An expressway pavement disease intelligent monitoring system based on digital twinning

Abstract

The application relates to the technical field of digital twinning, and discloses a highway pavement disease intelligent monitoring system based on digital twinning, which comprises a multidimensional data sensing module, a pavement disease feature extraction module, a disease identification model establishment module, a virtual mapping modeling module, a digital twinning disease prediction module and a real-time interaction feedback module; multidimensional data of different road sections of a highway pavement are acquired; a crack contour information of a pavement image is extracted by using an edge detection algorithm; a void area between a pavement and a base layer of different road sections is detected according to a sound wave detection signal; a virtual three-dimensional model which is synchronous with an actual pavement state and is updated is constructed by using a digital twinning technology based on a finite element analysis method; the risk of a highway pavement disease under the change trend of a traffic flow pressure distribution is predicted; and intelligent monitoring and risk prediction of the highway pavement disease are realized.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and more specifically to an intelligent monitoring system for highway pavement defects based on digital twins. Background Technology

[0002] With the rapid development of highway networks and the continuous growth of traffic flow, the demand for intelligent road maintenance is becoming increasingly prominent, and intelligent monitoring technology for road defects is gradually becoming a research hotspot. However, existing road defect monitoring technologies still have significant shortcomings in terms of real-time performance, accuracy, and system integration, making it difficult to fully meet the high-efficiency monitoring needs under the complex working conditions of highways. Relying on manual inspections or regular patrols by vehicle-mounted lidar is costly, has low coverage, and makes it difficult to detect early-stage hidden defects. If road defects such as cracks, ruts, and potholes are not treated in time, the repair costs will increase exponentially.

[0003] However, existing pavement distress monitoring technologies still have significant shortcomings in terms of real-time performance, accuracy, and system integration, making it difficult to fully meet the high-efficiency monitoring needs under complex highway conditions. For example, patent CN114740010B proposes a pavement distress monitoring method based on laser ranging and image analysis, which identifies distress conditions such as cracks, protrusions, and subsidence by generating distance-displacement curves and combining them with image processing technology. However, this technical solution mainly relies on single laser ranging and image analysis methods, lacking the ability to fuse and process multi-source data, making it difficult to achieve comprehensive and high-precision monitoring of pavement distress. In addition, this method does not incorporate digital twin technology, failing to construct dynamic interaction between the physical world and the virtual model, thus limiting its adaptability and predictive capabilities under complex road conditions.

[0004] Existing technologies still have significant shortcomings in areas such as multi-source data fusion, digital twin modeling, and comprehensive environmental factor analysis, making it difficult to meet the needs of intelligent maintenance of modern highways. Therefore, there is an urgent need for an intelligent monitoring system that can achieve real-time interaction between the physical world and virtual models through digital twin technology, combined with multi-source data fusion analysis and dynamic environmental factor modeling. This system would improve the accuracy, real-time performance, and predictive capabilities of pavement defect monitoring, providing technical support for efficient highway maintenance. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent monitoring system for highway pavement defects based on digital twins, so as to solve the problems existing in the background art.

[0006] The present invention provides the following technical solution: a digital twin-based intelligent monitoring system for highway pavement defects, comprising: a multi-dimensional data perception module, a pavement defect feature extraction module, a defect identification model establishment module, a virtual mapping modeling module, a digital twin defect prediction module, and a real-time interactive feedback module;

[0007] The multidimensional data perception module is used to acquire multidimensional data of different road sections of the highway, including a road image acquisition unit, a sound wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit.

[0008] The pavement distress feature extraction module uses an edge detection algorithm to extract crack contour information from pavement images, analyzes the first risk index of highway pavement distress, and analyzes the second risk index of highway pavement distress based on acoustic detection signals.

[0009] The disease identification model establishment module establishes a comprehensive early warning model based on the first risk index and the second risk index of highway pavement diseases, and performs intelligent monitoring of highway pavement diseases.

[0010] The virtual mapping modeling module constructs a virtual three-dimensional model that is synchronously updated with the actual road surface condition based on the finite element analysis method, using traffic flow pressure distribution data as the X-axis, the first risk index of highway pavement distress as the Y-axis, and the second risk index of highway pavement distress as the Z-axis.

[0011] The digital twin disease prediction module predicts the risk of highway pavement diseases under the changing trend of traffic flow pressure distribution based on a virtual three-dimensional model.

[0012] The real-time interactive feedback module sends key data from the virtual model to the remote monitoring terminal via a high-speed communication network, and displays the location and size change trends of the abnormal area through a graphical interface.

[0013] Preferably, the multi-dimensional data perception module is used to acquire road surface condition information from multiple sources, including a road surface image acquisition unit, an acoustic wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit. The total number of different road segments is n, i = 1, 2, 3, ..., n, where i represents the number of different road segments.

[0014] The road surface image acquisition unit acquires road surface images of different road sections by using a camera array installed along the highway.

[0015] The acoustic wave detection signal acquisition unit: emits high-frequency acoustic waves and receives reflected signals through an acoustic wave detection device to acquire acoustic wave detection signals of different sections of the highway and analyze whether there are cracks or voids in the internal structure of the road surface.

[0016] The traffic flow pressure distribution data acquisition unit uses piezoelectric sensors buried under the highway surface to collect traffic flow pressure distribution data for different road sections.

[0017] Preferably, the specific content of the pavement defect feature extraction module is as follows:

[0018] The analysis of the first risk index of highway pavement distress includes: highway pavement distress identification unit, pavement distress feature extraction unit, and highway pavement distress first risk index analysis unit.

[0019] The analysis of the second risk index of highway pavement distress includes: an acoustic detection signal extraction unit and a highway pavement distress second risk index analysis unit.

[0020] Preferably, the highway pavement distress identification unit: extracts pavement texture changes through image processing technology, identifies distress areas in different road sections, the distress areas include crack areas, rut areas and subsidence areas, and classifies and marks the distress areas through machine learning algorithms;

[0021] The pavement distress feature extraction unit uses edge detection to separate distressed areas from the background in different road sections, and then transmits the separated distressed areas to the highway pavement distress first risk index analysis unit for risk analysis.

[0022] The highway pavement distress first risk index analysis unit calculates the highway pavement distress first risk index using a quantitative evaluation method based on the area, length, width, and depth information of distressed areas in different road sections.

[0023] Preferably, the acoustic wave detection signal extraction unit: uses a piezoelectric crystal transducer to emit acoustic waves of a specific frequency, the acoustic waves propagate and reflect in the road structure, the receiver captures the reflected signal, generates an elastic wave signal through the reflected acoustic waves, and extracts acoustic wave detection signals from different road sections. The acoustic wave detection signal includes: acoustic wave frequency, acoustic wave amplitude, and acoustic wave velocity.

[0024] The second risk index analysis unit for highway pavement distress integrates acoustic detection signals from different road sections to construct a risk assessment index system, detects the void area between the pavement and the base layer in different road sections, and calculates the second risk index for highway pavement distress.

[0025] Preferably, the defect identification model establishment module analyzes highway surface defects and the separation between the highway pavement and the base layer based on the first risk index and the second risk index of highway pavement defects, respectively, and establishes a comprehensive early warning model. When the output result of the comprehensive early warning model is greater than the maximum value of the preset risk judgment interval, the judgment result is high risk; when the output result of the comprehensive early warning model is within the preset risk judgment interval, the judgment result is medium risk; and when the output result of the comprehensive early warning model is less than the minimum value of the preset risk judgment interval, the judgment result is low risk. The judgment result is then transmitted to the virtual mapping modeling module and the real-time interactive feedback module.

[0026] Preferably, the virtual mapping modeling module includes a judgment result presentation unit and a virtual 3D model establishment unit;

[0027] The judgment result presentation unit: according to the risk level, the risky road segments are color-coded, with red areas indicating road segments with a judgment result of high risk, yellow areas indicating road segments with a judgment result of medium risk, and green areas indicating road segments with a judgment result of low risk.

[0028] Virtual 3D Model Building Unit: Based on the finite element analysis method, a virtual 3D model is constructed and updated synchronously with the actual road surface condition. The traffic flow pressure distribution data is used as the X-axis, the first risk index of highway pavement distress is used as the Y-axis, and the second risk index of highway pavement distress is used as the Z-axis to show the impact of traffic flow pressure distribution on highway pavement distress risk in different road sections.

[0029] Preferably, the specific content of the digital twin disease prediction module is as follows:

[0030] Based on the time nodes of disease evolution prediction, the average daily traffic pressure value of different road sections is analyzed using historical traffic pressure data of different road sections. The time nodes of disease evolution prediction include daily prediction, weekly prediction and monthly prediction.

[0031] When the time node for predicting the evolution of road defects is daily, the average daily traffic pressure value of different road sections is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment.

[0032] When the time node for predicting the evolution of road defects is weekly, the product of the daily average traffic pressure value of different road sections and the number of days in a week is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment.

[0033] When the prediction time point for the evolution of road defects is monthly, the product of the average daily traffic pressure value of different road sections and the number of days in the month is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for prediction risk judgment.

[0034] Preferably, the real-time interactive feedback module is used to receive the risk assessment results from the disease identification model establishment module, and to provide risk warnings for highway pavement diseases based on the risk assessment results; to send key data in the virtual model to the remote monitoring terminal through a high-speed communication network, and to display the location and size change trends of abnormal areas through a graphical interface.

[0035] The technical effects and advantages of this invention are as follows:

[0036] This invention acquires multidimensional data of different sections of highway pavement by incorporating a multidimensional data perception module, a pavement distress feature extraction module, a distress identification model establishment module, a virtual mapping modeling module, a digital twin distress prediction module, and a real-time interactive feedback module. It uses an edge detection algorithm to extract crack contour information from pavement images, detects voids between the pavement and base layer in different sections based on acoustic detection signals, and constructs a virtual three-dimensional model that is synchronously updated with the actual pavement condition using digital twin technology based on finite element analysis. This model predicts the risk of highway pavement distress under changing traffic pressure distribution trends, thus achieving intelligent monitoring and risk prediction of highway pavement distress. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of a digital twin-based intelligent monitoring system for highway pavement defects. Detailed Implementation

[0038] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The intelligent monitoring system for highway pavement defects based on digital twins involved in the present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] like Figure 1 As shown, the present invention provides an intelligent monitoring system for highway pavement defects based on digital twins, including: a multi-dimensional data perception module, a pavement defect feature extraction module, a defect identification model establishment module, a virtual mapping modeling module, a digital twin defect prediction module, and a real-time interactive feedback module.

[0040] The multidimensional data perception module is used to acquire multidimensional data of different road sections of the highway, including a road image acquisition unit, a sound wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit.

[0041] The pavement distress feature extraction module uses an edge detection algorithm to extract crack contour information from pavement images, analyzes the first risk index of highway pavement distress, and analyzes the second risk index of highway pavement distress based on acoustic detection signals.

[0042] The disease identification model establishment module establishes a comprehensive early warning model based on the first risk index and the second risk index of highway pavement diseases, and performs intelligent monitoring of highway pavement diseases.

[0043] The virtual mapping modeling module constructs a virtual three-dimensional model that is synchronously updated with the actual road surface condition based on the finite element analysis method, using traffic flow pressure distribution data as the X-axis, the first risk index of highway pavement distress as the Y-axis, and the second risk index of highway pavement distress as the Z-axis.

[0044] The digital twin disease prediction module predicts the risk of highway pavement diseases under the changing trend of traffic flow pressure distribution based on a virtual three-dimensional model.

[0045] The real-time interactive feedback module sends key data from the virtual model to the remote monitoring terminal via a high-speed communication network, and displays the location and size change trends of the abnormal area through a graphical interface.

[0046] In this embodiment, it should be specifically noted that the multi-dimensional data perception module is used to acquire road surface condition information from multiple sources, including a road surface image acquisition unit, a sound wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit. The total number of different road segments is n, i = 1, 2, 3, ..., n, where i represents the number of different road segments.

[0047] The road surface image acquisition unit acquires road surface images of different road sections by installing a camera array along the highway. The camera uses a high-resolution industrial camera, which can capture road surface detail information with millimeter-level precision.

[0048] The acoustic wave detection signal acquisition unit: emits high-frequency acoustic waves and receives reflected signals through an acoustic wave detection device to acquire acoustic wave detection signals of different sections of the highway and analyze whether there are cracks or voids in the internal structure of the road surface.

[0049] The traffic flow pressure distribution data acquisition unit uses piezoelectric sensors buried under the highway surface to collect traffic flow pressure distribution data for different road sections.

[0050] In this embodiment, it should be specifically explained that the specific content of the road surface defect feature extraction module is as follows:

[0051] The analysis of the first risk index of highway pavement distress includes: highway pavement distress identification unit, pavement distress feature extraction unit, and highway pavement distress first risk index analysis unit.

[0052] The analysis of the second risk index of highway pavement distress includes: an acoustic detection signal extraction unit and a highway pavement distress second risk index analysis unit.

[0053] In this embodiment, it should be specifically explained that the highway pavement defect identification unit: extracts pavement texture changes through image processing technology, identifies defect areas in different road sections, the defect areas include crack areas, rut areas and subsidence areas, and classifies and marks the defect areas through machine learning algorithms. Transverse cracks: cracks perpendicular to the road driving direction, usually caused by temperature changes or foundation settlement; longitudinal cracks: cracks parallel to the road driving direction, mostly caused by uneven subgrade settlement or improper construction joint treatment; mesh cracks: small cracks that intersect each other to form a mesh pattern, usually related to pavement aging and insufficient base strength; rut areas: permanent pavement deformation caused by repeated vehicle rolling, usually appearing in wheel track areas; subsidence areas: local subsidence of pavement areas, possibly caused by insufficient subgrade compaction or underground cavities.

[0054] The pavement distress feature extraction unit uses edge detection to separate distressed areas from the background in different road sections, and then transmits the separated distressed areas to the highway pavement distress first risk index analysis unit for risk analysis.

[0055] The specific algorithm for edge detection is as follows: The gradient magnitude of each pixel is calculated based on the grayscale gradients of each pixel in the horizontal and vertical directions of the road surface image for different road segments. The calculation formula is: ,in This represents the gradient magnitude of each pixel in the road surface image of different road sections. This represents the grayscale gradient of each pixel in the road surface image of different road sections in the horizontal direction. This represents the grayscale gradient of each pixel in the vertical direction in the road surface image of different road sections;

[0056] By preserving local maxima along the gradient direction, suppressing non-edge points, segmenting edges using high and low thresholds, and connecting weak edges through hysteresis thresholding, the diseased areas of different road sections are separated from the background.

[0057] The highway pavement distress first risk index analysis unit calculates the highway pavement distress first risk index using a quantitative assessment method based on the area, length, width, and depth information of distressed areas in different road sections. The calculation formula is as follows: ,in This indicates the highest risk index for road surface defects on highways. This represents the correction factor. This indicates the area of ​​the damaged section in different road sections. This represents the total area of ​​different road sections. Indicates the length of road defects in different sections. This indicates the threshold for the length of road defects in different road sections. Indicates the width of road defects in different sections. This indicates the threshold width for road defects in different road sections. Indicates the depth of road damage in different sections. Indicates the critical depth of road defects in different road sections. , , as well as These represent the weighting coefficients, and their sum is 1.

[0058] In this embodiment, it should be specifically explained that the acoustic wave detection signal extraction unit: uses a piezoelectric crystal transducer to emit acoustic waves of a specific frequency, the acoustic waves propagate and reflect in the road structure, the receiver captures the reflected signal, generates an elastic wave signal through the reflected acoustic wave, and extracts acoustic wave detection signals from different road sections. The acoustic wave detection signal includes: acoustic wave frequency, acoustic wave amplitude, and acoustic wave velocity.

[0059] The second risk index analysis unit for highway pavement distress integrates acoustic detection signals from different road sections to construct a risk assessment index system, detects the void area between the pavement and the base layer in different road sections, and calculates the second risk index for highway pavement distress.

[0060] The sound frequency anomaly index for different road sections is calculated based on the measured sound frequency. The calculation formula is as follows: ,in This indicates the sound wave frequency anomaly index for different road sections. This indicates the measured sound wave frequencies for different road sections. This indicates the sound wave frequency of the preset healthy road surface;

[0061] The sound wave amplitude attenuation index for different road sections is calculated based on the measured sound wave amplitude. The calculation formula is as follows: ,in This represents the sound wave amplitude attenuation index for different road sections. This indicates the amplitude of the incident wave in different road sections. This indicates the amplitude of reflected waves in different road sections;

[0062] The sound wave velocity variation index for different road sections is calculated based on the measured sound wave velocity. The calculation formula is as follows: ,in This represents the index indicating the change in sound wave velocity across different road sections. This indicates the measured sound wave velocity on different road sections. This indicates the sound wave velocity of the preset healthy road surface;

[0063] The formula for calculating the second risk index of highway pavement distress is as follows: ,in This represents the second risk index for highway pavement distress, where e represents a constant. , and These represent the weighting coefficients, and their sum is 1.

[0064] In this embodiment, it should be specifically noted that the defect identification model establishment module analyzes highway surface defects and the separation between the highway pavement and the base layer based on the first risk index and the second risk index of highway pavement defects, respectively, and establishes a comprehensive early warning model. The expression of the comprehensive early warning model is as follows: ,in This represents the output value of the integrated early warning model. This represents the preset first decision function. This represents the preset second judgment function; when the output of the comprehensive early warning model is greater than the maximum value of the preset risk judgment interval, the judgment result is high risk; when the output of the comprehensive early warning model is within the preset risk judgment interval, the judgment result is medium risk; when the output of the comprehensive early warning model is less than the minimum value of the preset risk judgment interval, the judgment result is low risk, and the judgment result is transmitted to the virtual mapping modeling module and the real-time interactive feedback module.

[0065] In this embodiment, it should be specifically noted that the virtual mapping modeling module includes a judgment result presentation unit and a virtual three-dimensional model establishment unit;

[0066] The judgment result presentation unit: according to the risk level, the risky road segments are color-coded, with red areas indicating road segments with a judgment result of high risk, yellow areas indicating road segments with a judgment result of medium risk, and green areas indicating road segments with a judgment result of low risk.

[0067] Virtual 3D Model Building Unit: Based on the finite element analysis method, a virtual 3D model is constructed and updated synchronously with the actual road surface condition. The traffic flow pressure distribution data is used as the X-axis, the first risk index of highway pavement distress is used as the Y-axis, and the second risk index of highway pavement distress is used as the Z-axis to show the impact of traffic flow pressure distribution on highway pavement distress risk in different road sections.

[0068] In this embodiment, it should be specifically explained that the specific content of the digital twin disease prediction module is as follows:

[0069] Based on the time nodes of disease evolution prediction, the average daily traffic pressure value of different road sections is analyzed using historical traffic pressure data of different road sections. The time nodes of disease evolution prediction include daily prediction, weekly prediction and monthly prediction.

[0070] When the time node for predicting the evolution of road defects is daily, the average daily traffic pressure value of different road sections is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment.

[0071] When the time node for predicting the evolution of road defects is weekly, the product of the daily average traffic pressure value of different road sections and the number of days in a week is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment.

[0072] When the prediction time point for the evolution of road defects is monthly, the product of the average daily traffic pressure value of different road sections and the number of days in the month is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for prediction risk judgment.

[0073] In this embodiment, it should be specifically explained that the real-time interactive feedback module is used to receive the risk judgment result of the disease identification model establishment module, and to provide risk warning for highway pavement diseases based on the risk judgment result; the key data in the virtual model is sent to the remote monitoring terminal through the high-speed communication network, and the location and size change trend of the abnormal area are displayed through the graphical interface.

[0074] In this embodiment, it should be specifically noted that the main difference between this embodiment and the prior art lies in that this embodiment acquires multidimensional data of different road sections of the highway through a multidimensional data perception module, a road surface defect feature extraction module, a defect identification model establishment module, a virtual mapping modeling module, a digital twin defect prediction module, and a real-time interactive feedback module. It uses an edge detection algorithm to extract crack contour information from the road surface image, detects the void area between the road surface and the base layer in different road sections based on acoustic detection signals, and constructs a virtual three-dimensional model that is synchronously updated with the actual road surface condition based on the finite element analysis method using digital twin technology. This predicts the risk of highway road surface defects under the changing trend of traffic flow pressure distribution, thereby realizing intelligent monitoring and risk prediction of highway road surface defects.

[0075] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0076] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A digital twin-based intelligent monitoring system for highway pavement defects, characterized in that: include: The system includes a multi-dimensional data perception module, a road surface defect feature extraction module, a defect identification model establishment module, a virtual mapping modeling module, a digital twin defect prediction module, and a real-time interactive feedback module. The multidimensional data perception module is used to acquire multidimensional data of different road sections of the highway, including a road image acquisition unit, a sound wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit. The pavement distress feature extraction module uses an edge detection algorithm to extract crack contour information from pavement images, analyzes the first risk index of highway pavement distress, and analyzes the second risk index of highway pavement distress based on acoustic detection signals. The specific content of the road surface defect feature extraction module is as follows: The analysis of the primary risk index for highway pavement distress includes: Highway pavement distress identification unit: Extracts pavement texture changes through image processing technology, identifies distress areas in different road sections, including crack areas, rut areas and subsidence areas, and classifies and marks distress areas through machine learning algorithms; Road surface distress feature extraction unit: uses edge detection to separate distressed areas from the background of different road sections, and transmits the separated distressed areas to the first risk index analysis unit for highway road surface distress for risk analysis; The analysis unit for the first risk index of highway pavement distress: Based on the area, length, width and depth information of distressed areas in different road sections, the first risk index of highway pavement distress is calculated using a quantitative evaluation method; The disease identification model establishment module establishes a comprehensive early warning model based on the first risk index and the second risk index of highway pavement diseases, and performs intelligent monitoring of highway pavement diseases. The virtual mapping modeling module constructs a virtual three-dimensional model that is synchronously updated with the actual road surface condition based on the finite element analysis method, using traffic flow pressure distribution data as the X-axis, the first risk index of highway pavement distress as the Y-axis, and the second risk index of highway pavement distress as the Z-axis. The virtual mapping modeling module includes a judgment result presentation unit and a virtual 3D model establishment unit; The judgment result presentation unit: according to the risk level, the risky road segments are color-coded, with red areas indicating road segments with a judgment result of high risk, yellow areas indicating road segments with a judgment result of medium risk, and green areas indicating road segments with a judgment result of low risk. Virtual 3D Model Building Unit: Based on the finite element analysis method, a virtual 3D model is constructed and updated synchronously with the actual road surface condition. The traffic flow pressure distribution data is used as the X-axis, the first risk index of highway pavement distress is used as the Y-axis, and the second risk index of highway pavement distress is used as the Z-axis to show the impact of traffic flow pressure distribution on highway pavement distress risk in different road sections. The digital twin disease prediction module predicts the risk of highway pavement diseases under the changing trend of traffic flow pressure distribution based on a virtual three-dimensional model. The real-time interactive feedback module sends key data from the virtual model to the remote monitoring terminal via a high-speed communication network, and displays the location and size change trends of the abnormal area through a graphical interface.

2. The intelligent monitoring system for highway pavement defects based on digital twins according to claim 1, characterized in that: The multi-dimensional data perception module is used to acquire road surface condition information from multiple sources, including a road surface image acquisition unit, an acoustic wave detection signal acquisition unit, and a traffic flow pressure distribution data acquisition unit. The total number of different road segments is n, i = 1, 2, 3, ..., n, where i represents the number of different road segments. The road surface image acquisition unit acquires road surface images of different road sections by using a camera array installed along the highway. The acoustic wave detection signal acquisition unit: emits high-frequency acoustic waves and receives reflected signals through an acoustic wave detection device to acquire acoustic wave detection signals of different sections of the highway and analyze whether there are cracks or voids in the internal structure of the road surface. The traffic flow pressure distribution data acquisition unit uses piezoelectric sensors buried under the highway surface to collect traffic flow pressure distribution data for different road sections.

3. The intelligent monitoring system for highway pavement defects based on digital twins according to claim 1, characterized in that: The analysis of the second risk index for highway pavement distress includes: Acoustic wave detection signal extraction unit: It uses a piezoelectric crystal transducer to emit sound waves of a specific frequency. The sound waves propagate and reflect in the road structure. The receiver captures the reflected signal, generates an elastic wave signal through the reflected sound wave, and extracts the acoustic wave detection signals of different road sections. The acoustic wave detection signal includes: sound wave frequency, sound wave amplitude, and sound wave velocity. The second risk index analysis unit for highway pavement distress integrates acoustic detection signals from different road sections to construct a risk assessment index system, detects the void area between the pavement and the base layer in different road sections, and calculates the second risk index for highway pavement distress.

4. The intelligent monitoring system for highway pavement defects based on digital twins according to claim 1, characterized in that: The defect identification model building module analyzes highway surface defects and the separation between the highway pavement and the base layer based on the first risk index and the second risk index of highway pavement defects, respectively, and establishes a comprehensive early warning model. When the output of the comprehensive early warning model is greater than the maximum value of the preset risk judgment interval, the judgment result is high risk; when the output of the comprehensive early warning model is within the preset risk judgment interval, the judgment result is medium risk; and when the output of the comprehensive early warning model is less than the minimum value of the preset risk judgment interval, the judgment result is low risk. The judgment result is then transmitted to the virtual mapping modeling module and the real-time interactive feedback module.

5. The intelligent monitoring system for highway pavement defects based on digital twins according to claim 1, characterized in that: The specific contents of the digital twin disease prediction module are as follows: Based on the time nodes of disease evolution prediction, the average daily traffic pressure value of different road sections is analyzed using historical traffic pressure data of different road sections. The time nodes of disease evolution prediction include daily prediction, weekly prediction and monthly prediction. When the time node for predicting the evolution of road defects is daily, the average daily traffic pressure value of different road sections is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment. When the time node for predicting the evolution of road defects is weekly, the product of the daily average traffic pressure value of different road sections and the number of days in a week is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for predictive risk judgment. When the prediction time point for the evolution of road defects is monthly, the product of the average daily traffic pressure value of different road sections and the number of days in the month is input into the virtual three-dimensional model as X-axis data. The virtual three-dimensional model outputs the first risk index and the second risk index of road defects, and returns the output results to the defect identification model building module for prediction risk judgment.

6. The intelligent monitoring system for highway pavement defects based on digital twins according to claim 1, characterized in that: The real-time interactive feedback module is used to receive the risk assessment results from the disease identification model establishment module and to provide risk warnings for highway pavement diseases based on the risk assessment results. Key data from the virtual model is sent to a remote monitoring terminal via a high-speed communication network, and the location and size change trends of the abnormal area are displayed through a graphical interface.

Images