetration rod to ensure close contact between the test specimen and the soil sample. Then, adjust the pointers of the force and deformation dial indicators to integers and record the initial readings.

[0079] S84. Apply load so that the penetration rod is pressed into the specimen at a speed of 1 to 1.25 mm / min. Simultaneously record the readings of the three dial indicators. Record the penetration amount at certain integer readings (e.g., 20, 40, 60) of the dial indicator in the dynamometer, and make sure that the penetration amount is 250×10 -2 mm, there can be more than 5 readings; therefore, the first reading in the dynamometer should be the penetration of 30×10 -2 mm or so.

[0080] Furthermore, in the result compilation, a pl relationship curve is plotted with unit pressure (p) as the horizontal axis and penetration (l) as the vertical axis, as shown in Figure (1). Curve 1 in the figure is appropriate. The initial section of curve 2 is a concave curve and needs to be corrected. To do this, draw a tangent line at the point of variable curvature, intersecting the vertical axis at point O'. O' is the corrected origin.

[0081] Furthermore, in the result collation, the bearing ratio (CBR) when the penetration is 2.5 mm and 5 mm is calculated according to the formula;

[0082]

[0083] Where, CBR is load-bearing ratio, calculated to 0.1%, and p is unit pressure (kPa).

[0084] Furthermore, in the result collation, the wet density of the specimen is calculated as follows:

[0085]

[0086] Where, ρ is the wet density of the specimen, calculated to 0.01 g / cm 3 , m 2 -The combined mass of the test tube and the test piece (g), m 1 -mass of the test tube (g), 2177-volume of the test tube (cm 3 ).

[0087] The dry density of the specimen is calculated as follows:

[0088]

[0089] Where, ρ d -Dry density of the specimen, calculated to 0.01 g / cm 3 ;w-water content of specimen (%).

[0090] The water absorption of the specimen after soaking in water is calculated as follows:

[0091] w a =m3-m2;

[0092] Where w a -The amount of water absorbed by the test piece after soaking in water (g), m3-the combined mass of the test tube and the test piece after soaking in water (g), m2-the combined mass of the test tube and the test piece (g).

[0093] Furthermore, the collating and analyzing the mechanical property data obtained from the test includes: collating and analyzing the mechanical property data obtained from the mechanical property test, and calculating the mean value, standard deviation and variance of each mechanical property using statistical methods.

[0094] Specifically, (1) Calculate the mean: The mean is the sum of all data points in the dataset divided by the number of data points: in: is the mean, n is the number of data points, x iis the i-th data point; (2) Calculate the variance: Variance is the average of the squares of the differences between the data points and the mean, which is used to measure the degree of dispersion of the data: Where: 2 is the variance, is the mean, n is the number of data points, x i is the i-th data point; (3) Calculate the standard deviation: The standard deviation is the square root of the variance and is used to measure the degree of dispersion of the data. The unit is consistent with the original data: Where: σ is the standard deviation, σ 2 is the variance, is the mean, n is the number of data points x i is the i-th data point.

[0095] Furthermore, the mechanical strength evaluation model of road construction waste mixture is established based on the test data and the collated analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, including:

[0096] Pre-process the test data and analysis results, as well as the mix ratio, raw material properties, and curing condition data of the construction waste mixture;

[0097] The autoregressive integrated moving average model is used to predict mechanical strength;

[0098] The random forest algorithm is used to predict the mechanical strength by considering multiple independent variables, and the final mechanical strength prediction result is formed by combining the autoregressive integral moving average model;

[0099] Considering the maintenance conditions, the maintenance conditions are selected for scenario mechanical strength prediction.

[0100] In a specific embodiment, the preprocessing includes outlier processing and normalization processing:

[0101] The Lagrange interpolation method is used to deal with outliers in the test data and analysis results, as well as the mix ratio, raw material properties, and curing conditions of the construction waste mixture. The specific operations are as follows:

[0102] Input the time point X and corresponding data Y of the time series;

[0103] Calculate the basis function L of the n-th order Lagrangian algorithm i (x);

[0104]

[0105] Among them, L i (x) is the basis function of the interpolation polynomial, x is the basis function L iThe unknown number on (x), x i and x j are the i-th and j-th interpolation values ​​between two normal points, i = 0, 1, 2...n, j = 0, 1, 2...n, n is the number of differences between the two points, which is an integer; For the continuous multiplication operation, Multiply from j=0 to j=n.

[0106] Calculate the interpolation function L of the n-th order Lagrange interpolation algorithm n (x);

[0107] Interpolation function L n The calculation expression of (x) is:

[0108]

[0109] Among them, y i is the data corresponding to the interpolation point, L n (x) is y i The nth-order interpolation polynomial, L i (x) is the basis function of the interpolation polynomial, that is, L i (x)y i Add from i=0 to i=n.

[0110] Input the interpolation point and obtain the corresponding data interpolation through the n-order Lagrange formula.

[0111] The normalization method used is the maximum normalization method, and the calculation expression is:

[0112]

[0113] The ARIMA model, short for Autoregressive Integrated Moving Average (ARIMA), is a well-known time series forecasting approach proposed by Box and Jenkins in the early 1970s. It is also known as the Box-Jenkins model or the Box-Jenkins method. ARIMA (p, d, q) is an autoregressive moving average model with differences. AR stands for autoregressive, p is the autoregressive term; MA stands for moving average, q is the number of moving average terms, and d is the number of differences required to make the time series stationary. The ARIMA model transforms a non-stationary time series into a stationary one, then regresses the dependent variable only on its lagged values ​​and the present and lagged values ​​of the random error term. Depending on whether the original series is stationary and the components included in the regression, ARIMA models can be classified into moving average (MA), autoregressive (AR), autoregressive moving average (ARMA), and ARIMA processes.

[0114] Furthermore, the use of the autoregressive integrated moving average model to predict mechanical strength includes using the ARIMA model to predict mechanical strength:

[0115] Construct the mechanical strength time series matrix b;

[0116] Mechanical strength prediction based on ARIMA model, including:

[0117] The mechanical strength time series b after dimensionality reduction k , based on the ARIMA model, the mechanical strength value at future moments is predicted. The ARIMA (p, d, q) model is:

[0118]

[0119] Where p is the order of the autoregressive model, d is the number of differences made to make the series stationary, q is the order of the moving average model, O is the lag operator, and the random variable β is k (n) is white noise, t i is the coefficient of the AR model, b i are the coefficients of the MA model.

[0120] Specifically, it also includes the stationarity test, d-order difference Until the mechanical strength series is stable.

[0121] In a specific implementation, the stationarity test is performed by using an ADF check method to perform a stationarity test on the original sequence. If the sequence cannot meet the stationarity condition, a differential transformation is performed to convert the non-stationary time series into a stationary time series.

[0122] Furthermore, the order p+q of the ARIMA model is determined. By analyzing the partial autocorrelation plot PACF and the autocorrelation plot ACF, the parameters of the ARIMA model are identified. Based on the obtained parameters, the ARIMA model is constructed and the model parameters are estimated online:

[0123] The model of the mechanical strength time series at time n is:

[0124] The mechanical strength value at the future n+t moment is predicted by the mechanical strength time series matrix at the nth moment, that is,

[0125] in, d kh It represents the hth parameter of the EMA model, and the mechanical strength prediction of the ARIMA model is used to evaluate the mechanical strength of road construction waste mixture on a time scale.

[0126] Furthermore, the random forest algorithm is used to predict the mechanical strength including:

[0127] Mechanical strength was used as the dependent variable, and test data and analysis results, as well as the mix ratio and raw material properties of the construction waste mixture were used as independent variables and input into the random forest model.

[0128] Set the number of random forest trees K, the training ratio, use MSE as the node splitting evaluation criterion, set sampling with replacement, set the maximum depth of the tree to a, set the maximum number of leaf nodes to b, and the node partition impurity threshold to c;

[0129] After the predicted random forest model is run, the parameters are adjusted and the random forest model is retrained based on whether the observation evaluation criteria are reasonable;

[0130] The independent variable data to be predicted is input into the random forest model to obtain a prediction result. This embodiment uses mechanical strength as the dependent variable, and inputs test data and analysis results, as well as the mix ratio and raw material properties of the construction waste mixture, into the random forest algorithm to predict the mechanical strength. This achieves a microscopic evaluation of the mechanical strength of road construction waste mixtures.

[0131] Furthermore, it also includes: estimating accuracy verification of random forest model, including actual sample point verification, the actual sample point verification data comes from monitoring data at different time points, and the accuracy verification index includes the determination coefficient R 2, root mean square error RMSE and mean relative error MRE, R 2 The higher it is, the smaller the RMSE and MRE are, which means the better the performance of the random forest model.

[0132] Furthermore, the consideration of the maintenance conditions and the selection of the maintenance conditions for scene mechanical strength prediction include:

[0133] Considering the maintenance conditions, the scenario mechanical strength prediction is carried out according to the preset quarter, and the historical high temperature weather times, monthly maximum temperature average, non-rainy days times, and rainy days times in the preset quarter are collected according to the preset quarter;

[0134] Normalize the preset seasonal climate characteristics and give a comprehensive score;

[0135] Determine the climate characteristics of the scenario, and use the determined climate characteristics as search conditions to find similar days to the day to be predicted, and weight the mechanical strength prediction results of the similar days with the final mechanical strength prediction results as the predicted mechanical strength of the prediction day in the preset quarter.

[0136] In a specific embodiment, the consideration of curing conditions and the selection of curing conditions for scene mechanical strength prediction include:

[0137] Considering the maintenance conditions, the third quarter was selected to conduct scenario mechanical strength prediction based on the quarter, and the number of high temperature days, the average monthly maximum temperature, the number of non-rainy days, and the number of rainy days in the third quarter were counted respectively;

[0138] Normalize the climate characteristics of the third quarter and give a comprehensive score;

[0139] Determine the climate characteristics of the scenario plan, including: rapid growth of maintenance condition demand and slow growth of maintenance condition demand. Among them, the scenario of rapid growth of maintenance condition demand: take the climate conditions of the month with the highest historical annual score;

[0140] Moderate growth plan for maintenance condition requirements: take the climate conditions of the median month with a monthly historical annual score;

[0141] Based on the determined climate characteristics, similar days to the day to be predicted are searched as search conditions, and the mechanical strength prediction results of the similar days and the final mechanical strength prediction results are weighted as the predicted mechanical strength of the prediction day in the third quarter.

[0142] Furthermore, the weighting of the mechanical strength prediction results of similar days and the final mechanical strength prediction results also includes calculating the mechanical strength prediction results of similar days and the final mechanical strength prediction results through Euclidean distance, and performing weighted calculation when the calculation threshold is met.

[0143] In a specific embodiment, a multi-scale evaluation system for the mechanical strength of road construction waste mixture is provided. Figure 2 Shown, including:

[0144] Proportioning module: used to select aggregates from construction waste and mix them evenly with inorganic binders according to the preset proportions to prepare standard mechanical test specimens;

[0145] Testing module: used to test the mechanical properties of prepared specimens;

[0146] Arrangement module: used to organize and analyze the mechanical properties data obtained from the test;

[0147] Model building module: used to establish a mechanical strength evaluation model for road construction waste mixtures based on test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture;

[0148] Evaluation module: used for performing multi-scale evaluation of the mechanical strength of road construction waste mixture based on the mechanical strength evaluation model.

[0149] The specific implementation also includes a verification and optimization module, which is used to apply the evaluation model to actual road projects. By comparing actual use results with predicted results, the evaluation model is verified and optimized. Furthermore, the evaluation method and process are continuously improved and adjusted based on feedback from project practice.

[0150] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0151] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-scale evaluation method for the mechanical strength of road construction waste mixtures, characterized in that: include: Aggregates are screened from construction waste and evenly mixed with inorganic binders according to a preset mix ratio to prepare standard mechanical test specimens; Test the mechanical properties of the prepared specimens; Organize and analyze the mechanical properties data obtained from the test; Based on the test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, a mechanical strength evaluation model for road construction waste mixture was established; The mechanical strength evaluation model of road construction waste mixture is established based on the test data and collated analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, including: Pre-process the test data and analysis results, as well as the mix ratio, raw material properties, and curing condition data of the construction waste mixture; The autoregressive integrated moving average model is used to predict mechanical strength; The random forest algorithm is used to predict the mechanical strength by considering multiple independent variables, and the final mechanical strength prediction result is formed by combining the autoregressive integral moving average model; Considering the maintenance conditions, the maintenance conditions are selected to predict the mechanical strength of the scene; The prediction of mechanical strength using random forest algorithm includes: Mechanical strength was used as the dependent variable, and test data and analysis results, as well as the mix ratio and raw material properties of the construction waste mixture were used as independent variables and input into the random forest model. Set the number of random forest trees K, the training ratio, use MSE as the node splitting evaluation criterion, set sampling with replacement, set the maximum depth of the tree to a, set the maximum number of leaf nodes to b, and the node partition impurity threshold to c; After the predicted random forest model is run, the parameters are adjusted and the random forest model is retrained based on whether the observation evaluation criteria are reasonable; Input the independent variable data to be predicted into the random forest model to obtain the prediction results; It also includes: estimation accuracy verification of the random forest model, including actual sample point verification, the actual sample point verification data comes from monitoring data at different time points, and the accuracy verification indicators include the determination coefficient R 2 , root mean square error RMSE and mean relative error MRE, R 2 The higher it is, the smaller the RMSE and MRE are, which means the performance of the random forest model is better; The consideration of the curing conditions and the selection of curing conditions for scenario mechanical strength prediction include: Considering the maintenance conditions, the scenario mechanical strength prediction is carried out according to the preset quarter, and the historical high temperature weather times, monthly maximum temperature average, non-rainy days times, and rainy days times in the preset quarter are collected according to the preset quarter; Normalize the preset seasonal climate characteristics and give a comprehensive score; Determine the climate characteristics of the scenario, and use the determined climate characteristics as search conditions to find similar days to the day to be predicted, and weight the mechanical strength prediction results of the similar days with the final mechanical strength prediction results as the predicted mechanical strength of the prediction day in the preset quarter; The weighting of the mechanical strength prediction results of similar days and the final mechanical strength prediction results further includes calculating the mechanical strength prediction results of similar days and the final mechanical strength prediction results by Euclidean distance, and performing weighted calculation when a calculation threshold is met; A multi-scale evaluation of the mechanical strength of road construction waste mixture is performed based on the mechanical strength evaluation model.

2. The multi-scale evaluation method for mechanical strength of road construction waste mixture according to claim 1, characterized in that: The aggregate includes: granular materials processed from concrete, mortar, stone, bricks and tiles in construction waste, which are crushed and sieved to obtain particles of different particle sizes; The inorganic binder includes cement, lime and fly ash. The inorganic binder reacts chemically or physically with aggregate to form a road structure layer.

3. The multi-scale evaluation method for mechanical strength of road construction waste mixture according to claim 1, characterized in that: The mechanical property testing of the prepared specimens includes: performing compressive strength, flexural strength and tensile strength mechanical property testing on the prepared specimens.

4. The multi-scale evaluation method for mechanical strength of road construction waste mixture according to claim 1, characterized in that: The arranging and analyzing the mechanical property data obtained from the test includes: arranging and analyzing the mechanical property data obtained from the mechanical property test, and calculating the average value, standard deviation and variance of each mechanical property by using a statistical method.

5. A multi-scale evaluation system for the mechanical strength of road construction waste mixtures, characterized by: include: Proportioning module: used to select aggregates from construction waste and mix them evenly with inorganic binders according to the preset proportions to prepare standard mechanical test specimens; Testing module: used to test the mechanical properties of prepared specimens; Arrangement module: used to organize and analyze the mechanical properties data obtained from the test; Model building module: used to establish a mechanical strength evaluation model for road construction waste mixtures based on test data and analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture; The mechanical strength evaluation model of road construction waste mixture is established based on the test data and collated analysis results, as well as the mix ratio, raw material properties, and maintenance conditions of the construction waste mixture, including: Pre-process the test data and analysis results, as well as the mix ratio, raw material properties, and curing condition data of the construction waste mixture; The autoregressive integrated moving average model is used to predict mechanical strength; The random forest algorithm is used to predict the mechanical strength by considering multiple independent variables, and the final mechanical strength prediction result is formed by combining the autoregressive integral moving average model; Considering the maintenance conditions, the maintenance conditions are selected to predict the mechanical strength of the scene; The prediction of mechanical strength using random forest algorithm includes: Mechanical strength was used as the dependent variable, and test data and analysis results, as well as the mix ratio and raw material properties of the construction waste mixture were used as independent variables and input into the random forest model. Set the number of random forest trees K, the training ratio, use MSE as the node splitting evaluation criterion, set sampling with replacement, set the maximum depth of the tree to a, set the maximum number of leaf nodes to b, and the node partition impurity threshold to c; After the predicted random forest model is run, the parameters are adjusted and the random forest model is retrained based on whether the observation evaluation criteria are reasonable; Input the independent variable data to be predicted into the random forest model to obtain the prediction results; It also includes: estimation accuracy verification of the random forest model, including actual sample point verification, the actual sample point verification data comes from monitoring data at different time points, and the accuracy verification indicators include the determination coefficient R 2 , root mean square error RMSE and mean relative error MRE, R 2 The higher it is, the smaller the RMSE and MRE are, which means the performance of the random forest model is better; The consideration of the curing conditions and the selection of curing conditions for scenario mechanical strength prediction include: Considering the maintenance conditions, the scenario mechanical strength prediction is carried out according to the preset quarter, and the historical high temperature weather times, monthly maximum temperature average, non-rainy days times, and rainy days times in the preset quarter are collected according to the preset quarter; Normalize the preset seasonal climate characteristics and give a comprehensive score; Determine the climate characteristics of the scenario, and use the determined climate characteristics as search conditions to find similar days to the day to be predicted, and weight the mechanical strength prediction results of the similar days with the final mechanical strength prediction results as the predicted mechanical strength of the prediction day in the preset quarter; The weighting of the mechanical strength prediction results of similar days and the final mechanical strength prediction results further includes calculating the mechanical strength prediction results of similar days and the final mechanical strength prediction results by Euclidean distance, and performing weighted calculation when a calculation threshold is met; Evaluation module: used for performing multi-scale evaluation of the mechanical strength of road construction waste mixture based on the mechanical strength evaluation model.

Images