Bridge and tunnel intelligent monitoring method and system

Abstract

The invention relates to the field of infrastructure monitoring, and discloses a bridge and tunnel intelligent monitoring method and system, and the method comprises the steps: receiving the structural parameter data, collected by a fixed sensor network, of a bridge and a tunnel; performing anomaly detection on the structure parameter data to identify an abnormal region; generating an unmanned aerial vehicle inspection instruction based on the abnormal region; receiving visual image data and three-dimensional point cloud data returned by the unmanned aerial vehicle inspection module, and fusing the structural parameter data, the visual image data and the three-dimensional point cloud data to generate structural health assessment; predicting structure failure time based on the structure health assessment, and generating a maintenance suggestion; and outputting a visual report containing the risk level and the maintenance suggestion. According to the invention, through cooperative work of the fixed sensor network and the unmanned aerial vehicle inspection module, the efficiency and accuracy of bridge and tunnel structure health monitoring are significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of infrastructure monitoring, and in particular to an intelligent monitoring method and system for bridges and tunnels. Background Art

[0002] In recent years, structural health monitoring technologies for bridges and tunnels, as core components of transportation infrastructure, have garnered significant attention. Existing technologies include fixed sensor networks, which deploy strain, displacement, and temperature sensors to achieve real-time monitoring and are widely used in bridges and tunnels. Furthermore, drone inspection technology utilizes high-resolution cameras and lidar to perform visual inspections of hard-to-reach areas, generating 3D point cloud data. Artificial intelligence (AI) analyzes sensor data to identify structural anomalies and provide real-time alerts. These technologies each have their own advantages in bridge and tunnel monitoring, providing crucial support for structural safety.

[0003] However, the drawback of existing technologies is that fixed sensor networks and drone inspections typically operate independently and lack a collaborative mechanism, resulting in insufficient monitoring efficiency and accuracy. For example, while fixed sensor networks can collect structural parameters in real time, their coverage is limited, making it difficult to quickly verify anomalies in complex areas (such as under bridges or inside tunnels). Although drone inspections can provide high-resolution data, they require manual scheduling, the inspection process is time-consuming (e.g., several hours), and they cannot be automatically integrated with sensor data. Compared to the sensor and drone collaborative mechanism of the present invention, existing technologies do not implement an automated process for triggering drone inspections through anomaly detection, making it difficult to complete anomaly identification and verification in a short period of time (e.g., minutes). The accuracy of anomaly detection is also limited to a single data source (e.g., less than 90%), affecting the timeliness and reliability of bridge and tunnel maintenance. Summary of the Invention

[0004] In order to remedy the above shortcomings, the present invention provides an intelligent monitoring method and system for bridges and tunnels, aiming to improve the shortcomings of the existing technology, that is, fixed sensor networks and drone inspections usually operate independently and lack a collaborative mechanism, resulting in insufficient monitoring efficiency and accuracy.

[0005] In a first aspect, the present invention provides the following technical solution: an intelligent monitoring method for bridges and tunnels, applied to a central control system, comprising: receiving structural parameter data of bridges and tunnels collected by a fixed sensor network, wherein the structural parameter data includes strain, displacement, temperature and corrosion data; performing anomaly detection on the structural parameter data to identify abnormal areas; generating a drone inspection instruction based on the abnormal area, wherein the drone inspection instruction is used to trigger a drone inspection module to perform visual inspection and three-dimensional data acquisition on the abnormal area; receiving visual image data and three-dimensional point cloud data returned by the drone inspection module, fusing the structural parameter data, the visual image data, and the three-dimensional point cloud data to generate a structural health assessment; predicting structural failure time based on the structural health assessment and generating maintenance recommendations; Output is a visual report containing the risk level and stated maintenance recommendations.

[0006] Through the above technical solution, the central control system runs on the cloud server and communicates with the fixed sensor network and drone inspection module through the 5G network. The fixed sensor network deploys strain, displacement, temperature and corrosion sensors every 10 meters on bridges and tunnels, collects data at a frequency of 1Hz, and transmits JSON data packets containing timestamps, sensor IDs and measurement values. The central control system uses convolutional neural networks to detect anomalies, mark abnormal areas, and trigger drone inspection instructions containing coordinates and altitudes. The drone is equipped with a 20MP camera and lidar to collect images and point cloud data. Through 5G backhaul, the long short-term memory network fuses structural parameters, images and point cloud data to generate a health score. The system predicts failure time and generates maintenance recommendations. The visual report is output in HTML format through the web interface, including risk level, 3D model and recommendations, and is stored on the blockchain to ensure data security.

[0007] Preferably, the abnormality detection of the structural parameter data includes: using a convolutional neural network model to analyze the structural parameter data to identify patterns of excessive strain, abnormal displacement or increased corrosion.

[0008] Preferably, generating a drone inspection instruction based on the abnormal area includes: generating an optimized drone flight path according to the location and risk level of the abnormal area, wherein the flight path is used to guide the drone inspection module to perform the inspection task.

[0009] Preferably, the fusing of the structural parameter data, the visual image data and the three-dimensional point cloud data includes: using a long short-term memory network model to perform time series analysis on the structural parameter data, the visual image data and the three-dimensional point cloud data to generate the structural health assessment.

[0010] Preferably, the method further includes: generating a sensor deployment instruction, wherein the sensor deployment instruction is used to trigger the drone inspection module to deploy or replace a fixed sensor in the abnormal area.

[0011] Preferably, outputting the visual report including the risk level and the maintenance recommendation includes: generating the visual report through a natural language processing interface, and supporting an operator to query the structural health assessment through voice or text.

[0012] Preferably, the method further includes: storing the structural parameter data, the visual image data and the three-dimensional point cloud data in a blockchain database, wherein the blockchain database is used to ensure that the data cannot be tampered with.

[0013] In a second aspect, the present invention provides the following technical solution: an intelligent monitoring system for bridges and tunnels, including a central control system, wherein the central control system includes: A data receiving module is used to receive structural parameter data of bridges and tunnels collected by a fixed sensor network, including strain, displacement, temperature, and corrosion data, as well as visual image data and three-dimensional point cloud data returned by a drone inspection module; an anomaly detection module, configured to perform anomaly detection on the structural parameter data to identify abnormal areas; an instruction generation module, configured to generate a drone inspection instruction based on the abnormal area, wherein the drone inspection instruction is configured to trigger the drone inspection module to perform visual inspection and three-dimensional data acquisition on the abnormal area; a data fusion module, configured to fuse the structural parameter data, the visual image data, and the three-dimensional point cloud data to generate a structural health assessment; a prediction module for predicting structural failure time based on the structural health assessment and generating maintenance recommendations; The report output module is used to output a visual report including the risk level and the maintenance suggestion.

[0014] In the third aspect, the invention provides the following technical solution: a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned intelligent monitoring method for bridges and tunnels when executing the computer program.

[0015] In a fourth aspect, the present invention provides the following technical solution: a readable storage medium having a computer program stored thereon, and the computer program, when executed by a processor, implements the above-mentioned intelligent monitoring method for bridges and tunnels.

[0016] The present invention has the following beneficial effects: 1. In this invention, the efficiency and accuracy of health monitoring of bridge and tunnel structures are significantly improved through the collaborative work of a fixed sensor network and an unmanned aerial vehicle (UAV) inspection module. The central control system detects anomalies based on real-time strain, displacement, temperature, and corrosion data, automatically triggering UAVs to perform high-resolution visual inspections and three-dimensional data collection on abnormal areas. Health assessments are generated by fusing multi-source data. Compared with traditional single-source monitoring methods, this shortens anomaly identification time (for example, from hours to minutes) and improves anomaly detection accuracy (for example, to 95%).

[0017] 2. In the present invention, the ability to predict bridge and tunnel structural failures is enhanced by integrating multi-source data through an artificial intelligence analysis platform. The central control system uses a convolutional neural network to detect abnormal patterns and uses a long-short-term memory network to analyze the time series of structural parameters, visual images, and three-dimensional point cloud data to predict the failure time (for example, 6 months, with an error of ±10%). Compared with traditional real-time alarm systems, this system achieves early warning and optimized maintenance plans, reducing the risk of sudden failure (for example, by 50%).

[0018] 3. In this invention, monitoring data is stored in a blockchain database, which improves data security and traceability. The central control system encrypts and stores structural parameters, visual images, and three-dimensional point cloud data on the blockchain platform, generating tamper-proof transaction records (taking less than 0.1 seconds). Compared with traditional databases, this prevents the risk of data tampering (consistency reaches 100%) and supports efficient data query and auditing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of a method for intelligent monitoring of bridges and tunnels proposed by the present invention; Figure 2 This is a system architecture diagram of the intelligent bridge and tunnel monitoring method proposed in this invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] Example 1 Reference Figure 1 In a first embodiment of the present invention, the present invention provides an intelligent monitoring method for bridges and tunnels, which is applied to a central control system and includes: Receive structural parameter data of bridges and tunnels collected by fixed sensor networks, including strain, displacement, temperature and corrosion data; Perform anomaly detection on structural parameter data to identify abnormal areas; Generate drone inspection instructions based on abnormal areas. The drone inspection instructions are used to trigger the drone inspection module to perform visual inspection and three-dimensional data collection on the abnormal areas. Receive visual image data and 3D point cloud data returned by the UAV inspection module, fuse structural parameter data, visual image data and 3D point cloud data to generate a structural health assessment; Predict structural failure time based on structural health assessment and generate maintenance recommendations; Output a visual report containing risk levels and maintenance recommendations.

[0022] Specifically, in one embodiment, the central control system is deployed on a cloud computing server equipped with a high-performance processor (e.g., an Intel Xeon 2.4GHz) and high-capacity storage (e.g., a 1TB SSD). It communicates with a fixed sensor network and drone inspection modules via a high-speed network interface (e.g., 5G or fiber optic). The fixed sensor network includes sensors deployed on bridge girders, piers, and tunnel walls. Every 10 meters, a set of strain sensors (accuracy ±0.1με), displacement sensors (accuracy ±0.01mm), temperature sensors (accuracy ±0.1°C), and corrosion sensors (accuracy ±0.01mm) are installed, collecting data at a 1Hz frequency. The central control system receives structural parameter data via a LoRaWAN gateway. The data packets are formatted in JSON and contain a timestamp, sensor ID, and measurement value (e.g., strain value 100με, displacement 0.5mm). Anomaly detection is performed using a pretrained convolutional neural network (CNN) model trained on historical data (100,000 records) to identify excessive strain (>200με), abnormal displacement (>1mm), or increased corrosion (>0.1mm). Upon detecting an anomaly, the central control system generates a drone inspection command based on the anomaly's location (e.g., coordinates of a bridge's main girder [x=50m, y=10m]) and risk level (e.g., high risk, based on a threshold >80%). The command includes the target coordinates and flight altitude (e.g., 10m). The drone inspection module, equipped with a 20MP camera and lidar, captures visual images (1920x1080 resolution) and 3D point cloud data (with a density of 1000 points / m²), which are transmitted back to the central control system via a 5G network. Data fusion uses a long short-term memory (LSTM) network, taking as input a sequence of structural parameters (1000 frames), image features (extracted using ResNet), and point cloud features (extracted using PointNet), outputting a structural health score (ranging from 0 to 100). The prediction module, based on the LSTM, predicts the failure time (e.g., 6 months with an error of ±10%) and generates maintenance recommendations (e.g., "Reinforce the main girder"). Visual reports are output through a web interface, including risk levels (high / medium / low), 3D models (OBJ format) and maintenance recommendations, and are stored in a blockchain database (Hyperledger Fabric) to ensure that the data cannot be tampered with.

[0023] Anomaly detection of structural parameter data involves analyzing the structural parameter data using a convolutional neural network model to identify patterns of excessive strain, abnormal displacement, or increased corrosion.

[0024] Specifically, in one embodiment, the anomaly detection module of the central control system runs in a Python environment, using a convolutional neural network (CNN) model implemented using the TensorFlow framework. The CNN model consists of five convolutional layers (32 3x3 filters per layer), two pooling layers (2x2 max pooling), and three fully connected layers. Its input is a time series of structural parameter data (a 1000x4 matrix containing strain, displacement, temperature, and corrosion), and its output is an anomaly probability (0-1). The training dataset includes 100,000 bridge and tunnel data points, covering both normal (80%) and abnormal (20%) scenarios. Anomalies are defined as excessive strain (>200με), abnormal displacement (>1mm), or increased corrosion (>0.1mm). The model is trained using the Adam optimizer with a learning rate of 0.001 and a cross-entropy loss function. After 50 epochs, the accuracy reaches 95%. During operation, the central control system receives 100 pieces of structural parameter data per second. The CNN model processes these data in real time, detecting any anomalies (e.g., a strain of 300 με) and marking the abnormal area (e.g., the tunnel wall [x=20m, y=5m]) with a timestamp (e.g., 2025-04-22 10:00:00). These abnormality results are stored in an in-memory database (Redis) for subsequent use in generating drone commands.

[0025] Generating drone inspection instructions based on abnormal areas includes: generating an optimized drone flight path based on the location and risk level of the abnormal area, and the flight path is used to guide the drone inspection module to perform the inspection task.

[0026] Specifically, in one embodiment, the command generation module of the central control system uses a path planning algorithm (Algorithm A) to generate drone inspection instructions. The location of the abnormal area is obtained from the anomaly detection module (e.g., bridge main beam [x=50m, y=10m]), and the risk level is based on the probability of the abnormality (e.g., 0.9, with a threshold >0.8 indicating high risk). Algorithm A inputs a 3D map (based on OpenStreetMap, with 1m accuracy), the coordinates of the abnormality, and drone parameters (maximum speed 5m / s, 30-minute flight time), and outputs an optimized flight path (e.g., path points {[x=0,y=0,z=5],[x=50,y=10,z=10]}). The command format is XML and includes the path points, the inspection height (10m), and the acquisition mode (image + point cloud). The central control system sends the command to the drone inspection module via the 5G network. The drone, equipped with an RTK positioning system (accuracy ±5cm), follows the path, avoiding obstacles (e.g., bridge piers, with a detection range >2m). After the inspection mission is completed, the drone returns a confirmation signal (JSON format, including the mission ID and completion time), and the central control system updates the mission status (stored in the MySQL database).

[0027] The fusion of structural parameter data, visual image data and three-dimensional point cloud data includes: using a long short-term memory network model to perform time series analysis on the structural parameter data, visual image data and three-dimensional point cloud data to generate a structural health assessment.

[0028] Specifically, in one embodiment, the data fusion module of the central control system runs on the PyTorch framework and uses a long short-term memory (LSTM) model for time series analysis. This LSTM model consists of three layers of LSTM units (128 hidden units per layer) and one fully connected layer. Its input is a three-dimensional feature vector: a sequence of structural parameters (1000x4, including strain, displacement, temperature, and corrosion, normalized using Min-Max), image features (2048 dimensions, extracted using a ResNet-50 pre-trained model), and point cloud features (1024 dimensions, extracted using a PointNet). The model training dataset consists of 50,000 bridge and tunnel data points, and training is performed for 100 epochs. The loss function is mean squared error, and the optimizer is Adam (learning rate 0.0005). During operation, the central control system receives 100 pieces of structural parameter data per second, visual images (1920x1080 pixels per frame), and point cloud data (1000 points per second). The LSTM model generates a structural health assessment every 10 seconds, outputting a health score (e.g., 85 on a scale of 0-100) and anomaly type (e.g., crack). The assessment results are stored in a MongoDB database for use by the prediction module. The fusion process takes less than one second, ensuring real-time performance.

[0029] It also includes: generating a sensor deployment instruction, where the sensor deployment instruction is used to trigger the drone inspection module to deploy or replace fixed sensors in abnormal areas.

[0030] Specifically, in one embodiment, the command generation module of the central control system generates sensor deployment instructions in JSON format, including the target location (e.g., under a bridge [x=30m, y=8m]), sensor type (e.g., strain gauge), and mounting method (magnetic fixation). The anomaly detection module identifies the required deployment area (e.g., an area with increased corrosion, corrosion value >0.1mm). The command generation module plans the deployment mission based on the drone's payload capacity (maximum 2kg) and sensor size (5cmx5cm). The drone inspection module, equipped with a robotic arm (6 degrees of freedom, ±1mm accuracy), uses visual SLAM navigation (error <10cm) to reach the target location and perform sensor installation or replacement. After installation, the drone transmits installation confirmation data (including the sensor ID and new location coordinates), and the central control system verifies the sensor status (receiving initial data, such as a strain value of 50με, via LoRaWAN). Deployment mission records are stored in a blockchain database (a private Ethereum blockchain). Each deployment takes approximately 5 minutes, and a maximum of three sensors can be deployed in a single mission.

[0031] Outputting visual reports containing risk levels and maintenance recommendations includes: generating visual reports through a natural language processing interface and supporting operators to query structural health assessments via voice or text.

[0032] Specifically, in one embodiment, the central control system's report output module runs in a Node.js environment and integrates a natural language processing (NLP) interface. It uses a pre-trained BERT model (12-layer Transformer) to process operator queries. The report generation process includes obtaining a structural health assessment (e.g., a health score of 80, with unusual crack types) from the data fusion module, obtaining maintenance recommendations (e.g., "Reinforce the tunnel wall within 6 months") from the prediction module, and generating a visual report in HTML format. The report includes a risk level (high, based on a score <85), a 3D model (OBJ format, generated from point cloud data with 1cm accuracy), and maintenance recommendations (text, approximately 200 words). The NLP interface supports voice input (via the WebSpeech API, with a 95% recognition rate) and text input (via the REST API). For example, an operator querying "Show the latest report for Bridge A" returns a report link (URL format). The report is stored in the IPFS (internet distributed file system) and can be accessed via the web and mobile devices (iOS / Android). It generates reports in less than 2 seconds and supports multiple languages ​​(Chinese and English).

[0033] It also includes: storing structural parameter data, visual image data and three-dimensional point cloud data in a blockchain database, which is used to ensure that the data cannot be tampered with.

[0034] Specifically, in one embodiment, the central control system's data storage module uses the Hyperledger Fabric blockchain platform, deployed on a private chain (4 nodes, PBFT consensus algorithm). Structural parameter data (100 records per second, approximately 1KB per record), visual image data (approximately 2MB per frame), and 3D point cloud data (approximately 10MB per second) are stored using SHA-256 encryption. Before uploading the data to the blockchain, the central control system preprocesses the data (compressing it into ZIP format with a 50% compression ratio), generates a 256-bit data hash, and stores it in the blockchain ledger. Each storage operation generates a transaction record (including data ID, timestamp, and hash), which is verified by blockchain nodes (taking <0.1 seconds). The data itself is stored in a distributed database (Cassandra, with a capacity of 10TB), and the blockchain only stores the hash and metadata (approximately 100 bytes per record) to ensure efficient retrieval. Operators can query data through the API (for example, entering the time range 2025-04-22 00:00:00 to 10:00:00). The system returns the data and hash, verifying the data integrity (with a 100% match rate). Blockchain storage supports multi-bridge and tunnel data (up to 1,000 structures), with a single storage time of less than 0.5 seconds.

[0035] Example 2: Reference Figure 2 In a second embodiment of the present invention, the present invention provides an intelligent monitoring system for bridges and tunnels, including a central control system, the central control system including: The data receiving module is used to receive the structural parameter data of bridges and tunnels collected by the fixed sensor network, including strain, displacement, temperature and corrosion data, as well as the visual image data and 3D point cloud data returned by the drone inspection module; An anomaly detection module, used to perform anomaly detection on structural parameter data to identify abnormal areas; An instruction generation module is used to generate a drone inspection instruction based on the abnormal area. The drone inspection instruction is used to trigger the drone inspection module to perform visual inspection and three-dimensional data collection on the abnormal area; Data fusion module, used to fuse structural parameter data, visual image data and 3D point cloud data to generate structural health assessment; A prediction module is used to predict structural failure time based on structural health assessment and generate maintenance recommendations; Report output module, used to output visual reports containing risk levels and maintenance recommendations.

[0036] Specifically, in one embodiment, the central control system of the intelligent bridge and tunnel monitoring system is deployed on a high-availability server cluster (four servers, each equipped with NVIDIA A100 GPUs) running the Ubuntu 20.04 operating system. The data receiving module receives data from the fixed sensor network (throughput 1000 data points / second) and the drone inspection module (bandwidth 1Gbps) via a Kafka message queue, supporting data format conversion (JSON to Protobuf, improving efficiency by 30%). The anomaly detection module runs a CNN model (same as claim 2), processing 100 data points per second. The command generation module uses the A* algorithm (same as claim 3) to generate drone commands (latency <50ms). The data fusion module runs an LSTM model (same as claim 4), fusing multi-source data (structural parameters, images, and point clouds) to generate health assessments (scoring accuracy ±5%). The prediction module uses LSTM to predict failure times (error ±10%) and generate maintenance recommendations (approximately 200 words). The report output module integrates an NLP interface (same as claim 6), generates HTML reports, and supports real-time updates via WebSocket. All modules are integrated through a microservices architecture (SpringBoot) and deployed in a Kubernetes cluster, supporting parallel monitoring of multiple bridges and tunnels (up to 1,000 structures). The system supports fault recovery (downtime recovery time is less than 1 minute), and inter-module communication latency is less than 10ms.

[0037] Example 3 The third embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the intelligent monitoring method for bridges and tunnels in the above embodiment are implemented.

[0038] Example 4 The fourth embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer device, including: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory to execute the intelligent monitoring method of bridges and tunnels of the above embodiment.

[0039] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0040] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A bridge and tunnel intelligent monitoring method, characterized in that: Applied in central control systems, including: receiving structural parameter data of bridges and tunnels collected by a fixed sensor network, wherein the structural parameter data includes strain, displacement, temperature and corrosion data; performing anomaly detection on the structural parameter data to identify abnormal areas; generating a drone inspection instruction based on the abnormal area, wherein the drone inspection instruction is used to trigger a drone inspection module to perform visual inspection and three-dimensional data acquisition on the abnormal area; receiving visual image data and three-dimensional point cloud data returned by the drone inspection module, fusing the structural parameter data, the visual image data, and the three-dimensional point cloud data to generate a structural health assessment; predicting structural failure time based on the structural health assessment and generating maintenance recommendations; Output is a visual report containing the risk level and stated maintenance recommendations.

2. The intelligent monitoring method for bridges and tunnels according to claim 1, characterized in that: The abnormality detection of the structural parameter data includes: using a convolutional neural network model to analyze the structural parameter data and identify patterns of excessive strain, abnormal displacement or increased corrosion.

3. The intelligent monitoring method for bridges and tunnels according to claim 1, characterized in that: Generating a drone inspection instruction based on the abnormal area includes generating an optimized drone flight path according to the location and risk level of the abnormal area, wherein the flight path is used to guide the drone inspection module to perform the inspection task.

4. The intelligent monitoring method for bridges and tunnels according to claim 1, characterized in that: The fusing of the structural parameter data, the visual image data and the three-dimensional point cloud data includes: using a long short-term memory network model to perform time series analysis on the structural parameter data, the visual image data and the three-dimensional point cloud data to generate the structural health assessment.

5. The intelligent monitoring method for bridges and tunnels according to claim 1 is characterized in that: Also includes: A sensor deployment instruction is generated, where the sensor deployment instruction is used to trigger the drone inspection module to deploy or replace a fixed sensor in the abnormal area.

6. The intelligent monitoring method for bridges and tunnels according to claim 1, characterized in that: Outputting the visual report including the risk level and the maintenance recommendation includes: generating the visual report through a natural language processing interface, and supporting an operator to query the structural health assessment through voice or text.

7. The intelligent monitoring method for bridges and tunnels according to claim 1, characterized in that: Also includes: The structural parameter data, the visual image data, and the three-dimensional point cloud data are stored in a blockchain database, and the blockchain database is used to ensure that the data cannot be tampered with.

8. An intelligent monitoring system for bridges and tunnels, characterized in that: The intelligent monitoring method for bridges and tunnels according to any one of claims 1 to 7 includes a central control system, wherein the central control system includes: A data receiving module is used to receive structural parameter data of bridges and tunnels collected by a fixed sensor network, including strain, displacement, temperature, and corrosion data, as well as visual image data and three-dimensional point cloud data returned by a drone inspection module; an anomaly detection module, configured to perform anomaly detection on the structural parameter data to identify abnormal areas; an instruction generation module, configured to generate a drone inspection instruction based on the abnormal area, wherein the drone inspection instruction is configured to trigger the drone inspection module to perform visual inspection and three-dimensional data acquisition on the abnormal area; a data fusion module, configured to fuse the structural parameter data, the visual image data, and the three-dimensional point cloud data to generate a structural health assessment; a prediction module for predicting structural failure time based on the structural health assessment and generating maintenance recommendations; The report output module is used to output a visual report including the risk level and the maintenance suggestion.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the intelligent monitoring method for bridges and tunnels according to any one of claims 1 to 7 is implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the intelligent monitoring method for bridges and tunnels according to any one of claims 1 to 7 is implemented.

Images