A method and system for optimizing the deployment of embedded sensing units in smart roads

Abstract

This invention relates to the field of intelligent traffic detection, specifically to a method and system for optimizing the deployment of embedded sensing units in smart roads. The method includes: modeling and simulating the road; calculating the dynamic response of the road model under target conditions; determining scene segmentation thresholds; dividing the road into different scenes; selecting deployment points; constructing a monitoring index system; and constructing a deployment scheme for sensing units in each scene. This invention is based on mechanical simulation-driven quantification and can adapt to differences in road scenes, making the monitoring scheme more scientific and more aligned with engineering realities. Furthermore, this invention adjusts the sensing unit deployment density according to the segmented scenes, improving the representativeness of monitoring data and supporting accurate evaluation of road service status. It not only increases monitoring coverage but also reduces deployment costs and enables road detection under the same monitoring index system, improving the standardization of road monitoring.

Description

Technical Field

[0001] This invention relates to the field of intelligent traffic detection, and in particular to a method and system for optimizing the deployment of embedded sensing units in smart roads. Background Technology

[0002] With the development of information technologies such as the Internet, cloud computing, and big data, people's demand for intelligent transportation infrastructure has become more urgent. Intelligent road infrastructure, aimed at serving future smart transportation, has become an important direction for development in the contemporary transportation field. Among these, intelligent sensing is the foundation and forefront of research in intelligent road infrastructure. Intelligent sensing roads are road infrastructures that achieve service functions such as active sensing, automatic identification, autonomous adjustment, and dynamic indication through specific sensory communication, data network, and material structure system design. Research on the perception of road surface environment, mechanics, and traffic conditions in intelligent sensing roads has significant and far-reaching implications for fundamentally improving the service level of road infrastructure and for transportation construction and development.

[0003] The key to realizing intelligent road infrastructure is the accurate perception of complex and ever-changing road conditions, structural status, and traffic loads. Embedded sensing units, as an important component of intelligent sensing roads, are crucial for the long-term acquisition of intelligent road service information and can provide important guarantees for evaluating road service status and reducing safety risks.

[0004] The current deployment scheme for embedded sensing units in smart roads has a fixed deployment pattern. The deployment of longitudinal and cross sections, sensor selection, and monitoring indicators all adopt a fixed pattern without making personalized adjustments based on differences in road structure, environment, and load. This results in a higher risk of monitoring failure in special road sections.

[0005] The imbalance between deployment cost and coverage, reliance on manual experience or fixed-interval deployment, insufficient monitoring of key road sections and repeated deployment on ordinary road sections result in serious waste of resources and make it difficult to achieve the dual goals of low cost and high coverage. Consequently, the deployment of embedded sensing units in smart roads lacks unified scenario division, indicator construction, and scientific cross-section layout, leading to poor engineering implementation. Summary of the Invention

[0006] The purpose of this invention is to solve the four core problems of current smart road embedded sensing unit deployment schemes: poor scenario adaptability, high cost, low coverage, and lack of unified standards, and to provide a method and system for optimizing the deployment of smart road embedded sensing units.

[0007] In a first aspect, the present invention provides a method for optimizing the deployment of embedded sensing units in smart roads, comprising the following steps: S1. Model and simulate the road, divide the scene, and select layout points; including: Establish a road structure model and perform numerical simulation to calculate the stress, strain, and vertical deformation of the road model under target conditions; The dynamic response and strain field changes of each structural layer under different loads and pavement structures are compared; the standard deviation of three core indicators of each structural layer—peak stress, peak strain, and cumulative settlement deformation—is quantitatively analyzed to determine the scene division threshold; the road is divided into different scenes according to the scene division threshold; wherein, the scene division threshold includes at least the intra-scene division threshold and the inter-scene division threshold. Within the scene, the cross-sectional locations where the mechanical response is at the median level and the difference from the average response of the entire cross-section is less than the error threshold are selected as representative points. The scenarios are divided into key scenarios and ordinary scenarios; S2. Select monitoring equipment and construct a monitoring indicator system; S3. Based on the divided scenes, construct the layout scheme of the horizontal and vertical cross-sectional sensing units for each scene. Among them, the key scenarios are segmented and encrypted according to the characteristics of the scenarios, multi-source sensing units are deployed in a grid, and a hybrid topology network combining backbone nodes and edge nodes is constructed. For the general scene, key representative points are selected for the deployment of sensing units.

[0008] According to a preferred embodiment, the step of establishing a road structure model and performing numerical simulation to calculate the stress, strain, and vertical deformation of the road model under target conditions includes: A finite element structural model of the road was established by combining the topography, roadbed structure, and environmental hydrological conditions along the road. Conduct dynamic response numerical simulation to calculate the stress, strain, and vertical deformation of each structural layer of the road under the target traffic level and the target lifespan of the cumulative equivalent axle load.

[0009] According to a preferred embodiment, establishing the finite element structural model of the road further includes: A preliminary classification of road structure scene types is conducted. The structural scenario types include: general road sections, long longitudinal slopes, curved road sections, embankments, cuts, semi-fill and semi-cut sections, special roadbeds, bridge approach transition sections, and entrance and exit ramps.

[0010] According to a preferred embodiment, the scene segmentation threshold further includes an inter-scene segmentation threshold. Verification of the initially segmented scene types based on the scene segmentation threshold includes: Determine whether the difference in dynamic response between different scenes reaches the threshold for scene division; if not, adjust the scene boundaries. The same scene is divided into sub-scenes using the scene division threshold, and the number of monitoring sections is adjusted according to the number of sub-scenes.

[0011] According to a preferred embodiment, the scene segmentation threshold further includes an inter-scene segmentation threshold. The inter-scene segmentation threshold is used to clearly define the boundaries of each scene, classifying roads into different structural scene types. The same scene is divided into sub-scenes using the scene division threshold, and the number of monitoring sections is adjusted according to the number of sub-scenes.

[0012] The structural scenario types include: general road sections, long longitudinal slopes, curved road sections, embankments, cuts, semi-fill and semi-cut sections, special roadbeds, bridge approach transition sections, and entrance and exit ramps.

[0013] According to a preferred embodiment, the simulation process fixes structural layer parameters, environmental boundary conditions, gradient load levels, pavement structure combinations, and subgrade type conditions. The stress / strain cloud maps within each structural layer are sorted by numerical value, and the top 20% of high-value areas are selected as concentrated intervals. Based on historical disease statistics, sections where the disease density reaches a preset threshold within the same scenario are defined as high-risk, sensitive sections. Based on the simulation results, the spatial locations where stress, strain, or deformation within each structural layer reaches local maximum values ​​are determined as extreme points of structural dynamic response. Based on the simulation results, areas where the cumulative increase in plastic deformation exceeds 1.5 times the average value are identified as points prone to long-term deformation deterioration.

[0014] According to a preferred embodiment, a dual-dimensional evaluation model of unit monitoring cost and monitoring coverage effectiveness is used to determine the economic efficiency of monitoring points. The core indicators are the effective monitoring coverage mileage of a single sensor and the effectiveness of data-driven disease early warning. Redundant monitoring points and ineffective points with low response sensitivity are eliminated, balancing the total number of monitoring points with monitoring accuracy. Within the same scene and structural layer, if the difference in mechanical response between two monitoring points never exceeds the scene's classification threshold, then either monitoring point is considered a redundant monitoring point. Under standard load, if the difference between the mechanical response amplitude of a monitoring point and the average response amplitude of its layer exceeds the scene's classification threshold, then that monitoring point is considered an ineffective point with low response sensitivity.

[0015] According to a preferred embodiment, step S2, selecting monitoring equipment and constructing a monitoring index system, includes: constructing a differentiated, layered monitoring index system based on the unique stress characteristics, disease risk types, and structural deterioration patterns of each scenario after segmentation, and matching the embedded sensing device model and installation parameters accordingly. The scenario characteristics include: road section structural type, load distribution characteristics, hydrogeological conditions, types of long-term disease susceptibility, and structural layer deformation sensitivity, determined comprehensively through scenario mechanical simulation response results, geological survey data along the route, and historical disease statistics. The monitoring index system includes general basic monitoring indicators and scenario-specific supplementary indicators. The general basic monitoring indicators include: internal structural temperature and humidity, interlayer stress, structural strain, vertical displacement, traffic load, and vehicle flow. The specific supplementary indicators for the scenarios include: adding monitoring of subgrade earth pressure and subgrade moisture content for embankment subgrades; adding monitoring of lateral displacement and slope tilt deformation for cut slope sections; adding monitoring of uneven settlement difference for semi-fill and semi-cut sections; adding specific monitoring of bridge approach slab deformation and interlayer voids for bridge approach transition sections; and matching corresponding monitoring equipment according to the monitoring indicators.

[0016] According to a preferred embodiment, step S3, constructing a layout scheme for the horizontal and vertical cross-sectional sensing units of each scene based on the divided scenes, includes: Cross-section layout focuses on longitudinal layered monitoring, surface layer is buried with temperature and strain sensors to monitor asphalt fatigue damage, base layer is laid with earth pressure cells to assess bearing status, and deep subgrade is installed with displacement gauges to monitor settlement risk. The lateral layout deploys multi-parameter composite sensors at the lane dividers to simultaneously collect data on wheel track load and structural response.

[0017] In a second aspect, the present invention also provides a smart road embedded sensing unit optimization deployment system, comprising at least a processing unit. The processing unit stores a computer program, which, when executed by a processor, implements the smart road embedded sensing unit optimization deployment method provided by the present invention.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a method for optimizing the deployment of sensing units based on mechanical simulation quantification, which can adapt to differences in road scenarios. This makes the monitoring scheme more scientific and more in line with engineering practice. Furthermore, this invention improves the representativeness of monitoring data by quantifying the scenario division threshold through mechanical simulation and adjusting the deployment density of sensing units according to the divided scenarios. This enables accurate evaluation of the service status of roads, not only improving monitoring coverage but also reducing deployment costs. Moreover, it allows for road detection under the same monitoring index system, improving the standardization of road monitoring. Attached Figure Description

[0019] Figure 1This is a flowchart of the method for optimizing the deployment of embedded sensing units in smart roads according to the present invention.

[0020] Figure 2 This is a plan view of a key scenario encryption deployment example in this invention.

[0021] Figure 3 This is an elevation layout diagram of a key scenario encryption deployment case in this invention.

[0022] Figure 4 This is a plan view of a normal layout example in a typical scenario of this invention.

[0023] Figure 5 This is an elevation layout diagram of a typical scenario in this invention. Detailed Implementation

[0024] The present invention will now be described in further detail with reference to specific embodiments. However, this should not be construed as limiting the scope of the present invention to the following embodiments; all technologies implemented based on the content of the present invention fall within the scope of the present invention.

[0025] Unless otherwise specified, the terms "upper," "lower," "left," "right," "center," "inner," and "outer," etc., used in the description of specific embodiments of the present invention to indicate orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationship in which the product / equipment / device is usually placed during use. These terms are merely for the purpose of facilitating the description of the present invention or simplifying the description in specific embodiments, and for enabling those skilled in the art to quickly understand the solution, and do not indicate or imply that a particular device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, they should not be construed as limitations on the present invention.

[0026] Furthermore, the use of terms such as "horizontal," "vertical," "suspended," "parallel," and "coaxial" does not imply that the corresponding device / component / element must be absolutely horizontal, vertical, suspended, parallel, or coaxial. Slight tilt or deviation is permissible, as long as it does not affect the normal function of the relevant component. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," not that the structure must be perfectly horizontal; a slight tilt is acceptable. "Coaxial" means that two components are arranged as coaxially as possible, allowing them to move coaxially or approximately coaxially when their relative positions change. Alternatively, it can be simplified to mean that the corresponding device / component / element, when arranged in "horizontal," "vertical," "suspended," "parallel," or "coaxial" directions, can have an error / deviation of ±10% relative to the corresponding direction, more preferably within ±8%, more preferably within ±6%, more preferably within ±5%, and more preferably within ±4%. For example, the deviation in the "coaxial" direction is controlled within 0.2-1mm, preferably within 0.2-0.5mm. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its function in the solution of the present invention.

[0027] Furthermore, the use of terms such as "first," "second," and "third" in terminology is merely for distinguishing descriptions of identical or similar components and should not be interpreted as emphasizing or implying the relative importance of a particular component.

[0028] Furthermore, in the description of the embodiments of the present invention, "several", "more than", and "a number of" represent at least two. The number can be any number, such as two, three, four, five, six, seven, eight, or nine, and can even exceed nine.

[0029] Furthermore, in the description of the technical solution of this invention, unless otherwise explicitly specified / limited / restricted, the terms "set up," "install," "connect," "link," "provided with," "laid out," and "arranged" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to connection methods commonly used in the art, such as welding, riveting, bolting, and threaded connections. Such connections can be mechanical, electrical, or communication connections; they can be direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components.

[0030] Example 1 This embodiment provides a method for optimizing the deployment of embedded sensing units in smart roads. See also... Figure 1 The optimized deployment method for embedded sensing units in smart roads includes the following steps: S1. Model and simulate the road, divide the scene, and select the layout points; S2. Select monitoring equipment and construct a monitoring indicator system; S3. Based on the divided scenes, construct the layout scheme of the perception units in the horizontal and vertical sections of each scene.

[0031] Preferably, step S1 includes: Establish a road structure model and perform numerical simulation to calculate the stress, strain, and vertical deformation of the road model under target conditions; The dynamic response and strain field changes of each structural layer under different loads and pavement structures were compared; the standard deviation of three core indicators of each structural layer—peak stress, peak strain, and cumulative settlement deformation—was quantitatively analyzed to determine the scene division threshold; the road was divided into different scenes according to the scene division threshold; the scene division threshold includes the intra-scene division threshold and the inter-scene division threshold. Within the scene, the cross-sectional locations where the mechanical response is at the median level and the difference from the average response of the entire cross-section is less than the error threshold are selected as representative points. The scenarios are divided into key scenarios and ordinary scenarios.

[0032] Preferably, in step S3, when constructing the deployment scheme, multi-source sensing units are deployed in a grid-like manner for key scenarios based on their characteristics, while a hybrid topology network combining backbone nodes and edge nodes is constructed; for ordinary scenarios, key representative points are selected for sensing unit deployment.

[0033] This invention provides a method for optimizing the deployment of sensing units based on mechanical simulation quantification, which can adapt to differences in road scenarios. This makes the monitoring scheme more scientific and more in line with engineering practice. Furthermore, this invention improves the representativeness of monitoring data by quantifying the scenario division threshold through mechanical simulation and adjusting the deployment density of sensing units according to the divided scenarios. This enables accurate evaluation of the service status of roads, not only improving monitoring coverage but also reducing deployment costs. Moreover, it allows for road detection under the same monitoring index system, improving the standardization of road monitoring.

[0034] Example 2 This embodiment is a further improvement on embodiment 1, and the repeated content will not be described again.

[0035] Preferably, a road structure model is established and numerical simulation is performed to calculate the stress, strain, and vertical deformation of the road model under target conditions, including: A finite element structural model of the road was established by combining the topography, roadbed structure, and environmental hydrological conditions along the road. Conduct dynamic response numerical simulation to calculate the stress, strain, and vertical deformation of each structural layer of the road under the target traffic level and the target lifespan of the cumulative equivalent axle load.

[0036] Preferably, the present invention can classify road structure scene types in two ways.

[0037] Preferably, the first scenario division method is: When establishing a finite element structural model of a road, the road structural scene types are initially classified. Specifically, this can be done by combining factors such as the topography, roadbed structure type, and environmental hydrological conditions along the road to divide the road into different structural scenes. Structural scene types include: general road sections, long longitudinal slopes, curved road sections, embankments, cut sections, semi-fill / semi-cut sections, special roadbeds, bridge approach transition sections, and entrance / exit ramps.

[0038] After obtaining the scene segmentation threshold, the initially segmented scene types are validated.

[0039] Preferably, the initial scene classification is validated based on the scene classification threshold, including: Determine whether the differences in dynamic response between different scenes reach the threshold for scene segmentation; if not, adjust the scene boundaries. The same scene is divided into sub-scenes using the scene division threshold, and the number of monitoring sections is adjusted according to the number of sub-scenes.

[0040] Preferably, the second scenario division method is: The boundaries of each scene are clearly defined using scene segmentation thresholds. Combined with factors such as roadside topography, roadbed structure, and environmental hydrological conditions, roads are classified into different structural scene types. Then, scene-intra-scene segmentation thresholds are used to further divide the same scene into sub-scenes, and the number of monitoring sections is adjusted based on the number of sub-scenes. Structural scene types include: general road sections, long longitudinal slopes, curved road sections, embankments, cut sections, semi-fill / semi-cut sections, special roadbeds, bridge approach transition sections, and entrance / exit ramps.

[0041] Preferably, the simulation process fixes the structural layer parameters, environmental boundary conditions, gradient load levels, pavement structure combinations, and subgrade type conditions. The stress / strain cloud maps within each structural layer are sorted by numerical value, and the top 20% of high-value areas are selected as concentrated intervals. Based on historical disease statistics, sections where the disease density reaches a preset threshold within the same scenario are defined as high-risk, sensitive sections. According to the simulation results, the spatial locations where stress, strain, or deformation within each structural layer reaches local maximum values ​​are determined as extreme points of structural dynamic response. Based on the simulation results, areas where the cumulative increase in plastic deformation exceeds 1.5 times the average value are identified as points prone to long-term deformation deterioration.

[0042] Within a given scenario, due to the influence of the road structure, there may be situations where the mechanical response data of two cross-sections differ significantly, but this difference will not exceed the threshold for dividing the two scenarios. In order to better and more comprehensively monitor the scenario, after dividing the scenario in the first step, it is necessary to divide the same scenario into sub-scenarios and set at least one monitoring cross-section in each sub-scenarios to ensure that the monitoring effect is consistent with the actual road conditions.

[0043] Preferably, a dual-dimensional evaluation model of unit monitoring cost and monitoring coverage effectiveness is used to determine the economic efficiency of monitoring points: The effective monitoring coverage mileage of a single sensor and the effectiveness of data-driven disease early warning are used as core indicators. Redundant monitoring points and ineffective points with low response sensitivity are eliminated, balancing the total number of monitoring points with monitoring accuracy. Within the same scene and structural layer, if the difference in mechanical response between two monitoring points never exceeds the scene's classification threshold, either monitoring point is considered a redundant monitoring point. Under standard load, if the difference between the mechanical response amplitude of a monitoring point and the average response amplitude of its layer exceeds the scene's classification threshold, that monitoring point is considered an ineffective point with low response sensitivity.

[0044] Preferably, S2, select monitoring equipment and construct a monitoring indicator system, including: based on the unique stress characteristics, disease risk types, and structural deterioration patterns of each scenario after division, construct a differentiated and layered monitoring indicator system, and match the embedded sensing device model and installation parameters accordingly. Scenario characteristics include: road section structural type, load distribution characteristics, hydrogeological conditions, types of long-term disease susceptibility, and structural layer deformation sensitivity, determined comprehensively through scenario mechanical simulation response results, geological survey data along the route, and historical disease statistics. The monitoring indicator system includes general basic monitoring indicators and scenario-specific supplementary indicators. General basic monitoring indicators include: internal structural temperature and humidity, interlayer stress, structural strain, vertical displacement, traffic load, and vehicle flow. The additional indicators for specific scenarios include: for embankment subgrades, monitoring of subgrade earth pressure and subgrade moisture content is added; for cut slope sections, monitoring of slope lateral displacement and slope tilt deformation is added; for semi-fill and semi-cut sections, monitoring of uneven settlement difference is added; for bridge approach transition sections, special monitoring of bridge approach slab deformation and interlayer voids is added; and corresponding monitoring equipment is matched according to the monitoring indicators.

[0045] Preferably, S3, based on the divided scenes, constructs a layout scheme for the perception units in the horizontal and vertical sections of each scene, including: Cross-section layout focuses on longitudinal layered monitoring, surface layer is buried with temperature and strain sensors to monitor asphalt fatigue damage, base layer is laid with earth pressure cells to assess bearing status, and deep subgrade is installed with displacement gauges to monitor settlement risk. The lateral layout deploys multi-parameter composite sensors at the lane dividers to simultaneously collect data on wheel track load and structural response.

[0046] Example 3 This embodiment illustrates the specific implementation of the intelligent road embedded sensing unit optimization deployment method of the present invention with a specific application example: The optimized deployment method for embedded sensing units in smart roads includes the following steps: S1. Model and simulate the road, divide the scene, and select the layout points.

[0047] First, the road is modeled, and based on the topography, roadbed structure, and environmental hydrological conditions along the road, it is divided into nine scenarios: general road sections, long longitudinal slopes, curved road sections, fill sections, cut sections, semi-fill / semi-cut sections, special roadbeds, bridge approach transition sections, and entrance / exit ramps. These nine scenarios cover typical alignment changes (longitudinal slopes, curves), roadbed types (fill / cut / semi-fill / semi-cut / special roadbeds), structural abrupt changes (bridge approach transition sections), and traffic behavior characteristics (entrance / exit ramps) in highways. They are the main working conditions that lead to differences in the mechanical response of the pavement structure and are fully representative.

[0048] The ANSYS / ABAQUS finite element analysis software was used to conduct long-term dynamic response numerical simulation of the road model, calculating the stress, strain, and vertical deformation of each structural layer of the road under the target traffic level and the target cumulative equivalent axle load. Preferably, the target traffic level was selected as medium traffic level; the target cumulative equivalent axle load was selected as 15-year cumulative equivalent axle load.

[0049] Preferably, the medium traffic level can be defined according to the "JTG D50-2017 Highway Asphalt Pavement Design Specification" and is classified based on the cumulative equivalent axle load application times of the design lane: traffic levels are divided into four levels: light, medium, heavy, and extra-heavy. The medium traffic level corresponds to a cumulative equivalent axle load application time of 1×10⁻⁶. 7 ~2.5×10 7 This invention uses the midpoint of this interval (within the design life) for simulation. 7 This was used as the calculation benchmark. The medium traffic level was chosen because it covers the service load conditions of most highways in my country and is widely representative. Setting a 15-year design life conforms to the 15-year design reference period for asphalt pavement of highways stipulated in JTG D50-2017, which can simulate the fatigue damage accumulation process of the pavement structure during a typical service cycle, making the simulation results highly consistent with actual engineering conditions.

[0050] Preferably, simulation can be used to calculate the long-term mechanical theoretical response values ​​of each structural layer of the road pavement and subgrade under medium traffic level and 15-year cumulative equivalent axle cycles, including structural stress, strain, vertical settlement deformation, and interlayer voids.

[0051] Based on finite element simulation, the triaxial strain peak values ​​of each structural layer under various scenarios are extracted, and their standard deviations are calculated. Specifically, the standard deviations can be calculated separately for the three strain peak values ​​of two cross sections, and then the average value is taken.

[0052] The simulation core boundary strictly matches the actual service conditions of highways: Load parameters: strictly match the standard axle load of medium traffic level, simulating the cumulative axle load cycles over a 15-year design life, conforming to the actual laws of long-term fatigue deformation of roads; ("medium traffic level" is defined according to the "JTG D50-2017 Highway Asphalt Pavement Design Specification", and is divided according to the cumulative equivalent axle load cycles of the design lane: traffic level is divided into four levels: light, medium, heavy, and extra heavy. The cumulative equivalent axle load cycles corresponding to the medium traffic level are 1×10 7 ~2.5×10 7 (within the design life).

[0053] The simulation in this invention uses the midpoint of this interval, 1.75 × 10⁻⁶. 7 This was used as the calculation benchmark. The medium traffic level was chosen because it covers the service load conditions of most highways in my country and is widely representative. Setting a 15-year design life aligns with the 15-year design benchmark period for asphalt pavement in JTG D50-2017, simulating the fatigue damage accumulation process of the pavement structure during a typical service cycle, ensuring a high degree of consistency between the simulation results and actual engineering conditions. Structural parameters: The asphalt surface layer, water-stabilized base course, and fill subgrade layer thickness, material modulus, and interlayer contact bonding parameters are fully consistent with the on-site construction structure. Environmental parameters: Seasonal temperature and humidity cycles along the route and groundwater seepage boundary conditions are synchronously set to match the long-term service environment of outdoor roads. Parameters strictly adhere to JTG industry standards and the technical requirements of the long-term performance monitoring network pilot project, ensuring a high degree of consistency between the simulation results and actual on-site stress and deformation.

[0054] Preferably, this invention uses the response difference as the basis for setting the scene division threshold. It employs the standard deviations of three core indicators—peak stress, peak strain, and cumulative settlement deformation—to quantify the differences in dynamic response between scenes and sections. Furthermore, it statistically analyzes multi-condition cyclic simulation results to summarize the common patterns of road load transfer paths, inter-layer deformation evolution, and long-term deterioration development under different scenarios, clarifying the unique stress and service characteristics of each scenario. Finally, based on industry standards for the validity of highway structure monitoring data, mature engineering experience from similar road scientific observation networks, and the error tolerance range of 15 years of long-term service data, a 10% response difference threshold is determined within each scenario, ensuring that single-section monitoring data can fully represent the service status of the entire road section within the corresponding interval. This scene division threshold is derived from the laws of multi-condition dynamic simulation. Based on finite element simulation, the triaxial strain peak values ​​of each structural layer under each scenario are extracted, and their standard deviations are calculated.

[0055] The standard deviations of various parts of the road model are summarized and statistically analyzed to determine the standard deviation within the same scenario and the standard deviation between different scenarios.

[0056] Statistics show that the standard deviation within the same scenario (such as different cross sections of a general road) is generally between 0% and 8%, while the standard deviation between different scenarios (such as a general road section and a long longitudinal slope) generally exceeds 25%.

[0057] Therefore, setting the threshold for dividing between scenes to 30% can avoid misjudging essential differences as fluctuations within a scene; setting the threshold for dividing within a scene to 10% can effectively distinguish normal fluctuations.

[0058] Preferably, the scene division threshold is defined based on the principle that the difference in monitoring information within the same scene is ≤10%, and the difference in dynamic response across scenes is ≥30%. The determination of this threshold follows the following criteria: the standard deviations of three core indicators—peak stress, peak strain, and cumulative settlement deformation—are used to quantitatively analyze the results of multi-condition cyclic simulations. Statistics show that the standard deviation of response differences at different locations within the same scene is typically between 0% and 8%, and setting 10% as the upper limit can cover more than 95% of normal fluctuations; while the standard deviation of response differences across scenes generally exceeds 25%, and setting 30% as the lower limit can effectively distinguish the essential differences between scenes.

[0059] After dividing the scene using the inter-scene division threshold, the scene within the same scene is divided into sub-scenes using the intra-scene division threshold, and the parts with a standard deviation greater than 10% and less than 20% within the same scene are divided into different sub-scenes.

[0060] At least one monitoring section is set up in each sub-scene to ensure that the monitoring effect is consistent with the actual road conditions.

[0061] Preferably, after determining the threshold for different scenarios, the deployment points are selected to determine the monitoring range that a single cross-section can represent.

[0062] Preferably, the scenarios are divided into key scenarios and ordinary scenarios.

[0063] Preferably, the key scenarios are the stress-strain concentration ranges and high-risk sensitive areas for defects in each scenario.

[0064] The stress-strain concentration zone is defined in the finite element simulation results by sorting the stress / strain contour maps of each structural layer according to their numerical values ​​and selecting the top 20% of the highest values ​​as the concentration zone. This zone is typically located directly below the wheel track zone and at the edge of abrupt changes in interlayer stiffness (such as the junction between the base course and the subbase course).

[0065] High-incidence sensitive areas for diseases are defined as areas where the disease density in the same scene reaches a preset threshold, based on historical disease statistics.

[0066] Preferably, based on the stress and strain concentration ranges and high-incidence sensitive sections of each scenario output by simulation, sensing units are densely deployed at extreme points of structural dynamic response and points prone to long-term deformation and deterioration.

[0067] The extreme point of structural dynamic response refers to the spatial location where the stress, strain or deformation in each structural layer reaches its local maximum value in finite element simulation (such as directly below the wheel track, the contact edge between layers, etc.).

[0068] Long-term deformation and deterioration-prone areas refer to regions where the cumulative increase in plastic deformation exceeds 1.5 times the average value after 15 years of cumulative equivalent axle cycle simulation based on moderate traffic. These areas are usually located at the bottom of the base course, the top surface of the subgrade, and other layers prone to fatigue damage.

[0069] Preferably, all road sections other than those in key scenarios are ordinary homogeneous road sections in ordinary scenarios.

[0070] For ordinary homogeneous road sections, characteristic points with representative mechanical responses across the entire cross section are selected as unified representative monitoring points.

[0071] Preferably, a dual-dimensional evaluation model of unit monitoring cost and monitoring coverage effectiveness is used to determine the economic efficiency of monitoring sites: With the effective monitoring coverage mileage of a single sensor and the effectiveness of data-driven disease early warning as the core indicators, duplicate and redundant monitoring points and invalid points with low response sensitivity are eliminated to balance the total number of deployments and monitoring accuracy.

[0072] Redundant monitoring points are defined as follows: if the difference in mechanical response between two sections is always ≤10% within the same scene and structural layer, then either section is considered redundant and can be combined for deployment.

[0073] Low-response-sensitivity invalid sites are defined as sites where, under standard load, the mechanical response amplitude is more than 30% less than the average response of the layer (such as the hard shoulder edge far from the wheel track), or the response change at the site is not clearly correlated with the evolution of major defects.

[0074] By eliminating these two types of locations, the total number of locations can be controlled within the economically optimal range while ensuring the integrity of monitoring information. Ultimately, the optimal cost-effective locations for all road sections and scenarios can be selected, achieving full coverage of key risk areas and streamlined and efficient deployment on regular road sections.

[0075] Preferably, in step S2, the construction of the monitoring indicator system and the selection of monitoring equipment include: Based on the unique stress characteristics, disease risk types, and structural deterioration patterns of each key scenario after division, a differentiated and hierarchical monitoring indicator system is constructed to match the model and installation parameters of the embedded sensing equipment. The characteristics of key scenarios include: road section structural type, load distribution characteristics, hydrogeological conditions, types of long-term diseases, and structural layer deformation sensitivity. These are determined comprehensively through scenario mechanical simulation response results, geological survey data along the route, and historical disease statistics.

[0076] General basic monitoring indicators: internal temperature and humidity of the structure, interlayer stress, structural strain, vertical displacement, traffic load and traffic flow; scenario-specific supplementary indicators: for embankment subgrade, add subgrade soil pressure and subgrade moisture content monitoring; for cut slope sections, add slope lateral displacement and slope tilt deformation monitoring; for semi-fill and semi-cut sections, add uneven settlement differential monitoring; for bridge approach transition sections, add bridge approach slab deformation and interlayer void monitoring.

[0077] For the corresponding monitoring indicators, embedded fiber optic temperature and humidity sensors, vibrating wire strain sensors, static level displacement sensors, embedded earth pressure gauges, road surface embedded dynamic weighing systems, embedded moisture content sensors, pore water pressure gauges, and slope inclinometers are matched respectively.

[0078] The monitoring indicator system and sensor deployment types for each scenario are shown in Table 1.

[0079] Table 1: Monitoring indicator system and sensor deployment types for each scenario

[0080] The monitoring content and methods for the structural morphology of various types of sensing units are shown in Table 2.

[0081] Table 2: Monitoring Content and Methods of Structural Morphology of Various Types of Sensing Units

[0082] Preferably, in step S3, the deployment scheme of the perception units in the horizontal and vertical sections of each scene is constructed based on the divided scenes, including: for key scenes, segmentation and densification are adopted according to the characteristics of the scene, and multi-source perception units are deployed in a grid, while constructing a hybrid topology network combining backbone nodes and edge nodes; for the ordinary scenes, key representative points are selected for perception unit deployment.

[0083] Preferably, key scenarios are densely deployed and representative points are set up on ordinary road sections to achieve a planar coverage rate of ≥90%.

[0084] Planar coverage is defined as the percentage of road segment length whose mechanical response characteristics can be effectively represented by the data monitored by the sensing units across the entire route.

[0085] The method for calculating the plane coverage rate is as follows: taking the single section monitoring point as the center, extend to both sides along the driving direction until the position where the difference in response with the section first exceeds 10% is the effective representative range of the section.

[0086] After optimization, key scenarios achieve 100% independent coverage, while ordinary road sections are covered by representative points, ensuring that the overall coverage rate is no less than 90%.

[0087] Preferably, based on the characteristics of each key scenario after division and the actual engineering situation, a corresponding cross-sectional and longitudinal section layout scheme for sensing units is constructed.

[0088] The cross-sectional layout plan includes the location and number of each type of sensing unit, as well as the detection content.

[0089] The locations and quantities of sensors deployed in key scenarios in this invention are shown in Table 3.

[0090] Table 3: Sensor Deployment Locations and Quantities in Key Scenarios

[0091] Preferably, see Figure 2 and Figure 3 The specific deployment scheme for key cross-sections and key scenarios is as follows: The layout scheme of the dynamic weighing system is as follows: At the top of the upper layer of the structure, 4-5m away from the strain sensor test matrix, it is arranged in a single lane perpendicular to the vehicle driving direction for measuring vehicle axle load, axle type and vehicle speed.

[0092] The arrangement of pressure sensors is as follows: at the top of the roadbed and at the location of the wheel track, two pressure sensors are arranged longitudinally at a distance of 1m from the bottom of each layer for measuring the pressure at the bottom of each layer of the road surface. A total of four pressure sensors are arranged in each cross section.

[0093] The horizontal strain sensor arrangement is as follows: At the bottom of the lower layer, six longitudinal horizontal strain sensors and six transverse horizontal strain sensors are arranged symmetrically around the wheel track area. They are arranged in four rows: longitudinal, transverse, longitudinal, and transverse; three sensors per row with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m, forming a horizontal strain testing matrix for measuring the longitudinal and transverse strain of the road surface. At the bottom of the flexible base layer, three longitudinal horizontal strain sensors and three transverse horizontal strain sensors are arranged symmetrically around the wheel track area. They are arranged in two rows: longitudinal and transverse; three sensors per row with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m, used for measuring the longitudinal strain of the road surface and determining the transverse offset of traffic loads. At the bottom of the semi-rigid base layer, six longitudinal horizontal strain sensors and six transverse horizontal strain sensors are arranged symmetrically around the wheel track area. They are arranged in four rows: longitudinal, transverse, longitudinal, and transverse; three sensors per row with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m. A total of 30 transverse strain sensors and 30 longitudinal strain sensors are installed in each cross section.

[0094] The vertical strain sensor arrangement is as follows: At the bottom of each layer, six vertical strain sensors are arranged symmetrically in two rows of three sensors along the wheel track. The transverse center-to-center spacing of the sensors is 0.4m, and the longitudinal spacing is 0.5m. At the bottom of the flexible base layer, three vertical strain sensors are arranged symmetrically in one row of three sensors along the wheel track. The transverse center-to-center spacing of the sensors is 0.4m. At the top of the roadbed, six vertical strain sensors are arranged symmetrically in two rows of three sensors along the wheel track. The transverse center-to-center spacing of the sensors is 0.4m, and the longitudinal spacing is 0.5m. A total of 30 vertical strain sensors are deployed at each cross-section.

[0095] The temperature sensors are arranged as follows: two temperature sensors are placed longitudinally at 1m intervals on the bottom of the lower layer, the bottom of the flexible base layer, the bottom of the semi-rigid subbase layer, and the top of the roadbed, and are placed on the hard shoulder 0.3m away from the outer edge of the driving lane. A total of eight temperature sensors are installed in each cross section.

[0096] The humidity sensors are arranged as follows: two temperature sensors are placed longitudinally at 1m intervals on the bottom of the lower layer, the bottom of the flexible base layer, the bottom of the semi-rigid subbase layer, and the top of the roadbed, and are placed on the hard shoulder 0.3m away from the outer edge of the driving lane. A total of 8 humidity sensors are installed in each section.

[0097] In addition to the conventional deployment of sensing units, a differential settlement meter is added at the bottom of the roadbed 5m from the tail of the bridge abutment in the transition section.

[0098] The locations and quantities of sensors for normal deployment in typical scenarios are shown in Table 4.

[0099] Table 4: Sensor Deployment Locations and Quantities in Typical Scenarios

[0100] See Figure 4 and Figure 5 For normal layout schemes of cross-sections in common scenarios: The arrangement of pressure sensors is as follows: at the top of the roadbed and at the location of the wheel track, two pressure sensors are arranged longitudinally at a distance of 1m from the bottom of each layer for measuring the pressure at the bottom of each layer of the road surface. A total of four pressure sensors are arranged in each cross section.

[0101] The horizontal strain sensor arrangement is as follows: At the bottom of the lower layer, three longitudinal horizontal strain sensors and three transverse horizontal strain sensors are arranged symmetrically at the wheel track area. Two rows are formed, one longitudinal and one transverse; three sensors are arranged in each row, with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m, forming a horizontal strain testing matrix for measuring the longitudinal and transverse strain of the road surface. At the bottom of the flexible base layer, three longitudinal horizontal strain sensors and three transverse horizontal strain sensors are arranged symmetrically at the wheel track area. Two rows are formed, one longitudinal and one transverse; three sensors are arranged in each row, with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m, used for measuring the longitudinal strain of the road surface and determining the transverse offset of traffic loads. At the bottom of the semi-rigid subbase layer, three longitudinal horizontal strain sensors and three transverse horizontal strain sensors are arranged symmetrically at the wheel track area. Two rows are formed, one longitudinal and one transverse; three sensors are arranged in each row, with a transverse spacing of 0.4m and a longitudinal spacing of 0.5m. A total of 18 transverse strain sensors and 18 longitudinal strain sensors are installed in each cross section.

[0102] The vertical strain sensor arrangement is as follows: At the bottom of each layer, three vertical strain sensors are arranged symmetrically in one row along the wheel track area, with a center-to-center spacing of 0.4m and a longitudinal spacing of 0.5m. At the bottom of the flexible base layer, three vertical strain sensors are also arranged symmetrically in one row along the wheel track area, with a center-to-center spacing of 0.4m. At the top of the roadbed, three vertical strain sensors are also arranged symmetrically in one row along the wheel track area, with a center-to-center spacing of 0.4m and a longitudinal spacing of 0.5m. A total of 18 vertical strain sensors are deployed at each cross-section.

[0103] The temperature sensors are arranged as follows: one temperature sensor is placed longitudinally at 1m intervals on the bottom of the lower layer, the bottom of the flexible base layer, the bottom of the semi-rigid subbase layer, and the top of the roadbed, and is placed on the hard shoulder 0.3m away from the outer edge of the driving lane. A total of 4 temperature sensors are installed in each section.

[0104] The humidity sensors are arranged as follows: one temperature sensor is placed longitudinally at 1m intervals on the bottom of the lower layer, the bottom of the flexible base layer, the bottom of the semi-rigid subbase layer, and the top of the roadbed, and placed on the hard shoulder 0.3m away from the outer edge of the driving lane. A total of 4 humidity sensors are installed in each section.

[0105] The present invention provides a low-cost, high-coverage method for optimizing the deployment of embedded sensing units in smart roads. Following the deployment principles of "practicality, economy, and low disturbance," the method achieves this by quantifying stress differences through refined long-term mechanical simulation, dividing key scenarios into categories using standardized threshold grading, optimizing deployment points based on mechanical response, constructing a scenario-specific monitoring indicator system, and customizing a multi-dimensional, layered deployment scheme across horizontal and vertical sections. This multi-dimensional collaboration results in a low-cost, high-coverage overall deployment scheme for embedded sensing units in smart roads.

[0106] This invention overcomes the drawbacks of traditional uniform, experience-based deployment by systematically optimizing the deployment of embedded sensing units across the entire road and structural layers. This significantly improves the coverage of road service status monitoring while substantially reducing the total number of sensors deployed and engineering construction and maintenance costs, maintaining excellent economic efficiency in the long term. It accurately captures the stress deformation and potential defects of the road structure throughout its entire lifecycle, effectively ensuring road operation safety and comprehensively improving the level of intelligent operation and maintenance services for the entire lifecycle of road infrastructure.

[0107] Example 4 This embodiment provides a smart road embedded sensing unit optimization deployment system, which includes at least a processing unit. The processing unit stores a computer program, which, when executed by a processor, implements a smart road embedded sensing unit optimization deployment method according to Embodiments 1 and 2.

[0108] 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, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the deployment of embedded sensing units in smart roads, characterized in that, Includes the following steps: S1. Model and simulate the road, divide the scene, and select layout points; including: Establish a road structure model and perform numerical simulation to calculate the stress, strain, and vertical deformation of the road model under target conditions; The dynamic response and strain field changes of each structural layer under different loads and pavement structures are compared; the standard deviation of three core indicators of each structural layer—peak stress, peak strain, and cumulative settlement deformation—is quantitatively analyzed to determine the scene division threshold; the road is divided into different scenes according to the scene division threshold; wherein, the scene division threshold includes at least the scene division threshold. Within the scene, the cross-sectional locations where the mechanical response is at the median level and the difference from the average response of the entire cross-section is less than the error threshold are selected as representative points. The scenarios are divided into key scenarios and ordinary scenarios; S2. Select monitoring equipment and construct a monitoring indicator system; S3. Based on the divided scenes, construct the layout scheme of the horizontal and vertical cross-sectional sensing units for each scene. Among them, the key scenarios are segmented and encrypted according to the characteristics of the scenarios, multi-source sensing units are deployed in a grid, and a hybrid topology network combining backbone nodes and edge nodes is constructed. For the general scene, key representative points are selected for the deployment of sensing units.

2. The method for optimizing the deployment of embedded sensing units in smart roads according to claim 1, characterized in that, The process of establishing a road structure model and performing numerical simulations to calculate the stress, strain, and vertical deformation of the road model under target conditions includes: A finite element structural model of the road was established by combining the topography, roadbed structure, and environmental hydrological conditions along the road. Conduct dynamic response numerical simulation to calculate the stress, strain, and vertical deformation of each structural layer of the road under the target traffic level and the target lifespan of the cumulative equivalent axle load.

3. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 2, characterized in that, The establishment of the finite element structural model of the road also includes: A preliminary classification of road structure scene types is conducted. The structural scenario types include: general road sections, long longitudinal slopes, curved road sections, embankments, cuts, semi-fill and semi-cut sections, special roadbeds, bridge approach transition sections, and entrance and exit ramps.

4. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 3, characterized in that, The scene segmentation threshold also includes the scene segmentation threshold; The preliminary scene classification is validated based on the scene classification threshold, including: Determine whether the difference in dynamic response between different scenes reaches the threshold for scene division; if not, adjust the scene boundaries. The same scene is divided into sub-scenes using the scene division threshold, and the number of monitoring sections is adjusted according to the number of sub-scenes.

5. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 2, characterized in that, The scene segmentation threshold also includes the scene segmentation threshold; The boundaries of each scene are clearly defined using the scene segmentation threshold, and the road is divided into different structural scene types; The same scene is divided into sub-scenes using the scene segmentation threshold; the number of monitoring sections is adjusted according to the number of sub-scenes. The structural scenario types include: general road sections, long longitudinal slopes, curved road sections, embankments, cuts, semi-fill and semi-cut sections, special roadbeds, bridge approach transition sections, and entrance and exit ramps.

6. A method for optimizing the deployment of embedded sensing units in smart roads according to claim 4 or 5, characterized in that, The simulation process uses fixed structural layer parameters, environmental boundary conditions, gradient-varying load levels, pavement structure combinations, and subgrade type working conditions. The stress / strain contour maps within each structural layer are sorted by numerical value, and the top 20% of high-value areas are taken as the concentration intervals. Based on historical disease statistics, the section where the disease density in the same scene reaches a preset threshold is defined as a high-incidence sensitive section of disease. Based on the simulation results, the spatial locations where the stress, strain, or deformation in each structural layer reaches the local maximum value are determined as the extreme points of the structural dynamic response. Based on the simulation results, areas where the cumulative increase in plastic deformation exceeds 1.5 times the average value are identified as points prone to long-term deformation and deterioration.

7. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 6, characterized in that, The economic efficiency of monitoring points is determined by a dual-dimensional evaluation model of unit monitoring cost and monitoring coverage effectiveness: the effective monitoring coverage mileage of a single sensor and the effectiveness of data disease early warning are the core indicators. Duplicate and redundant monitoring points and ineffective points with low response sensitivity are eliminated to balance the total number of points deployed and the monitoring accuracy. Within the same scene and the same structural layer, if the difference in mechanical response between two monitoring points never exceeds the threshold for dividing the scene, then either monitoring point is considered a redundant monitoring point. Under standard load, if the difference between the mechanical response amplitude of the monitoring point and the average response amplitude of the layer exceeds the threshold for dividing the scene, the monitoring point is considered an invalid point with low response sensitivity.

8. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 7, characterized in that, S2, Selecting monitoring equipment and constructing a monitoring indicator system, including: Based on the unique stress characteristics, disease risk types, and structural deterioration patterns of each scenario after division, a differentiated and hierarchical monitoring indicator system is constructed to match the model and installation parameters of the embedded sensing equipment. The characteristics of the scenario include: road section structure type, load distribution characteristics, hydrogeological conditions, types of long-term defects, and structural layer deformation sensitivity. Furthermore, the characteristics of the scenario are determined comprehensively based on the scenario mechanical simulation response results, geological survey data along the route, and historical disease statistics. The monitoring indicator system includes general basic monitoring indicators and scenario-specific supplementary indicators; The general basic monitoring indicators include: internal temperature and humidity of the structure, inter-story stress, structural strain, vertical displacement, traffic load and traffic flow. The specific supplementary indicators for the aforementioned scenarios include: Additional monitoring equipment for subgrade earth pressure and subgrade moisture content was installed in embankment subgrades. For excavated slope sections, additional monitoring equipment will be installed to detect lateral displacement and slope tilt deformation. For sections of road that are partially filled and partially cut, additional monitoring of uneven settlement differentials will be installed. Special monitoring equipment for bridge approach slab deformation and inter-layer voiding has been added to the bridge approach transition section; Match the corresponding monitoring equipment according to the monitoring indicators.

9. The method for optimizing the deployment of embedded sensing units in a smart road according to claim 8, characterized in that, S3, based on the divided scenes, constructs a layout scheme for the horizontal and vertical cross-section sensing units of each scene, including: Cross-section layout focuses on longitudinal layered monitoring, surface layer is buried with temperature and strain sensors to monitor asphalt fatigue damage, base layer is laid with earth pressure cells to assess bearing status, and deep subgrade is installed with displacement gauges to monitor settlement risk. The lateral layout deploys multi-parameter composite sensors at the lane dividers to simultaneously collect data on wheel track load and structural response.

10. A smart road embedded sensing unit optimized deployment system, characterized in that, It includes at least a processing unit; The processing unit stores a computer program, which, when executed by the processor, implements the method for optimizing the deployment of embedded sensing units in smart roads as described in any one of claims 1 to 9.

Images