Intelligent compaction monitoring method and system for highway subgrade and pavement

Abstract

The invention relates to an intelligent compaction monitoring method and system for a highway subgrade pavement. The method comprises the following steps: collecting construction site data of the highway subgrade pavement; wherein the construction site data comprises soil mass density, water content and temperature; inputting the construction site data into a compactness prediction model to obtain the compactness of the roadbed and the pavement; wherein the compaction degree prediction model is constructed based on a quantitative relation between construction site data and radar signals; and carrying out early warning on the compaction degree. According to the invention, intelligent and precise monitoring of the compaction degree of the roadbed and the pavement can be realized, the accuracy and the reliability of compaction quality evaluation are improved, data support is provided for optimization of a compaction construction process, and the roadbed and pavement engineering quality is effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of highway monitoring, and in particular to a method and system for monitoring the intelligent compaction of a highway subgrade and pavement. Background Art

[0002] In highway engineering, the compaction of the roadbed and pavement directly affects the performance and service life of the highway. At present, the monitoring of the compaction of the roadbed and pavement mainly relies on manual detection and single equipment detection, which has problems such as low efficiency and incomplete data. How to achieve real-time, comprehensive and accurate monitoring of the compaction process of the roadbed and pavement is a technical problem that needs to be solved urgently.

[0003] Traditional compaction monitoring methods cannot take into account both the roadbed and the pavement at the same time, and the monitoring data is relatively single and lacks comprehensive analysis capabilities. To address this problem, it is urgent to develop an intelligent monitoring method that can simultaneously monitor the compaction of the roadbed and the pavement. This method needs to be able to collect multi-source compaction data such as soil parameters and construction parameters in real time, and perform comprehensive analysis through data fusion algorithms. The data fusion algorithm needs to be able to fully consider the influence of environmental factors such as soil dielectric constant, density, temperature, and water content to improve the accuracy of compaction calculation.

[0004] In addition, how to achieve real-time collection and transmission of multi-source data in a complex construction environment is also a technical challenge. At the same time, the data fusion algorithm needs to be able to adapt to different soil conditions and construction processes, and have strong adaptability and robustness. Finally, how to combine machine learning technology with data fusion algorithms to achieve intelligent compaction calculation and pattern recognition is also a problem worth exploring. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring the intelligent compaction of a highway roadbed and pavement, so as to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for monitoring intelligent compaction of a highway subgrade and pavement, comprising:

[0008] Collecting construction site data of highway subgrade and pavement; wherein the construction site data includes: soil density, water content, and temperature;

[0009] Input the construction site data into a compaction prediction model to obtain the compaction of the roadbed and the road surface; wherein the compaction prediction model is constructed based on the quantitative relationship between the construction site data and the radar signal;

[0010] An early warning is given for the compaction degree.

[0011] Optionally, the compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

[0012] Optionally, inputting the construction site data into a compaction prediction model includes:

[0013] Input the construction site data into the soil dielectric constant calculation sub-model to obtain the initial value of the soil dielectric constant;

[0014] Determine whether the initial value of the soil dielectric constant is within a preset threshold range; if not, reacquire the construction site data and input it into the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant;

[0015] Inputting the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant;

[0016] The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of the compaction degree to obtain an estimated compaction degree distribution.

[0017] Optionally, constructing the soil dielectric constant calculation submodel includes:

[0018] Based on the construction site data and historical data of soil dielectric constant, the support vector machine is trained to obtain the soil dielectric constant calculation model.

[0019] Optionally, constructing a soil dielectric constant inversion sub-model includes:

[0020] Collect radar signals from highway roadbed and pavement;

[0021] Based on the historical data of the soil dielectric constant and the radar signal, a regression analysis method is used to construct an inversion sub-model of the soil dielectric constant.

[0022] Optionally, constructing an empirical relationship sub-model of compaction degree includes:

[0023] Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, the empirical relationship sub-model of the compaction degree is constructed using a regression analysis method; wherein the soil physical and mechanical parameters include: construction site data and compaction degree.

[0024] A highway roadbed and pavement intelligent compaction monitoring system, comprising: a data acquisition module, a model building module, a compaction prediction module and an early warning module;

[0025] The data acquisition module is used to collect construction site data of the highway subgrade and pavement; wherein the construction site data includes: soil density, water content, and temperature;

[0026] The model building module is used to build a compaction prediction model;

[0027] The compaction prediction module is used to input the construction site data into the compaction prediction model to obtain the compaction of the roadbed and the road surface;

[0028] The early warning module is used to issue an early warning for the compaction degree.

[0029] Optionally, the compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

[0030] Optionally, the model building module trains a support vector machine based on the construction site data and historical data of the soil dielectric constant to obtain a soil dielectric constant calculation model;

[0031] Collecting radar signals of the roadbed and pavement, and constructing an inversion sub-model of the soil dielectric constant by using a regression analysis method based on the soil dielectric constant and historical data of the radar signals;

[0032] Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, the empirical relationship sub-model of the compaction degree is constructed using a regression analysis method; wherein the soil physical and mechanical parameters include: construction site data and compaction degree.

[0033] Optionally, the compaction degree prediction module inputs the construction site data into the soil dielectric constant calculation sub-model to obtain an initial value of the soil dielectric constant;

[0034] Determine whether the initial value of the soil dielectric constant is within a preset threshold range; if not, reacquire the construction site data and input it into the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant;

[0035] Inputting the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant;

[0036] The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of the compaction degree to obtain an estimated compaction degree distribution.

[0037] The beneficial effects of the present invention are:

[0038] The present invention first collects construction site data of the roadbed and pavement; wherein the construction site data includes: soil density, water content, and temperature; secondly, the construction site data is input into a compaction prediction model to obtain the compaction of the roadbed and pavement; wherein the compaction prediction model is constructed based on the quantitative relationship between the construction site data and the radar signal; finally, an early warning is issued for the compaction. The present invention can realize intelligent and precise monitoring of the compaction of the roadbed and pavement, improve the accuracy and reliability of compaction quality assessment, provide data support for the optimization of compaction construction technology, and effectively ensure the quality of the roadbed and pavement engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 The present invention is a flowchart of a method for monitoring the intelligent compaction of a highway roadbed and pavement according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] Compactness, also known as tamping degree, refers to the ratio of the dry density of soil or other road construction materials after compaction to the standard maximum dry density, expressed as a percentage. For roadbeds, semi-rigid bases of pavements and flexible bases of granular materials, compaction refers to the ratio of the dry density actually achieved on site to the maximum dry density obtained by indoor standard compaction tests; for asphalt surface layers and asphalt stabilized bases, compaction refers to the ratio of the density achieved on site to the indoor standard density.

[0044] The degree of compaction is directly related to the overall performance of the material. The higher the degree of compaction, the greater the density, which means that the strength, rigidity, stability and flatness of the roadbed and pavement are guaranteed, thus extending the service life of the roadbed and pavement.

[0045] The determination of compaction mainly includes the determination of indoor standard density (maximum dry density) and on-site density test. Specific methods include ring knife method, sand filling method, water filling method, etc. These methods are suitable for different types of soil and road base.

[0046] The knife ring method is a classic compaction test method, which is suitable for fine-grained soil without aggregate. Its operation steps include: using the knife ring to take samples and measuring the dry density of the sample. The compaction degree is obtained by calculating the ratio of the dry density to the standard density. Advantages: simple equipment and easy operation. Disadvantages: limited by soil quality, and the soil may become loose during the sampling process, affecting the test results.

[0047] The sand filling method is a destructive testing method applicable to all types of soil. The basic steps include: filling standard sand into a hole of known volume to measure the density of the soil. The compaction degree is obtained by calculating the volume of the soil and the volume of the filled sand. Advantages: The measured value is accurate. Disadvantages: The operation is complicated and is greatly affected by environmental and human factors.

[0048] Nuclear density meter detection method: The nuclear density meter uses the principle of nuclear physics to determine the density of the soil by measuring the radioactive decay in the soil. The steps include: placing the nuclear density meter in the area to be tested and measuring. Comparing the measured density value with the standard value to determine the compaction degree. Advantages: It can be measured quickly and is suitable for large-area testing. Disadvantages: The equipment cost is high and strict safety regulations must be followed.

[0049] Vibration harmonic theory: A compaction degree detection method based on vibration harmonic theory, which evaluates the compaction degree of soil by analyzing the vibration signal generated by the roller during the compaction process. Its operation steps include: installing an acceleration sensor to monitor the vibration signal of the roller in real time. By analyzing the spectrum of the vibration signal, characteristic parameters related to the compaction degree are extracted. Advantages: It can realize online monitoring and real-time feedback of the compaction effect. Disadvantages: It has high requirements for equipment installation and data processing.

[0050] Infrared camera detection technology: Infrared camera technology evaluates compaction by monitoring the temperature of concrete paving and compaction at different locations. The steps include: using infrared cameras to obtain the temperature distribution map of the road surface. Determine the compaction effect based on the temperature difference. Advantages: non-destructive testing, able to quickly obtain large-scale data. Disadvantages: sensitive to environmental conditions and may be affected by climate.

[0051] Visual inspection method: Visual inspection method relies on the experience of the observer and evaluates the compaction by judging the appearance of the roadbed and pavement. Its steps include: judging the compaction state of the roadbed by visual inspection or using simple measuring tools. Advantages: simple operation and low cost. Disadvantages: highly subjective and easily affected by the observer's experience.

[0052] Although traditional compaction monitoring methods meet engineering needs to a certain extent, they generally have the following limitations: Low efficiency: Manual inspection and single equipment inspection often take a lot of time and cannot meet the needs of modern engineering for real-time monitoring. Incomplete data: Traditional methods can usually only provide local data and lack a comprehensive assessment of the overall compaction status. Environmental impact: Many traditional methods are sensitive to environmental conditions and may lead to inaccurate measurement results. High subjectivity: Methods such as visual inspection rely on the experience of the operator, and the results may vary greatly.

[0053] With the advancement of science and technology, more and more modern technologies are being introduced into compaction monitoring to overcome the shortcomings of traditional methods. For example: Radar ground penetration technology: using electromagnetic waves to detect the dielectric constant of soil and then evaluate compaction. Machine learning algorithm: by learning from historical data, a compaction prediction model is established to achieve intelligent monitoring. Sensor technology: combining multiple sensors to collect data in real time and provide a comprehensive compaction assessment. The development of these modern technologies will provide more efficient and accurate solutions for compaction monitoring and promote the quality control of civil engineering projects towards intelligence and automation.

[0054] Traditional compaction monitoring methods have played an important role in civil engineering, but with the expansion of engineering scale and the development of technology, their limitations have gradually become apparent. The introduction of modern technology has provided new ideas and methods for compaction monitoring. Future research should continue to explore how to combine these new technologies with traditional methods to achieve more efficient and accurate compaction monitoring.

[0055] like Figure 1 As shown, this embodiment proposes a method for monitoring the intelligent compaction of a highway subgrade and pavement, which includes:

[0056] Collect construction site data of highway subgrade and pavement; construction site data includes: soil density, water content, and temperature;

[0057] Input the construction site data into the compaction prediction model to obtain the compaction of the roadbed and road surface; wherein the compaction prediction model is constructed based on the quantitative relationship between the construction site data and the radar signal;

[0058] Provide early warning of compaction.

[0059] Specifically, in this embodiment, multiple sensors are deployed at the roadbed and pavement construction sites to obtain multi-source compaction data including soil density, water content, temperature, etc., and the data is transmitted to the monitoring center in real time using wireless transmission technology; the soil dielectric constant is calculated based on the obtained soil density, water content and other parameters, combined with the pre-established relationship model between the soil dielectric constant and the density and water content; radar ground penetration technology is used to emit high-frequency electromagnetic waves to the roadbed and pavement, and the distribution of the soil dielectric constant is determined based on the time and intensity changes of the reflected signal, thereby determining the compaction distribution of the roadbed and pavement; and a machine learning algorithm is used to extract and fuse the multi-source compaction data, establish a compaction prediction model, and estimate and predict the overall compaction of the roadbed and pavement in real time.

[0060] Furthermore, the compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

[0061] Furthermore, the construction site data is input into the compaction prediction model including:

[0062] Input the construction site data into the soil dielectric constant calculation sub-model to obtain the initial value of the soil dielectric constant;

[0063] Determine whether the initial value of the soil dielectric constant is within the preset threshold range; if not, reacquire the construction site data and input the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant

[0064] Input the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant;

[0065] The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of compaction degree to obtain the estimated compaction degree distribution.

[0066] Furthermore, the construction of the soil dielectric constant calculation sub-model includes:

[0067] Based on the construction site data and historical data of soil dielectric constant, the support vector machine is trained to obtain the soil dielectric constant calculation model.

[0068] Furthermore, the construction of the soil dielectric constant inversion sub-model includes:

[0069] Collect radar signals from highway roadbed and pavement;

[0070] Based on the historical data of soil dielectric constant and radar signal, the soil dielectric constant inversion sub-model is constructed using regression analysis method.

[0071] Furthermore, the empirical relationship sub-model of compaction degree is constructed including:

[0072] Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, an empirical relationship sub-model of compaction degree is constructed using regression analysis method; the soil physical and mechanical parameters include: construction site data and compaction degree.

[0073] Specifically, in this embodiment, first, by deploying a variety of sensors at the roadbed and pavement construction site, multi-source compaction data including soil density, water content, temperature, etc. are obtained, and the data is transmitted to the monitoring center in real time using wireless transmission technology.

[0074] In roadbed construction and pavement construction, the deployment of sensor arrays is a crucial step. For example, soil density sensors can be installed at various points on the predetermined roadbed to monitor the density of the compacted soil to ensure that it meets the engineering standards. Moisture content sensors can help monitor the moisture status of the soil, which is of great significance for controlling the compaction effect and preventing roadbed water damage. The deployment of temperature sensors, especially in areas with large temperature changes, can monitor the temperature changes of the soil in real time to avoid the impact of high or low temperatures on the compaction quality. The collected data needs to be processed by data preprocessing technology to improve the availability and accuracy of the data. For example, data cleaning can remove erroneous data caused by sensor failure or external interference. Data normalization can unify data from different sensors and different ranges to the same scale for subsequent analysis. These preprocessing steps are crucial to eliminate data noise and outliers, and can significantly improve the accuracy of data analysis. After data preprocessing, the data is sent to the monitoring center in real time through wireless transmission technology, such as using Wi-Fi or 4G networks. This remote transmission not only improves the efficiency of data processing, but also reduces the complexity of field operations.

[0075] Secondly, based on the obtained parameters such as soil density and water content, combined with the pre-established relationship model between soil dielectric constant, density and water content, the soil dielectric constant is calculated.

[0076] The relationship model is based on historical data and physical theory to predict the dielectric constant of soil, which is an important indicator for evaluating soil compaction and water content. If the calculated initial value of the soil dielectric constant is not within a reasonable range, this may be due to sensor errors or data transmission errors. In this case, it is necessary to re-acquire the density and water content parameters of the soil to ensure the accuracy of the data. For example, if the soil density measured initially is abnormally low, it may be due to the wrong sensor position or loose soil, and the sensor needs to be repositioned and measured. Once the initial value of the dielectric constant is confirmed to be within a reasonable range, this value can be used for further analysis. To improve the accuracy of the prediction, the relationship model can be trained and optimized using a support vector machine algorithm. This machine learning method can find a more accurate relationship between density, water content and dielectric constant by analyzing a large amount of historical data. For example, by analyzing past construction records, the model may find that the relationship between the dielectric constant of the soil and the density and water content will be different under specific temperature and humidity conditions. In addition, the ambient temperature will also affect the measurement of the dielectric constant, so it is necessary to obtain the ambient temperature parameter and compensate for the calculation result of the dielectric constant. For example, at higher temperatures, the dielectric constant of the soil may be higher, so the calculation results need to be adjusted according to the temperature to ensure the accuracy of the data. Finally, the calculated soil dielectric constant results are stored in the database and associated with the corresponding soil density parameters and water content parameters. This not only facilitates subsequent data analysis and quality control, but also provides reference data for future construction, helping the construction team optimize the construction plan and improve construction quality and efficiency.

[0077] Then, radar ground penetration technology is used to emit high-frequency electromagnetic waves to the roadbed and road surface. Based on the time and intensity changes of the reflected signal, the distribution of the soil dielectric constant is judged, and then the compaction distribution of the roadbed and road surface is determined.

[0078] When detecting roadbed and pavement structures, it is crucial to select the appropriate radar antenna frequency and detection parameters. For example, for deep roadbed detection, selecting a low-frequency antenna (such as 100MHz) can ensure that electromagnetic waves penetrate deeper, while for shallow pavement detection, a high-frequency antenna (such as 1GHz) is selected to improve resolution. Ensuring that electromagnetic waves effectively penetrate the roadbed and pavement structure is to obtain accurate internal information. The radar antenna is controlled to move along the detection path, continuously emitting high-frequency electromagnetic waves and receiving reflected echo signals. Assuming that detection is carried out on a certain section of highway, the antenna moves at a speed of 5km / h and emits electromagnetic waves every 1 meter. Such continuous detection can ensure that high-density data points are obtained and improve detection accuracy. The received reflected echo signal is time- and amplitude-corrected to eliminate interference and distortion. For example, noise is removed by time-domain filtering, and the influence of uneven antenna gain is eliminated by amplitude correction. Assuming that there is 50dB of noise in the original signal, the noise is reduced to 30dB after filtering, and the signal quality is significantly improved. Time-frequency analysis methods, such as wavelet transform, are used to extract the time-frequency characteristics of the reflected signal. Assuming that the Daubechies wavelet basis is used for transformation, it is found that there is a high-frequency component in the reflected signal at 50ns, which corresponds to the interface between the pavement and the roadbed. In this way, the distribution law of signal energy in the time and frequency domain can be revealed, which helps to identify the interface of different structural layers. According to the time-frequency characteristics of the reflected signal, a quantitative relationship model between the soil dielectric constant and the signal characteristic parameters is established. For example, through regression analysis, it is found that there is a linear relationship between the peak frequency of the reflected signal and the soil dielectric constant. The model expression is: ε=a*f+b, where ε is the dielectric constant, f is the peak frequency, and a and b are regression coefficients. Through model inversion, the spatial distribution of the soil dielectric constant in the detection area is calculated. Under the condition of known soil type and water content, the dielectric constant distribution is converted into the compaction distribution of the roadbed and pavement using the empirical relationship between the dielectric constant and the physical and mechanical parameters of the soil. Assuming that the dielectric constant distribution range of a section of the roadbed is 4 to 8, it is converted into the compaction distribution through the empirical formula K=c*ε+d (where K is the compaction, c and d are empirical coefficients). It was found that the compaction degree of a certain area was less than 90%, indicating that the area was insufficiently compacted. The compaction distribution results were visualized and analyzed to identify areas with insufficient or abnormal compaction. For example, a compaction distribution map was generated using GIS software, and the depth of color indicated the degree of compaction. It was found that there were many red areas in a certain section of road, indicating that the compaction degree was generally low and needed special attention. The implementation of this method not only improved the detection accuracy, but also provided an intuitive compaction distribution map, providing a scientific basis for the maintenance and repair of roadbed and pavement. By identifying areas with insufficient compaction, targeted reinforcement treatment can be carried out to extend the service life of the road and reduce maintenance costs.Selecting the appropriate radar antenna frequency and detection parameters to ensure effective penetration of electromagnetic waves is to obtain comprehensive data; time and amplitude correction to eliminate interference is to improve signal quality; time-frequency analysis to reveal signal characteristics is to establish an accurate model; conversion of dielectric constant and compaction is to convert physical parameters into engineering parameters; visualization and analysis are to intuitively identify problem areas. Each step is closely linked to each other and together constitutes a complete detection and analysis process.

[0079] This embodiment also proposes a highway roadbed and pavement intelligent compaction monitoring system, including: a data acquisition module, a model building module, a compaction prediction module and an early warning module;

[0080] The data acquisition module is used to collect construction site data of the highway subgrade and pavement; the construction site data includes: soil density, water content, and temperature;

[0081] A model building module, used to build a compaction prediction model;

[0082] The compaction prediction module is used to input the construction site data into the compaction prediction model to obtain the compaction of the roadbed and pavement;

[0083] The early warning module is used to warn of compaction degree.

[0084] Furthermore, the compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

[0085] Furthermore, the model building module trains the support vector machine based on the construction site data and the historical data of the soil dielectric constant to obtain the soil dielectric constant calculation model;

[0086] Collect radar signals from highway subgrades and pavements, and build a soil dielectric constant inversion submodel using regression analysis methods based on historical data of soil dielectric constants and radar signals;

[0087] Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, an empirical relationship sub-model of compaction degree is constructed using regression analysis method; the soil physical and mechanical parameters include: construction site data and compaction degree.

[0088] Furthermore, the compaction degree prediction module inputs the construction site data into the soil dielectric constant calculation sub-model to obtain the initial value of the soil dielectric constant;

[0089] Determine whether the initial value of the soil dielectric constant is within the preset threshold range; if not, reacquire the construction site data and input it into the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant;

[0090] Input the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant;

[0091] The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of compaction degree to obtain the estimated compaction degree distribution.

[0092] The above embodiments are only descriptions of the preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for monitoring the intelligent compaction of a highway subgrade and pavement, characterized in that: include: Collecting construction site data of highway subgrade and pavement; wherein the construction site data includes: soil density, water content, and temperature; Input the construction site data into a compaction prediction model to obtain the compaction of the roadbed and the road surface; wherein the compaction prediction model is constructed based on the quantitative relationship between the construction site data and the radar signal; An early warning is given for the compaction degree.

2. The method for monitoring the intelligent compaction of a highway subgrade and pavement according to claim 1, characterized in that: The compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

3. The method for monitoring the intelligent compaction of a highway subgrade and pavement according to claim 2, characterized in that: Inputting the construction site data into the compaction prediction model includes: Input the construction site data into the soil dielectric constant calculation sub-model to obtain the initial value of the soil dielectric constant; Determine whether the initial value of the soil dielectric constant is within a preset threshold range; if not, reacquire the construction site data and input it into the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant; Inputting the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant; The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of the compaction degree to obtain an estimated compaction degree distribution.

4. The method for monitoring the intelligent compaction of a highway subgrade and pavement according to claim 2, characterized in that: Constructing the soil dielectric constant calculation sub-model includes: Based on the construction site data and historical data of soil dielectric constant, the support vector machine is trained to obtain the soil dielectric constant calculation model.

5. The method for monitoring the intelligent compaction of a highway subgrade and pavement according to claim 2, characterized in that: The construction of the soil dielectric constant inversion sub-model includes: Collect radar signals from highway roadbed and pavement; Based on the historical data of the soil dielectric constant and the radar signal, a regression analysis method is used to construct an inversion sub-model of the soil dielectric constant.

6. The method for monitoring the intelligent compaction of a highway subgrade and pavement according to claim 2, characterized in that: The empirical relationship sub-model for building compaction degree includes: Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, the empirical relationship sub-model of the compaction degree is constructed using a regression analysis method; wherein the soil physical and mechanical parameters include: construction site data and compaction degree.

7. A highway roadbed and pavement intelligent compaction monitoring system, characterized in that: A method for monitoring the intelligent compaction of a highway subgrade and pavement according to any one of claims 1 to 6, comprising: a data acquisition module, a model building module, a compaction prediction module and an early warning module; The data acquisition module is used to collect construction site data of the highway subgrade and pavement; wherein the construction site data includes: soil density, water content, and temperature; The model building module is used to build a compaction prediction model; The compaction prediction module is used to input the construction site data into the compaction prediction model to obtain the compaction of the roadbed and the road surface; The early warning module is used to issue an early warning for the compaction degree.

8. The highway roadbed and pavement intelligent compaction monitoring system according to claim 7 is characterized in that: The compaction degree prediction model includes: a soil dielectric constant calculation sub-model, a soil dielectric constant inversion sub-model and a compaction degree empirical relationship sub-model.

9. The highway roadbed and pavement intelligent compaction monitoring system according to claim 8, characterized in that: The model building module trains the support vector machine based on the construction site data and the historical data of the soil dielectric constant to obtain the soil dielectric constant calculation model; Collecting radar signals of the roadbed and pavement, and constructing an inversion sub-model of the soil dielectric constant by using a regression analysis method based on the soil dielectric constant and historical data of the radar signals; Based on the soil dielectric constant and soil physical and mechanical parameters in historical data, the empirical relationship sub-model of the compaction degree is constructed using a regression analysis method; wherein the soil physical and mechanical parameters include: construction site data and compaction degree.

10. The highway roadbed and pavement intelligent compaction monitoring system according to claim 8, characterized in that: The compaction degree prediction module inputs the construction site data into the soil dielectric constant calculation sub-model to obtain the initial value of the soil dielectric constant; Determine whether the initial value of the soil dielectric constant is within a preset threshold range; if not, reacquire the construction site data and input it into the soil dielectric constant calculation submodel; if within the threshold range, output the final soil dielectric constant; Inputting the soil dielectric constant into the soil dielectric constant inversion sub-model to obtain the spatial distribution of the soil dielectric constant; The spatial distribution of the soil dielectric constant is input into the empirical relationship sub-model of the compaction degree to obtain an estimated compaction degree distribution.

Images