Method for adjusting mixing uniformity of modulus agent in asphalt mixture

Abstract

The invention relates to the technical field of road material detection, in particular to a method for adjusting the mixing uniformity degree of a modulus agent in an asphalt mixture, and the method comprises the following steps: S1, weighing aggregate and a high modulus agent in proportion, preheating the aggregate, and stirring; s2, uniformly scattering a high-modulus agent into the preheated aggregate, mixing and stirring the high-modulus agent and the preheated aggregate to obtain an intermediate material I, and calculating the wrapping degree of the intermediate material I; s3, asphalt and aluminum powder are weighed in proportion, the asphalt is heated, then the aluminum powder is added, mixing is conducted, and a second intermediate material is obtained; s4, adding the intermediate material II into the intermediate material I, uniformly stirring, adding a proper amount of mineral powder, and continuously stirring to obtain a high-modulus asphalt mixture; s5, detecting the aluminum powder in the high-modulus asphalt mixture by using industrial CT (Computed Tomography) to obtain an image; s6, constructing a mixing uniformity prediction model, and predicting the AU; by predicting the optimal blending process parameters, the modulus adjusting agent is ensured to be uniformly blended in the asphalt mixture.

Description

Technical Field

[0001] The invention relates to the technical field of road material detection, in particular to a method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture. Background Art

[0002] With the development of transportation infrastructure construction, the quality requirements of road projects are increasing. Asphalt pavement has been widely used at home and abroad due to its good construction performance and economy. However, traditional asphalt mixtures have certain defects, such as easy softening and deformation at high temperatures, and becoming brittle and easy to crack at low temperatures. These problems not only affect the service life of the road surface, but may also cause traffic safety accidents. Therefore, how to improve the mechanical properties of asphalt mixtures and improve their durability and fatigue resistance has become the focus of research.

[0003] In order to overcome the shortcomings of traditional asphalt mixtures, researchers have begun to try to add various modified materials to asphalt mixtures in recent years. Among them, high modulus agents, as an additive that can significantly improve the modulus of asphalt mixtures, have received widespread attention. High modulus agents can enhance the rigidity and bearing capacity of asphalt mixtures, while improving their high temperature stability and low temperature crack resistance. However, the key issue that needs to be solved for the effective use of high modulus agents is their uniform distribution in asphalt mixtures. If the mixture is uneven, not only will the role of the high modulus agent not be fully exerted, but it may also lead to local performance degradation and even early failure.

[0004] At present, there are few studies on the mixing technology of high modulus agents in asphalt mixtures. The existing mixing methods often rely on experience and trial and error, and lack a scientific and effective evaluation system. In particular, there is a lack of precise quantitative standards and prediction methods in terms of the degree of coating between high modulus agents and aggregates, and the degree of mixing between asphalt and aluminum powder. In addition, the lack of understanding of the influencing mechanism of various factors in the mixing process also limits the application effect of high modulus agents in actual engineering. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture.

[0006] A method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture of the present invention comprises the following steps: S1: Weigh aggregate and high modulus agent in proportion, preheat the aggregate, and stir during the preheating process; Aggregates can be crushed stone, natural sand, machine-made sand, etc. The purpose of preheating the aggregate is to increase the temperature of the aggregate to make it easier to mix with the high modulus agent. Stirring can not only ensure that the aggregate is heated evenly, but also remove moisture from the surface of the aggregate, further improving the performance of the mixture; S2: evenly sprinkle the high modulus agent on the aggregate preheated in step S1, mix and stir the two to obtain intermediate material 1, and calculate the coating degree of the intermediate material 1; collect data, build a coating degree prediction model, apply the coating degree prediction model to actual production, and adjust the parameters according to the predicted coating degree value; When spreading the high modulus agent, ensure that the high modulus agent can be evenly distributed on the entire aggregate surface to avoid local accumulation or omission; fully stir the aggregate sprinkled with the high modulus agent to ensure that the high modulus agent is evenly distributed in the aggregate. Appropriate equipment should be used during the mixing process, such as a forced mixer or a rotary mixer, and the process parameters during mixing should be recorded to provide a data basis for the establishment of a coating degree prediction model; S3: Weigh asphalt and aluminum powder according to proportion, heat the asphalt to make it have good fluidity, then add aluminum powder according to proportion, and place the mixture of asphalt and aluminum powder in a high-speed shearing machine for high-speed shearing. After the shearing is completed, take it out and place it in a 175°C oven for standby use to obtain intermediate material 2; The high shearing speed of the high-speed shearing machine can effectively break up the aluminum powder agglomerates and make it evenly dispersed in the asphalt, which helps to improve the overall performance of the mixture and avoid local performance degradation; S4: adding the intermediate material 2 to the intermediate material 1, stirring evenly, and then adding an appropriate amount of mineral powder and continuing to stir to obtain a high modulus asphalt mixture; Mineral powder can be limestone powder, quartz powder, diatomaceous earth, talcum powder, fly ash and steel slag powder. Mineral powder can fill the gaps in the mixture, improve the density and stability of the mixture, improve its mechanical properties, increase the cohesion between asphalt and aggregate, and improve the overall strength and durability of the mixture. At the same time, recording process parameters such as stirring speed, stirring time and stirring temperature can provide a scientific basis for subsequent production optimization and quality control; S5: Use industrial CT to detect aluminum powder in high modulus asphalt mixture, obtain images, and calculate the area of ​​aluminum powder blank area around the aggregate, recorded as A N The area of ​​the aggregate zone is recorded as A a , the total area of ​​the image is A O The mixing uniformity is recorded as , with A U As an indicator for quantitative analysis of the uniformity of mixing of high modulus agents; Since aluminum powder is evenly distributed in the gaps between aggregates in high modulus asphalt mixture along with the asphalt with good fluidity, and the initial state of high modulus agent is granular, it does not have the good fluidity of hot asphalt, and the market for dry mixing of high modulus agent and aggregate is relatively short, and it is mainly softened by the temperature of aggregate, so a large amount of evenly distributed aluminum powder appears in the image, and there will be clumps and clusters of aluminum powder blank areas around the aggregate. The number and size of aluminum powder blank areas are used as the basis for qualitative analysis of the uniformity of high modulus agent in asphalt mixture mixing; S6: Construct a mixing uniformity prediction model and apply it to actual production. U Make a prediction and use the predicted A U Adjust the parameters.

[0007] As a preferred embodiment of the present invention, the calculation method of the wrapping degree is: a. Randomly take a certain amount of sample from the intermediate material 1 and place it on a microscope slide; b. Adjust the focus and magnification of the microscope until the image is clear; c. Use a digital camera of the microscope to take multiple images of the high modulus agent distribution on the aggregate surface, and then import the high modulus agent distribution images into image processing software; d. Convert the high modulus agent distribution image into a grayscale image and then remove the noise; e. Determine the gray value threshold range, and divide the image into aggregate, softened high modulus agent and unsoftened high modulus agent according to the gray value threshold range; f. Calculate the area k1 of the softened high modulus agent and the area k2 of the unsoftened agglomerates in the high modulus agent distribution image according to the determined threshold range; e. Calculate the wrapping degree e = k1 / (k1+k2).

[0008] Direct-cast high modulus agent is generally used in the preparation of high modulus asphalt mixture by dry mixing, that is, it is directly mixed with hot aggregate for 30 seconds. However, the high modulus agent is generally in the form of particles with a diameter of about 3mm and a high softening point. When mixed with aggregate in the mixing pot, it mainly relies on the aggregate temperature to soften it, and the dry mixing time is also short. It is often not evenly coated and mixed with the aggregate. If it cannot be evenly dispersed when producing high modulus asphalt mixture, it will have a direct impact on its road performance. On the other hand, the addition of high modulus agent often weakens the low-temperature performance of asphalt mixture. If the high modulus agent exists in an unsoftened agglomerated state, it cannot achieve a uniform and good degree of fusion with asphalt, and an obvious high modulus agent-asphalt weak interface will be formed, which will have a negative impact on the low-temperature performance of asphalt mixture. Therefore, detecting the uniform dispersion degree of direct-cast high modulus agent in high modulus asphalt mixture has important research significance and engineering value for effectively controlling the road performance of high modulus asphalt mixture; The above-mentioned method for calculating the coating degree uses a microscope to capture the distribution image of the high modulus agent on the aggregate surface, and uses image processing software for analysis. It can measure the distribution of the high modulus agent very accurately. This method avoids the subjective error of naked eye observation and provides objective data support. The image is converted into a grayscale image and threshold segmentation is performed to clearly distinguish between the aggregate, softened high modulus agent and unsoftened high modulus agent. This method can accurately calculate the area of ​​each region, thereby improving the accuracy of the coating degree calculation.

[0009] As a preferred embodiment of the present invention, the high modulus agent is one or more of a polymer modifier, a fiber reinforcement agent, a nano material, a resin modifier and a composite modifier.

[0010] As a high modulus agent, polymer modifier can significantly increase the modulus of asphalt mixture, enhance its toughness and crack resistance. Polymer modifier has a high softening point and can reduce rutting under high temperature conditions. Fiber reinforcement can increase the tensile strength and compressive strength of asphalt mixture and enhance the overall mechanical properties. Fiber can absorb and disperse stress, reduce the expansion of cracks and improve the crack resistance of the mixture. Nanomaterials have extremely high specific surface area and activity, which can improve the microstructure of asphalt mixture and improve its stability and strength. Resin modifier can significantly increase the modulus and hardness of asphalt mixture and enhance its bearing capacity. Composite modifier combines the advantages of multiple modifiers and can comprehensively improve the mechanical properties, durability and stability of asphalt mixture.

[0011] As a preferred embodiment of the present invention, the preheating temperature of the aggregate is between 160-180°C.

[0012] The temperature of 160-180℃ can effectively soften the high modulus agent, making it easier to wrap on the aggregate surface and improve the coating degree. The increase in coating degree means that the contact between the high modulus agent and the aggregate is closer, which helps to improve the mechanical properties of the mixture.

[0013] As a preferred embodiment of the present invention, the mass of aluminum powder is 1% of the mass of asphalt.

[0014] Too high an aluminum powder content may cause a decline in the performance of the mixture, such as increased brittleness, reduced low-temperature performance, etc. A ratio of 1% is a relatively safe and effective range, which can play a tracing role without negatively affecting the performance of the mixture.

[0015] As a preferred embodiment of the present invention, the aluminum powder has an extremely fine particle size of 0.5 to 10 μm. The aluminum powder with a particle size of 0.5 to 10 μm can be more easily dispersed evenly in the asphalt to avoid agglomeration, which helps to improve the overall uniformity of the mixture and ensure that the aluminum powder can be evenly distributed in the gaps between the aggregates.

[0016] As a preferred embodiment of the present invention, the shearing speed of the high-speed shearing machine is 2000-6000 rpm. This speed range can provide sufficient shear force to ensure that the agglomerates of aluminum powder are effectively broken up without causing excessive burden on the equipment. The shearing time is 5-20 minutes. This time range can ensure that the aluminum powder is fully mixed in the asphalt to avoid local agglomeration, and will not cause excessive shearing to affect the performance of the mixture. During the shearing process, the temperature of the intermediate material 2 is controlled between 170-180°C. This temperature range can ensure that the asphalt has good fluidity, which is conducive to the uniform dispersion of aluminum powder, and will not cause overheating of the asphalt to affect its performance.

[0017] As a preferred solution of the present invention, the shearing speed of the high-speed shearing machine is 3000 rpm; the shearing time is 15 min; during the shearing process, the temperature of the intermediate material 2 is controlled at 175°C.

[0018] The shearing speed of the high-speed shearing machine is set to 3000 rpm, the shearing time is set to 15 minutes, and the temperature of the intermediate material 2 is controlled at 175°C during the shearing process. This can significantly improve the uniform dispersion of aluminum powder in asphalt, improve the performance of asphalt mixture, optimize the construction process, and improve the stability and durability of the mixture. At the same time, it has good economic and environmental benefits.

[0019] As a preferred solution of the present invention, the specific method of collecting data, constructing a wrapping degree prediction model, applying the wrapping degree prediction model to actual production, and adjusting parameters according to the predicted wrapping degree value includes: S211, collecting the production process parameters of the next batch of intermediate material 1 and the coating degree calculated for the current batch to obtain a data set, wherein the production process parameters include: characteristic data of the aggregate and the high modulus agent, mixing process parameters of the intermediate material 1 and environmental parameters; S212, extracting features from the data in the data set, organizing the extracted features into a two-dimensional matrix, and normalizing the two-dimensional matrix; S213, constructing a convolutional neural network model as a wrapping degree prediction model, using the normalized two-dimensional matrix as the input of the wrapping degree prediction model, and obtaining the next batch of wrapping degree prediction values; S214. Optimize the production process parameters of the intermediate material 1 according to the predicted value of the coating degree.

[0020] As a preferred solution of the present invention, the method of constructing a mixing uniformity prediction model, applying the mixing uniformity prediction model to actual production, predicting AU, and adjusting parameters according to the predicted AU includes: S611, constructing a mixing uniformity prediction model, collecting the production process parameters of the next batch of high modulus asphalt mixture and the mixing uniformity calculated for the current batch, inputting them into the mixing uniformity prediction model for prediction, and obtaining a predicted value of the mixing uniformity for the next batch; S612, setting a target value for mixing uniformity, defining an optimization target as minimizing the gap between the predicted value for mixing uniformity and the target value for mixing uniformity, and constructing an objective function; S613, selecting a suitable optimization algorithm, setting upper and lower bounds for each of the production process parameters, and running the optimization algorithm to obtain the optimal production process parameters.

[0021] Compared with the prior art, the present invention has the following beneficial effects: First, the coating degree of the intermediate material is accurately measured using image processing technology, and based on these data, coating degree prediction models and mixing uniformity prediction models are constructed. These models can not only accurately evaluate the mixing quality, but also predict the mixing effect under different process parameters, providing a scientific basis for optimizing the mixing process. The optimal mixing process parameters can be predicted in the design stage, reducing the number of experiments and reducing R&D costs; Second, considering that industrial CT technology can significantly detect and distinguish metal and asphalt mixture, the present invention dry mixes the aggregate with the high modulus agent, and then uses high-speed shearing to achieve uniform mixing of fine aluminum powder and hot asphalt in the hot asphalt, and adds the mixture of hot asphalt and aluminum powder and mineral powder to a mixing pot for continuous stirring to obtain a high modulus asphalt mixture. The high modulus asphalt mixture is tested by industrial CT. Most of the aluminum powder is fully coated on the aggregate due to the good fluidity of the hot asphalt, and exists between the aggregate and the asphalt filling. Aluminum powder is intuitively displayed in industrial CT scanning images due to its rich metallic properties. High modulus agents that are not fully mixed and softened and agglomerated will not contain aluminum powder. Blank areas without aluminum powder are shown in the image. Based on the shape, size and quantity of the blank areas without aluminum powder, a qualitative judgment can be made on whether the high modulus agent is evenly mixed in the asphalt mixture, and the mixing degree of the high modulus agent can be quantitatively analyzed through area quantification, which provides a new method and idea for the quality control of high modulus asphalt mixture, and has high research significance and road engineering material detection value. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram of the distribution of aluminum powder, aggregate and high modulus agent in high modulus asphalt mixture detected by industrial CT technology in the embodiment; Figure 2 The schematic diagram of the distribution of aluminum powder, aggregate and high modulus agent in high modulus asphalt mixture detected by industrial CT technology for comparison; Markings in the attached figure: 1. Aggregate; 2. Aluminum powder; 3. Aggregates in which the high modulus agent has not been completely melted; 4. The surface of the aggregate is not coated with the high modulus agent; 5. The surface of the aggregate is coated with the softened high modulus agent. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0026] The present invention provides a method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture, comprising the following steps: S1: Weigh aggregate and high modulus agent in proportion, preheat the aggregate, and stir during the preheating process; Specifically, aggregates can be crushed stone, natural sand, machine-made sand, etc.; The high modulus agent is one or more of a polymer modifier, a fiber reinforcement agent, a nanomaterial, a resin modifier and a composite modifier; as a high modulus agent, the polymer modifier can significantly increase the modulus of the asphalt mixture, enhance its toughness and crack resistance, and the polymer modifier has a high softening point and can reduce rutting under high temperature conditions; Fiber reinforcement can improve the tensile strength and compressive strength of asphalt mixture and enhance the overall mechanical properties. Fiber can absorb and disperse stress, reduce the expansion of cracks and improve the crack resistance of the mixture. Nanomaterials have extremely high specific surface area and activity, which can improve the microstructure of asphalt mixture and increase its stability and strength; Resin modifiers can significantly increase the modulus and hardness of asphalt mixtures and enhance their bearing capacity; Composite modifiers combine the advantages of multiple modifiers and can comprehensively improve the mechanical properties, durability and stability of asphalt mixtures; The purpose of preheating aggregate is to increase the temperature of aggregate, making it easier to mix with high modulus agent. Stirring can not only ensure that the aggregate is heated evenly, but also remove moisture on the surface of aggregate, further improving the performance of the mixture. The preheating temperature of 160-180℃ can effectively soften the high modulus agent, making it easier to wrap on the surface of aggregate and improve the coating degree. The increase in coating degree means that the contact between high modulus agent and aggregate is closer, which helps to improve the mechanical properties of the mixture. S2: evenly sprinkle the high modulus agent on the aggregate preheated in step S1, mix and stir the two to obtain intermediate material 1, and calculate the coating degree of the intermediate material 1; collect data, build a coating degree prediction model, apply the coating degree prediction model to actual production, and adjust the parameters according to the predicted coating degree value; Specifically, when spreading the high modulus agent, ensure that the high modulus agent can be evenly distributed on the entire aggregate surface to avoid local accumulation or omission; fully stir the aggregate sprinkled with the high modulus agent to ensure that the high modulus agent is evenly distributed in the aggregate. Appropriate equipment should be used during the mixing process, such as a forced mixer or a rotary mixer, and the process parameters during mixing should be recorded to provide a data basis for the establishment of a coating degree prediction model; The calculation method of wrapping degree is: a. Randomly take a certain amount of sample from the intermediate material 1 and place it on a microscope slide; b. Adjust the focus and magnification of the microscope until the image is clear; c. Use a digital camera of the microscope to take multiple images of the high modulus agent distribution on the aggregate surface, and then import the high modulus agent distribution images into image processing software. The image processing software can be ImageJ, Adobe Photoshop, GIMP, OpenCV or QGIS; d. Convert the high modulus agent distribution image into a grayscale image and then remove the noise; e. Determine the gray value threshold range, and divide the image into aggregate, softened high modulus agent and unsoftened high modulus agent according to the gray value threshold range; f. Calculate the area k1 of the softened high modulus agent and the area k2 of the unsoftened agglomerates in the high modulus agent distribution image according to the determined threshold range; e. Calculate the wrapping degree e = k1 / (k1+k2); By taking images of the distribution of the high modulus agent on the aggregate surface under a microscope and analyzing them with image processing software, the distribution of the high modulus agent can be measured very accurately. This method avoids the subjective error of naked eye observation and provides objective data support. By converting the image into a grayscale image and performing threshold segmentation, the aggregate, softened high modulus agent and unsoftened high modulus agent can be clearly distinguished. This method can accurately calculate the area of ​​each region, thereby improving the accuracy of the calculation of the coating degree. The specific method of collecting data, constructing a wrapping degree prediction model, applying the wrapping degree prediction model to actual production, and adjusting parameters according to the predicted wrapping degree value includes: S211, collecting the production process parameters of the next batch of intermediate material 1 and the coating degree calculated for the current batch to obtain a data set, wherein the production process parameters include: characteristic data of aggregate and high modulus agent, mixing process parameters of the intermediate material 1 and environmental parameters; In this embodiment, the collected aggregate properties include: particle size distribution, shape, surface area, water absorption, etc.; the collected high modulus agent properties include: viscosity, softening point, needle penetration, etc.; the collected mixing process parameters include: mixing temperature, time, stirring speed, addition ratio of aggregate and high modulus agent, etc., and the collected environmental parameters include: temperature, humidity, etc.; the collected data can be saved in a structured file or database to ensure that each record is clearly marked with the batch and time sequence; data quality checks can be performed before using the data, including missing value processing, outlier processing and data distribution analysis. Missing value processing can fill or delete missing coating degrees or process parameters, outlier processing can detect and eliminate obviously unreasonable data points, and data distribution analysis can analyze the distribution of each variable to ensure that the data covers common working conditions.

[0027] S212, extracting features from the data in the data set, organizing the extracted features into a two-dimensional matrix, and normalizing the two-dimensional matrix; Based on the above embodiments, the collected data are all numerical data. In order to show the possible local correlation or spatial pattern between features and enable the model to efficiently process complex data, the extracted features can be matrix mapped, and the numerical features can be mapped to a two-dimensional matrix. The rows of the two-dimensional matrix represent the features: each row corresponds to an input variable, such as temperature, time, etc.; the columns represent samples: each column represents a batch of feature data; matrix values: the value of each position represents the value of the feature in the batch. After matrix mapping, in order to avoid the influence of different feature dimension differences on the matrix, the data can be normalized to the same range, such as [0, 1].

[0028] S213, constructing a convolutional neural network model as a wrapping degree prediction model, using the normalized two-dimensional matrix as the input of the wrapping degree prediction model, and obtaining the next batch of wrapping degree prediction values; Specifically, the data has been visualized, and the local patterns or global trends in the image may have an important impact on the prediction of the wrapping degree. The convolutional neural network model can extract the local features of the image through the convolution operation, which is very suitable for processing image data with complex nonlinear relationships. Therefore, a convolutional neural network model can be constructed as a wrapping degree prediction model. The convolutional neural network model essentially extracts local patterns through convolution operations. Regardless of whether the input is the original two-dimensional matrix or the image, the core process is to slide the convolution kernel on the two-dimensional grid to capture patterns and features. The normalized two-dimensional matrix already contains numerical information, and its numerical value can be directly processed by the convolutional neural network model without additional conversion to an image. Directly using the normalized matrix can reduce the computational cost and additional format conversion, avoid potential distortion or errors in the visualization process, and retain numerical accuracy. The convolutional neural network model structure designed in this step may include: input layer: input image data; convolution layer: extract local features; pooling layer: reduce dimensionality and retain key information; fully connected layer: perform feature combination and prediction; output layer: output the next batch of wrapping degrees.

[0029] S214. Optimize the production process parameters of the intermediate material 1 according to the predicted value of the coating degree.

[0030] The coating degree prediction model is applied to actual production. By collecting the characteristic data of each batch of aggregate and high modulus agent, recording the mixing process parameters and environmental parameters, and performing data cleaning and standardization, the pre-processed data is input into the trained coating degree prediction model for prediction. The predicted coating degree value is used to determine whether the coating degree of the mixture meets the expected standard. If it does not meet the expected standard, the production process parameters are adjusted. In order to continuously improve the prediction model, production data can be regularly collected to update and optimize the coating degree prediction model, and a feedback mechanism can be established. This series of methods can significantly improve production efficiency and quality control level, ensure the consistency and performance stability of high modulus asphalt mixture, reduce human errors, reduce production costs, improve resource utilization, and ultimately improve the service life of the road and driving safety.

[0031] The method for constructing the wrapping degree prediction model comprises: S221, collecting characteristic data of several batches of aggregates and high modulus agents, recording mixing process parameters and environmental parameters of each batch, and calculating the coating degree of each batch using microscope photography and image processing software; S222, cleaning the collected data, removing outliers and missing values, and standardizing the data; S223, performing correlation analysis, selecting important features that are highly correlated with the wrapping degree, and organizing the important features into a two-dimensional matrix; S224, selecting a convolutional neural network model as the basic architecture of the wrapping degree prediction model; S225. Divide the two-dimensional matrix into training set data and test set data, use the training set data to train the selected convolutional neural network model, adjust the model parameters to optimize the performance, and use the cross-validation method to evaluate the generalization ability of the model; use the mean square error and determination coefficient to evaluate the predictive performance of the machine learning model.

[0032] Through the above steps, an accurate and reliable coating degree prediction model can be constructed. The model can effectively predict the coating degree of high modulus asphalt mixture, improve production efficiency and quality control level, ensure the consistency and performance stability of the mixture, reduce human errors, reduce production costs, improve resource utilization, and ultimately improve the service life of the road and driving safety.

[0033] S3: Weigh asphalt and aluminum powder according to proportion, heat the asphalt to 175°C to make it have good fluidity, then add aluminum powder according to proportion, and place the mixture of asphalt and aluminum powder in a high-speed shearing machine for high-speed shearing. After the shearing is completed, take it out and place it in a 175°C oven for standby use to obtain intermediate material 2; Specifically, the mass of aluminum powder is 1% of the mass of asphalt. Too high aluminum powder content may lead to a decline in the performance of the mixture, such as increasing brittleness and reducing low-temperature performance. The proportion of 1% is a relatively safe and effective range, which can play a tracing role without having a negative impact on the performance of the mixture. Aluminum powder with a particle size of 0.5 to 10 μm can be more easily dispersed evenly in asphalt to avoid agglomeration, which helps to improve the overall uniformity of the mixture and ensure that the aluminum powder can be evenly distributed in the gaps between the aggregates; The shearing speed of the high-speed shearing machine is 2000-6000 rpm. This speed range can provide sufficient shearing force to ensure that the agglomerates of aluminum powder are effectively broken up without placing too much burden on the equipment. The shearing time is 5-20 minutes. This time range can ensure that the aluminum powder is fully mixed in the asphalt to avoid local agglomeration, while not causing excessive shearing and affecting the performance of the mixture. During the shearing process, the temperature of the intermediate material 2 is controlled between 170-180°C. This temperature range can ensure that the asphalt has good fluidity, which is conducive to the uniform dispersion of aluminum powder, while not causing overheating of the asphalt and affecting its performance.

[0034] The high shear speed of the high-speed shearing machine can effectively break up the aluminum powder agglomerates and make it evenly dispersed in the asphalt, which helps to improve the overall performance of the mixture and avoid local performance degradation; controlling the temperature between 170-180℃ can ensure that the asphalt maintains good fluidity while avoiding the degradation of material performance due to excessive temperature; S4: adding the intermediate material 2 to the intermediate material 1, stirring evenly, and then adding an appropriate amount of mineral powder and continuing to stir to obtain a high modulus asphalt mixture; Specifically, the mineral powder can be limestone powder, quartz powder, diatomaceous earth, talcum powder, fly ash and steel slag powder, etc. The mineral powder can fill the voids in the mixture, improve the density and stability of the mixture, and improve its mechanical properties. The mineral powder can increase the cohesion between asphalt and aggregate, and improve the overall strength and durability of the mixture. At the same time, recording process parameters such as stirring speed, stirring time and stirring temperature can provide a scientific basis for subsequent production optimization and quality control; S5: Use industrial CT to detect aluminum powder in high modulus asphalt mixture, obtain images, and calculate the area of ​​aluminum powder blank area around the aggregate, recorded as A N The area of ​​the aggregate zone is recorded as A a , the total area of ​​the image is A O , the mixing uniformity is recorded as , with A U As an indicator for quantitative analysis of the uniformity of mixing of high modulus agents; Through industrial CT testing of high modulus asphalt mixture, most of the aluminum powder is fully coated with aggregates due to the good fluidity of hot asphalt, and exists between the aggregates with asphalt filling. Aluminum powder is intuitively displayed in the industrial CT scanning image due to its rich metallic properties. High modulus agent that is not fully mixed and softened and agglomerated will not contain aluminum powder. In the image, blank areas without aluminum powder are shown. Based on the shape, size and number of blank areas without aluminum powder, it is possible to qualitatively determine whether the high modulus agent is evenly mixed in the asphalt mixture, and quantitatively analyze the mixing degree of the high modulus agent through area quantification, which provides a new method and idea for the quality control of high modulus asphalt mixture, and has high research significance and road engineering material detection value; S6: Construct a mixing uniformity prediction model and apply it to actual production. U Make a prediction and use the predicted A U Adjust the parameters.

[0035] Specifically, the mixing uniformity prediction model is constructed and applied to actual production. U Make a prediction and use the predicted A U Methods for adjusting parameters include: S611, constructing a mixing uniformity prediction model, collecting the production process parameters of the next batch of high modulus asphalt mixture and the mixing uniformity calculated for the current batch, inputting them into the mixing uniformity prediction model for prediction, and obtaining a predicted value of the mixing uniformity for the next batch; Specifically, the choice of model depends on the complexity, scale and relationship between features of the data. In this step, random forest, support vector regression model, neural network model, etc. can be selected as the mixing uniformity prediction model. In this step, the production process parameters of the next batch of high modulus asphalt mixture are collected, including: asphalt properties (bitumen viscosity, softening point, needle penetration), asphalt ratio, mineral powder properties (type of mineral powder, particle size distribution, density, shape of mineral powder), mineral powder ratio, mixing temperature, mixing time, etc.; in order to make the collected raw data suitable for the model, the data can be cleaned and normalized to improve the quality of the data, laying a good data foundation for the prediction of the model. It should be noted that by inputting the production process parameters of the next batch and the mixing uniformity of the current batch, the mixing uniformity of the next batch can be predicted more accurately. Therefore, if there is a large randomness or sudden event in the production process, such as material fluctuations or equipment failures, the uncertainty of the prediction results of the mixing uniformity needs to be evaluated.

[0036] S612, setting a target value for mixing uniformity, defining an optimization target as minimizing the gap between a predicted value for mixing uniformity and a target value for mixing uniformity, and constructing an objective function; S613. Select a suitable optimization algorithm, set upper and lower bounds for each production process parameter, run the optimization algorithm, and obtain the optimal production process parameters.

[0037] On the basis of the above embodiments, the target value of the mixing uniformity can be set according to historical data analysis, experimental verification or industry standards, and an objective function with minimizing the gap between the predicted value and the target value as the optimization goal can be constructed. For complex objective functions and multivariable parameters, the performance of the optimization algorithm directly affects the final result. Therefore, a suitable optimization algorithm can be selected according to the characteristics of the objective function. In order to ensure that the optimization result meets the actual production capacity and physical limitations, reasonable upper and lower limits can be set for each production process parameter, and then the prediction model and objective function are input, and the optimization algorithm is run to search for the optimal process parameters. The optimization algorithms that can be selected include: (1) Genetic algorithm: It is applicable to nonlinear and non-convex optimization problems. It searches for the global optimal solution by simulating natural selection and is suitable for optimizing high-dimensional input parameters.

[0038] (2) Particle swarm optimization is suitable for continuous variable optimization and can simulate swarm intelligence, such as bird foraging. Through the collaborative search of particle swarms, solutions close to the target value can be quickly found, and the search speed is relatively fast.

[0039] (3) Gradient descent: It is applicable when the objective function is differentiable and can quickly converge to the local optimal solution through gradient information.

[0040] (4) Hybrid optimization algorithm: combines global search algorithms (such as genetic algorithms and particle swarm optimization) with local optimization algorithms (such as gradient descent) to improve search efficiency.

[0041] The above method can significantly improve the production efficiency and quality control level of high modulus asphalt mixture. This method can ensure that the mixing uniformity of the mixture is always in the optimal state, reduce human errors, reduce production costs, improve resource utilization, and ultimately improve the service life of the road and driving safety.

[0042] The method for constructing the mixing uniformity prediction model comprises: S621, collect the characteristic data of asphalt and mineral powder of the same batch, the ratio of asphalt and mineral powder, record the mixing process parameters and environmental parameters of the batch, and collect the coating degree of the intermediate material 1 involved in the mixing of this batch, and collect the A U value; S622, clean the data, remove outliers and missing values, and standardize the data; S623, perform correlation analysis and select U Important features with highly correlated values; S624, selecting a machine learning model as the basic architecture of the model; S625. Divide the relevant data of important features into training set data and test set data, use the training set data to train the selected machine learning model, adjust the model parameters to optimize the performance, and use the cross-validation method to evaluate the generalization ability of the model; use the mean square error and determination coefficient to evaluate the predictive performance of the machine learning model.

[0043] The mixing uniformity prediction model constructed through the above steps can effectively predict the mixing uniformity of high modulus asphalt mixture, improve production efficiency and quality control level, ensure the consistency and performance stability of the mixture, reduce human errors, reduce production costs, improve resource utilization, and ultimately improve the service life of the road and driving safety.

[0044] Example: like Figure 1 As shown, a method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture comprises the following steps: S1. Determine the characteristic data, temperature, humidity and coating threshold of aggregate and high modulus agent, wherein the coating threshold is the minimum value to be achieved; use the coating prediction model to adjust the mixing process parameters so that the predicted coating value is not less than the coating threshold; the data are as follows: the aggregate is crushed stone with a particle size range of 2.36mm-19mm, the high modulus agent is polymer modifier SBS, the ambient temperature is 25°C, the humidity is 72%, and the coating threshold is 0.9; input the above data into the coating prediction model, adjust the mixing process parameters to a mixing temperature of 175°C, a stirring time of 30s, and predict that the encountered coating value is 0.93; S2. For aggregate 1 and high modulus agent, take aggregate 1 and weighed high modulus agent, put aggregate 1 and high modulus agent into a mixing pot at 175°C and dry mix for 30 seconds; S3, evenly sprinkle the high modulus agent on the aggregate preheated in step S1, mix and stir the two to obtain an intermediate material 1, and calculate the coating degree of the intermediate material 1 to be 0.92; S4. Use a stainless steel container to weigh asphalt and a trace amount of aluminum powder 2, where the mass of the aluminum powder 2 is 1% of the mass of the asphalt and the particle size of the aluminum powder 2 is an extremely fine powder of 0.5 to 10 μm; S5. Heat the asphalt in the stainless steel container to 175°C to make it have good fluidity, add aluminum powder 2 to the hot asphalt, and place the mixture of asphalt and aluminum powder in a high-speed shearing machine for high-speed shearing. The shear rate and shear rate are set to 15min and 3000r / min respectively. After the shearing is completed, take it out and put it in a 175°C oven for standby use to obtain intermediate material 2; S6. Determine the asphalt characteristic data, the mineral powder characteristic data, the asphalt, mineral powder ratio and preset A U , where A is preset U is the desired value; the mixing uniformity prediction model is used to adjust the mixing process parameters to make the predicted A U With preset A U The data are as follows: No. 70 asphalt, the mineral powder is limestone powder that passes through a 0.075mm sieve, the mineral powder addition ratio is 0.3% of the asphalt weight, the ambient temperature is 25℃, and the humidity is 72%; the above data is input into the mixing uniformity prediction model, and the mixing process parameters are adjusted to a mixing temperature of 175℃ and a mixing time of 90s. The predicted A U With preset A U same; S7, adding the mixture of asphalt and aluminum powder into the intermediate material 1, starting the mixing pot to stir for 90 seconds, the mixing temperature is 175°C, and then adding mineral powder and continuing to stir to obtain a high modulus asphalt mixture; S8. Take out the high-temperature asphalt mixture from the mixing pot to a stainless steel tray, separate it using the quartering method, and then randomly take out a portion of high modulus asphalt mixture; S9. Use industrial CT to detect aluminum powder in high modulus asphalt mixture and obtain accurate images. Since aluminum powder 2 is evenly distributed in the gaps of aggregate 1 in high modulus asphalt mixture along with asphalt with good fluidity, the image shows more aluminum powder 2 evenly distributed in large quantities, and there are high modulus agents 4 that are not coated on the surface of aggregates and high modulus agents 5 that are coated and softened on the surface of aggregates, and the mixing is relatively uniform. The initial state of high modulus agent is granular, and it does not have the good fluidity of hot asphalt. The market for dry mixing of high modulus agent and aggregate is relatively short, and it is mainly softened by the temperature of aggregates. Therefore, the image shows that high modulus agent has not completely melted agglomerates 3, and high modulus agent has not completely melted agglomerates 3 do not contain aluminum powder, which is an aluminum powder blank area. Aluminum powder blank areas will appear around aggregate 1. After observation, there are 2 aluminum powder blank areas in the figure, and the number is small. It is qualitatively judged that the mixing is well uniform.

[0045] S10. Use the image processing software Image-J to calculate the blank area of ​​aluminum powder around the aggregate, recorded as A N The area of ​​the aggregate zone is recorded as A a , the total area of ​​the image is A O The mixing uniformity is recorded as , after calculation, A N 1.77mm 2 , A O -A a Area: 189.3mm 2 , then A U It is 0.9%, which indicates that the degree of bit fusion is good.

[0046] Comparative Example: like Figure 2 As shown, a method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture comprises the following steps: Aggregate 1, high modulus agent 4, mineral powder, asphalt and aluminum powder 2 are the same as those in the embodiment; S1. The ambient temperature is 20℃ and the humidity is 70%. For aggregate 1 and high modulus agent, take aggregate 1 and weighed high modulus agent, put aggregate 1 and high modulus agent into a 175℃ mixing pot for dry mixing for 30s; S2, evenly sprinkle the high modulus agent on the aggregate preheated in step S1, mix and stir the two to obtain intermediate material 1; S3. Use a stainless steel container to weigh asphalt and a trace amount of aluminum powder 2, where the mass of the aluminum powder 2 is 1% of the mass of the asphalt and the particle size of the aluminum powder 2 is an extremely fine powder of 0.5 to 10 μm; S5. Heat the asphalt in the stainless steel container to 175°C to make it have good fluidity, add aluminum powder 2 to the hot asphalt, and place the mixture of asphalt and aluminum powder in a high-speed shearing machine for high-speed shearing. The shear rate and shear rate are set to 15min and 3000r / min respectively. After the shearing is completed, take it out and put it in a 175°C oven for standby use to obtain intermediate material 2; S6, the ambient temperature is 20℃, the humidity is 70%; the mixing process parameters are adjusted to mixing temperature 176℃, and stirring time 100s; S7, adding the mixture of asphalt and aluminum powder into the intermediate material 1, starting the mixing pot to stir for 100 seconds, the mixing temperature is 176°C, and then adding mineral powder and continuing to stir to obtain a high modulus asphalt mixture; S8. Take out the high-temperature asphalt mixture from the mixing pot to a stainless steel tray, separate it using the quartering method, and then randomly take out a portion of high modulus asphalt mixture; S9. Use industrial CT to detect aluminum powder in high modulus asphalt mixture and obtain accurate images. Since aluminum powder 2 is evenly distributed in the gaps of aggregate 1 in high modulus asphalt mixture along with asphalt with good fluidity, the image shows more aluminum powder 2 evenly distributed in large quantities, and there are high modulus agents 4 that are not coated on the surface of aggregates and high modulus agents 5 that are coated and softened on the surface of aggregates, and the mixing is relatively uniform. However, the initial state of high modulus agent is granular, and it does not have the good fluidity of hot asphalt. The market for dry mixing of high modulus agent and aggregate is relatively short, and it is mainly softened by the temperature of aggregates. Therefore, high modulus agent incompletely melted agglomerates 3 appear in the image, and high modulus agent incompletely melted agglomerates 3 do not contain aluminum powder, which is an aluminum powder blank area, and aluminum powder blank areas will appear around the aggregates. After observation, there are 12 aluminum powder blank areas in the figure, which is a large number, and there are several large-shaped incompletely melted agglomerates 3, which are qualitatively judged to be poor in mixing uniformity.

[0047] S10. Use the image processing software Image-J to calculate the blank area of ​​aluminum powder around the aggregate, recorded as A N The area of ​​the aggregate zone is recorded as A a , the total area of ​​the image is A O The mixing uniformity is recorded as , after calculation, A N 25.93mm 2 , A O -A a Area: 176.3mm 2 , then A U It is 14.71%, and the degree of bit fusion is poor.

[0048] In the comparative example, since the coating degree prediction model and the mixing uniformity prediction model were not adopted, when the environmental factors changed, adjustments were made according to experience, resulting in poor indicators of the obtained high modulus asphalt mixture, which did not meet the use requirements.

[0049] It will be appreciated that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but will be a routine task of design, fabrication, and production for those of ordinary skill having the benefit of this disclosure without undue experimentation.

[0050] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for adjusting the mixing uniformity of a modulus agent in an asphalt mixture, characterized in that: The following steps are involved: S1: Weigh aggregate and high modulus agent in proportion, preheat the aggregate, and stir during the preheating process; S2: evenly sprinkle the high modulus agent on the aggregate preheated in step S1, mix and stir the two to obtain intermediate material 1, calculate the coating degree of the intermediate material 1; collect data, build a coating degree prediction model, and adjust the parameters according to the coating degree value; S3: Weigh asphalt and aluminum powder according to proportion, heat the asphalt to make it have good fluidity, then add aluminum powder according to proportion, and place the mixture of asphalt and aluminum powder in a high-speed shearing machine for shearing. After the shearing is completed, take it out and place it in a 175°C oven for standby use to obtain intermediate material 2; S4: adding the intermediate material 2 to the intermediate material 1, stirring evenly, and then adding the mineral powder and continuing to stir to obtain a high modulus asphalt mixture; S5: Use industrial CT to detect aluminum powder in high modulus asphalt mixture, obtain images, and calculate the area of ​​aluminum powder blank area around the aggregate, recorded as A N The area of ​​the aggregate zone is recorded as A a , the total area of ​​the image is A O , the mixing uniformity is recorded as , with A U As an indicator for quantitative analysis of the uniformity of mixing of high modulus agents; S6: Construct a mixing uniformity prediction model and apply it to actual production. U Make a prediction and use the predicted A U Adjust the parameters.

2. The method for adjusting the uniformity of mixing the modulus agent in the asphalt mixture according to claim 1, characterized in that: The calculation method of the wrapping degree is: a. Randomly take a certain amount of sample from the intermediate material 1 and place it on a microscope slide; b. Adjust the focus and magnification of the microscope until the image is clear; c. Use a digital camera of the microscope to take multiple images of the high modulus agent distribution on the aggregate surface, and then import the high modulus agent distribution images into image processing software; d. Convert the high modulus agent distribution image into a grayscale image and then remove the noise; e. Determine the gray value threshold range, and divide the image into aggregate, softened high modulus agent and unsoftened high modulus agent according to the gray value threshold range; f. Calculate the area k1 of the softened high modulus agent and the area k2 of the unsoftened agglomerates in the high modulus agent distribution image according to the determined threshold range; e. Calculate the wrapping degree e = k1 / (k1+k2).

3. The method for adjusting the uniformity of mixing the modulus agent in asphalt mixture as claimed in claim 1, characterized in that: The high modulus agent is one or more of a polymer modifier, a fiber reinforcement agent, a nano material, a resin modifier and a composite modifier.

4. The method for adjusting the uniformity of mixing the modulus agent in asphalt mixture as claimed in claim 1, characterized in that: The preheating temperature of the aggregate is between 160-180°C.

5. The method for adjusting the uniformity of mixing the modulus agent in asphalt mixture as claimed in claim 1, characterized in that: The mass of the aluminum powder is 1% of the mass of the asphalt.

6. The method for adjusting the uniformity of mixing the modulus agent in the asphalt mixture as claimed in claim 5, characterized in that: The aluminum powder is extremely fine powder with a particle size of 0.5 to 10 μm.

7. The method for adjusting the uniformity of mixing the modulus agent in asphalt mixture according to claim 1, characterized in that: The shearing speed of the high-speed shearing machine is 2000-6000 rpm; the shearing time is 5-20 min; during the shearing process, the temperature of the intermediate material 2 is controlled between 170-180°C.

8. The method for adjusting the uniformity of mixing the modulus agent in the asphalt mixture as claimed in claim 7, characterized in that: The shearing speed of the high-speed shearing machine is 3000 rpm; the shearing time is 15 min; during the shearing process, the temperature of the intermediate material 2 is controlled at 175°C.

9. The method for adjusting the uniformity of mixing the modulus agent in asphalt mixture as claimed in claim 2, characterized in that: The specific method of collecting data, building a wrapping degree prediction model, and adjusting parameters according to the wrapping degree value includes: S211, collecting the production process parameters of the next batch of intermediate material 1 and the coating degree calculated for the current batch to obtain a data set, wherein the production process parameters include: characteristic data of the aggregate and the high modulus agent, mixing process parameters of the intermediate material 1 and environmental parameters; S212, extracting features from the data in the data set, organizing the extracted features into a two-dimensional matrix, and normalizing the two-dimensional matrix; S213, constructing a convolutional neural network model as a wrapping degree prediction model, using the normalized two-dimensional matrix as the input of the wrapping degree prediction model, and obtaining the next batch of wrapping degree prediction values; S214. Optimize the production process parameters of the intermediate material 1 according to the predicted value of the coating degree.

10. The method for adjusting the uniformity of mixing the modulus agent in the asphalt mixture as claimed in claim 2, characterized in that: The mixing uniformity prediction model is constructed and applied to actual production. U Make a prediction and use the predicted A U Methods for adjusting parameters include: S611, constructing a mixing uniformity prediction model, collecting the production process parameters of the next batch of high modulus asphalt mixture and the mixing uniformity calculated for the current batch, inputting them into the mixing uniformity prediction model for prediction, and obtaining a predicted value of the mixing uniformity for the next batch; S612, setting a target value for mixing uniformity, defining an optimization target as minimizing the gap between the predicted value for mixing uniformity and the target value for mixing uniformity, and constructing an objective function; S613, selecting a suitable optimization algorithm, setting upper and lower bounds for each of the production process parameters, and running the optimization algorithm to obtain the optimal production process parameters.

Images