Bridge health monitoring system

Abstract

The invention discloses a bridge health monitoring system which comprises a sensing layer, a transmission layer, a processing layer and an application layer which are in communication connection in sequence. The sensing layer synchronously collects structural mechanics, environmental influence and traffic load data through multi-source heterogeneous sensing nodes, and the sensing nodes adopt a solar energy and vibration energy double-source energy supply and low-power-consumption mechanism; the transmission layer adopts a'wireless private network + edge gateway 'mixed mode, and has data caching and breakpoint resuming functions; the processing layer analyzes data through a deep learning fusion algorithm based on an attention mechanism, and realizes dynamic early warning threshold adaptive adjustment in combination with a Bayesian reasoning model; and the application layer provides a visual interface, graded early warning and targeted maintenance decision suggestions. The system achieves the precise monitoring of the full life cycle of the bridge, improves the evaluation accuracy and system stability, reduces the operation and maintenance cost, provides scientific support for the safe operation and maintenance of the bridge, is suitable for the health monitoring scenes of various bridge types, and greatly improves the accuracy of the evaluation of the health state of the bridge.

Description

Technical Field

[0001] This invention relates to the field of bridge health monitoring technology, specifically a bridge health monitoring system. Background Technology

[0002] As a core component of transportation infrastructure, the structural safety of bridges directly impacts traffic safety and transportation efficiency. With increasing service life, traffic loads, and environmental erosion, bridges are prone to defects such as cracks, deformation, and fatigue damage. Failure to monitor and address these issues promptly can lead to safety accidents. Therefore, bridge health monitoring systems have become a crucial technological means to ensure the safety of bridges throughout their entire lifecycle.

[0003] Existing bridge health monitoring systems have many technical shortcomings: First, the data collection dimension is singular. Most systems only focus on the collection of structural mechanical parameters (such as strain and vibration), ignoring the collaborative collection of environmental impact data (temperature, humidity, wind speed) and traffic load data. This results in a lack of comprehensiveness in health status assessment and makes the system susceptible to misjudgment due to interference data. Secondly, there are limitations in the power supply methods for sensor nodes. Traditional wired power supply is costly and inflexible. Single solar power supply is greatly affected by weather, while battery power supply requires frequent replacement, which is difficult to meet long-term monitoring needs. Third, the early warning mechanism is rigid and uses fixed early warning thresholds, without taking into account the differences in different bridge types, spans, service life, and real-time traffic and environmental conditions, resulting in frequent false alarms and missed alarms; Fourth, data transmission stability is insufficient. Single wireless transmission is easily affected by obstruction and interference, resulting in significant data loss or delay issues. Fifth, maintenance decisions lack specificity. They can only provide simple early warnings and cannot combine the characteristics of the bridge structure with historical data to generate accurate maintenance plans, making it difficult to support scientific operation and maintenance. Summary of the Invention

[0004] The purpose of this invention is to provide a bridge health monitoring system that aims to solve the above problems.

[0005] To address the aforementioned technical problems, this invention provides a bridge health monitoring system, comprising: (i) system component deployment and connection. Sensing layer deployment: Multi-source heterogeneous sensor nodes are installed using detachable fixing components at key stress-bearing parts of the bridge (such as mid-span and supports), connection nodes (such as hinge joints and expansion joints), and environmentally sensitive areas (such as the windward side of the bridge and areas with drastic temperature changes). The spacing between sensor nodes is adjusted according to the bridge type and span; the spacing is 5-8m for small and medium-span bridges and 3-5m for large-span bridges to ensure comprehensive monitoring coverage.

[0006] Transmission layer setup: Edge gateways are deployed at both ends or in the middle of the bridge to establish bidirectional communication with the sensing nodes of the perception layer through wireless communication interfaces (such as LoRa and 5G); the wireless private network unit adopts industrial-grade wireless communication base stations to cover the entire bridge and data transmission path, and the edge gateways establish a stable connection with the processing layer server through the wireless private network.

[0007] Processing layer configuration: The processing layer server is deployed in the bridge operation and maintenance management center. It has a built-in bridge structural parameter database (storing standard health parameters and basic early warning thresholds for different bridge types such as beam bridges, arch bridges, and cable-stayed bridges, different spans of ≤50m, 50-100m, and >100m, and different service lives of 0-10 years, 10-20 years, and >20 years). It integrates a multi-source data fusion processing module, a dynamic early warning threshold adjustment module, and a health status prediction module. It establishes data interaction with the transmission layer edge gateway and application layer terminal through data interfaces.

[0008] Application layer deployment: The application layer includes the operation and maintenance center visualization terminal (computer, large screen), the management personnel mobile terminal (mobile APP), and the on-site sound and light early warning equipment (installed at the entrances at both ends of the bridge), which realizes data synchronization with the processing layer server through the network interface.

[0009] (II) Implementation of System Workflow Data acquisition phase: The sensor nodes are woken up at a preset frequency (5 minutes / time under normal operating conditions, 1 minute / time during busy traffic periods) through a low-power sleep wake-up mechanism. They simultaneously acquire structural mechanics data such as strain, vibration, and displacement; environmental data such as temperature, humidity, and wind speed; and traffic data such as vehicle load and traffic volume. After acquisition, they automatically enter sleep mode. In the power supply module, solar panels convert light energy into electrical energy, and piezoelectric conversion modules convert the mechanical energy generated by bridge vibration into electrical energy. Both are stored in the energy storage unit to continuously power the sensor nodes.

[0010] Data transmission phase: The sensor node sends the collected data to the edge gateway through the wireless communication interface. The edge gateway performs preliminary filtering of the data (removing obviously abnormal data) and format standardization conversion (unifying to JSON format). If the wireless private network communication is normal, the data is directly uploaded to the processing layer server. If the communication is interrupted, the data is temporarily stored in the gateway cache module and automatically resumed through the breakpoint resume module after the communication is restored.

[0011] Data processing stage: After receiving the data, the processing layer server uses a multi-source data fusion processing module to extract features and perform correlation modeling on structural mechanics data, environmental data, and traffic data through a deep learning fusion algorithm based on an attention mechanism. It removes interference data such as environmental temperature fluctuations and instantaneous wind loads, and outputs accurate bridge health characteristic parameters. The dynamic early warning threshold adjustment module obtains the basic threshold and real-time operating condition data from the database through the parameter input submodule. After correction by the Bayesian inference calculation submodule, it outputs the dynamic early warning threshold. The health status assessment module compares the health characteristic parameters with the dynamic early warning threshold to generate the bridge health status level (safe, concern, risk, emergency) and assessment report.

[0012] Application Output Phase: The application layer's visual interaction module displays health status levels, real-time data, historical trend curves, and risk location markers, supporting access from multiple terminals such as computers and mobile phones; the tiered early warning module triggers corresponding early warnings based on the health status level. Level 1 emergency risks simultaneously trigger audible and visual warnings, SMS notifications to management personnel, and platform push notifications; Level 2 risks trigger SMS notifications and platform push notifications; Level 3 concerns only trigger platform push notifications, and the warning information includes the risk location, exceeding data, and preliminary handling suggestions; the maintenance decision module combines the bridge structure parameter database, historical monitoring data, and operation and maintenance records to generate targeted solutions for maintenance timing, maintenance locations, and maintenance methods.

[0013] In summary, due to the adoption of the above-mentioned technologies, the beneficial effects of this invention are: 1. By synchronously collecting structural mechanics data, environmental impact data, and traffic load data through multi-source heterogeneous sensing nodes, and combining them with a deep learning fusion algorithm based on attention mechanism, the correlation analysis and interference removal of multi-dimensional data can be realized, which greatly improves the accuracy of bridge health status assessment and avoids one-sided judgment caused by single data collection.

[0014] 2. The sensing node adopts a dual-source power supply structure of "solar energy + vibration energy" and is equipped with a low-power sleep wake-up mechanism. This not only solves the limitations of single power supply, but also minimizes energy consumption. It eliminates the need for frequent battery replacements or reliance on wired power supply, adapts to the long-term monitoring needs in complex field environments, and significantly reduces operation and maintenance costs.

[0015] 3. A dynamic early warning threshold adaptive adjustment mechanism based on a Bayesian inference model, which combines a bridge structural parameter database with real-time operating data, dynamically corrects the early warning threshold to avoid false alarms and missed alarms caused by fixed thresholds. At the same time, a hierarchical early warning mechanism is used to quickly transmit risk information and shorten emergency response time.

[0016] 4. The hybrid transmission mode of wireless private network and edge gateway, combined with the caching and breakpoint resume function of edge gateway, effectively resists communication interference, ensures the real-time performance and integrity of data transmission, and solves the data loss problem of single wireless transmission.

[0017] 5. The application layer combines the bridge structural parameter database, historical monitoring data, and operation and maintenance records to generate maintenance plans adapted to the current bridge status, upgrading from "simple early warning" to "early warning + precise decision-making," providing scientific support for the operation and maintenance of the bridge throughout its entire life cycle and extending the bridge's service life. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention, making other features, objects, and advantages of the invention more apparent. The illustrative embodiments of the invention illustrated in the drawings and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a system diagram of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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 a part of the embodiments of the present invention, not all of them. 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. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. 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.

[0020] In the description of this invention, it should be understood that the terms indicating orientation or positional relationship are based on the orientation or positional relationship shown in the drawings and are only for the convenience of describing the invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention.

[0021] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific context of the specification.

[0022] This invention provides a bridge health monitoring system.

[0023] Application scenarios It is suitable for concrete simply supported beam bridges with a span of 30m and a service life of 8 years on urban main roads, with an average daily traffic volume of about 8,000 vehicles, and large fluctuations in temperature and humidity in the surrounding environment, with no extreme wind loads.

[0024] Specific implementation Sensing Layer: Twelve multi-source heterogeneous sensing nodes are deployed, installed at the mid-span of the beam, supports, hinge joints, and environmentally sensitive areas on both sides of the bridge, with a spacing of 6m. The sensing nodes integrate strain sensors (measurement range 0-2000με), vibration sensors (frequency range 0-500Hz), displacement sensors (accuracy ±0.01mm), temperature and humidity sensors (temperature -40℃~85℃, humidity 0-100%RH) and vehicle load monitoring modules (measurement range 0-100t). The power supply module uses a 10W solar panel and a PZT-5H piezoelectric conversion module, paired with a 5000mAh lithium polymer energy storage unit. The sleep-wake mechanism is set to timed wake-up (5 minutes / time) + strain mutation trigger wake-up (mutation value > 50με).

[0025] Transmission layer: One industrial-grade LoRa edge gateway is deployed at each end of the bridge, with a communication distance of ≥3km, supporting a data cache capacity of ≥16GB, and having the function of automatically filtering instantaneous invalid data (such as temperature and humidity changes of ±20%); the wireless private network adopts the city's public 5G network, and the edge gateway communicates with the processing layer server through the 5G module.

[0026] Processing layer: The bridge structural parameter database stores basic parameters such as standard strain threshold (≤1500με) and vibration frequency threshold (5-30Hz) for concrete simply supported beam bridges (span 20-50m, service life 0-10 years); the attention weight allocation of the multi-source data fusion algorithm is 0.6 for structural mechanics data, 0.3 for traffic load data, and 0.1 for environmental data; the Bayesian inference parameters of the dynamic early warning threshold adjustment module are set to traffic volume weight 0.4, temperature and humidity weight 0.2, and basic threshold weight 0.4.

[0027] Application layer: The visual interface displays the strain, displacement data and health status of each sensor node in real time. The tiered early warning audio-visual equipment is installed at the entrances at both ends of the bridge. The maintenance decision module prioritizes outputting daily maintenance suggestions such as hinge joint sealing and bearing maintenance based on the characteristics of concrete beam bridges.

[0028] Application scenarios This method is applicable to cable-stayed bridges spanning 300m and with a service life of 12 years. The daily traffic volume is approximately 15,000 vehicles, and the bridge is significantly affected by river winds (maximum wind speed 15m / s). The key monitoring points are the stay cables, the connection between the tower and the beam, and the mid-span of the main beam.

[0029] Specific implementation Sensing Layer: 30 multi-source heterogeneous sensing nodes are deployed, of which 15 are installed at the anchor ends of the stay cables and in the middle of the cable body (spaced 10m apart), 10 are installed at the tower-beam connection and the mid-span and quarter-span positions of the main beam, and 5 are deployed on the top of the bridge tower and the windward side of the bridge deck; the sensing nodes have added wind speed sensors (measurement range 0-60m / s), the vibration sensor frequency range has been expanded to 0-1000Hz, and the vehicle load monitoring module supports simultaneous monitoring of multiple lanes; the power supply module adopts a 30W solar panel and a high-power piezoelectric conversion module (adapted to the large vibration amplitude of the cable-stayed bridge), the energy storage unit has a capacity of 10000mAh, and the sleep wake-up mechanism is set to wake up every 2 minutes + triggering by sudden changes in vibration frequency.

[0030] Transmission layer: "LoRa+5G" dual-mode hybrid transmission is adopted. Two high-performance edge gateways are deployed on the top of the bridge tower, which can support the simultaneous access of 30 sensor nodes. The cache capacity is ≥32GB. The interruption resumption trigger condition is that the communication interruption exceeds 30 seconds. The wireless private network adopts a dedicated 5G base station to ensure the stability of cross-river transmission. The edge gateway communicates with the bridge surface sensor nodes through LoRa and connects to the processing layer server through 5G.

[0031] Processing layer: The database adds parameters such as the cable stress threshold (≤1200MPa) and the main beam deflection threshold (≤L / 1000, where L is the span) for cable-stayed bridges (span 200-400m, service life 10-20 years); the attention weight of the multi-source data fusion algorithm is adjusted to 0.5 for structural mechanics data, 0.3 for environmental data (mainly wind speed), and 0.2 for traffic load data; the dynamic early warning threshold adjustment module adds a wind speed correction coefficient, which automatically increases the safety margin of vibration and displacement early warning thresholds when the wind speed is >10m / s.

[0032] Application Layer: The visualization interface adds stress variation curves of stay cables and dynamic simulation of main beam deflection, and links the graded early warning system with the traffic control system. When a level one early warning is issued, traffic closure suggestions are automatically pushed. The maintenance decision module focuses on outputting targeted solutions such as stay cable tension adjustment and anti-corrosion treatment at the tower-beam connection.

[0033] Application scenarios This method is applicable to highway concrete T-beam bridges with a span of 40m and a service life of 25 years. The daily traffic volume is approximately 6,000 vehicles. The bridge has minor cracks and bearing aging issues, requiring long-term monitoring of structural damage development to support decisions on reconstruction and reinforcement.

[0034] Specific implementation Sensing layer: 18 multi-source heterogeneous sensing nodes are deployed in a high density, focusing on covering the cracked areas of the beam, the bearings, and key parts of the old bridge renovation (4m spacing); the accuracy of the strain sensor of the sensing node is improved to ±1με, the displacement sensor range is extended to 0-50mm, and a crack width sensor (measurement range 0-5mm) is added; the power supply module enhances the stability of solar power supply (equipped with a solar tracking device), the piezoelectric conversion module is adapted to the small vibration amplitude of old bridges and adopts a high-sensitivity model, and the energy storage unit supports low-temperature charging (-20℃~60℃).

[0035] Transport layer: Deploy a multi-functional edge gateway in the middle of the bridge, which supports customizable data filtering accuracy (the filtering threshold is set to ±15% to address the large data fluctuations of older bridges), cache capacity ≥24GB, and supports incremental data retransmission for breakpoint resume; the wireless private network adopts 4G+LoRa dual-mode to ensure communication coverage in remote areas.

[0036] Processing layer: The database stores the attenuated foundation thresholds (e.g., strain threshold ≤ 1200με) for old concrete T-beam bridges (span 30-50m, service life > 20 years), and adds a foundation threshold for crack width (≤ 0.3mm); the multi-source data fusion algorithm strengthens the correlation analysis between crack data and strain data, with attention weights allocated as follows: 0.7 for structural mechanics data, 0.1 for environmental data, and 0.2 for traffic load data; the dynamic early warning threshold adjustment module reduces the safety margin of the foundation threshold and improves the early warning sensitivity, and the Bayesian inference model adds a service life correction coefficient.

[0037] Application Layer: The visualization interface highlights the trend of crack width changes and bearing settlement data, and the graded early warning SMS notifications are extended to bridge renovation construction units; the maintenance decision module combines the damage characteristics of old bridges to output renovation plan suggestions such as crack grouting, bearing replacement, and beam reinforcement, and marks the priority and implementation window period.

Claims

1. A bridge health monitoring system, characterized in that: The system comprises a perception layer, a transmission layer, a processing layer, and an application layer, all interconnected sequentially. These layers collaborate to monitor the bridge's health status throughout its entire lifecycle. The perception layer deploys several multi-source heterogeneous sensor nodes, positioned at key stress points, connection points, and sensitive areas of the surrounding environment to simultaneously collect bridge structural mechanics data, environmental impact data, and traffic load data. The transmission layer employs a hybrid transmission mode, with the edge gateway establishing communication connections with both the perception layer's sensor nodes and the processing layer to ensure stable, real-time data transmission. The processing layer incorporates a bridge structural parameter database and a multi-source data fusion algorithm. This algorithm analyzes and processes the transmitted data, and, combined with a dynamic early warning threshold adaptive adjustment mechanism, generates bridge health status assessment results. The application layer communicates with the processing layer, providing a visual monitoring interface, tiered early warning notifications, and targeted maintenance decision suggestions, and supports multi-terminal access.

2. The bridge health monitoring system according to claim 1, characterized in that, The multi-source heterogeneous sensing node integrates a strain sensor, a vibration sensor, a displacement sensor, a temperature and humidity sensor, and a vehicle load monitoring module. The sensing node adopts a dual-source power supply structure. The piezoelectric conversion module is used to convert the mechanical energy generated by bridge vibration into electrical energy. The energy storage unit is electrically connected to the solar panel and the piezoelectric conversion module respectively. The sensing node is equipped with a low-power sleep-wake mechanism, in which only the power supply of the data acquisition trigger module is retained in the sleep state.

3. The bridge health monitoring system according to claim 1, characterized in that, The multi-source data fusion algorithm is a deep learning fusion algorithm based on the attention mechanism. By extracting features and modeling the correlation between structural mechanics data, environmental impact data and traffic load data, it eliminates invalid interference data caused by environmental temperature fluctuations and instantaneous wind loads, and outputs accurate bridge structural health characteristic parameters.

4. The bridge health monitoring system according to claim 1, characterized in that, The dynamic early warning threshold adaptive adjustment mechanism includes a parameter input module, a Bayesian inference calculation module, and a threshold output module. The parameter input module is used to obtain the basic threshold from the bridge structure parameter database and the real-time collected traffic volume, load level, temperature, humidity, and wind speed data. The Bayesian inference calculation module performs probabilistic inference based on the above input parameters and dynamically corrects the basic threshold. The threshold output module sends the corrected threshold to the comparison unit of the processing layer.

5. The bridge health monitoring system according to claim 1, characterized in that, The edge gateway of the transmission layer has functions such as preliminary data filtering, format standardization conversion, breakpoint resume transmission and data caching. When the wireless private network communication is interrupted, the edge gateway stores the untransmitted data and automatically retransmits it after the communication is restored, ensuring the integrity of data transmission.

6. The bridge health monitoring system according to claim 1, characterized in that, The application layer's hierarchical early warning notification is divided into three levels. Level 1 early warning corresponds to emergency risks and simultaneously triggers audible and visual warnings, SMS notifications, and platform push notifications; Level 2 early warning corresponds to potential risks and triggers SMS notifications and platform push notifications. Level 3 warnings correspond to risks that require attention. They only trigger platform notifications, and each level of warning includes the location of the risk, the data exceeding the standard, and preliminary handling suggestions.

7. The bridge health monitoring system according to claim 1, characterized in that, The bridge structural parameter database stores standard health parameters and basic early warning thresholds corresponding to different bridge types, spans, and service lives.

8. The bridge health monitoring system according to claim 1, characterized in that, The maintenance decision recommendations in the application layer are generated by the historical data mining module in the processing layer. This module performs correlation analysis on long-term monitoring data, bridge structural parameters and historical operation and maintenance records, and outputs recommendations on maintenance timing, maintenance locations and maintenance methods that are suitable for the current bridge status.

9. The bridge health monitoring system according to claim 1, characterized in that, The processing layer also includes a health status prediction module, which uses time series analysis algorithms to predict the trend of bridge health status changes over the next 1-3 years based on long-term collected bridge health characteristic parameters and identify potential safety hazards.

10. The bridge health monitoring system according to claim 2, characterized in that, The low-power sleep-wake mechanism is triggered by a timed wake-up command or a data mutation. After wake-up, the sensor node collects data at a preset frequency. After the data transmission is completed, it automatically enters sleep mode, reducing energy consumption during non-working periods.

Images