A method for predicting the skid resistance of asphalt mixtures that combines aggregate mineralogy and texture characteristics

Abstract

The application relates to a method for predicting the skid resistance of asphalt mixture by fusing the aggregate mineral and texture characteristics, and aims to solve the problems of time-consuming and laborious long-term skid resistance test of the existing asphalt mixture, human and environmental influence and insufficient accuracy.The prediction method comprises the following steps: 1, obtaining the aggregate average hardness parameter AHP and measuring the SiO2 / Al2O3 index in the mineral composition of the aggregate; 2, obtaining the aggregate texture parameter root mean square height, kurtosis and fractal dimension; 3, performing correlation analysis on the aggregate characteristic parameters, and selecting the parameters without correlation as the aggregate characteristic parameters; and 4, establishing a multiple linear regression model of the skid resistance of the asphalt mixture through the aggregate characteristic parameters, and calculating the skid resistance of the asphalt mixture through the multiple linear regression model.The application introduces the aggregate mineral chemical parameters on the basis of considering the texture index, establishes a multiple linear regression model by means of SPSS software to predict the skid resistance of the asphalt mixture, and improves the accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of road materials, and particularly relates to a method for predicting the skid resistance of asphalt mixture by fusing the mineral and texture characteristics of aggregate. BACKGROUND

[0002] As an important component of asphalt mixture, the performance of aggregate is crucial to the skid resistance of asphalt mixture. Compared with short-term performance, the long-term skid resistance of asphalt pavement is more critical to driving safety. Although the skid resistance of asphalt mixture can generally be characterized by indoor / field tests, the accuracy and repeatability of the test are inevitably affected by changes in operating conditions (human factors, environmental conditions, etc.). Therefore, more and more attention is paid to predicting the long-term skid resistance of asphalt mixture based on short-term data. The commonly used skid resistance prediction model is composed of a tire prediction model and a pavement prediction model. From the perspective of pavement, this study predicts the skid resistance of asphalt pavement based on the mineral, chemical and texture characteristics of aggregate by means of a multiple linear regression model. SUMMARY

[0003] The present application is to solve the problem of time-consuming and laborious long-term skid resistance test of existing asphalt mixture, which is affected by human factors and environmental conditions, and has insufficient accuracy, and provides a method for predicting the skid resistance of asphalt mixture by fusing the mineral and texture characteristics of aggregate.

[0004] The method for predicting the skid resistance of asphalt mixture by fusing the mineral and texture characteristics of aggregate according to the present application is realized according to the following steps:

[0005] I. Obtaining aggregate mineral parameters

[0006] The mineral characteristics of single aggregate asphalt mixture are evaluated by the aggregate average hardness parameter AHP, and the calculation formula of the average hardness parameter AHP is shown as formula (1)-(3);

[0007] AHP = dmp + Cd (1)

[0008] dmp = ådvi * pi (2)

[0009] Cd = dvp / min(dvi) (3)

[0010] In the formula: AHP is the average hardness parameter of aggregate, dmp is the average hardness of aggregate, Cd is the hardness ratio of aggregate, dvi is the Mohs hardness of each mineral constituting the aggregate, pi is the mass percentage of each mineral constituting the aggregate, dvp is the Mohs hardness of the hardest mineral in the aggregate, and min(dvi) is the smallest Mohs hardness of the minerals constituting the aggregate;

[0011] The SiO2 / Al2O3 index in the mineral composition of the aggregate is determined by X-ray fluorescence spectroscopy (XRF);

[0012] II. Obtaining the aggregate texture parameters

[0013] The aggregate texture parameters include the root mean square height R q , the kurtosis R ku , and the fractal dimension D.

[0014] The root mean square height R q represents the root mean square of the average values of the heights of all aggregate profiles along the reference length direction, and the calculation formula of the root mean square height R q is shown in equation (4).

[0015]

[0016] where N is the total number of cross-section profile sampling points, and Z n is the height (μm) of the aggregate profile in the reference length direction.

[0017] The kurtosis R ku uses the fourth power of the root mean square height to show the dimensionless fourth power of the reference length Z(x) (the reference length is consistent with the reference length in the root mean square height), and the calculation formula of the kurtosis R ku is shown in equation (5).

[0018]

[0019] The fractal dimension D is used to quantitatively evaluate the roughness of the aggregate surface, and the calculation formula of the fractal dimension D is shown in equation (6).

[0020]

[0021] In the formula, r is the edge length of a cube, and N(r) is the number of small cubes required to cover the object.

[0022] III. Correlation analysis of aggregate characteristic parameters

[0023] The correlation analysis is performed on the aggregate average hardness parameter AHP and the SiO2 / Al2O3 index, and the correlation analysis is performed between the root mean square height R q , the kurtosis R ku , and the fractal dimension D. From the aggregate average hardness parameter AHP, the SiO2 / Al2O3 index, the root mean square height R q , the kurtosis R ku , and the fractal dimension D, the parameter without correlation is selected as the aggregate characteristic parameter.

[0024] IV. Establishment of a multiple linear regression model

[0025] The aggregate, mineral powder and asphalt are mixed to prepare asphalt mixture test pieces, the dynamic friction coefficient mu of the asphalt mixture test pieces is tested, the multiple linear regression model of the skid resistance of the asphalt mixture is established through the aggregate characteristic parameters in step three, then the skid resistance of the asphalt mixture is estimated and calculated through the multiple linear regression model, so that the estimation method of the skid resistance of the asphalt mixture fusing the mineral and texture characteristics of the aggregate is completed.

[0026] The estimation method of the skid resistance of the asphalt mixture fusing the mineral and texture characteristics of the aggregate mainly includes the processes of aggregate mineral parameter acquisition, aggregate texture parameter acquisition, aggregate characteristic parameter correlation analysis and multiple linear regression model establishment, and the parameter without correlation is selected from the aggregate average hardness parameter AHP, SiO2 / Al2O3 index, root mean square height R q , kurtosis R ku and fractal dimension D as the aggregate characteristic parameter.

[0027] The traditional road surface skid resistance prediction model is mainly based on texture and does not consider the influence of the chemical composition of the aggregate, the present application introduces the chemical parameters of the aggregate on the basis of considering the texture index, and establishes a multiple linear regression model by means of SPSS software to predict the skid resistance of the asphalt mixture, which provides a new way for saving manpower and material resources and quickly obtaining the skid resistance of the asphalt pavement. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 It is a root mean square height R q test graph of four different aggregates in the embodiment;

[0029] Figure 2 It is a fractal dimension D test graph of four different aggregates in the embodiment;

[0030] Figure 3 It is a kurtosis R ku test graph of four different aggregates in the embodiment;

[0031] Figure 4 It is a skid resistance test graph of different types of asphalt mixtures in the embodiment. DETAILED DESCRIPTION

[0032] Detailed implementation one: the estimation method of the skid resistance of the asphalt mixture fusing the mineral and texture characteristics of the aggregate in the embodiment is implemented according to the following steps:

[0033] I. Aggregate mineral parameter acquisition

[0034] The aggregate mineral characteristics of single aggregate asphalt mixture are evaluated through the aggregate average hardness parameter AHP, and the calculation formula of the average hardness parameter AHP is shown as formula (1)-(3);

[0035] AHP = dmp + Cd (1)

[0036] dmp = ∑dvi * pi (2)

[0037] Cd = dvp / min(dvi) (3)

[0038] wherein AHP is the aggregate average hardness parameter, dmp is the average hardness of the aggregate, Cd is the aggregate hardness ratio, dvi is the Mohs hardness of each mineral that constitutes the aggregate, pi is the mass percentage of each mineral that constitutes the aggregate, dvp is the Mohs hardness of the hardest mineral in the aggregate, and min(dvi) is the smallest Mohs hardness of the minerals that constitute the aggregate;

[0039] The SiO2 / Al2O3 index of the mineral composition of the aggregate is determined by X-ray fluorescence spectroscopy (XRF);

[0040] II. Aggregate texture parameter acquisition

[0041] The aggregate texture parameters include the root mean square height R q , the kurtosis R ku , and the fractal dimension D.

[0042] The root mean square height R q represents the root mean square of the average height of all aggregate profiles along the reference length direction, and the calculation formula of the root mean square height R q is shown in equation (4).

[0043]

[0044] wherein N is the total number of cross-section profile sampling points, and Z n is the height (μm) of the aggregate profile in the reference length direction.

[0045] The kurtosis R ku uses the fourth power of the root mean square height to show the dimensionless fourth power of the reference length Z(x) (the reference length is consistent with the reference length in the root mean square height), and the calculation formula of the kurtosis R ku is shown in equation (5).

[0046]

[0047] The fractal dimension D is used to quantitatively evaluate the roughness of the aggregate surface, and the calculation formula of the fractal dimension D is shown in equation (6).

[0048]

[0049] wherein r is the side length of a cube, and N(r) is the number of small cubes required to cover the object.

[0050] Three, aggregate characteristic parameter correlation analysis

[0051] Correlation analysis is performed on the aggregate average hardness parameter AHP and the SiO2 / Al2O3 index, and correlation analysis is performed between the root mean square height R q , the kurtosis R ku and the fractal dimension D, and the parameter without correlation is selected from the aggregate average hardness parameter AHP, the SiO2 / Al2O3 index, the root mean square height R q , the kurtosis R ku and the fractal dimension D as the aggregate characteristic parameter.

[0052] Four, establishment of a multiple linear regression model

[0053] The aggregate, mineral powder and asphalt are mixed to prepare asphalt mixture specimens, the dynamic friction coefficient μ of the asphalt mixture specimens is tested, the aggregate characteristic parameters in step three are used to establish a multiple linear regression model of the asphalt mixture skid resistance, and then the multiple linear regression model is used to estimate and calculate the asphalt mixture skid resistance, thereby completing the method for estimating the skid resistance of asphalt mixture that integrates the mineral and texture characteristics of aggregate.

[0054] Specific implementation method two: the difference between this implementation method and the specific implementation method one is that the aggregate in step one is artificial aggregate or natural aggregate.

[0055] Specific implementation method three: the difference between this implementation method and the specific implementation method two is that the artificial aggregate is calcined bauxite, and the natural aggregate is basalt or limestone.

[0056] Specific implementation method four: the difference between this implementation method and any one of the specific implementation methods one to three is that the SiO2 / Al2O3 index in step two is obtained by X-ray fluorescence spectrometer.

[0057] Specific implementation method five: the difference between this implementation method and any one of the specific implementation methods one to four is that the texture image of the aggregate surface is obtained by a laser microscope in step two to obtain the root mean square height R q , the kurtosis R ku and the fractal dimension D.

[0058] Specific implementation method six: the difference between this implementation method and any one of the specific implementation methods one to five is that in the correlation analysis process in step three, when the correlation coefficient R 2 > 0.7, it indicates that the two parameters have correlation, and when the correlation coefficient R 2 < 0.7, it indicates that the two parameters have no correlation.

[0059] Specific embodiment seven: the difference between this embodiment and one of the specific embodiments one to six is that the aggregate, mineral powder and asphalt are mixed in step four, and the asphalt mixture test piece is obtained after compaction treatment.

[0060] Specific embodiment eight: the difference between this embodiment and one of the specific embodiments one to seven is that the dynamic friction coefficient μ of the asphalt mixture test piece is tested by a dynamic friction coefficient tester in step four.

[0061] Specific embodiment nine: the difference between this embodiment and one of the specific embodiments one to eight is that the aggregate characteristic parameters obtained in step three are used as independent variables for regression by using the multiple linear regression module in the SPSS software (Statistical Product and Service Solutions software) in step four.

[0062] Specific embodiment ten: the difference between this embodiment and specific embodiment nine is that when the statistical value p of the independent variable is less than 0.05, it indicates that the independent variable has a significant effect on the skid resistance of the asphalt mixture.

[0063] Embodiment: the asphalt mixture skid resistance prediction method of this embodiment fuses the mineral and texture characteristics of the aggregate, and is implemented according to the following steps:

[0064] I. Aggregate mineral parameter acquisition

[0065] In this embodiment, four different aggregates are selected, and the four aggregates are 88# and 75# calcined bauxite two kinds of artificial aggregates and two kinds of natural aggregates basalt and limestone;

[0066] The aggregate mineral characteristics of the single aggregate asphalt mixture are evaluated by the aggregate average hardness parameter AHP, and the calculation formula of the average hardness parameter AHP is shown as formula (1)-(3);

[0067] AHP=dmp+Cd (1)

[0068] dmp=∑dvi*pi (2)

[0069] Cd=dvp / min(dvi) (3)

[0070] In the formula: AHP is the aggregate average hardness parameter, dmp is the average hardness of the aggregate, Cd is the aggregate hardness ratio, dvi is the Mohs hardness of each mineral constituting the aggregate, pi is the mass percentage of each mineral constituting the aggregate, dvp is the Mohs hardness of the hardest mineral in the aggregate, and min(dvi) is the smallest Mohs hardness of the minerals constituting the aggregate;

[0071] The SiO2 / Al2O3 index of the mineral composition of the aggregate is obtained by X-ray fluorescence spectroscopy (XRF) measurement;

[0072] II. Aggregate texture parameter acquisition

[0073] The aggregate texture parameters include root mean square height R q , kurtosis R ku and fractal dimension D.

[0074] The root mean square height R q represents the root mean square of the average value of the profile height of all aggregates along the reference length direction, and the calculation formula of the root mean square height R q is shown in equation (4).

[0075]

[0076] Where N is the total number of profile sampling points, and Z n is the height of the aggregate profile in the reference length direction.

[0077] The kurtosis R ku uses the fourth power of the root mean square height to show the dimensionless fourth power of the reference length Z(x), and the calculation formula of the kurtosis R ku is shown in equation (5).

[0078]

[0079] The fractal dimension D is used to quantitatively evaluate the roughness of the aggregate surface, and the calculation formula of the fractal dimension D is shown in equation (6).

[0080]

[0081] In the formula, r is the edge length of the cube, and N(r) is the number of small cubes required to cover the object.

[0082] III. Correlation analysis of aggregate characteristic parameters

[0083] The Origin software is used to perform correlation analysis on the aggregate average hardness parameter AHP and the SiO2 / Al2O3 index, and correlation analysis is performed between the root mean square height R q , the kurtosis R ku and the fractal dimension D.

[0084] The AHP and the SiO2 / Al2O3 index are both mineral and chemical parameters of the aggregate, and after correlation analysis, the adjusted R 2 is -0.09, indicating that the AHP and the SiO2 / Al2O3 index have no significant correlation and are two independent variables.

[0085] The root mean square height R q , the kurtosis R ku and the fractal dimension D are all texture parameters of the aggregate, and the correlation results are shown in Table 3. As can be seen from Table 3, R ku and Rq The correlation coefficient was 0.034, indicating that R... ku With R q There is no significant correlation; they are two independent variables. R ku The correlation coefficient with D is 0.234, indicating that R... ku There is no significant correlation between D and R, indicating they are two independent variables; however, D and R... q The correlation coefficient is 0.894, indicating that D and R... q There is a significant correlation between D and R. q For the same aggregate texture feature parameters;

[0086] The average hardness parameter AHP, SiO2 / Al2O3 index, and root mean square height R of the aggregate were analyzed. q and peak state R ku Four parameters serve as aggregate characteristic parameters;

[0087] IV. Establishment of a Multiple Linear Regression Model

[0088] Four different aggregates were mixed with mineral powder and asphalt respectively, poured into molds, and compacted using a hand-push roller to obtain asphalt mixture specimens. The controlled asphalt-aggregate ratios for the four specimens were 6.51%, 6.58%, 6.02%, and 6.04%, respectively. The dynamic friction coefficient μ of the asphalt mixture specimens was then measured. Based on the aggregate characteristic parameters from step three, a multiple linear regression model for the skid resistance of the asphalt mixture was established. The skid resistance of the asphalt mixture was then predicted using this model, thus completing a method for predicting the skid resistance of asphalt mixtures that integrates aggregate mineral and texture characteristics.

[0089] Table 1 shows the chemical composition of different types of aggregates. As can be seen from the table, the chemical composition of calcined bauxite is significantly different from that of natural aggregates. Calcined bauxite is mainly composed of SiO2 and Al2O3, while natural aggregates are mainly composed of CaO and SiO2.

[0090] Table 2 shows the main mineral parameters of different aggregates. Similar to their chemical composition, calcined bauxite is mainly composed of corundum and mullite, while natural aggregates contain plagioclase, amphibole, calcite, and dolomite, among others.

[0091] Figures 1-3 The figure shows the texture parameters of different aggregates. As can be seen from the figure, in most cases, the texture parameters of calcined bauxite are better than those of natural aggregates.

[0092] The present embodiment selects the root mean square height Rq, the peak state Rku, AHP and SiO2 / Al2O3 as the independent variables for representing the aggregate morphology and mineral and chemical characteristics. The multiple linear regression module in the SPSS software is used to regress the four independent variables affecting the skid resistance of the asphalt mixture, and the dynamic friction coefficient μ is used to represent the skid resistance of the asphalt mixture, as shown in Table 4. Generally, if the statistical value (p) of the independent variable is less than 0.05, it indicates that the factor has a significant effect on the skid resistance of the asphalt mixture. As can be seen from the table, the coefficients of AHP, SiO2 / Al2O3 and Rku are 0.000, 0.001 and 0.000 respectively, all of which are less than 0.05, indicating that AHP, SiO2 / Al2O3 and Rku have a significant effect on the value of μ. However, in terms of Rq, the coefficient is too large and should be excluded. Therefore, the multiple linear regression model is shown in equation (7). The statistical p value of each independent variable is less than 0.05, which has a significant effect on the friction coefficient of the pavement.

[0093] The multiple linear regression model obtained in the present embodiment is shown in equation (7) as follows:

[0094] μ = -0.894 + 0.004AHP - 0.009 SiO2 / Al2O3 + 0.525 Rku (7)

[0095] Table 1 Main chemical composition of different aggregates

[0096]

[0097] Table 2 Main mineral parameters of different aggregates

[0098]

[0099] Table 3 Pearson correlation coefficient matrix of different texture indexes

[0100]

[0101] Table 4 Multiple linear regression model

[0102]

Claims

1. A method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics, characterized in that... The estimation method is implemented according to the following steps: I. Acquisition of aggregate mineral parameters The aggregate mineral properties of a single aggregate asphalt mixture are evaluated by the average hardness parameter (AHP). The calculation formulas for the average hardness parameter (AHP) are shown in equations (1)-(3). AHP=dmp+Cd (1) dmp=∑dvi*pi (2) Cd=dvp / min(dvi) (3) In the formula: AHP is the average hardness parameter of the aggregate, dmp is the average hardness of the aggregate, Cd is the hardness ratio of the aggregate, dvi is the Mohs hardness of each mineral that makes up the aggregate, pi is the mass percentage of each mineral that makes up the aggregate, dvp is the Mohs hardness of the mineral with the highest hardness in the aggregate, and min(dvi) is the minimum Mohs hardness of the mineral that makes up the aggregate. The SiO2 / Al2O3 ratio in the mineral composition of the aggregate was determined by X-ray fluorescence spectroscopy. II. Obtaining Aggregate Texture Parameters Aggregate texture parameters include root mean square height R q , peak state R ku and fractal dimension D; Root mean square height R q The root mean square height R represents the average height of all aggregate profiles along the reference length direction. q The calculation formula is shown in equation (4); Where N is the total number of cross-sectional profile sampling points, Z n The height of the aggregate profile along the reference length; Peak state R ku The dimensionless fourth power of the reference length Z(x) is represented by the fourth power of the root mean square height, and the kurtosis R is also shown. ku The calculation formula is shown in equation (5); The fractal dimension D is used to quantitatively evaluate the roughness of the aggregate surface. The formula for calculating the fractal dimension D is shown in equation (6). In the formula: r is the side length of the cube, and N(r) is the number of small cubes required to cover the object; III. Correlation Analysis of Aggregate Characteristic Parameters Correlation analysis was performed on the average hardness parameter AHP and the SiO2 / Al2O3 index of the aggregate, and the root mean square height R was also analyzed. q , peak state R ku Correlation analysis was performed between each pair of fractal dimensions D, based on the aggregate average hardness parameter AHP, SiO2 / Al2O3 index, and root mean square height R. q , peak state R ku And select uncorrelated parameters from the fractal dimension D as aggregate characteristic parameters; IV. Establishment of a Multiple Linear Regression Model Asphalt mixture specimens are prepared by mixing aggregates, mineral powder and asphalt. The dynamic friction coefficient μ of the asphalt mixture specimens is measured. Based on the aggregate characteristic parameters in step three, a multiple linear regression model for the skid resistance of asphalt mixtures is established. Then, the skid resistance of asphalt mixtures is predicted by the multiple linear regression model, thus completing the prediction method for the skid resistance of asphalt mixtures that integrates aggregate mineral and texture characteristics.

2. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... The aggregate mentioned in step one is either artificial or natural aggregate.

3. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 2, characterized in that... The artificial aggregate is calcined bauxite, and the natural aggregate is basalt or limestone.

4. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step two, the SiO2 / Al2O3 ratio was obtained by X-ray fluorescence spectroscopy.

5. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step two, a laser microscope is used to obtain a texture image of the aggregate surface in order to obtain the root mean square height R. q , peak state R ku and fractal dimension D.

6. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step three, during the correlation analysis, when the correlation coefficient R... 2 A correlation coefficient greater than 0.7 indicates that the two parameters are correlated. 2 If the value is less than 0.7, it indicates that the two parameters are not correlated.

7. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step four, the aggregates, mineral powder and asphalt are mixed and compacted to obtain asphalt mixture specimens.

8. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step four, the dynamic friction coefficient μ of the asphalt mixture specimen is obtained by using a dynamic friction coefficient tester.

9. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 1, characterized in that... In step four, the multiple linear regression module in SPSS software is used to perform regression on the aggregate characteristic parameters obtained in step three as independent variables.

10. The method for predicting the skid resistance of asphalt mixtures that integrate aggregate mineral and texture characteristics according to claim 9, characterized in that... In step four, if the statistical value p of the independent variable is less than 0.05, it indicates that the independent variable has a significant impact on the skid resistance of asphalt mixture.

Images