Prestressed cable displacement intelligent monitoring and early warning system for highway slope

Abstract

This invention discloses an intelligent monitoring and early warning system for prestressed wire-type displacement of highway slopes, relating to the field of highway slope displacement monitoring and early warning technology. It includes a monitoring module, an early warning module, and a database. The monitoring module comprises a vehicle sensing component and a soil sensing component. The vehicle sensing component is connected to the database via signal transmission. The vehicle sensing component includes a weighing unit, a vibration and stress coupling monitoring unit, and a model building unit. The weighing unit collects vehicle information and transmits it to the database. The vibration and stress coupling monitoring unit collects the vibration frequency, energy distribution, and dynamic changes in soil stress caused by vehicle traffic and transmits this information to the database. The model building unit constructs a vehicle load-slope stress transfer model. This invention, by designing a system for collecting vehicle driving information and constructing a vehicle load-slope stress transfer model, achieves the function of quantifying the impact of vehicle loads on slopes.

Description

Technical Field

[0001] This invention relates to the field of highway slope displacement monitoring and early warning technology, specifically a highway slope prestressed wire-type intelligent monitoring and early warning system for displacement. Background Technology

[0002] Traditional geological exploration methods rely on on-site reconnaissance and drilling sampling by professionals, which have problems such as long monitoring cycles, priority of coverage, and difficulty in nighttime operations. They are difficult to meet the needs of high-speed, all-weather monitoring. Early monitoring technologies often focus on single parameters and lack the ability to perceive multiple factors of slope stability in a coordinated manner, resulting in delayed early warning.

[0003] Vibration loads generated by vehicle traffic can cause instantaneous displacement fluctuations on the slope surface, which may lead to misjudgments. At the same time, the fatigue effect of repeated rolling of heavy vehicles on the slope cannot be quantified separately through displacement data, which masks the true deformation trend and leads to delayed early warning. Natural factors such as rainfall and geological activity show similar characteristics to vehicle vibration in displacement data. Without information on the time of vehicle traffic, it is impossible to effectively separate the interference source and reduce the accuracy of early warning.

[0004] Patent CN118675073B discloses a method for predicting highway slope displacement. This patent improves the accuracy of prediction by comprehensively utilizing multi-source data and using advanced deep learning models.

[0005] The aforementioned patent constructs a comprehensive data acquisition system by integrating data from multiple sensors, satellite remote sensing, and UAV images. It uses convolutional networks combined with long short-term memory neural networks to extract spatiotemporal features from images, and graph neural networks combined with long short-term memory neural networks to extract spatiotemporal features from sensor data. The features are fused through an attention mechanism to form comprehensive spatiotemporal feature data. Integrated empirical mode decomposition is applied to decompose and reconstruct the fused features, remove noise, and retain core features. The main spatiotemporal features are then input into a Transformer model for prediction. By utilizing the powerful spatiotemporal dependency capture capability of the Transformer model, the prediction of highway slope displacement can be achieved. There is room for optimization in the collection of vehicle driving information.

[0006] Therefore, this application proposes an intelligent monitoring and early warning system for prestressed wire displacement of highway slopes, which collects vehicle driving information to construct a vehicle load-slope stress transfer model. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent monitoring and early warning system for prestressed wire displacement of highway slopes, so as to solve the technical problem mentioned in the background art that the monitoring and early warning system does not refer to vehicle driving information.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a prestressed wire displacement intelligent monitoring and early warning system for highway slopes, comprising a monitoring module, an early warning module, and a database. The monitoring module includes a vehicle sensing component and a soil sensing component, and the vehicle sensing component is connected to the database via signal transmission.

[0009] The vehicle sensing component includes: a weighing unit, a vibration and stress coupling monitoring unit, and a model building unit. The weighing unit collects vehicle axle load, vehicle speed, vehicle type information, real-time vehicle trajectory, and vehicle aggregation status, and transmits the information to the database.

[0010] The vibration and stress coupling monitoring unit is used to collect the vibration frequency, energy distribution and dynamic changes in soil stress caused by vehicle passage, and transmit the information to the database.

[0011] The model building unit establishes a three-dimensional transfer function by integrating vehicle load, vibration field, and stress field. Based on Abaqus, it builds a slope solid model and uses the strength reduction method to calculate stress patching under different working conditions. The mesh division follows the principle of densification at the slope toe to construct a vehicle load-slope stress transfer model.

[0012] Preferably, the vehicle perception component further includes a visual sensor and an infrared sensor. The visual sensor is used to collect vehicle model, license plate and body color information and transmit it to the database. The infrared sensor is used to collect vehicle traffic frequency and speed and transmit the information to the database.

[0013] Preferably, the model building unit constructs a displacement-humidity-water level coupling equation, calculates the displacement field distribution using elastic layering theory to establish a vehicle load-displacement transfer matrix, obtains a three-dimensional real-scene model of the slope terrain from the database, establishes a parameterized model of the slope soil layer using CAE software, maps the information collected by the soil sensing component to the three-dimensional real-scene model, calculates the stress propagation path of the vehicle load in three-dimensional space using the finite element method, dynamically updates the soil layer parameters based on a Bayesian network, and introduces an LSTM network to predict the hysteresis effect of water level changes on displacement to construct a vehicle-slope-environment model.

[0014] Preferably, the early warning module is connected to the database via signal transmission. The early warning module is equipped with an early warning model, which is based on the vehicle load-slope stress transfer model and the vehicle-slope-environment model. The early warning threshold is dynamically corrected by the displacement trend predicted by the LSTM network, and a Bayesian network is introduced to evaluate the probability of multiple indicators being jointly triggered. Based on the prediction results, an early warning is issued to the management personnel.

[0015] Preferably, the monitoring and early warning system further includes a recovery module, which is connected to the monitoring module and the early warning module via a signal line;

[0016] The recovery module includes: a piezoelectric energy harvesting unit, a wind power generation unit, a heat energy recovery unit, an electromagnetic power generation unit, and an energy storage unit. The energy storage unit distributes electrical energy to the monitoring module and the early warning module through a database.

[0017] The piezoelectric energy harvesting unit is deployed on the surface of the road and in the roadbed. It triggers the piezoelectric effect by the road deformation caused by the decompression of the vehicle, converting mechanical energy into electrical energy and storing it in the energy storage unit.

[0018] The wind power generation unit is deployed on the slope guardrail and isolation belt. It captures the turbulent wind energy generated by the vehicle's high-speed travel and adjusts the angle according to the wind direction to drive the generator to generate electricity.

[0019] The heat recovery unit recovers heat energy by utilizing the heat gradient in the contact area between the tire and the ground.

[0020] The electromagnetic power generation unit generates induced current by embedding magnets and coils into the flexible road surface and slope structure layer, and periodically releasing the magnetic field under vehicle load.

[0021] Preferably, the weighing unit includes a piezoelectric sensor and a vehicle positioning terminal. The piezoelectric sensor is used to collect vehicle axle load, vehicle speed and vehicle type information in real time and transmit the information to the database. The vehicle positioning terminal is used to track the real-time trajectory of the vehicle, monitor the aggregation status of the vehicle and transmit the information to the database.

[0022] Preferably, the vibration and stress coupling monitoring unit includes a fiber optic grating vibration sensor and a digital earth pressure gauge. The fiber optic grating vibration sensor is deployed along the toe of the slope to detect the vibration frequency caused by vehicle passage and transmit the information to the database. The digital earth pressure gauge is deployed on the surface and deep layers of the slope to record the dynamic changes in soil stress when vehicles pass.

[0023] Preferably, the soil sensing component is connected to the database via signal transmission.

[0024] The soil sensing components include a GNSS displacement monitoring station, a wire displacement gauge, an inclinometer, a soil moisture sensor, a groundwater level gauge, and a rain gauge. The GNSS displacement monitoring station measures the surface displacement of the slope using satellite positioning technology. The wire displacement gauge is used to measure the local displacement of the slope. The inclinometer is used to monitor the deep horizontal displacement inside the slope. The soil moisture sensor is used to measure the moisture content of the surface soil. The groundwater level gauge is used to monitor groundwater level fluctuations. The rain gauge is used to collect rainfall data in real time.

[0025] Preferably, the early warning is divided into three levels: a level one warning is issued when a single indicator exceeds the limit and the LSTM prediction trend is stable; a level two warning is issued when two indicators exceed the limit together for 30 minutes; and a level three warning is issued when the sliding surface is shown to be connected in the three-dimensional model.

[0026] Preferably, the piezoelectric energy acquisition unit provides energy to the monitoring module, the wind power generation unit provides energy to the visual sensor and the infrared sensor, the heat recovery unit is used to maintain the operating temperature of the monitoring module and the early warning module, and the electromagnetic power generation unit provides energy to the early warning module.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. This invention constructs a vehicle load-slope stress transfer model by collecting vehicle driving information, thereby realizing the function of quantifying the impact of vehicle load on slope. It solves the problems of being unable to capture the instantaneous impact of vehicle dynamic load on slope and the difficulty in quantifying the coupling relationship between vibration frequency and soil stress change. It can finely map dynamic vehicle load, broaden the application scenarios of the system, and improve the accuracy of early warning.

[0029] 2. This invention constructs a vehicle-slope-environment model by collecting vehicle, slope, and environmental information, realizing the function of multi-source sensing data fusion. It solves the problems of isolated data analysis, static data collection, and single early warning threshold. It can monitor vehicle, slope, and environmental data comprehensively, improve the accuracy of early warning, extend the service life of the three-dimensional model, shorten the early warning response time, and reduce the system operation and maintenance costs.

[0030] 3. This invention, by incorporating an early warning module, achieves the function of dynamically adjusting the early warning threshold, solving the problems of single information collection, large judgment error, and low prediction accuracy. It can match the early warning threshold with environmental load and structural response, shortening the early warning time and improving the early warning sensitivity and accuracy.

[0031] 4. This invention, by incorporating an energy recovery module, realizes the function of recovering dissipated energy, solving the problems of limitations, dependence on external power supply, and high cost of a single energy recovery mode. It can avoid monitoring interruption caused by power outages, improve energy recovery efficiency and coverage, enhance system autonomy and stability, reduce costs, and improve the system's environmental performance. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the intelligent monitoring and early warning system of the present invention;

[0033] Figure 2 This is a schematic diagram of the vehicle sensing component of the present invention;

[0034] Figure 3 This is a schematic diagram of the soil sensing component of the present invention;

[0035] Figure 4 This is a schematic diagram of the early warning module of the present invention;

[0036] Figure 5 This is a schematic diagram of the recycling module of the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Example 1: Please refer to Figure 1 and Figure 2 A highway slope prestressed wire displacement intelligent monitoring and early warning system includes a monitoring module, an early warning module, and a database. The monitoring module includes a vehicle sensing component and a soil sensing component. The vehicle sensing component is connected to the database through signal transmission.

[0039] The vehicle sensing component includes: a weighing unit, a vibration and stress coupling monitoring unit, and a model building unit. The weighing unit collects vehicle axle load, vehicle speed, vehicle type information, real-time vehicle trajectory, and vehicle aggregation status, and transmits the information to the database.

[0040] The vibration and stress coupling monitoring unit is used to collect the vibration frequency, energy distribution and dynamic changes in soil stress caused by vehicle passage, and transmit the information to the database.

[0041] The model building unit establishes a three-dimensional transfer function by integrating vehicle load, vibration field and stress field, builds a slope solid model based on Abaqus, calculates stress patching under different working conditions using the strength reduction method, and the mesh division follows the principle of densification at the slope toe to construct a vehicle load-slope stress transfer model.

[0042] The vehicle perception component also includes a visual sensor and an infrared sensor. The visual sensor is used to collect vehicle model, license plate and body color information and transmit it to the database. The infrared sensor is used to collect vehicle passage frequency and speed and transmit the information to the database.

[0043] The weighing unit includes a piezoelectric sensor and a vehicle positioning terminal. The piezoelectric sensor is used to collect vehicle axle load, vehicle speed and vehicle type information in real time and transmit the information to the database. The vehicle positioning terminal is used to track the real-time trajectory of the vehicle, monitor the aggregation status of the vehicle and transmit the information to the database.

[0044] The vibration and stress coupling monitoring unit includes a fiber optic grating vibration sensor and a digital earth pressure gauge. The fiber optic grating vibration sensor is deployed along the toe of the slope to detect the vibration frequency caused by vehicle passage and transmit the information to the database. The digital earth pressure gauge is deployed on the surface and deep layers of the slope to record the dynamic changes in soil stress when vehicles pass.

[0045] Furthermore, by utilizing dynamic weighing technology with piezoelectric sensors, the total weight of passing vehicles is acquired in real time, accurately measuring the weight of each axle and the total number of axles. This identifies whether the axle load distribution is compliant, infers vehicle type based on wheelbase and number of axles, collects real-time vehicle speed data, and combines this with vehicle information from visual sensors to improve the vehicle information in the database. At night or in low visibility conditions, infrared sensors collect vehicle traffic frequency and density. Based on the collected vehicle information, combined with vibration frequency collected by fiber optic vibration sensors and dynamic changes in soil stress collected by digital earth pressure gauges, a three-dimensional transfer function is established. This function is then used to build an edge-to-edge transfer function based on Abaqus. The slope solid model uses the strength reduction method to calculate stress joints under different working conditions. The mesh generation follows the principle of densification at the slope toe. A vehicle load-slope stress transfer model is constructed. After adding vehicle information as a parameter to the intelligent monitoring and prediction system, the vehicle speed limit and flow restriction values ​​can be dynamically adjusted based on the critical traffic density threshold, which helps traffic control and management. It realizes the function of quantifying the impact of vehicle load on slope, and solves the problems of not being able to capture the instantaneous impact of vehicle dynamic load on slope and the difficulty in quantifying the coupling relationship between vibration frequency and soil stress change. It can finely map dynamic vehicle load, broaden the application scenarios of the system, and improve the accuracy of early warning.

[0046] Example 2: Please refer to Figure 1 and Figure 3 The highway slope prestressed wire displacement intelligent monitoring and early warning system, the model building unit constructs a displacement-humidity-water level coupling equation, uses elastic layer theory to calculate the displacement field distribution to establish a vehicle load-displacement transfer matrix, obtains a three-dimensional real-scene model of the slope topography in the database, establishes a parameterized model of the slope soil layer using CAE software, maps the information collected by the soil sensing component to the three-dimensional real-scene model, calculates the stress propagation path of the vehicle load in three-dimensional space using the finite element method, dynamically updates the soil layer parameters based on a Bayesian network, and introduces an LSTM network to predict the hysteresis effect of water level changes on displacement to construct a vehicle-slope-environment model;

[0047] The soil sensing component is connected to the database via signal transmission.

[0048] The soil sensing components include a GNSS displacement monitoring station, a wire displacement gauge, an inclinometer, a soil moisture sensor, a groundwater level gauge, and a rain gauge. The GNSS displacement monitoring station measures the surface displacement of the slope using satellite positioning technology. The wire displacement gauge is used to measure the local displacement of the slope. The inclinometer is used to monitor the deep horizontal displacement inside the slope. The soil moisture sensor is used to measure the moisture content of the surface soil. The groundwater level gauge is used to monitor groundwater level fluctuations. The rain gauge is used to collect rainfall data in real time.

[0049] Furthermore, GNSS displacement monitoring stations are deployed along the slope surface with a sampling frequency greater than or equal to 10Hz. Millimeter-level surface displacement monitoring is achieved using BeiDou / GPS dual-mode differential positioning technology. Wire-type displacement gauges are deployed on the potential sliding surface, and inclinometers are installed via vertical drilling. Soil moisture sensors are deployed in a double layer at depths of 0.5m and 1.5m. Based on the vehicle load-slope stress transfer model, slope displacement data and environmental parameters are collected and integrated into the model. The coordinates of the GNSS displacement monitoring stations are used as a reference to transform the local coordinates of the wire-type displacement gauges and inclinometers. A 3D real-world model from the database is used as a spatial reference for the equipment deployment locations. By constructing a displacement-humidity-water level coupling equation and combining inclinometer data, the deep sliding surface level is inverted. The system employs elastic layered theory to calculate displacement field distribution, establishes a vehicle load-displacement transfer matrix, acquires a 3D real-world model of the slope topography from the database, and builds a parameterized model of the slope soil layer using CAE software. Information collected by the soil sensing component is mapped into the 3D real-world model. The stress propagation path of the vehicle load in 3D space is calculated using the finite element method. Soil layer parameters are dynamically updated based on a Bayesian network. An LSTM network is introduced to predict the hysteresis effect of water level changes on displacement, constructing a vehicle-slope-environment model. This achieves the function of multi-source sensing data fusion, solving the problems of isolated data analysis, static data acquisition, and single early warning thresholds. It can comprehensively monitor vehicle, slope, and environmental data, improving early warning accuracy, extending the lifespan of the 3D model, shortening early warning response time, and reducing system operation and maintenance costs.

[0050] Example 3: Please refer to Figure 1 and Figure 4 A prestressed wire displacement intelligent monitoring and early warning system for highway slopes. The early warning module is connected to the database via signal transmission. The early warning module is equipped with an early warning model. The early warning model is based on the vehicle load-slope stress transfer model and the vehicle-slope-environment model. The early warning threshold is dynamically corrected by the displacement trend predicted by the LSTM network. A Bayesian network is introduced to evaluate the probability of multiple indicators being jointly triggered. The system issues early warnings to management personnel based on the prediction results.

[0051] The warning is divided into three levels: a Level 1 warning is issued when a single indicator exceeds the limit and the LSTM prediction trend is stable; a Level 2 warning is issued when two indicators exceed the limit together for 30 minutes; and a Level 3 warning is issued when the three-dimensional model shows that the sliding surface has been connected.

[0052] Furthermore, based on information collected by the monitoring module, combined with the vehicle load-slope stress transfer model and the vehicle-slope-environment model, the displacement trend is predicted using an LSTM network. A Bayesian network is introduced to evaluate the joint trigger probability of multiple indicators. The first-level warning trigger condition is one of the following: the proportion of vehicle load-induced displacement exceeds 15% for the first time; the LSTM model predicts a displacement trend increase of less than 5% in the next 24 hours; the horizontal or vertical surface displacement rate is greater than 5 mm / day; the cumulative rainfall is greater than 50 mm / 24 hours; and the soil moisture content increases by more than 20%. The second-level warning trigger condition is one of the following: the proportion of vehicle load exceeds 25% for more than 35 minutes; the landslide probability assessed by the Bayesian network is 0.5–0.7; the deep horizontal displacement mutation rate is greater than 0.1 mm / h; the surface displacement rate is greater than 8 mm / day; the groundwater level rises by more than 1 m / 24 hours; and the pore water pressure is greater than 0.3 MPa. The third-level warning trigger condition is one of the following: the proportion of vehicle load-induced displacement exceeds 30%; and the slope stress-strain ratio is greater than 0.3 MPa. The system incorporates one of the following criteria: a sliding surface displacement rate greater than 150% of the design value, a Bayesian network joint probability greater than 0.7, a sliding surface displacement rate greater than 0.5 mm / h with the 3D model showing the sliding surface connected, or rainfall greater than 100 mm / 24h with soil moisture content greater than 95%. It integrates slope, soil, environmental, and vehicle information to predict and warn of potential safety risks. During warning, an LSTM neural network is used to dynamically model the displacement time-series data monitored by GNSS. A sliding window is used to predict the displacement change trend over the next 6 hours. Combined with dynamic adjustment of the threshold range based on displacement acceleration, the Bayesian network calculates the joint trigger probability using multiple indicators. When the joint trigger probability is greater than 0.7, the single indicator threshold is automatically reduced by 15%–20%, achieving dynamic adjustment of the warning threshold. This solves the problems of single information collection, large judgment errors, and low prediction accuracy. It can match the warning threshold with environmental loads and structural responses, shortening the warning time and improving warning sensitivity and accuracy.

[0053] Example 4: Please refer to Figure 1 and Figure 5 A prestressed wire-type intelligent monitoring and early warning system for highway slope displacement, wherein the monitoring and early warning system also includes a recovery module, which is connected to the monitoring module and the early warning module via a signal line;

[0054] The recovery module includes: a piezoelectric energy harvesting unit, a wind power generation unit, a heat energy recovery unit, an electromagnetic power generation unit, and an energy storage unit. The energy storage unit distributes electrical energy to the monitoring module and the early warning module through a database.

[0055] The piezoelectric energy harvesting unit is deployed on the surface of the road and in the roadbed. It triggers the piezoelectric effect by the road deformation caused by the decompression of the vehicle, converting mechanical energy into electrical energy and storing it in the energy storage unit.

[0056] The wind power generation unit is deployed on the slope guardrail and isolation belt. It captures the turbulent wind energy generated by the vehicle's high-speed travel and adjusts the angle according to the wind direction to drive the generator to generate electricity.

[0057] The heat recovery unit recovers heat energy by utilizing the heat gradient in the contact area between the tire and the ground.

[0058] The electromagnetic power generation unit generates induced current by embedding magnets and coils into the flexible road surface and slope structure layer, and periodically releasing the magnetic field under vehicle load.

[0059] The piezoelectric energy acquisition unit provides energy to the monitoring module, the wind power generation unit provides energy to the visual sensor and the infrared sensor, the heat recovery unit is used to maintain the operating temperature of the monitoring module and the early warning module, and the electromagnetic power generation unit provides energy to the early warning module.

[0060] Furthermore, piezoelectric energy harvesting units deployed in the road base and subgrade trigger the piezoelectric effect through dynamic pressure changes when vehicles pass, converting mechanical energy into electrical energy. A single vehicle roll can generate 0.5W to 1.2W of instantaneous power, which can meet the real-time power supply requirements of micro sensors. The energy harvested during vehicle passage is used to power the monitoring module. Wind power generation units deployed on slope guardrails and medians can adjust their angle according to wind direction to improve the efficiency of wind power generation. A single wind power generation unit can stably output 20W to 50W of power at a traffic density of 40km / h, providing power for monitoring equipment such as visual sensors and infrared sensors. When vehicles pass on highways, the load on the ground varies due to differences in vehicle model, speed, weight, and driver operation. In areas with frequent vehicle starts and stops, such as long downhill braking sections, curves, slope changes, and service areas, heat recovery units are installed to recover the heat generated by the friction between the wheels and the ground. This heat is then used to generate electricity using the temperature difference, forming a closed loop of heat and electricity. Magnets and coils are embedded in the flexible pavement and slope structure layer. The magnetic field is periodically compressed and released by the vehicle load, generating an induced current. A single vehicle passing through can generate a 3V to 5V pulse voltage, which, after rectification, can provide energy to the warning module. This achieves the function of recovering energy dissipated during vehicle movement, solving the limitations, dependence on external power supply, and high costs of single energy recovery modes. It can avoid monitoring interruptions caused by power outages, improve energy recovery efficiency and coverage, enhance system autonomy and stability, reduce costs, and improve the system's environmental performance.

[0061] Example 5: Please refer to Figure 1 , Figure 2 and Figure 5 A highway slope prestressed wire displacement intelligent monitoring and early warning system includes a monitoring module, an early warning module, and a database. The monitoring module includes a vehicle sensing component and a soil sensing component. The vehicle sensing component is connected to the database through signal transmission.

[0062] The vehicle sensing component includes: a weighing unit, a vibration and stress coupling monitoring unit, and a model building unit. The weighing unit collects vehicle axle load, vehicle speed, vehicle type information, real-time vehicle trajectory, and vehicle aggregation status, and transmits the information to the database.

[0063] The vibration and stress coupling monitoring unit is used to collect the vibration frequency, energy distribution and dynamic changes in soil stress caused by vehicle passage, and transmit the information to the database.

[0064] The model building unit establishes a three-dimensional transfer function by integrating vehicle load, vibration field and stress field, builds a slope solid model based on Abaqus, calculates stress patching under different working conditions using the strength reduction method, and the mesh division follows the principle of densification at the slope toe to construct a vehicle load-slope stress transfer model.

[0065] The monitoring and early warning system also includes a recovery module, which is connected to the monitoring module and the early warning module via signal lines;

[0066] The recovery module includes: a piezoelectric energy harvesting unit, a wind power generation unit, a heat energy recovery unit, an electromagnetic power generation unit, and an energy storage unit. The energy storage unit distributes electrical energy to the monitoring module and the early warning module through a database.

[0067] The piezoelectric energy harvesting unit is deployed on the surface of the road and in the roadbed. It triggers the piezoelectric effect by the road deformation caused by the decompression of the vehicle, converting mechanical energy into electrical energy and storing it in the energy storage unit.

[0068] The wind power generation unit is deployed on the slope guardrail and isolation belt. It captures the turbulent wind energy generated by the vehicle's high-speed travel and adjusts the angle according to the wind direction to drive the generator to generate electricity.

[0069] The heat recovery unit recovers heat energy by utilizing the heat gradient in the contact area between the tire and the ground.

[0070] The electromagnetic power generation unit generates induced current by embedding magnets and coils into the flexible road surface and slope structure layer, and periodically releasing the magnetic field under vehicle load.

[0071] Furthermore, during vehicle passage, the energy recovery module recovers energy to support the monitoring and early warning modules. When the vehicle sensing components detect that the axle load or vehicle type exceeds the relevant threshold, or when vehicle aggregation causes a surge in local load, the piezoelectric energy harvesting unit and electromagnetic power generation unit in that area are shut down in advance to prevent excessive road surface deformation due to overload, which could lead to piezoelectric material breakage or coil damage. When there is abnormal deep soil pressure on the slope, the buried electromagnetic power generation unit is shut down. When the vehicle load-slope stress transfer module calculation shows that the slope stress is close to the material yield strength, the slope deformation may cause the guardrail structure to become unstable, so the wind power generation unit needs to be shut down. When the fiber optic grating sensor detects that the vibration frequency is close to the slope's natural frequency and there is a risk of resonance, the piezoelectric energy harvesting unit, electromagnetic power generation unit, and heat recovery unit need to be shut down to avoid resonance exacerbating mechanical fatigue and damaging the heat transfer material. This realizes the function of dynamic protection recovery module, solves the problem of relying on physical protection and static threshold protection, reduces damage to recovery module, extends the service life of recovery module, and improves energy recovery efficiency.

[0072] Working principle: Through dynamic weighing technology of piezoelectric sensors, the total weight of passing vehicles is obtained in real time, the weight of each axle and the total number of axles are accurately measured, the compliance of axle load distribution is identified, the vehicle type is inferred from the wheelbase and number of axles, the real-time speed of passing vehicles is collected, and the vehicle information in the database is improved by combining the vehicle information collected by the vision sensor. At night or in low visibility, the frequency and density of vehicle passage are collected by the infrared sensor. Based on the collected vehicle information, combined with the vibration frequency collected by the fiber optic grating vibration sensor and the dynamic change of soil stress collected by the digital earth pressure gauge, a three-dimensional transfer function is established. A slope solid model is built based on Abaqus, and the stress compensation under different working conditions is calculated by the strength reduction method. The mesh division follows the principle of densification at the slope toe. A vehicle load-slope stress transfer model is constructed. After adding vehicle information as parameters to the intelligent monitoring and prediction system, the values ​​of vehicle speed limit and flow restriction can be dynamically adjusted by the critical traffic flow density threshold, which helps traffic control and management.

[0073] GNSS displacement monitoring stations are deployed along the slope surface with a sampling frequency greater than or equal to 10Hz. Millimeter-level surface displacement monitoring is achieved using BeiDou / GPS dual-mode differential positioning technology. Wire-type displacement gauges are deployed on the potential sliding surface, and inclinometers are installed in vertical boreholes. Soil moisture sensors are deployed in a double layer at depths of 0.5m and 1.5m. Based on the vehicle load-slope stress transfer model, slope displacement data and environmental parameters are collected and integrated into the model. The coordinates of the GNSS displacement monitoring stations are used as a reference to transform the local coordinates of the wire-type displacement gauges and inclinometers, combining this with 3D real-world data from the database. The model serves as a spatial reference for the equipment deployment location. By constructing a coupled equation of displacement, humidity, and water level, and combining inclinometer data to invert the location of the deep sliding surface, the displacement field distribution is calculated using elastic layered theory. A vehicle load-displacement transfer matrix is ​​established, a three-dimensional real-scene model of the slope terrain is obtained from the database, and a parameterized model of the slope soil layer is established using CAE software. The information collected by the soil sensing component is mapped to the three-dimensional real-scene model. The stress propagation path of the vehicle load in three-dimensional space is calculated using the finite element method. Soil layer parameters are dynamically updated based on a Bayesian network. An LSTM network is introduced to predict the hysteresis effect of water level changes on displacement to construct a vehicle-slope-environment model.

[0074] Information collected by the monitoring module, combined with the vehicle load-slope stress transfer model and the vehicle-slope-environment model, is used to predict displacement trends via an LSTM network. A Bayesian network is introduced to evaluate the probability of multiple indicators jointly triggering the warning. The first-level warning trigger conditions are: the proportion of vehicle load-induced displacement exceeding 15% for the first time; the LSTM model predicts a displacement trend increase of less than 5% over the next 24 hours; a horizontal or vertical surface displacement rate greater than 5 mm / day; a cumulative rainfall greater than 50 mm / 24 hours; and a soil moisture content increase greater than 20%. The second-level warning trigger conditions are: the proportion of vehicle load exceeding 25% for more than 35 minutes; a landslide probability assessed by the Bayesian network of 0.5–0.7; a deep horizontal displacement mutation rate greater than 0.1 mm / h; a surface displacement rate greater than 8 mm / day; and a groundwater level rise greater than or equal to 1 m / 24 hours with pore water pressure greater than 0.3. One of the MPa is used for the Level 3 early warning triggering conditions, which are: vehicle load-induced displacement accounting for more than 30%, slope stress and strain exceeding 150% of the design value, Bayesian network joint probability greater than 0.7, sliding surface displacement rate greater than 0.5 mm / h and 3D model showing sliding surface penetration, and rainfall greater than 100 mm / 24h and soil moisture content greater than 95%. It integrates slope information, soil information, environmental information, and vehicle information to predict and warn of potential safety risks. When issuing an early warning, it uses an LSTM neural network to dynamically model the displacement time series data monitored by GNSS, predicts the displacement change trend in the next 6 hours through a sliding window, and dynamically adjusts the threshold range based on displacement acceleration. The Bayesian network calculates the joint trigger probability through multiple indicators. When the joint trigger probability is greater than 0.7, the threshold of a single indicator is automatically reduced by 15% to 20%.

[0075] Piezoelectric energy harvesting units deployed in the road base and subgrade trigger the piezoelectric effect through dynamic pressure changes when vehicles pass by, converting mechanical energy into electrical energy. A single vehicle roll can generate 0.5W to 1.2W of instantaneous power, which can meet the real-time power supply needs of micro sensors. The energy harvested during vehicle passage is used to power the monitoring module. Wind power generation units deployed on slope guardrails and medians can adjust their angle according to wind direction to improve the efficiency of wind power generation. A single wind power generation unit can stably output 20W to 50W of power at a traffic density of 40km / h, providing power for monitoring equipment such as visual sensors and infrared sensors. When vehicles pass by on highways, the load on the ground varies due to differences in vehicle model, speed, weight, and driver operation. In areas with frequent vehicle starts and stops, such as long downhill braking sections, curves, slope changes, and service areas, heat recovery units are installed to recover the heat generated by the friction between the wheels and the ground, using the temperature difference to generate electricity, forming a heat-electricity closed loop. Magnetic energy is embedded in the flexible pavement and slope structure layer. Iron and coils are used to periodically compress and release magnetic fields under vehicle load, generating induced current. A single vehicle passage can generate a 3V to 5V pulse voltage, which, after rectification, can provide energy for the early warning module. During vehicle passage, the energy is recovered by the recovery module to provide energy support for the monitoring and early warning modules. When the vehicle sensing component detects that the axle load or vehicle type exceeds the relevant threshold, or when vehicle aggregation causes a surge in local load, the piezoelectric energy acquisition unit and electromagnetic power generation unit in that area are shut down in advance to avoid excessive deformation of the road surface due to overload, which could lead to breakage of the piezoelectric material or damage to the coil. When there is abnormal deep soil pressure on the slope, the buried electromagnetic power generation unit is shut down. When the vehicle load-slope stress transfer module calculation shows that the slope stress is close to the material yield strength, the slope deformation may cause instability of the guardrail structure, so the wind power generation unit needs to be shut down. When the fiber optic grating sensor detects that the vibration frequency is close to the natural frequency of the slope and there is a risk of resonance, the piezoelectric energy acquisition unit, electromagnetic power generation unit, and heat recovery unit need to be shut down to avoid resonance exacerbating mechanical fatigue and damaging the heat conduction material.

[0076] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A highway slope prestressed cable displacement intelligent monitoring and early warning system, comprising a monitoring module, an early warning module and a database, characterized in that: The monitoring module comprises a vehicle perception component and a soil perception component, the vehicle perception component transmits signals to a database; The vehicle perception component comprises a weighing unit, a vibration and stress coupling monitoring unit and a model building unit, the weighing unit collects vehicle axle load, vehicle speed, vehicle type information, vehicle real-time trajectory and vehicle aggregation state, and transmits the information to the database; The vibration and stress coupling monitoring unit is used for collecting vibration frequency, energy distribution and soil stress dynamic change caused by vehicle passing, and transmitting the information to the database; The model building unit establishes a three-dimensional transfer function by comprehensively considering vehicle load, vibration field and stress field, establishes a slope entity model based on Abaqus, calculates stress distribution under different working conditions by using the strength reduction method, and constructs a vehicle load-slope stress transfer model by following the slope foot encryption principle in grid division; The vehicle perception component further comprises a visual sensor and an infrared sensor, the visual sensor is used for collecting vehicle model, license plate and vehicle body color information and transmitting the information to the database, and the infrared sensor is used for collecting vehicle passing frequency and speed and transmitting the information to the database; On the basis of the vehicle load-slope stress transfer model, by collecting slope displacement data and environmental parameters, the two are integrated into the vehicle load-slope stress transfer model, the coordinates of the GNSS displacement monitoring station are used as the reference, the local coordinates of the tensioned wire displacement meter and the inclinometer equipment are converted, the three-dimensional real scene model in the database is combined as the spatial reference of the equipment layout position, the displacement-humidity-water level coupling equation is constructed, the deep sliding surface position is inversed combined with the inclinometer data, the displacement field distribution is calculated by using the elastic layered theory, the vehicle load-displacement transfer matrix is established, the three-dimensional real scene model of the slope terrain in the database is obtained, the soil layer parameterization model of the slope is established by using the CAE software, the information collected by the soil perception component is mapped into the three-dimensional real scene model, the stress propagation path of the vehicle load in the three-dimensional space is calculated by using the finite element method, the soil layer parameters are dynamically updated based on the Bayesian network, the vehicle-slope-environment model is constructed by introducing the LSTM network to predict the hysteresis effect of water level change on displacement; The early warning module transmits signals to the database, and an early warning model is arranged in the early warning module, the early warning model is based on the vehicle load-slope stress transfer model and the vehicle-slope-environment model, the displacement trend predicted by the LSTM network is used for dynamically correcting the early warning threshold, the Bayesian network is introduced to evaluate the joint triggering probability of multiple indexes, and the management personnel are warned according to the prediction result; The soil perception component transmits signals to the database; The soil perception component comprises a GNSS displacement monitoring station, a tensioned wire displacement meter, an inclinometer, a soil humidity sensor, a groundwater level meter and a rain gauge, the GNSS displacement monitoring station measures the slope surface displacement by using satellite positioning technology, the tensioned wire displacement meter is used for measuring the local displacement of the slope, the inclinometer is used for monitoring the deep horizontal displacement of the slope, the soil humidity sensor is used for measuring the water content of the surface soil, the groundwater level meter is used for monitoring the fluctuation of the underground water level, and the rain gauge is used for collecting rain data in real time.

2. The highway slope prestressed cable displacement intelligent monitoring and early warning system according to claim 1, characterized in that: The monitoring and early warning system further comprises a recycling module connected with the monitoring module and the early warning module through signal lines; The recycling module comprises a piezoelectric energy harvesting unit, a wind power generation unit, a thermal energy recycling unit, an electromagnetic power generation unit and an energy storage unit, and the energy storage unit distributes electric energy to the monitoring module and the early warning module through a database; The piezoelectric energy harvesting unit is arranged in the road surface layer and the roadbed, and triggers the piezoelectric effect through the road surface deformation caused by the pressure reduction of the vehicle, converts mechanical energy into electric energy, and stores the electric energy in the energy storage unit; The wind power generation unit is arranged on the slope guardrail and the separation belt, captures the turbulent wind energy generated by the vehicle driving at a high speed, adjusts the angle according to the wind direction, drives the generator to generate electricity; The thermal energy recycling unit recycles thermal energy through the heat gradient of the tire contact area with the ground; The electromagnetic power generation unit embeds magnets and coils in the flexible road surface and the slope structure layer, releases the magnetic field under the periodic compression of the vehicle load, and generates induced current.

3. The highway slope prestressed cable displacement intelligent monitoring and early warning system according to claim 1, characterized in that: The weighing unit comprises a piezoelectric sensor and a vehicle-mounted positioning terminal, the piezoelectric sensor is used to collect vehicle axle load, vehicle speed and vehicle type information in real time, and transmit the information to the database, and the vehicle-mounted positioning terminal is used to track the real-time track of the vehicle and monitor the aggregation state of the vehicle, and transmit the information to the database.

4. The highway slope prestressed cable displacement intelligent monitoring and early warning system according to claim 1, characterized in that: The vibration and stress coupling monitoring unit comprises a fiber grating vibration sensor and a digital soil pressure gauge, the fiber grating vibration sensor is arranged along the slope toe, and is used to detect the vibration frequency caused by the vehicle passing, and transmit the information to the database, and the digital soil pressure gauge is arranged in the slope surface layer and the deep layer, and is used to record the dynamic change of soil stress when the vehicle passes.

5. The highway slope prestressed cable displacement intelligent monitoring and early warning system according to claim 1, characterized in that: The early warning is divided into three levels, one-level early warning is issued when single index exceeds the limit and LSTM prediction trend is stable, two-level early warning is issued when two indexes exceed the limit and last for 30 minutes, and three-level early warning is issued when the three-dimensional model shows that the sliding surface penetrates.

6. The highway slope prestressed cable displacement intelligent monitoring and early warning system according to claim 2, characterized in that: The piezoelectric energy harvesting unit provides energy for the monitoring module, the wind power generation unit provides energy for the visual sensor and the infrared sensor, the thermal energy recycling unit is used to maintain the working temperature of the monitoring module and the early warning module, and the electromagnetic power generation unit provides energy for the early warning module.

Images